Week 2 Ethical Health Promotion Paper Health promotion and Disease prevention

Find a scholarly, peer-reviewed article no more than four years old that discusses an ethical health promotion-related issue. Use the WCU library databases to search for appropriate articles.

In your paper:

• Briefly summarize the presented issue. 
• Describe your thoughts on the role health care professionals should play in resolving the ethical issue. 
• Provide specific theories and refer to specific ethical codes to support your position. 

Your paper should be 2–3 pages long. Use APA to cite and reference the article and any other optional sources you use. Adhere to APA formatting throughout your paper.

127

REFUSING TO TREAT NONCOMPLIANT PATIENTS IS
BAD MEDICINE

Jessica Mantel†

Government health programs and private payors have adopted various reforms
that fundamentally transform the physician-patient relationship. Public reporting on
how well physicians perform on various quality and cost metrics, as well as payment
reforms that link physicians’ reimbursements to their performance on these metrics,
incentivize physicians to improve the quality and efficiency of care they provide to
patients. Less appreciated, however, is that these reforms also create strong incentives
for physicians to reject patients who do not abide by their physician’s medical
opinion, including recommendations that the patient adopt healthier behaviors.
These noncompliant patients increasingly will find themselves rejected by physicians,
as current legal and ethical standards generally grant physicians full autonomy in
deciding which patients to treat. This Article evaluates whether the law and
standards of professional conduct should afford physicians broad discretion in
deciding whether to treat noncompliant patients. It concludes that they should not
and calls upon lawmakers and professional associations to place legal and ethical
restraints on physicians’ ability to reject noncompliant patients.

TABLE OF CONTENTS

INTRODUCTION …………………………………………………………………………………………………..128

I. AVOIDING THE NONCOMPLIANT PATIENT …………………………………………………….132
A. Physicians’ Incentives Under Value-Based Purchasing ……………………….133
B. Reputational Concerns ………………………………………………………………………138

II. PHYSICIANS’ LEGAL AND ETHICAL OBLIGATIONS …………………………………………..140

III. JUSTIFICATIONS FOR PROHIBITING PHYSICIANS FROM DISCRIMINATING
AGAINST NONCOMPLIANT PATIENTS …………………………………………………………….142

† Associate Professor, University of Houston Law Center. Thank you to Jessica Roberts
and participants in the University of Houston’s faculty workshop for their useful discussion and
suggestions; and to Emily Lawson, Robert Clark, Matthew Mantel, Muneeza Ilahi, Bobby Dale
Joe, and Nicholas Tolat for their research assistance.

128 C A R D O Z O L A W R E V I E W [Vol. 39:127

A. Reinforcing the Policy Goals Behind Performance Incentive
Programs …………………………………………………………………………………………..142

1. Lowering Barriers to Patient Compliance and Healthy
Behaviors ………………………………………………………………………………..143

2. Preserving Incentives for Physicians to Lower Barriers to
Compliance and Healthy Behaviors ………………………………………..151

B. Honoring Professional Norms of Beneficence and Nonmaleficence ……..152
1. Eliminating the Adverse Consequences from Discontinuity

in Care ……………………………………………………………………………………153
2. Protecting Patients’ Trust in Physicians ………………………………….155
3. Avoiding the Stigmatization and Shaming of Patients …………….157

C. Honoring Patient Autonomy ……………………………………………………………..159
D. Reducing Disparities in Health …………………………………………………………..161

1. The Disproportionate Impact of Social and Environmental
Factors ……………………………………………………………………………………162

2. Physicians’ Implicit Bias …………………………………………………………165

IV. JUSTIFICATIONS FOR THE STATUS QUO ………………………………………………………….171
A. The Personal Responsibility Rationale ……………………………………………….171

1. Noncompliant Patients’ Moral Culpability ……………………………..173
2. Holding Noncompliant Patients Accountable ………………………..175

B. The Fairness and Autonomy Rationales …………………………………………….178
1. Physicians’ Ethical Obligations Limit Their Autonomy ………….179
2. Policy Justifications for Limiting Physicians’ Right to

Refuse Treatment to Noncompliant Patients ………………………….183
C. The Beneficence Rationale …………………………………………………………………187

1. The Paternalistic Rationale ……………………………………………………..188
2. The Alternative Physician Rationale ……………………………………….188

V. SUPPORTING PROVIDERS’ EFFORTS TO IMPROVE PATIENT COMPLIANCE …………191

CONCLUSION……………………………………………………………………………………………………….194

INTRODUCTION

Every physician has encountered the noncompliant patient—the
diabetic patient who fails to consistently take her prescribed
medications, self-monitor her blood glucose, and abide by dietary
restrictions; the patient with chronic obstructive pulmonary disease
(COPD) who continues smoking, does not exercise regularly, and quits
pulmonary rehabilitation; the orthopedic surgical patient who ignores
postoperative weight-bearing instructions and misses follow-up

2017] N O N C O M P L I A N T P A T I E N T S 129

appointments. In all such cases, the patient’s behavior fails to coincide
with their physician’s medical advice and recommendations for health.1

Patient noncompliance presents “a major obstacle” to the effective
and efficient delivery of health care.2 Studies consistently find that a
large number of patients fail to take their medications as prescribed,
with as many as fifty percent of patients on long-term medication
therapies failing to do so.3 Compliance with physician-recommended
lifestyle changes (e.g., diet modification, smoking cessation) is even
lower, with only twenty to thirty percent of patients changing their
behavior.4 Not surprisingly, patients’ noncompliance leads to poorer
treatment outcomes and higher costs, including higher rates of
mortality and morbidity and excess urgent care visits and
hospitalizations.5

Many physicians express frustration with their noncompliant
patients.6 Some physicians even terminate the physician-patient
relationship in egregious cases of noncompliance. Most, though,
continue treating their noncompliant patients, doing their best to
educate patients about the importance of adhering to recommended
treatments and healthy lifestyles. However, physicians’ tolerance for

1 Patients’ behavior may diverge from their physicians’ recommendations for two
reasons—rejecting physicians’ advice or nonadherence. In the first scenario the patient
affirmatively refuses to follow the physician’s recommendations. For example, a patient may
reject a recommended medical intervention because she fears its risks or side effects or a patient
may decide she simply does not want to change her diet, quit smoking, or otherwise modify her
behavior. In contrast, the scenario of the nonadherent patient entails a patient who accepts the
physician’s advice but “fails to adhere to the regimens needed to implement it.” David B.
Resnik, The Patient’s Duty to Adhere to Prescribed Treatment: An Ethical Analysis, 30 J. MED. &
PHIL. 167, 168–69 (2005). For example, a patient may fill her prescription but fail to take the
medication as prescribed, or she may attempt to lose weight or quit smoking but fail in her
attempts to do so. See generally WORLD HEALTH ORG., ADHERENCE TO LONG-TERM THERAPIES:
EVIDENCE FOR ACTION 3–11 (2003) (defining nonadherence as “the extent to which a person’s
behavior—taking medication, following a diet, and/or executing lifestyle changes—corresponds
with agreed upon recommendations from a health care provider”). For purposes of this Article,
I refer to both types of scenarios as “noncompliance.”
2 Nadia Sciberras et al., Ethics in Practice: The Ethical and Practical Challenges of Patient
Noncompliance in Orthopedic Surgery, 61 J. BONE & JOINT SURGERY e61(1) (2013).
3 See Jing Jin et al., Factors Affecting Therapeutic Compliance: A Review from the Patient’s
Perspective, 4 THERAPEUTICS & CLINICAL RISK MGMT. 269, 269 (2008); Leigh Page, Why Should
Your Noncompliance Harm My Income?, MEDSCAPE (Oct. 9, 2014), http://www.medscape.com/
features/content/6006314.
4 See Jin et al., supra note 3, at 269.
5 See id. at 270–71 (explaining that noncompliance is directly associated with poor
treatment outcomes in patients with a range of chronic conditions and poses financial burdens
to society due to excess urgent care visits, hospitalizations, and higher treatment costs); see also
WORLD HEALTH ORG., supra note 1, at 11–14 (describing the health and economic costs of
nonadherence among patients with COPD, asthma, diabetes and other chronic conditions);
Neil Chesanow, The Noncompliance Epidemic: Why Are So Many Patients Noncompliant?,
MEDSCAPE (Jan. 16, 2014) http://www.medscape.com/viewarticle/818850 (“Poor medication
compliance is implicated in over 125,000 US deaths per year.”).
6 See Bruce G. Bender, Motivating Patient Adherence to Allergic Rhinitis Treatments, 15
CURRENT ALLERGY ASTHMA REP. 10, 12 (2015) (“Health-care providers frequently report
frustration over the nonadherence of their patients.”).

130 C A R D O Z O L A W R E V I E W [Vol. 39:127

their most noncompliant patients will wane as emerging health policy
reforms give physicians a new reason to avoid noncompliant patients—
profitability.

New payment models that tie physicians’ reimbursements to their
patients’ health status and treatment costs mean noncompliant patients
will reduce physicians’ income.7 In addition, physicians may fear that
their noncompliant patients, with their poorer health outcomes, will
harm the physicians’ performance on publicly available “report cards,”
thereby damaging their reputation and limiting their contract
opportunities with private payors.8 As treating noncompliant patients
increasingly becomes a financial burden for physicians, noncompliant
patients may find themselves fired by their physicians.9 In addition,
physicians may refuse to treat prospective patients who are likely (or
perceived as likely) to be noncompliant.10 With limited exceptions, both
the law and standards of professional conduct permit physicians to
refuse to treat noncompliant patients.11

The prospect of widespread avoidance of noncompliant patients by
physicians raises the question of whether the law and standards of
professional conduct should afford physicians this discretion. Most
basically, should the physician-patient relationship be treated like any
other consumer transaction, with either party free to terminate the
relationship at will? Or do moral and policy considerations justify a
departure from traditional principles of freedom-of-contract in the case
of the noncompliant patient? Scholars and policymakers have largely
neglected this important issue.12 This Article fills in this gap by
exploring the arguments for and against granting physicians broad
discretion on whether to treat noncompliant patients. It concludes that
on balance the arguments favor limiting physicians’ ability to reject
noncompliant patients and calls upon regulators and professional
associations to place legal and ethical restraints on physicians’ ability to

7 See infra Section I.A.
8 See infra Section I.B.
9 See infra Part I.
10 See Bender, supra note 6, at 12. The practice of favoring healthier, more compliant
patients over less healthy, less compliant patients is sometimes referred to as “cherry-picking.”
See, e.g., Amar A. Desai et al., Is There “Cherry Picking” in the ESRD Program? Perceptions from
a Dialysis Provider Survey, 4 CLINICAL J. AM. SOC’Y NEPHROLOGY 772 (2009) (defining “cherry
picking” in the health care context as insurers and providers protecting themselves
economically by avoiding sicker patients or preferentially attracting healthier patients); Judith
H. Hibbard et al., Does Compensating Primary Care Providers to Produce Higher Quality Make
Them More or Less Patient Centric?, 72 MED. CARE RES. & REV. 481, 482 (2015) (describing the
practice of selecting healthier and more compliant patients as “cherry-picking”).
11 See infra Part II.
12 Two commentators have discussed whether physicians should be permitted to reject
noncompliant patients. See David Orentlicher, Denying Treatment to the Noncompliant Patient,
265 JAMA 1579, 1581 (1991) (arguing that ethical rules governing physicians generally should
not reject patients for noncompliance); Mark Wicclair, Dismissing Patients for Health-Based
Reasons, 22 CAMBRIDGE Q. HEALTHCARE ETHICS 308 (2013) (same).

2017] N O N C O M P L I A N T P A T I E N T S 131

do so.
Part I describes how recent policy reforms create strong incentives

for physicians to reject patients who engage in unhealthy behaviors and
fail to follow their physician’s medical advice. Part II summarizes
current legal and ethical standards governing the physician-patient
relationship and explains that they permit physicians to refuse to treat
noncompliant patients.

Part III calls upon Congress and state legislatures to enact
legislation prohibiting physicians from firing or otherwise refusing to
treat noncompliant patients and urges professional organizations and
medical boards to revise their ethical standards to do the same.
Imposing on physicians an obligation to treat noncompliant patients
would serve important policy goals. First, it would strengthen the
incentives recent policy reforms give providers to tackle the social,
environmental, and behavioral health barriers to good health. Second,
when physicians reject noncompliant patients, they experience
discontinuity in care, a loss of trust in physicians, and stigma and
shame. Requiring physicians to treat noncompliant patients eliminates
these harms, thereby reinforcing physicians’ ethical commitment to
serving their patients’ best interest. Third, prohibiting physicians from
rejecting noncompliant patients protects the autonomy of patients who
might feel pressured to agree to treatment recommendations, even if
inconsistent with their personal preferences, in order to avoid
termination of the physician-patient relationship. Finally, members of
vulnerable groups are more likely to be rejected by physicians seeking to
avoid noncompliant patients, leading to greater disparities in health.
Prohibiting physicians from refusing to care for noncompliant patients
would guard against this inequitable outcome.

Part IV considers justifications for affording physicians the
discretion to fire or otherwise refuse to treat noncompliant patients: (1)
physicians are justified in terminating the physician-patient relationship
because noncompliant patients should be held accountable for failing to
exercise self-care; (2) physicians should not be forced to assume
financial liability for poor patient outcomes that stem from patients’
lifestyle choices and nonadherence; (3) requiring physicians to treat
noncompliant patients unduly compromises physicians’ personal
autonomy and freedom of association rights; and (4) firing
noncompliant patients serves the patients’ best interest by motivating
the patient to improve their treatment adherence or, alternatively,
allowing the patient to find another physician better suited to treat the
patient. While these rationales are not without merit, ultimately they do
not justify rules that permit physicians to reject noncompliant patients
given competing policy considerations, societal norms of fairness and
compassion, and professional norms of benevolence.

Finally, in recognition of the challenges faced by physicians caring

132 C A R D O Z O L A W R E V I E W [Vol. 39:127

for noncompliant patients, Part V outlines policy reforms that would
support physicians’ efforts to improve patients’ health-related behaviors
and adherence to medical advice.

I. AVOIDING THE NONCOMPLIANT PATIENT

Physicians have long dealt with patients who fail to follow
treatment recommendations. Sometimes physicians elect to terminate
their most noncompliant patients out of frustration13 or a belief that
these patients waste physicians’ time and medical resources.14 Other
physicians may hope that firing or threatening to fire a patient will
compel the patient to modify his or her behavior.15 And some
physicians may avoid noncompliant patients because they fear that they
will face more lawsuits given noncompliant patients’ higher-risk of
morbidity and mortality.16 But physicians’ increasingly have a new
reason to avoid noncompliant patients—their bottom-line.

Emerging payment models link physicians’ reimbursements to
patient outcomes and aggregate health care costs. Consequently,
noncompliant patients who are at higher-risk of poor health and require
more costly services threaten to lower a physician’s income. Physicians
also may fear that noncompliant patients will drag down the physicians’
scores on various patient outcome and efficiency metrics, causing harm
to the physicians’ reputation and limiting their professional
opportunities. I refer to these reforms collectively as “performance
incentive programs.” As noncompliant patients become a pressing
financial concern for physicians under performance incentive programs,
physicians will increasingly resort to firing their noncompliant patients
and will refuse to take on prospective patients likely (or perceived as
likely) to be noncompliant.

13 See Resnik, supra note 1, at 170–71 (“When a patient fails to adhere to a prescription, a
doctor may feel betrayed or exploited because the patient is not upholding his or her end of the
bargain. . . . Non-adherence eventually takes its toll. At some point, many doctors decide that
they would rather not treat non-adherent patients.”); Wicclair, supra note 12, at 312 (explaining
that a physician may fire a patient in response to “feelings of frustration and moral distress”).
14 See Resnik, supra note 1, at 170–71 (“[I]f the pattern of nonadherence continues despite
the doctor’s concerted efforts to help the patient implement the treatment, the doctor may feel
that he or she is wasting time and society’s resources.”); Wicclair, supra note 12, at 311 (stating
that one motivation for firing a patient is a desire on the part of the provider to
“prevent . . . wasting time and medical resources”).
15 See Wicclair, supra note 12, at 314 (“When persistent efforts at persuasion have failed [to
change a patient’s behavior], firing or threating to fire adult patients may be perceived as a last
resort to get them to make betters choices and modify their behavior.”).
16 See Page, supra note 3, at 4 (“Physicians are also concerned about malpractice. Even
though noncompliant patients are often responsible for bad outcomes, [one physician stated
that he] is concerned they still might sue him if something went wrong . . . .”).

2017] N O N C O M P L I A N T P A T I E N T S 133

A. Physicians’ Incentives Under Value-Based Purchasing

Traditionally, Medicare, Medicaid, and private insurers paid for
their enrollees’ health care on a fee-for-service basis, with health care
providers receiving a separate payment for each unit of service provided
to their Medicare patients.17 Fee-for-service thus rewarded physicians
who provided patients a higher volume of care and favored high-tech,
more intensive treatments that garnered higher payment rates.18 These
incentives fueled a health care inflation rate that for the past fifty years
has almost always exceeded general inflation, sometimes by as much as
five percentage points.19 Fee-for-service also did not promote patients
receiving high quality care, as physicians were compensated based on
what they did and not on whether their patients’ health improved.20

Concerned with both the cost and quality of care, payors have
shifted away from fee-for-service in favor of new payment models that
reward providers for improving health outcomes and lowering costs.21
Collectively known as value-based purchasing (VBP), these payment
strategies link providers’ compensation to their success in raising the
quality and lowering the cost of care. Under the first category of VBP
programs, pay-for-performance,22 providers are rewarded with higher
payment rates or bonuses if they perform well on selected measures of
quality and/or efficiency, such as reducing the rate of post-operative
complications, lowering diabetic patients blood glucose (or A1c)
levels,23 decreasing the rate of hospital admissions and readmissions, or
lowering average treatment costs. Pay-for-performance initiatives also

17 See Harold D. Miller, From Volume to Value: Better Ways to Pay for Health Care, 28
HEALTH AFF. 1418, 1419 (2009), http://content.healthaffairs.org/content/28/5/1418.full
+html (defining fee-for-service as paying providers a predetermined amount for each discrete
service provided).
18 See generally Jessica Mantel, Accountable Care Organizations: Can We Have Our Cake
and Eat It Too?, 42 SETON HALL L. REV. 1392, 1405 (2012) (describing how fee-for-service
encourages “the provision of care of marginal or uncertain benefits, as doing so increases
providers’ incomes and satisfies patient demands that providers do everything possible to
improve a patient’s health,” and “skews the system toward more costly interventions”).
19 See Jim Dolmas, Health Care Services Depress Recent PCE Inflation Readings, 11
DALLASFED 1, 2 (2016) (using data from the Bureau of Economic Analysis).
20 See Steven A. Schroeder & William Frist, Phasing Out Fee-for-Service Payment, 368 NEW
ENG. J. MED. 2029 (2013).
21 See infra text accompanying notes 28–30.
22 See CHERYL L. DAMBERG ET AL., RAND CORP., MEASURING SUCCESS IN HEALTH CARE
VALUE-BASED PURCHASING PROGRAMS xi (2014); NAT’L ACADS. OF SCIENCES ENGINEERING
MED., ACCOUNTING FOR SOCIAL RISK FACTORS IN MEDICARE PAYMENT 25–26 (Leslie Y. Kwan,
Kathleen Stratton & Donald M. Steinwachs et al. eds., 2017) [hereinafter Kwan & Steinwachs]
(describing VBP payment models that include financial or quality incentives).
23 An A1c test measures a patient’s average blood glucose during the previous two to three
months. Lower A1c values signal better diabetes control and reduce a patient’s risk of
developing complications such as eye, heart, and kidney disease. See Effects of the Medicare
Modernization Act on Clinicians Involved in Diabetes Care, AM. DIABETES ASS’N: CLINICAL
DIABETES (Jan. 2006), http://clinical.diabetesjournals.org/content/24/1/12.

134 C A R D O Z O L A W R E V I E W [Vol. 39:127

may penalize providers with lower payment if they perform poorly.24
For example, under Medicare’s Physician Value-Based Payment
Modifier, higher performing physicians receive an upward adjustment
to their rates under Medicare’s physician fee schedule while poorer
performing physicians receive lower payments.25

The second category of VBP payment models include risk-based
alternative payment models that hold providers accountable for the
quality and cost of care by shifting financial risk to providers.26 One of
the more prominent examples of risk-based alternative payment models
is shared savings for accountable care organizations (ACOs). ACOs are
local organizations comprised of primary care physicians and other
providers that agree to be jointly accountable for the cost and quality of
care delivered to a patient population.27 For example, under Medicare’s
Shared Savings Program, providers participating in ACOs that
successfully lower the aggregate annual cost of caring for their Medicare
patients receive a percentage of the savings, provided that the ACO also
satisfies certain quality metrics.28

24 See DAMBERG ET AL., supra note 22, at ix, xiv.
25 Patient Protection and Affordable Care Act, 42 U.S.C. § 1395w-4(p) (2012). Physicians
classified as low cost/high quality, low cost/average quality, and average cost/higher quality
receive an upward adjustment in their payment rates; physicians classified as low cost/low
quality, average cost/average quality, or high cost/high quality receive no adjustment in their
payment rates; and physicians classified as average cost/low quality, high cost/average quality,
and high cost/low quality receive a downward adjustment in their payment rates. See CTRS. FOR
MEDICARE & MEDICAID SERVS., SUMMARY OF 2015 PHYSICIAN VALUE-BASED PAYMENT
MODIFIER POLICIES 15, table 4 (2016), http://www.cms.gov/Medicare/Medicare-Fee-for-
Service-Payment/PhysicianFeedbackProgram/Downloads/CY2015ValueModifierPolicies .
Implemented in January of 2015, the Physician VBP Modifier initially applied only to
physicians and “eligible professionals” in groups of one hundred or more professionals but
applies to all physicians in 2017. See Medicare Program; Revisions to Payment Policies Under
the Physician Fee Schedule, 42 C.F.R. §§ 410, 414, 415, 423, 425, 486, 495 (2012). The Medicare
statute defines an “eligible professional” to include physicians, physician assistants, nurse
practitioners, clinical nurse specialists, certified registered nurse anesthetists, certified nurse
midwives, clinical social workers, clinical psychologists, registered dieticians, nutrition
professionals, audiologists, and physical, occupational, and qualified speech-language
therapists. 42 U.S.C. § 1395w-4(k)(3)(B) (2012).
26 See Kwan & Steinwachs, supra note 22, at I–2.
27 Patient Protection and Affordable Care Act, 42 U.S.C. § 1395jjj (2012).
28 Under the shared savings payment model, the ACO continues to receive fee-for-service
based payments, but Medicare also rewards an ACO that meets or exceeds its targeted cost
savings with a bonus equal to a percentage of the savings. See Medicare Program; Medicare
Shared Savings Program: Accountable Care Organizations, 76 Fed. Reg. 67,802, 67,927 (Nov. 2,
2011) (to be codified at 42 C.F.R. pt. 425). The Medicare Shared Savings Program also includes
economic incentives for ACOs to improve quality by tying a portion of an ACO’s
reimbursement to its performance on quality benchmarks. For example, an ACO that performs
poorly on the relevant quality measures may be ineligible for any bonus payment under the
shared savings or shared savings and risk payment models, even if the ACO lowers the cost of
care. See 42 C.F.R. § 425.100(b) (2012) (stating that ACOs participating in the Medicare Shared
Savings Program are eligible for shared savings only if they meet the minimum quality
performance standards, among other requirements). After completing their initial term in the
program, providers participating in an ACO will continue to receive a percentage of any
Medicare savings but also will be penalized with a downward adjustment in their Medicare

2017] N O N C O M P L I A N T P A T I E N T S 135

Experimentation with the VBP payment models began among
private payors and state Medicaid programs in the mid-1990s, with the
Medicare program joining the trend over ten years ago.29 Today, payors
have fully embraced VBP payment models. In 2015, the U.S.
Department of Health and Human Services (HHS) set the ambitious
goal of shifting ninety percent of traditional Medicare payments to these
new payment models by 2018.30 State Medicaid programs and private
payors similarly are adopting VBP payment models.31

As payors continue the shift to VBP payment models, physicians
will see a growing percentage of their compensation tied to performance
metrics. For example, physicians participating in Medicare’s
forthcoming Merit-Based Incentive Payment System (MIPS) will see the
portion of their fees linked to performance measures increase from four
percent in 2019 to nine percent in 2022.32 Private payors also are

reimbursement rates if the ACO does not meet targeted cost savings. See 42 C.F.R. § 425.600(b)
(providing that for subsequent agreement periods, an ACO may not operate under the one-
sided model described at 42 C.F.R. § 425.604, leaving available only the two-sided model
described at 42 C.F.R. § 425.606). ACOs also may elect to enroll in the shared savings and risk
model during their initial term. See 42 C.F.R. § 425.600(a) (providing that during its initial
agreement period, an ACO may elect to operate under either the one-sided model or two-sided
model). A second example of alternative risk-based payment models is bundled payments,
where the payor makes a single payment for an episode of care that then is allocated among
multiple providers treating the patient. Bundled payment arrangements typically include
payment adjustments based on the providers’ collective performance on quality and efficiency
measures. See DAMBERG ET AL., supra note 22, at xi.
29 See DAMBERG ET AL., supra note 22,. at 1 (describing the history of value-based
purchasing).
30 See Press Release, U.S. Dep’t of Health & Human Servs., Better, Smarter, Healthier: In
Historic Announcement, HHS Sets Clear Goals and Timeline for Shifting Medicare
Reimbursements from Volume to Value (Jan. 26, 2015), https://wayback.archive-it.org/3926/
20170127185400/https://www.hhs.gov/about/news/2015/01/26/better-smarter-healthier-in-
historic-announcement-hhs-sets-clear-goals-and-timeline-for-shifting-medicare-
reimbursements-from-volume-to-value.html.
31 See CTR. FOR HEALTH CARE STRATEGIES, INC., MEDICAID ACCOUNTABLE CARE
ORGANIZATIONS: STATE UPDATE 1 (2015), https://www.chcs.org/media/ACO-Fact-Sheet-06-
13-17 (“Many states have begun to implement Medicaid accountable care organizations
(ACOs) . . . .”); DAMBERG ET AL., supra note 22, at ix; VERNON K. SMITH ET AL., MEDICAID
REFORMS TO EXPAND COVERAGE, CONTROL COSTS AND IMPROVE CARE 35–38 (2015), http://
www.kff.org/report-section/medicaid-reforms-to-expand-coverage-control-costs-and-improve-
care-introduction (reporting that thirty-seven state Medicaid programs in either fiscal year
2015 or fiscal year 2016 are adopting or expanding their initiatives to reward quality and
encourage integrated care, including patient-centered medical homes, health homes, and
ACOs, with some states also implementing episode of care (e.g., bundled payment) initiatives
and/or value-based purchasing initiatives); David Muhlestein, Growth and Dispersion of
Accountable Care Organizations in 2015, HEALTH AFF. BLOG (Mar. 31, 2015), http://
healthaffairs.org/blog/2015/03/31/growth-and-dispersion-of-accountable-care-organizations-
in-2015-2 (reporting that Medicaid ACOs have grown significantly since 2014, and that the
growth in people included in accountable care arrangements since 2014 is primarily from the
commercial and Medicaid sectors).
32 Under the Medicare Access and CHIP Reauthorization Act (“MACRA”) of 2016,
Medicare physicians must choose between two tracks beginning in 2019. The first track, MIPS,
combines three existing Medicare performance incentives programs—the Physician Quality
Reporting Program, the Physician Value-Based Modifier, and the Electronic Health Records

136 C A R D O Z O L A W R E V I E W [Vol. 39:127

expected to increase the portion of physicians’ compensation tied to
performance measures.33

Many physicians have responded to these financial incentives by
restructuring their practices in ways that advance their patients’ health
at lower costs. They have improved coordination of patient care across
different providers and clinical settings, ensured that patients receive
clinically-based preventive care, and avoided providing medical care of
limited value.34 Unfortunately, linking physicians’ payments to the
overall health of their patient populations invites a less welcome
development—physicians avoiding noncompliant patients.

Physicians will achieve financial success under VBP payment
models only if they improve patient outcomes and lower costs. These
goals cannot be realized simply by physicians improving the quality and
efficiency of clinical care, but they also depend on patients adopting
healthier lifestyles and adhering to physicians’ recommendations. For
example, lowering diabetic patients’ A1c levels requires that the diabetic
patient eat a healthier diet, lose weight, and exercise more.35 Greater
patient adherence to medication regimens lowers the risk of
complications, thereby avoiding costly emergency care, hospitalizations,
or other treatments.36 And patients who adhere to post-surgical

Incentive Program. The second track pays an annual incentive payment equal to five percent of
Medicare base payments to those physicians participating in alternative payment models
(payment models that require the physicians to bear financial risk, link payments to various
quality measures and satisfy other requirements specified by CMS). See Medicare Access and
CHIP Reauthorization Act of 2016, Pub. L. No. 114-10, §§ 101(c), (e),129 Stat 87 (2015).
33 Cf. Dave Barkholz, Changing How Doctors Get Paid, MOD. HEALTHCARE (Mar. 11, 2017),
http://www.modernhealthcare.com/article/20170311/magazine/303119983 (quoting various
health care executives who note that a higher percentage of their organizations’ payor contracts
will be based on cost and quality outcomes, with some organizations responding by linking a
higher share of their physicians’ compensation to performance metrics).
34 See MAIA CRAWFORD ET AL., POPULATION HEALTH IN MEDICAID DELIVERY SYSTEM
REFORMS 3–4 (2015), http://www.milbank.org/uploads/documents/papers/CHCS_
PopulationHealth_IssueBrief (discussing the types of health services that improve
population health, such as immunizations, screening for disease, and counseling for tobacco
use, obesity, and other risky behaviors); John V. Jacobi, Multiple Medicaid Missions: Targeting,
Universalism, or Both?, 15 YALE J. HEALTH POL’Y L. & ETHICS 89, 90 (2015) (explaining how
ACOs reduce fragmentation among providers); Lauris Christopher Kaldjian, Patient Care and
Population Health: Goals, Roles and Costs, 3 J. PUB. HEALTH RES. 81, 81 (2014) (“Much of the
current emphasis on cost control is appropriately directed at avoiding tests and treatments that
do not improve health.”).
35 See generally Standards of Medicare Care in Diabetes—2017, 40 J. CLINICAL & APPLIED
RES. & EDUC. 51 (2017) (recommending various drug therapies, nutritional counseling, and
physical activities to help diabetic patients control their condition).
36 See Leslie R. Martin et al., The Challenge of Patient Adherence, 1 THERAPEUTICS &
CLINICAL RISK MGMT. 189, 189 (2005) (explaining that medication nonadherence is a risk
factor for a variety of subsequent health outcomes, including hospitalizations and even death);
Meghan E. McGrady & Kevin A. Hommel, Medication Adherence and Health Care Utilization
in Pediatric Chronic Illness: A Systematic Review, 132 PEDIATRICS 730, 730, 737 (2013)
(reporting that a systematic review found that nine of ten studies demonstrated a relationship
between medication nonadherence and increased health care use among children and
adolescents with chronic medical conditions); Michael C. Sokol et al., Impact of Medication

2017] N O N C O M P L I A N T P A T I E N T S 137

instructions and rehabilitation have higher functional outcomes and
lower morbidity.37 Accordingly, a provider with a high percentage of
noncompliant patients may have difficulty performing well on quality
and cost metrics, resulting in the provider receiving lower payments
under VBP payment models.38 Patients who fail to follow treatment
recommendations therefore threaten physicians’ profitability.39 Given
the challenges physicians face in coaxing patients into sustained lifestyle
changes,40 refusing to treat the noncompliant patient offers physicians
an easy way to protect their bottom-line.

Although there is little empirical evidence documenting the extent
to which VBP payment models lead physicians to reject noncompliant
patients, several physician surveys suggest a real risk of physicians doing
so. In a 2015 survey of primary care physicians who had forty percent of
their compensation linked to quality metrics, fifteen percent reported
“that the compensation model increased the frequency that they
suggested to noncompliant patients that they see a different [primary
care physician].”41 In a 2006 survey of physicians, not only did eighty-

Adherence on Hospitalization Risk and Healthcare Cost, 43 MED. CARE 521, 521 (2005) (finding
that for patients with diabetes and hypercholesterolemia, a high level of medication adherence
was associated with lower disease-related medical costs, including lower hospitalization rates).
37 See Sciberras et al., supra note 2, at e61(1) (“The functional outcome and morbidity after
many orthopaedic surgical procedures are closely dependent on patient compliance with
postoperative instructions and rehabilitation.”).
38 See Page, supra note 3, at 1 (“As payers begin to shift to outcomes-based reimbursements,
physicians with high percentages of nonadherent patients stand to potentially see payments
fall.”).
Providers who treat patients who are sicker on average could perform worse on quality and cost
metrics due to patient factors that are not under the provider’s control (e.g., age, severity of
illness), rather than due to lower quality care. See CTRS. FOR MEDICARE & MEDICAID SERVS.,
FACT SHEET: RISK ADJUSTMENT 1 (2015), https://www.cms.gov/Medicare/Medicare-Fee-for-
Service-Payment/PhysicianFeedbackProgram/Downloads/Risk-Adjustment-Fact-Sheet .
Payors therefore use risk adjustment to adjust scores on quality measures and ensure that
comparisons are fair across providers. See id.; ERIC SCHONE & RANDALL S. BROWN, ROBERT
WOOD JOHNSON FOUND., RISK ADJUSTMENT: WHAT IS THE CURRENT STATE OF THE ART, AND
HOW CAN IT BE IMPROVED? 1 (2013), http://www.rwjf.org/content/dam/farm/reports/reports/
2013/rwjf407046. Risk adjustment, when done properly, thereby deters providers from avoiding
sicker patients. Risk adjustment, however, does not account for differences in the extent to
which a provider’s patients comply with medical advice or adopt healthier behaviors. See, e.g.,
CTRS. FOR MEDICARE & MEDICAID SERVS., supra note 38 (describing the factors CMS takes into
account when performing risk adjustment under the Physician Value-Based Payment Modifier
program). Consequently, risk adjustment does not account for the impact of patient
noncompliance on providers’ quality and cost metric scores.
39 See Page, supra note 3, at 1 (“Physicians have always had to deal with patients who refuse
to follow treatment recommendations, but this age-old quality and patient-care issue is about to
become a pressing financial concern for doctors.”).
40 See Rhonda Dailey et al., Challenges in Making Therapeutic Lifestyle Changes Among
Hypercholesterolemic African-American Patients and Their Physicians, 98 J. NAT’L MED. ASS’N
1895, 1895 (2006) (physicians participating in focus groups reported that barriers to therapeutic
lifestyle changes among their African-American patients with hypercholesterolemia include
lack of patient readiness and responsibility for change, as well as physicians’ lack of time for
and inadequate knowledge about patient counseling).
41 Hibbard et al., supra note 10, at 482–83.

138 C A R D O Z O L A W R E V I E W [Vol. 39:127

two percent respond that quality measures could lead to physicians
avoiding high-risk patients, but in written comments many stated that
poorly compliant patients also would be avoided.42 For example, one
physician commented that “If my pay depended on A1c values [of my
diabetic patients], I have 10–15 patients whom I would have to fire.”43
And in a 2008 survey of California physicians participating in pay-for-
performance programs, some physicians reported that they had “forced
disenrollment of noncompliant patients.”44 Moreover, among the four
physician organizations surveyed, physicians affiliated with the
organization with the largest financial rewards (thirty percent of
physician remuneration) expressed greater resentment toward
noncompliant patients and were more likely to have disenrolled patients
who did not change their behavior.45 This finding suggests that the risk
of physicians firing noncompliant patients likely will increase over time
as a larger percentage of physicians’ incomes are linked to quality and
efficiency metrics.46

B. Reputational Concerns

Various changes in the health care marketplace also incentivize
physicians to avoid noncompliant patients. In an effort to support
patients making informed decisions when choosing among physicians,
government agencies and payors have made available to the public data
comparing physicians’ performance on various quality and cost
measures.47 For example, Medicare’s Physician Compare website allows
patients to compare physician groups’ performance on various
performance measures.48 About half of states also have implemented

42 See Lawrence P. Casalino et al., General Internists’ Views on Pay-for-Performance and
Public Reporting of Quality Scores: A National Survey, 26 HEALTH AFF. 492, 495 (2007).
43 Id.
44 Ruth McDonald & Martin Roland, Pay for Performance in Primary Care in England and
California: Comparison of Unintended Consequences, 7 ANNALS FAM. MED. 121, 121 (2009).
45 See id. at 122–23.
46 Experts similarly believe that under VBP payment models some physicians will resort to
firing patients who fail to follow proffered treatment advice. See Nellie Bristol, Patient
Engagement, Physician Performance Measures Could Be at Odds, Experts Say, CONG. Q.
HEALTHBEAT, Nov. 4, 2011, at 1 (reporting that participants in an Agency for Healthcare
Research and Quality National Advisory Council forum speculated that “[p]hysicians could
resort to ‘firing’ their patients to avoid missing quality-performance measures as patients
become more assertive about making their own health care decisions . . .”).
47 See JULIA JAMES, HEALTH POLICY BRIEF: PUBLIC REPORTING ON QUALITY AND COSTS: DO
REPORT CARDS AND OTHER MEASURES OF PROVIDERS’ PERFORMANCE LEAD TO IMPROVED
CARE AND BETTER CHOICES BY CONSUMERS? 1 (Mar. 8, 2012), http://healthaffairs.org/
healthpolicybriefs/brief_pdfs/healthpolicybrief_65 .
48 See U.S. CTRS. FOR MEDICARE & MEDICAID SERVS., Find Physicians & Other Clinicians,
MEDICARE, https://www.medicare.gov/physiciancompare/search.html (last visited Aug. 3,
2017).

2017] N O N C O M P L I A N T P A T I E N T S 139

public reporting programs, some of which profile physicians,49 and
some payors highlight on their websites physicians identified as
providing high quality and cost-effective care.50 Physicians who perform
poorly on publicly reported data may fear that they will lose patients to
competitors if they develop a reputation for providing low quality,
costly care.51

Low-scoring physicians also may find themselves at a competitive
disadvantage when contracting with private payors. Health insurers
increasingly are adopting narrow networks comprised of higher quality,
lower cost providers, with enrollees receiving care from “out of
network” providers paying significantly higher cost-sharing or footing
their bill entirely.52 In addition, health insurers are making greater use of
tiered provider networks, with plan enrollees paying lower cost-sharing
when they select efficient, high-value providers and higher cost-sharing
when treated by less efficient providers.53 Consequently, physicians with
a poor track record for patient outcomes or efficiency risk exclusion

49 See JAMES, supra note 47, at 2.
50 For example, United Healthcare’s premium designation program collects data on its
network physicians’ performance on quality and efficiency measures, with higher scoring
physicians receiving a premium designation that is displayed publicly on United Healthcare’s
consumer and physician web sites. See UnitedHealth Premium Designation Program,
UNITEDHEALTHCARE, https://www.myuhc.com/content/myuhc/Member/Assets/Pdfs/
Geoaccess/UnitedHealth_Premium_Overview (last visited Sept. 11, 2017).
51 Although the difficulties faced by patients in evaluating the care they receive may limit
their ability to make informed choices among physicians, some physicians nevertheless are
motivated to provide high quality care in order to protect their reputations. See FED. TRADE
COMM’N & DEP’T OF JUSTICE, IMPROVING HEALTH CARE: A DOSE OF COMPETITION 17 (2004),
https://www.ftc.gov/sites/default/files/documents/reports/improving-health-care-dose-
competition-report-federal-trade-commission-and-department-justice/
040723healthcarerpt (stating that although there exists informational and payment barriers
to effective competition, competition can play an important role in enhancing quality of care);
see also Anne Frølich et al., A Behavioral Model of Clinician Responses to Incentives to Improve
Quality, 80 HEALTH POL’Y 179, 187 (2007) (discussing a study of Wisconsin hospitals finding
that public reporting of quality performance made hospitals more likely to adopt quality
improvement programs); David Hyman, The Poor State of Health Care Quality in the U.S.: Is
Malpractice Liability Part of the Problem or Part of the Solution?, 90 CORNELL L. REV. 893, 957
n.364 (2005) (stating that one motive of providers for improving quality may be concern for
their reputation).
52 See Joseph Burns, Narrow Networks Found to Yield Substantial Savings, MANAGED CARE
(Feb. 2012), https://www.managedcaremag.com/archives/2012/2/narrow-networks-found-
yield-substantial-savings (summarizing comments from health insurance executives, who
described adopting narrow networks built around cost-effective, high-quality providers);
Merrill Goozner, Building Narrow Networks That Work, MOD. HEALTHCARE (Dec. 21, 2013),
http://www.modernhealthcare.com/article/20131221/magazine/312219986. (same); see also
AM.’S HEALTH INS. PLANS, ISSUE BRIEF: HIGH-VALUE PROVIDER NETWORKS 3 (2013), https://
www.ahip.org/wp-content/uploads/2016/04/High-Value-Provider-Networks-AHIP-Issue-
Brief (describing trends of insurers adopting smaller provider networks comprised of
providers with a track record of providing high-quality, cost-efficient care).
53 See PAUL FRONSTIN, EMP. BENEFIT RESEARCH INST., TIERED NETWORKS FOR HOSPITAL
AND PHYSICIANS HEALTH CARE SERVICES 3 (2003), http://www.ebri.org/pdf/briefspdf/
0803ib (explaining that payors with tiered networks may assign providers to tiers based on
their quality and efficiency of care). Providers also may be assigned to tiers based on their
charges. See id.

140 C A R D O Z O L A W R E V I E W [Vol. 39:127

from plans’ networks or preferred tiers.54 In an effort to avoid these
competitive disadvantages, physicians may reject noncompliant patients
in order to raise their performance on quality and cost measures.55

II. PHYSICIANS’ LEGAL AND ETHICAL OBLIGATIONS

The law generally treats physicians and patients no differently than
other sellers and buyers in the marketplace for goods and services—free
agents who can choose with whom they wish to do business. Courts
have long afforded physicians the freedom to select their patients.56 The
common law therefore permits a physician to refuse to enter into a
physician-patient relationship with a prospective patient for virtually
any reason. Although federal statutes have carved out exceptions to the
common law rules, these exceptions have been narrowly drawn. Most
notably, anti-discrimination laws generally prohibit physicians and
other health care providers from refusing to treat patients on the basis of
the patient’s race, gender, religion, national origin, age, disability, and
sexual orientation.57 The Emergency Medical Treatment and Active
Labor Act (EMTALA) also requires physicians to screen and stabilize
any individual with an emergency condition accessing a hospital’s
emergency department.58 Beyond these limited exceptions, however,
physicians have no duty to treat any individual with whom they do not
have a physician-patient relationship. Accordingly, a physician can
decline to accept as a new patient any individual the physician believes
will be noncompliant.

Once the physician and patient have entered into a treatment

54 See AM. HOSP. ASS’N, HOSPITALS IN PURSUIT OF EXCELLENCE: A COMPENDIUM OF
ACTION GUIDES 27 (2013), http://www.hpoe.org/Reports-HPOE/2013_
HPOE_Compendium (explaining that less efficient, lower quality providers may “be
excluded from the narrow and tiered networks that are being formed nationwide”); see also
Bryan A. Liang, Deselection Under Harper v. Healthsource: A Blow for Maintaining Patient-
Physician Relationships in the Era of Managed Care?, 72 NOTRE DAME L. REV. 799, 799–803
(1997) (explaining that deselection is driven by managed care plan’s desire to minimize costs);
Suzanne Delbanco, The Payment Reform Landscape: Benefit and Network Design Strategies to
Complement Payment Reform, HEALTH AFF. BLOG (Nov. 4, 2014), http://healthaffairs.org/blog/
2014/11/04/the-payment-reform-landscape-benefit-and-network-design-strategies-to-
complement-payment-reform (stating that higher cost, lower quality providers “do not make it
into the preferred tiers”).
55 See generally supra notes 37–38 and accompanying text (explaining how noncompliant
patients can hurt a provider’s performance on quality and cost measures).
56 See Findlay v. Bd. of Supervisors, 230 P.2d 526, 531 (Ariz. 1951) (stating that a physician
is under “no obligation to engage in practice or to accept professional employment . . .”);
Hurley v. Eddingfield, 59 N.E. 1058 (Ind. 1901) (discussing the discretion given to physicians in
choosing the terms on which they practice medicine); Rice v. Ronaldo, 119 N.E.2d 657, 659
(Ohio Ct. App. 1951) (explaining that a physician is under no legal obligation to render services
to everyone who seeks to engage him); Limbaugh v. Watson, 12 Ohio Law Abs. 150, 151 (Ohio
Ct. App. 1932) (stating that physicians have a right to select their patients).
57 See infra notes 289–92 and accompanying text.
58 See infra note 288 and accompanying text.

2017] N O N C O M P L I A N T P A T I E N T S 141

relationship, the physician can terminate the physician-patient
relationship at any time subject only to the prohibitions under the
aforementioned anti-discrimination laws and the common law rules on
patient abandonment. The law of patient abandonment requires a
physician to provide her patient with all necessary care until
termination of the physician-patient relationship.59 The doctrine further
provides that physicians can unilaterally terminate the physician-patient
relationship only after giving the patient sufficient notice so as to afford
the patient a reasonable opportunity to find another physician.60 Thus,
the patient abandonment doctrine only imposes a procedural restraint
on a physician’s ability to terminate the treatment relationship; it does
not limit the substantive reasons for a physician’s doing so.61 As long as
a physician provides proper notice, she is free to fire a noncompliant
patient.

Ethical guidelines issued by professional physician associations,
including the American Medical Association (AMA) and American
College of Physicians (ACP), echo these legal standards. Consistent with
the common law rule granting physicians discretion on whom to accept
as patients, the AMA’s Principles of Medical Ethics state that “[a]

59 See Groce v. Myers, 29 S.E.2d 553, 557 (N.C. 1944) (“[The physician-patient relationship]
cannot be terminated at the mere will of the physician, but must last until the treatment is no
longer required, or until it is dissolved by the consent of the parties, or until reasonable notice
is given in order that the patient may have an opportunity to engage the services of another.”);
Ricks v. Budge, 64 P.2d 208, 211–12 (Utah 1937) (“The [physician’s] obligation of continuing
attention can be terminated only by the cessation of the necessity which gave rise to the
relationship, or by the discharge of the physician by the patient, or by the withdrawal from the
case by the physician after giving the patient reasonable notice to enable the patient to secure
other medical attention.”); Gray v. Davidson, 130 P.2d 341, 345 (Wash. 1942) (“[W]hen a
physician undertakes to treat a patient, it is his duty to continue to devote his best attention to
the case either until medical attention is no longer needed, he is discharged by the patient, or he
has given the patient reasonable notice of his intention to cease to treat the patient, so that
another physician may be obtained.”).
60 See Hammonds v. Aetna Cas. & Sur. Co., 237 F. Supp. 96 (N.D. Ohio 1965) (discussing
the need for reasonable notice so that another physician may be procured); Scripps Clinic v.
Superior Court, 134 Cal. Rptr. 2d 101, 109 (Cal. Ct. App. 2003) (explaining that a physician can
abandon a patient only after due notice and the opportunity to secure other medical personnel);
Lyons v. Grether, 239 S.E.2d 103, 106 (Va. 1977) (explaining that a physician has the right to
withdraw but needs to afford the patient a reasonable opportunity to find another physician).
Some courts also allow the physician to terminate the physician-patient relationship if she
arranges for another physician to care for the patient. See Norton v. Hamilton, 89 S.E.2d 809,
812 (Ga. Ct. App. 1955) (holding that prior to termination a physician must provide either
notice or a replacement physician). In communities where the physician is the only physician
available to treat the patient, some courts may not allow the physician to terminate the
physician patient relationship even with notice to the patient. See Angela R. Holder, Physician’s
Abandonment of Patient, in 3 AMERICAN JURISPRUDENCE PROOF OF FACTS 117, § 3 (2017) (“If,
of course, as is true in many communities in this country, the physician is the only one
available, it is probable that regardless of the provocation given him by the patient he is not free
to withdraw.”).
61 See MARK A. HALL ET AL., HEALTH CARE LAW AND ETHICS IN A NUTSHELL 112 (3d ed.
1999) (“The only explicit restraint on a doctor’s (or hospital’s) freedom to abandon a patient is
the procedural one of notice. As classically conceived, there is no real substantive content to
abandonment law because the law does not scrutinize the reasons for abandonment . . . .”).

142 C A R D O Z O L A W R E V I E W [Vol. 39:127

physician shall, in the provision of appropriate patient care, except in
emergencies, be free to choose whom to serve . . . .”62 Similarly, the ACP
Ethics Manual provides that a physician-patient relationship arises only
upon the “mutual agreement” of both the physician and patient, and
that “[i]n the absence of a preexisting relationship, the physician is not
ethically obliged to provide care to an individual” absent an emergency
or the unavailability of another physician.63 The AMA’s Code of
Medical Ethics and the ACP Ethics Manual also permit physicians to
terminate the physician-patient relationship with proper notice64;
although, the latter states that physicians should do so only “[u]nder
rare circumstances” and only if “adequate care is available elsewhere and
the patient’s health is not jeopardized” by the dismissal.65 With limited
exceptions, then, the medical profession’s standards of professional
conduct thus permit physicians to refuse to treat noncompliant patients.

III. JUSTIFICATIONS FOR PROHIBITING PHYSICIANS FROM
DISCRIMINATING AGAINST NONCOMPLIANT PATIENTS

The prospect of widespread avoidance of noncompliant patients by
physicians raises fundamental questions regarding whether to afford
physicians this discretion. This Part examines this important issue and
concludes that both policy and moral considerations support legal and
ethical prohibitions against physicians dismissing noncompliant
patients or rejecting prospective patients that a physician believes will be
noncompliant.

A. Reinforcing the Policy Goals Behind Performance Incentive
Programs

In rewarding improved patient outcomes, performance incentive
programs push physicians and their affiliated organizations66 toward

62 See AM. MED. ASS’N, CODE OF MEDICAL ETHICS 1 (2016), https://www.ama-assn.org/
sites/default/files/media-browser/code-of-medical-ethics-chapter-1 [hereinafter CODE OF
MEDICAL ETHICS].
63 Lois Snyder, American College of Physicians Ethics Manual: Sixth Edition, 156 ANNALS
INTERNAL MED. 73, 75 (2012).
64 The AMA’s Code of Medical Ethics provides as follows: “Physicians’ fiduciary
responsibility to patients entails an obligation to support continuity of care for their
patients. . . . When considering withdrawing from a case, physicians must: [n]notify the patient
(or authorized decision maker) long enough in advance to permit the patient to secure another
physician . . . .” CODE OF MEDICAL ETHICS, supra note 62, at 12.
65 Snyder, supra note 63, at 76.
66 While some physicians own their own practices, physicians increasingly are employed by
or otherwise affiliated with larger organizations. See, e.g., Stephen L. Isaacs et al., The
Independent Physician—Going, Going . . . ., 360 NEW ENG. J. MED. 655, 655–57 (2009) (stating
that the percentage of physicians who own their own practices has been declining at a rate of

2017] N O N C O M P L I A N T P A T I E N T S 143

better management of patients’ care.67 In particular, performance
incentive programs encourage physicians and their affiliated
organizations to provide more preventive care, follow evidence-based
guidelines, and improve the coordination of care across providers and
clinical settings.68 Performance incentives also motivate physicians and
their affiliated organizations to address the barriers to patients adopting
healthier behaviors, including social, environmental, and behavioral
health factors, low health literacy, and poor physician-patient
communications. Legal and ethical standards that allow physicians to
reject noncompliant patients weaken these incentives and thereby
frustrate the goals underlying performance incentive programs.

1. Lowering Barriers to Patient Compliance and Healthy
Behaviors

Social and environmental factors play a significant role in shaping
individuals’ health.69 Behavioral health issues, health literacy, and the
quality of physician-patient communications also impact health,
including whether patients comply with their physicians’
recommendations and adopt healthy behaviors.70 In rewarding
providers for improved patient outcomes, performance incentive
programs encourage physicians and their affiliated organizations to
address these barriers to patient compliance.

The social determinants of health include “the circumstances in
which people are born, grow up, live, work and age . . . .”71 They include

approximately two percent for the past twenty-five years, and that the percentage of physicians
in small practices (i.e., practices with ten or fewer physicians) decreased by nearly fifteen
percent between 1996 and 2004). The types of organizations physicians are joining vary but
include integrated delivery systems, multispecialty group practices, and accountable care
organizations. Many physicians also are becoming employees of hospitals. See SUZANNE M.
KIRCHHOFF, CONG. RESEARCH SERV., R42880, PHYSICIAN PRACTICES: BACKGROUND,
ORGANIZATION, AND MARKET CONSOLIDATION (2013), https://fas.org/sgp/crs/misc/
R42880 .
67 See KIRCHHOFF, supra note 66; supra text accompanying note 34.
68 See supra text accompanying note 34; see also Aparna Higgins et al., Provider
Performance Measures in Private and Public Programs: Achieving Meaningful Alignment with
Flexibility to Innovate, 32 HEALTH AFF. 1453, 1456 (2013) (listing the performance
measurement domains and subdomains commonly used by health plans as including care
coordination, patient safety, and preventives services and screenings).
69 See David A. Asch & Kevin G. Volpp, What Business Are We in? The Emergence of
Health as the Business of Health Care, 367 NEW ENG. J. MED. 888, 888 (2012) (“An enormous
body of literature supports the view that differences in health are determined as much by the
social circumstances that underlie them as by the biologic processes that mediate them.”);
Jacobi, supra note 34, at 97 (“[Nonmedical factors] can be more powerfully determinative of the
health of a population than the delivery of traditional health services.”).
70 See Geoffrey R. Swain et al., Health Care Professionals: Opportunities to Address Social
Determinants of Health, 113 WIS. MED. J. 218, 218, 221–22 (2014) (describing the different
types of social determinants affecting health).
71 Social Determinants of Health: Key Concepts, WORLD HEALTH ORG., http://www.who.int/

144 C A R D O Z O L A W R E V I E W [Vol. 39:127

various financial and nonfinancial factors that influence whether
individuals receive appropriate medical care in a timely manner,
including whether they adhere to recommended medication regimens
and obtain follow-up care. For example, financial considerations such as
the inability to pay cost-sharing obligations lead some individuals to
delay or forego needed medical care.72 Transportation challenges,73 lack
of paid sick leave,74 and an inability to arrange for child care during
appointment times may lead patients to forego follow-up care.75 Living
and working conditions also contribute to unhealthy lifestyles. Many
individuals consume less healthy foods because they cannot afford or
lack access to healthier options, increasing their risk for obesity or
malnutrition.76 Similarly, lack of green space or safe neighborhoods

social_determinants/thecommission/finalreport/key_concepts/en (last visited Sept. 15, 2016).
72 See Jeffrey T. Kullgren et al., Nonfinancial Barriers and Access to Care for U.S. Adults, 47
HEALTH SERVICES RES. 462, 467 (2007) (reporting the results of a survey finding that “barriers
in the affordability dimension were the most common reasons for unmet need or delayed care”
(emphasis omitted)). While the insurance reforms and subsidies put in place by the Affordable
Care Act (ACA) lower these barriers, they do not completely eliminate them. See Benjamin D.
Sommers, Health Care Reform’s Unfinished Work—Remaining Barriers to Coverage and Access,
373 NEW ENG. J. MED. 2395, 2395–96 (2015) (stating that for people with higher incomes who
do not qualify for subsidies under the ACA, cost remains a significant barrier to obtaining
health insurance, and even among insured individuals high cost-sharing can limit access to
timely and affordable care). For example, a survey of 10,000 patients found that seventeen
percent identified cost issues as a reason for their not taking their medications as directed by
their physicians. See FROST & SULLIVAN, PATIENT NONADHERENCE: TOOLS FOR COMBATING
PERSISTENCE AND COMPLIANCE ISSUES 4, http://www.frost.com/prod/servlet/cpo/
115071625 .
73 See Kullgren et al., supra note 72, at 470 (identifying transportation problems as a reason
for unmet need or delayed care); Richard Wallace et al., Access to Health Care and
Nonemergency Medical Transportation: Two Missing Links, 1924 J. TRANSP. RES. BOARD 76, 76
(2005) (reporting that approximately 3.6 million Americans do not obtain medical care in a
given year because of lack of transportation).
74 See KEVIN MILLER, CLAUDIA WILLIAMS & YOUNGMIN YI, INST. FOR WOMEN’S POLICY
RESEARCH, PAID SICK DAYS AND HEALTH: COST SAVINGS FROM REDUCED EMERGENCY
DEPARTMENT VISITS iii, 7–8 (2011), https://iwpr.org/wp-content/uploads/wpallimport/files/
iwpr-export/publications/B301-PSD&ED (finding that workers with paid sick days are less
likely to delay seeking care for themselves and their families). For example, the percentage of
workers who underwent mammograms, Pap tests, and endoscopies at recommended intervals,
who had seen a doctor during the previous twelve months, or who had at least one visit to a
health care provider during the previous twelve months was significantly lower among those
lacking paid sick leave as compared to those with sick leave (even after controlling for
sociodemographic and health care-related factors). See Lucy A. Peipins et al., The Lack of Paid
Sick Leave as a Barrier to Cancer Screening and Medical Care-Seeking: Results from the National
Health Interview Survey, 12 BMC PUB. HEALTH 520, 520, 523–24 (2012).
75 See Jason R. Woloski et al., Childcare Responsibilities and Women’s Medical Care, J.
WOMEN’S HEALTH ISSUES & CARE, Jan. 2014, at 4–5 (linking patients foregoing and delaying
care to logistical challenges associated with childcare responsibilities).
76 See Patti Neighmond, People with Low Incomes Say They Pay a Price in Poor Health,
NPR: SHOTS (Mar. 2, 2015, 4:05 AM), http://www.npr.org/sections/health-shots/2015/03/02/
389347123/people-with-low-incomes-say-they-pay-a-price-in-poor-health (profiling the story
of Anna Beer, who after losing her job could no longer afford fresh fruits, vegetables, and
poultry and instead purchased less expensive canned and frozen foods with more salt and
preservatives, which she believed had contributed to her deteriorating health); Michele Ver
Ploeg, Access to Affordable, Nutritious Food is Limited in “Food Deserts,” U.S. DEP’T AGRIC.

2017] N O N C O M P L I A N T P A T I E N T S 145

limit opportunities for exercise.77
Behavioral health issues also influence patient compliance. Patients

who cope with the psychological stress of their disease through denial or
otherwise exhibit a negative attitude toward therapy are more likely to
reject their physicians’ recommendations.78 In addition, patients with
substance abuse problems and mental illness such as depression may
have less capacity for self-management of their disease.79

Studies also link patient compliance to health literacy,80 defined by
the National Institutes of Health as the “degree to which individuals

(Mar. 1, 2010), https://www.ers.usda.gov/amber-waves/2010/march/access-to-affordable-
nutritious-food-is-limited-in-food-deserts (explaining that “food deserts” are neighborhoods
where residents do not live near supermarkets or other food retailers that carry affordable and
nutritious food, and where residents instead rely on smaller stores that often do not carry
healthy foods and/or charge higher prices). These problems can be especially problematic for
self-management of chronic conditions such as diabetes and hypertension. For example,
hospitals have observed a spike in admissions for hypoglycemia among low-income diabetics
during the last week of the month when Supplemental Nutrition Assistance Program (SNAP)
allocations are exhausted. See Hilary K. Seligman et al., Exhaustion of Food Budgets at Month’s
End and Hospital Admissions for Hypoglycemia, 33 HEALTH AFF. 116 (2014).
77 See Gary G. Bennett et al., Safe to Walk? Neighborhood Safety and Physical Activity
Among Public Housing Residents, 4 PLOS MED. 1599 (2007) (discussing the connection between
neighborhood safety and physician activity); Ross C. Brownson et al., Environmental and Policy
Determinants of Physical Activity in the United States, 12 AM. J. PUB. HEALTH 1995 (2001)
(finding that neighborhood characteristics such as sidewalks, enjoyable scenery, and heavy
traffic were positively associated with physical activity).
78 See Jin et al., supra note 3, at 277 (“Fifteen studies showed an association between
patients’ negative attitude toward therapy (e[.]g[.], depression, anxiety, fears of anger about
their illness) and their compliance.”); Orentlicher, supra note 12, at 1580 (“Dialysis patients
typically respond to the psychological stresses of their disease and its treatment with the coping
mechanisms of denial. Too much denial by a patient of his or her condition may cause rejection
of therapeutic recommendations . . . .”).
79 See Elizabeth H.B. Lin et al., Relationship of Depression and Diabetes Self-Care,
Medication Adherence, and Preventive Care, 27 DIABETES CARE 214 (2004) (finding that among
diabetics, patients with major depression had poorer self-care for patient-initiated behaviors
that are difficult to maintain, including exercise, diet, and medication adherence); Randy A.
Sansone & Lori A. Sansone, Alcohol/Substance Misuse and Treatment Nonadherence: Fatal
Attraction, 5 PSYCHIATRY 19, 43 (2008) (summarizing studies finding that alcohol and
substance abuse reduces patient adherence). Numerous studies have found an association
between patient’s compliance and their mental health and substance abuse problems. For
example, a meta-analysis of studies examining medication adherence found that chronically ill
patients who suffered from depression were seventy-six percent less likely than non-depressed
patients to adhere to their medication regimen. See Jerry L. Grenard et al., Depression and
Medication Adherence in the Treatment of Chronic Diseases in the United States: A Meta-
Analysis, 26 J. GEN. INTERNAL MED. 1175, 1177–78 (2011). Similarly, studies show that
depression in elderly patients with coronary artery disease “affected compliance markedly,” Jin
et al., supra note 3, at 277; and mental illness and substance abuse impacts the extent to which
surgical patients comply with postoperative instructions. See Sciberras et al., supra note 2, at 3.
80 See generally Aleda M.H. Chen et al., Health Literacy and Self-Care of Patients with
Health Failure, 26 J. CARDIOVASCULAR NURSING 447, 448–49 (2011) (finding that health
literacy was positively related to self-care maintenance among patients with heart failure
symptoms); Jin et al., supra note 3, at 277 (summarizing the results of empirical studies
examining the association between health literacy and compliance); Voncella McCleary-Jones,
Health Literacy and Its Association with Diabetes Knowledge, Self-Efficacy and Disease Self-
Management Among African Americans with Diabetes Mellitus, 22 ABNF J. 30 (2011)
(reporting that among African Americans with diabetes mellitus, health literacy was positively

146 C A R D O Z O L A W R E V I E W [Vol. 39:127

have the capacity to obtain, process, and understand basic health
information and services needed to make appropriate health
decisions.”81 Nearly half of all adults in the United States have poor
health literacy.82 Patients with lower health literacy often do not
understand the health education materials provided by physicians or
pharmacists and may have greater difficulty monitoring their
symptoms.83 They also may fear that they will become dependent on
long-term medications and have mistaken beliefs regarding the
effectiveness of medications over time.84

Finally, poor physician-patient communication contributes to
patients lacking full knowledge about their disease and the role that
recommended therapies and behavior changes play in improving their
health. Too often physicians fail to give their patients sufficient and
clear information.85 Nor do all physicians take the time to explore a
patient’s beliefs, concerns, or preferences,86 and rarely do physicians
evaluate their patients’ comprehension.87 The problem of poor
physician-patient communication is particularly acute among patients
with poor health literacy, as the combination of physicians’ use of

associated with diabetes knowledge, and that those with lower levels of diabetes knowledge had
lower levels of dietary self-care activities); Chandra Y. Osborn et al., Health Literacy Explains
Racial Disparities in Diabetes Medication Adherence, 16 J. HEALTH COMM. 268 (2011)
(summarizing research finding an association between health literacy and medication
adherence).
81 CATHERINE SELDEN ET AL., NAT’L INSTS. OF HEALTH, HEALTH LITERACY vi (2000),
https://www.nlm.nih.gov/archive//20061214/pubs/cbm/hliteracy .
82 See MARK KUTNER ET AL., AM. INSTS. FOR RESEARCH, THE HEALTH LITERACY OF
AMERICA’S ADULTS v (2006), https://nces.ed.gov/pubs2006/2006483 (reporting rates of
health literacy in the United States).
83 See Michael K. Paasche-Orlow & Michael S. Wolf, The Causal Pathways Linking Health
Literacy to Health Outcomes, 31 AM. J. HEALTH BEHAV. S1, S19, S23 (2007) (discussing the ways
in which health literacy impacts patient compliance).
84 See Jin et al., supra note 3, at 276–77 (describing patient attitudes and beliefs associated
with lower medication compliance). Patients’ knowledge (or lack thereof) about their disease
and treatment significantly impacts patients’ compliance. For example, patients who believe
their medications are necessary and benefit their health have higher rates of adherence than
those lacking a clear understanding of why their physician prescribed a medication or the
consequences of not taking the medication. See Bender, supra note 6, at 4 (stating that various
studies show that patients are less likely to adhere to their treatment regimens if they lack a
clear understanding of both their illness and the treatment regimen recommended by their
physician, or are confused or have doubts about the prescribed medication regimen); Harry
Chummun & David Bolan, How Patient Beliefs Affect Adherence to Prescribed Medication
Regimens, 22 BRITISH J. NURSING 270, 273 (2013).
85 See R.C. Chaurasia, Compliance—The Root of Therapy, 109 J. INDIAN MED. ASS’N 339,
339 (2011) (“A doctor may be responsible for poor compliance in many ways like by giving
little or insufficient information, poor or less explanation . . . .”).
86 See Chesanow, supra note 5, at 4 (“[Physicians] are often unable to understand
differences in patient preferences regarding information and participation during
consultations. They often fail to listen to patients and explore their views on their disease and
medication.”).
87 See Paasche-Orlow & Wolf, supra note 83, at S22 (“Oftentimes, a great amount of
information is relayed to patients, but providers seldom evaluate patient comprehension in any
meaningful manner.”).

2017] N O N C O M P L I A N T P A T I E N T S 147

medical terms and these patients’ limited health vocabulary leads to
limited patient understanding of the information conveyed.88 As a
result, a significant number of patients, from forty to sixty percent,
cannot correctly report what their physicians expect of them, such as the
physician’s directions for prescribed medications.89 Not surprisingly,
studies evaluating the link between patient adherence and physician
communication repeatedly find higher rates of nonadherence when the
communication between physicians and patients is poor.90

Performance incentive programs push physicians and their
affiliated organizations to address the social, environmental, and
behavioral health conditions that impact patients’ health behaviors.
Many providers assist their patients with income and food insecurity
needs by helping them enroll in public assistance programs that address
their income and food insecurity needs, such as the Supplemental
Nutrition Assistance Program (SNAP) (formerly known as food
stamps), the Women, Infants, and Children (WIC) program, or the
Social Security Disability Insurance program.91 Providers also can
arrange transportation for patients needing transport to and from their
medical appointments.92 Several providers have launched initiatives

88 See Mark V. Williams et al., The Role of Health Literacy in Patient-Physician
Communication, 34 FAM. MED. 383, 384 (2002) (discussing the challenges of patients’
understanding physician’s vocabulary); see also Osborn et al., supra note 80, at 273 (finding that
patients with lower health literacy “report inadequate provider communication across domains
critical to successful chronic disease care and self-management”). For example, diabetic
patients with poor health literacy are more likely than those with adequate health literacy to
rate their physician’s communication as low in the domains of general clarity, explanation of
the patient’s condition, and explanation of processes of care. See Dean Schillinger et al.,
Functional Health Literacy and the Quality of Physician-Patient Communication Among
Diabetes Patients, 52 PATIENT EDUC. & COUNSELING 315 (2004).
89 See Beena Jimmy & Jimmy Jose, Patient Medication Adherence: Measures in Daily
Practice, 26 OMAN MED. J. 155, 156 (2011).
90 See Lut Berben et al., An Ecological Perspective on Medication Adherence, 34 WESTERN J.
NURSING RES. 635, 639–40 (2012) (reporting that a meta-analysis on patient adherence to
treatment recommendations “found that the risk for nonadherence was 19% higher when there
was poor communication with the physician”).
91 See STUART BUTLER, JONATHAN GRABINSKY & DOMITILLA MASI, BROOKINGS INST.,
HOSPITALS AS HUBS TO CREATE HEALTH COMMUNITIES: LESSONS FROM WASHINGTON
ADVENTIST HOSPITAL 7 (2015), https://www.brookings.edu/research/hospitals-as-hubs-to-
create-healthy-communities-lessons-from-washington-adventist-hospital (describing a
hospital’s program to help patients enroll in “social services and benefits for which they are
eligible”); Christopher J. Gearon, Treating Hunger as a Health Issue, U.S. NEWS & WORLD REP.
(Feb. 13, 2014, 10:28 AM), http://health.usnews.com/health-news/hospital-of-tomorrow/
articles/2014/02/13/treating-hunger-as-a-health-issue (stating in response to “payment carrots
and sticks” under the ACA that reward “keeping people well and preventing disease,” some
hospitals are going beyond care-delivery transformations and “teaming up with local
stakeholders . . . in hopes of moving the needle on social determinants of health in
communities”).
92 See, e.g., BUTLER, GRABINSKY & MASI, supra note 91, at 6, 9–10 (describing the efforts of
Washington Adventist Hospital to coordinate transportation to and from appointments);
Imran Cronk, The Transportation Barrier, ATLANTIC (Aug. 9, 2015), http://
www.theatlantic.com/health/archive/2015/08/the-transportation-barrier/399728 (noting that
some health care providers employ community health workers to coordinate transportation for

148 C A R D O Z O L A W R E V I E W [Vol. 39:127

designed to address patients’ food insecurity issues, providing meals to
at-risk individuals and pushing for produce-filled grocery stores in
urban food deserts.93 Recent years also have seen growth in provider-
sponsored medical-legal partnerships94 that assist patients with legal
problems that contribute to poor health.95

Performance incentive programs also encourage physicians,
particularly primary care providers, to more effectively identify and
address their patients’ behavioral health problems. Primary care and
behavioral health care largely operate separately from one another.96
This has contributed to far too many patients receiving little or no
treatment for their behavioral health conditions, as medical providers
often do not recognize their patients’ behavioral health needs or do little
to help patients obtain effective behavioral health treatment.97 Greater
coordination and integration between primary care and behavioral
health providers, however, can improve treatment of behavioral health
conditions,98 which in turn can increase patient compliance.99
Specifically, primary care providers can utilize evidence-based screening
tools to help identify patients with behavioral health needs; establish
stronger referral relationships with behavioral health providers or co-

patients to and from appointments).
93 See Gearon, supra note 91.
94 Whereas there were few medical-legal partnerships five to ten years ago, today
approximately 300 hospitals and health centers have medical-legal partnerships. See NAT’L CTR.
FOR MED. LEGAL PARTNERSHIP, http://medical-legalpartnership.org/partnerships (last visited
Jan. 14, 2016) (presenting statistics on the current number of medical-legal partnerships); Tina
Rosenberg, When Poverty Makes You Sick, a Lawyer Can Be the Cure, N.Y. TIMES:
OPINIONATOR (July 17, 2014, 9:30 PM), http://opinionator.blogs.nytimes.com/2014/07/17/
when-poverty-makes-you-sick-a-lawyer-can-be-the-cure/?_r=0 (“There were few medical-legal
partnerships until about five or 10 years ago . . . .”).
95 Medical-legal partnership (MLP) embeds lawyers and paralegals alongside health care
teams to detect, address, and prevent health-harming social conditions for people and
communities. See The MLP Response, NAT’L CTR. FOR MED. LEGAL PARTNERSHIP, http://
medical-legalpartnership.org/mlp-response (last visited Feb. 5, 2016); KATE MARPLE ET AL.,
NAT’L CTR. FOR MED. LEGAL PARTNERSHIP, FRAMING LEGAL CARE AS HEALTH CARE: A GUIDE
TO HELP CIVIL LEGAL AID PRACTITIONERS MESSAGE THEIR WORK TO HEALTH CARE
AUDIENCES 3 (2015), http://medical-legalpartnership.org/wp-content/uploads/2015/01/
Framing-Legal-Care-as-Health-Care-Messaging-Guide . For example, MLPs may assist
patients with housing issues, such as preventing evictions or suing landlords for noncompliance
with local housing standards, helping victims of domestic violence obtain restraining orders, or
assisting with appeals of public benefit denials. See id.
96 See Sarah Klein & Martha Hostetter, In Focus: Integrating Behavioral Health and Primary
Care, COMMONWEALTH FUND, http://www.commonwealthfund.org/publications/newsletters/
quality-matters/2014/august-september/in-focus (last visited Sept. 15, 2017).
97 See id. (stating that as many as sixty to seventy percent of patients with behavioral health
problems leave medical settings without receiving treatment for their behavioral health
conditions).
98 See CHRIS COLLINS ET AL., MILBANK MEMORIAL FUND, EVOLVING MODELS OF
BEHAVIORAL HEALTH INTEGRATION IN PRIMARY CARE 3–4 (2010), https://www.milbank.org/
wp-content/uploads/2016/04/EvolvingCare (making the case for greater coordination and
integration among primary care and behavioral health providers).
99 See supra notes 80–81 and accompanying text (discussing the association between
behavioral health problems and patient compliance).

2017] N O N C O M P L I A N T P A T I E N T S 149

locate behavioral health care in the primary care setting; and regularly
consult with behavioral health providers and jointly develop a treatment
plan that includes both medical and behavioral elements.100

Performance incentive programs also reward those physicians who
increase patient compliance by improving their communication style,
particularly among those patients with poor health literacy.101 Most
obviously, physicians can provide simple and clear information and
instructions102 and take the time to address patient’s concerns about
risks and side effects.103 In addition, physicians can improve compliance
by supplementing oral communications with written information and
directions, information leaflets, pictures, and medication charts, all of
which have been shown to improve patient understanding.104 Physicians
also can improve patient compliance by replacing a didactic approach to
communicating health information with shared decision-making, a
collaborative approach where patients and providers jointly identify
therapeutic goals and treatment options.105

Physicians and their affiliated organizations also can improve
patient compliance by implementing intensive patient education
interventions that target risky behaviors among high-need patients. For
example, participants in a comprehensive diabetes education initiative
reduced their A1c levels and increased their physical activity.106

100 See COLLINS, supra note 98, at 12–13 (identifying approaches to increasing coordination
and integration among primary care and behavioral health providers); Klein & Hostetter, supra
note 96 (same).
101 More effective patient-provider communication can lead to improved patient
compliance. For example, a study of diabetic patients found that effective patient-provider
communication (both general communication processes and diabetes-specific
communications) led to better self-care among patients, even when controlling for patients’
sociodemographic characteristics, health status, and other characteristics of the health care
context. See John O. Piette et al., Dimensions of Patient-Provider Communication and Diabetes
Self-Care in an Ethnically Diverse Population, 18 J. GEN. INTERNAL MED. 624, 632 (2003). As
noted previously, increased patient compliance improves physicians’ performance under
performance incentive programs. See supra Section III.A.
102 See Felicity C. Blackstock et al., Why Don’t Our Patients with Chronic Obstructive
Pulmonary Disease Listen to Us?: The Enigma of Nonadherence, 13 ANNALS AM. THORACIC
SOC’Y 317, 320 (2016) (stating that potential strategies for improving medication adherence
include “providing simple and clear instructions”); Chaurasia, supra note 85, at 340 (“Give
precise and clear instructions to the patient.”).
103 See Jimmy & Jose, supra note 89, at 157 (“Patients’ fears and concerns about adverse drug
reactions can be alleviated by educating patients regarding common side effects of the drugs
which they are taking, how to prevent an adverse drug reaction, if possible, and also convincing
the patient of the need for treatment.”).
104 See generally Jimmy & Jose, supra note 89, at 156–57; Jin et al., supra note 3, at 277;
Paasche-Orlow & Wolf, supra note 83, at S22.
105 See Piette et al., supra note 101, at 624 (stating that diabetic patients have higher rates of
compliance when given a central role in setting self-care goals); Sandra R. Wilson et al., Shared
Treatment Decision Making Improves Adherence and Outcomes in Poorly Controlled Asthma,
181 AM. J. RESPIRATORY & CRITICAL CARE MED. 566 (2010) (finding that patients of providers
who received training in shared decision-making demonstrated dramatically higher adherence
than patients of other providers).
106 See Michael S. Spencer et al., Effectiveness of a Community Health Worker Intervention

150 C A R D O Z O L A W R E V I E W [Vol. 39:127

Intensive smoking cessation interventions also have been shown to
improve long-term smoking cessation rates, which in turn reduces
hospitalizations and mortality rates.107

Finally, performance incentive programs promote physicians and
their affiliated organizations to improve patient compliance by
developing less complex treatment regimens, particularly in the area of
prescribed medications. Studies show that medication adherence rates
fall as the medication regimen becomes more complex and involve
numerous medications with varying dosages and schedules.108
Medication adherence also declines if the cost of the prescribed therapy
is a financial burden for a patient.109 Physicians and their affiliated
organizations therefore can improve medication adherence by
simplifying medication regimens and prescribing lower cost therapies
when available. In addition, prescribing a longer supply (e.g., twelve
months rather than three months) improves compliance given the
inconvenience of obtaining refills.110 Studies also show that sending
automated reminders to patients who have not refilled their
prescriptions, such as telephone calls, emails, text messages, and letters,
improves patients’ medication adherence.111

Among African American and Latino Adults with Type 2 Diabetes: A Randomized Controlled
Trial, 101 AM. J. PUB. HEALTH 2253–60 (2011) (reporting the results of the REACH Detroit
Partnership, which used community health workers to conduct diabetes education class and
two home visits).
107 See Syed M. Mohiuddin et al., Intensive Smoking Cessation Intervention Reduces
Mortality in High-Risk Smokers with Cardiovascular Disease, 131 CHEST 446 (2007). The
authors reported that thirty-three percent of smokers with acute cardiovascular disease who
received twelve weeks of behavior modification counseling and individualized
pharmacotherapy after a hospitalization had continuous smoking cessation rates at twenty-four
months, as compared to only nine percent of those who had received printed educational
materials and counseling prior to discharge. Moreover, during the two-year follow-up period,
those receiving the intensive smoking cessation interventions were less likely to have been
hospitalized and had a lower mortality rate than those who only had received printed
educational materials and pre-discharge counseling. See id.
108 See Jimmy & Jose, supra note 89, at 156 (“Barriers to the effective use of medicines
specifically include . . . complex regimens that require numerous medications with varying
dosing schedules.”); Jin et al., supra note 3, at 278 (citing studies that found a correlation
between medication compliance and the number of dosing times per day). For example, for
patients prescribed a single medication to be taken once daily, the average compliance rates is
approximately eighty percent, whereas the average compliance rate drops to fifty percent for
medications that must be taken four times per day. See Chesanow, supra note 5, at 5.
109 See Patricia Anne O’Malley, Medication Adherence and Patient Outcomes Part 1: Why
Patients Fail to Take Prescribed Medications, 27 CLINICAL NURSE SPECIALIST 227, 228 (2013)
(recommending that physicians prescribe “affordable medications to minimize interrupted
supply or nonfilling of prescriptions due to financial burden”).
110 See John F. Steiner, Rethinking Adherence, 157 ANNALS INTERNAL MED. 580, 583 (2012)
(reporting that women who were prescribed a twelve month supply of oral contraceptives had a
thirty percent decrease in the risk for unintended pregnancies and a forty-six percent decrease
in the risk of abortion as compared to those prescribed a three month supply).
111 For example, among parents identified as not refilling their asthmatic children’s inhaled
corticosteroid medication, those who received computer-generated calls were 25.4% more
adherent than those who did not. See Bruce G. Bender et al., Pragmatic Trial of Health Care
Technologies to Improve Adherence to Pediatric Asthma Treatment: A Randomized Clinical

2017] N O N C O M P L I A N T P A T I E N T S 151

2. Preserving Incentives for Physicians to Lower Barriers to
Compliance and Healthy Behaviors

As the above discussion illustrates, for many patients simply giving
them competent medical advice and treatments will do little to improve
their health if providers then send these patients back to the same living
conditions that made them unhealthy.112 Similarly, physicians will have
limited success in improving patients’ health if they fail to address
patients’ behavioral health issues or poor health literacy. Performance
incentives change this dynamic by rewarding physicians who, together
with their affiliated organizations, improve patients’ health knowledge
and tackle the social, environmental, and behavioral health factors that
contribute to unhealthy behavior and noncompliance.113 In fact, many
physicians and their affiliated organizations have responded to
performance incentives by implementing various programs that address
obstacles to their patients living healthier lives and adhering to medical
advice.114

Legal standards and standards of professional conduct that allow
physicians to reject noncompliant patients, however, weaken the
incentives for providers to tackle the barriers to patient compliance.
Specifically, current legal and ethical standards give physicians an

Trial, 169 JAMA PEDIATRICS 317 (2015). Another study found that among patients receiving
treatment for allergic rhinitis, those receiving text message reminders reported adherence rates
at double of those not receiving text messages. See Kuiji Wang et al., A Randomized Controlled
Trial to Assess Adherence to Allergic Rhinitis Treatment Following a Daily Short Message Service
(SMS) via the Mobile Phone, 163 INT’L ARCHIVES ALLERGY & IMMUNOLOGY 51 (2014). And a
group of Kaiser Permanente patients who received automated telephone call reminders
followed by a reminder letter were 1.6 times more likely to refill their statin prescriptions. See
Bender, supra note 6, at 12–13.
112 See Steven Ross Johnson, Getting to the Root of the Problem: Systems Tackle Social
Conditions to Improve Outcomes Despite the Financial Risks, MOD. HEALTHCARE (Feb. 1, 2014),
http://www.modernhealthcare.com/article/20140201/magazine/302019986 (quoting David
Williams, a professor of public health at Harvard University).
113 See CRAWFORD ET AL., supra note 34, at 2–4, 11 (“[T]here are a growing number of
examples of population health services that extend literally and/or figuratively beyond the
traditional walls of the a clinical setting, . . . [including initiatives] promoting community or
public health services.”); Gearon, supra note 91 (stating in response to “payment carrots and
sticks” under the ACA that reward “keeping people well and preventing disease,” some
hospitals are going beyond care-delivery transformations and “teaming up with local
stakeholders . . . in hopes of moving the needle on social determinants of health in
communities”); Johnson, supra note 112 (“[A] growing number of health systems across the
country . . . have begun tackling the social, economic and environmental conditions in the
communities they serve as part of their programs to reduce hospital readmissions and improve
outcomes.”); see also Jennifer DeCubellis & Leon Evans, Investing in the Social Safety Net:
Health Care’s Next Frontier, HEALTH AFF. BLOG (July 7, 2014), http://healthaffairs.org/blog/
2014/07/07/investing-in-the-social-safety-net-health-cares-next-frontier (“Across the country, a
growing number of innovators in the health care sector are designing care coordination
programs to better serve low-income, high-need populations and begin to address the relevant
social issues.”). See generally Jessica Mantel, Taking Aim at the Social Determinants of Health: A
Central Role for Providers, 33 GA. ST. U. L. REV. 217, 239–42 (2017).
114 See Mantel, supra note 113, at 246–49.

152 C A R D O Z O L A W R E V I E W [Vol. 39:127

escape hatch from the penalties performance incentive programs
impose on providers who do not improve their patients’ health
behaviors. Rather than simplify the medication regimen or provide
automated refill reminders to a patient having trouble with medication
adherence, the physician can simply dismiss the patient. Rather than
arrange transportation for the patient who otherwise will not obtain
regular follow-up care, the physician can terminate the physician-
patient relationship. Rather than provide an intensive education
program for the diabetic patient with continuously high A1c levels, the
physician can refuse to treat the patient.

Physicians can and should do more to promote their patients
leading healthier lives. Indeed, policymakers adopted performance
incentive programs for the very purpose of motivating physicians and
other providers to improve their patients’ health. Legal and professional
standards that allow physicians to reject noncompliant patients thwart
this vitally important policy. Accordingly, if policymakers and
professional associations are serious about health promotion, they must
constrain physicians’ ability to avoid noncompliant patients.

B. Honoring Professional Norms of Beneficence and
Nonmaleficence

The medical ethics literature underscores the special nature of the
physician-patient relationship, a relationship characterized by patients
who depend on physicians using their skills and knowledge to promote
their patients’ well-being.115 The professional norms governing
physicians therefore require that physicians show fidelity to their
patients. Specifically, the norm of beneficence requires that “[t]he
physician’s primary commitment must always be to the patient’s welfare
and best interests . . . [regardless of] patient characteristics, such as
decision-making capacity, behavior, or social status.”116 The corollary
norm of nonmaleficence requires that the physician refrain from any
action that would unnecessarily harm a patient.117

Firing or otherwise avoiding noncompliant patients, however,
flouts the norms of beneficence and nonmaleficence. As described
below, dismissing noncompliant patients results in discontinuity in care,

115 See Marc A. Rodwin, Strains in the Fiduciary Metaphor: Divided Physician Loyalties and
Obligations in a Changing Health Care System, 21 AM. J.L. & MED. 241, 245, 247 (1995)
(characterizing the physician-patient relationship as one where physicians have specialized
knowledge and expertise and patients are dependent on physicians given their illness and
anxiety, and therefore “[c]ontemporary literature in medicine and medical ethics assumes that
physicians are indeed fiduciaries . . . .”).
116 Snyder, supra note 63, at 75.
117 See ELIZABETH MARTIN, Nonmaleficence, CONCISE MEDICAL DICTIONARY (8th ed. 2010)
(defining nonmaleficence as the principle that “doctors should avoid causing harm to
patients”).

2017] N O N C O M P L I A N T P A T I E N T S 153

and this in turn leads to poorer health among noncompliant patients
and higher medical spending. The rejection of noncompliant patients
also lowers these patients’ trust in physicians and can leave patients
feeling stigmatized and ashamed. A prohibition against physicians
rejecting noncompliant patients would curtail these harms, thereby
ensuring that physicians honor their professional norms of beneficence
and nonmaleficence.

1. Eliminating the Adverse Consequences from
Discontinuity in Care

Experts have long-recognized the importance of continuity of care,
defined as “a sustained partnership between patient and clinician.”118
Noncompliant patients rejected by physicians, particularly those fired
by their current provider, experience discontinuity in care. In particular,
these patients may incur an extended interruption in their care, as
physicians may be reluctant to accept as a new patient an individual
dismissed by their prior physician for noncompliance.119 As explained
below, the discontinuity in care experienced by these noncompliant
patients leads to poorer health and higher health costs. Prohibiting
physicians from avoiding noncompliant patients would eliminate these
problems.

An interruption in care can lead to unmet medical needs or a delay
in care. During the time a patient has no usual source of care,120
prescriptions may go un-refilled, diagnostics tests and medical
procedures may be delayed, and medical counseling may be foregone.
For example, one survey found that respondents who experienced a
change in their usual source of care121 during the prior twelve months
were more likely than those with a usual source of care to report unmet
medical needs (12.5% versus 6.6%) and postponement of needed
medical care (32.8% versus 22%).122 Similarly, a 2005 study found that

118 Dong Wook Shin, Impact of Continuity of Care on Mortality and Health Care Costs: A
Nationwide Cohort Study in Korea, 12 ANNALS FAM. MED. 534, 534 (2014).
119 Cf. Wicclair, supra note 12, at 316 (discussing a Sun-Sentinel article describing the
phenomenon of ob-gyns refusing to treat overweight women who are or may become pregnant
given their increased risk of morbidity and mortality).
120 “Usual source of care” means “the particular medical professional, doctor’s office, clinic,
heath center, or other place where a person would usually go if sick or in need of advice about
his or her health.” MEPS Topics: Usual Source of Care, AGENCY FOR HEALTHCARE RES. &
QUALITY, https://meps.ahrq.gov/data_stats/MEPS_topics.jsp?topicid=44Z-1 (last visited Sept.
12, 2017).
121 In addition to being dismissed by a provider, individuals may change their usual source
of care due to changes in health insurance, changing medical needs, quality concerns,
geographic moves, or personal preference. See Maureen A. Smith & Jessica M. Bartell, Changes
in Usual Source of Care and Perceptions of Health Care Access, Quality, and Use, 42 MED. CARE
975, 975 (2004).
122 See id. at 977.

154 C A R D O Z O L A W R E V I E W [Vol. 39:127

relative to those with a stable usual source of care, low-income children
in Oregon who changed their usual source of care reported higher rates
of unmet medical need (25.5% versus 10.6%), unmet prescription need
(32.1% versus 17.3%), problems getting immediate care (23.5% versus
16.7%), problems getting specialty care (38.6% versus 21.8%), unmet
dental need (34.1% versus 19.9%), and unmet counseling need (37.3%
versus 13.9%).123

Although many noncompliant patients eventually find a new
source of care, the discontinuity in their care nevertheless may adversely
impact the quality of care received from their new provider. As
explained by one commentator, “[w]hen patients concentrate their care
with a single physician, these physicians are more likely to develop an
accumulated knowledge about their patients’ medical conditions,”
including “a finer understanding of the severity of each medical
problem and how multiple medical problems interact.”124 In addition,
patients who have continuity with the same physician are more likely to
receive ongoing preventive care.125 Patients with a sustained relationship
with a physician also may develop greater trust in their physician’s
expertise and judgment and thus may be more likely to seek out and
abide by their physician’s opinion.126 In contrast, when a patient fired
for noncompliance switches physicians, the physician-patient bond of
trust and understanding “must laboriously be re-created.”127 In the
interim, the quality of care received by the patient may suffer.

Studies confirm that continuity of care promotes higher quality
care and enhanced patient-adherence and self-management, while
discontinuity in care results in poorer health outcomes and higher
utilization and costs. For example, a 2014 study found that among
Medicare patients with congestive heart failure, chronic obstructive
pulmonary disease, and type 2 diabetes, those with higher levels of
continuity of care were less likely to be hospitalized, require emergency
room visits, or experience complications; they also had lower episode of
care costs.128 Similarly, a study of HMO129 patients with arthritis,

123 See Jennifer E. DeVoe et al., A Medical Home Versus Temporary Housing: The
Importance of a Stable Usual Source of Care Among Low-Income Children, 124 PEDIATRICS
1363, 1369 (2009).
124 James M. Gill et al., The Effect of Continuity of Care on Emergency Department Use, 9
ARCHIVES FAM. MED. 333, 336–37 (2000).
125 See Shin, supra note 118, at 539 (“In addition to cardiovascular mechanisms, patients
[with cardiovascular conditions] who have a sustained relationship with a physician may
receive other preventive services . . . .”).
126 See Gill et al., supra note 124, at 337 (“When patients have a continuity relationship with
their physician, it is likely that they will develop a sense of trust in the physician’s knowledge
and medical judgment.”). See generally infra notes 137–41 and accompanying text (discussing
the association between patient trust and adherence).
127 Mark A. Hall & Carl E. Schneider, Patients as Consumers: Courts, Contracts, and the New
Medical Marketplace, 106 MICH. L. REV. 643, 653 (2008).
128 See Peter S. Hussey et al., Continuity and the Costs of Care for Chronic Disease, 174 JAMA
742 (2014).

2017] N O N C O M P L I A N T P A T I E N T S 155

asthma, epigastric pain/peptic ulcer disease, hypertension, and otitis
media found that patients with greater continuity of care used fewer
health care resources.130 Other studies find similar results.131

These studies indicate that when physicians refuse to treat
noncompliant patients, the resulting discontinuity in care leads to
poorer health for these individuals, as well as higher costs for the health
insurers and government health programs that pay for their care. Such
actions hardly demonstrate a commitment to patients’ best interests. A
prohibition against physicians rejecting noncompliant patients,
however, would promote greater continuity of care, thereby ensuring
that physicians’ honor the professional norms of beneficence and
nonmaleficence.

2. Protecting Patients’ Trust in Physicians

The beneficence norm, with its requirement that a physician act in
her patient’s best interests, preserves patient trust in physicians.132
Patient trust is the cornerstone of successful physician-patient
relationships. Patient trust promotes patients sharing with their
physicians sensitive and confidential information that may be crucial to
the effective diagnosis and treatment of a patient’s ailments.133 In
addition, with trust patients are more willing to seek care from a
physician and follow her medical advice.134 Physicians therefore must

129 HMO is the commonly used abbreviation of “health maintenance organization.”
130 See Michele Raddish et al., Continuity of Care: Is It Cost Effective?, 5 AM. J. MANAGED
CARE 727 (1999) (finding that increasing the number of primary care or specialty care
providers a patient saw was significantly associated with increased prescriptions and
prescription costs, more outpatients visits, and increased hospital admissions).
131 See, e.g., Gill et al., supra note 124, at 336–37 (finding that Medicaid patients with high
provider continuity make fewer emergency department visits). International studies also have
found that patients with greater continuity of care have better health incomes and lower
utilization of expensive care. See Shin, supra note 118, at 535–38 (examining Korean National
Health Insurance enrollees with new diagnosed cardiovascular risk factors and finding that
those with continuity of care scores above the median had increased risk of mortality, incident
myocardial infarction, and incident ischemic stroke, fewer inpatient and outpatients days, and
lower inpatient and outpatient costs relative to those with continuity of care scores below the
median); Jan M. De Maeseneer et al., Provider Continuity in Family Medicine: Does it Make a
Difference for Total Health Care Costs?, 1 ANNALS FAM. MED. 144 (2003) (finding that Belgian
patients who visited the same family physicians during the measured time period had higher
functional status and lower total health care costs than those who visited multiple physicians).
132 See Jessica Mantel, A Defense of Physicians’ Gatekeeping Role: Balancing Patients’ Needs
with Society’s Interests, 42 PEPP. L. REV. 633, 648 (2015) (“Demanding that physicians act as the
patient’s fiduciary . . . reinforc[es] patients’ trust in their physicians.”).
133 See Mark A. Hall, Law, Medicine, and Trust, 55 STAN. L. REV. 463, 498–500 (2002); David
Orentlicher, Health Care Reform and the Patient-Physician Relationship, 5 HEALTH MATRIX
141, 147–48 (1995).
134 See Hall, supra note 133, at 478 (describing the instrumental value of trust in medical
relationships); Orentlicher, supra note 133, at 147 (“The willingness of patients to turn to
physicians for care, to speak openly about intimate and potentially embarrassing information,

156 C A R D O Z O L A W R E V I E W [Vol. 39:127

take care to promote a culture of patient trust. Physicians’ rejection of
noncompliant patients, however, threatens to do the opposite.

Noncompliant patients, fearful of being fired, may withhold
important information from their physicians.135 In particular, patients
may minimize their self-care failures and exaggerate their self-care
successes when communicating with their physician. For example, a
previously fired diabetic patient may not reveal to her new physician
that she has not adhered to the prescribed medication regimen,
continues to eat unhealthy foods, and/or fails to regularly monitor her
glucose levels. Although no studies have examined whether
noncompliant patients do in fact withhold important information from
physicians, both physicians and patients report that patients who fear
being judged by their physicians are reluctant to discuss their true self-
care behaviors.136 Presumably noncompliant patients’ fear of rejection
by physicians would similarly chill physician-patient communications
given the parallels between feeling judged and fears of being rejected by
one’s physician.

Patients fired for noncompliance also may lose trust in the medical
profession generally, leading to even lower levels of treatment adherence
and reduced use of future medical services. When patients do not trust
their physician to act in their best interest, they have less confidence in
the physician’s treatment recommendations.137 Indeed, numerous
studies have found a positive association between treatment adherence
and the degree to which a patient trusts her physician.138 Moreover,

and to rely on their physicians’ recommendations depends in large part on the ability of
patients to trust that physicians are acting primarily to advance the interests of their patients.”).
135 See Lois Shepherd, HIV, the ADA, and the Duty to Treat, 37 HOUS. L. REV. 1055, 1097
(2000) (arguing that physicians should have a duty to treat, as patients “[s]ecure in the
knowledge that they cannot be legally denied medical care” will then “reveal those aspects
about themselves that doctors need to know”).
136 See Marilyn D. Ritholz et al., Barriers and Facilitators to Self-Care Communication
During Medical Appointments in Adults with Type 2 Diabetes, 10 CHRONIC ILLNESS 303 (2014).
137 See Sannisha K. Dale et al., Medical Mistrust Is Related to Lower Longitudinal Medication
Adherence Among African-American Males with HIV, 21 J. HEALTH PSYCHOL. 1311, 1312
(2014) (“[M]edical mistrust may result in suspicions about physicians’
recommendations . . . and consequently lead to lower medication adherence.”); Arch G.
Mainous III et al., Continuity of Care and Trust in One’s Physician: Evidence from Primary Care
in the United States and the United Kingdom, 33 FAM. MED. 22, 22–23, 26 (2001) (“The trust
that patients have in their physician to act in their best interest may also contribute to the
effectiveness in medical care . . . [whereas a] lack of trust by patients may lead to conflict
between the patient and the physician about the cause of the patient’s medical program or
appropriate methods of treatment. Such conflict may, in turn, lead to a lack of patient
adherence . . . .”); Hayley S. Thompson et al., The Group-Based Medical Mistrust Scale:
Psychometric Properties and Association with Breast Cancer Screening, 38 PREVENTIVE MED.
209, 210 (2004) (stating that individuals with greater distrust in the health care system are
expected to have more doubts about the benefits and effectiveness of recommended treatments,
including cancer screening).
138 See Berben et al., supra note 90, at 640 (“In our recent systematic review, we . . . found
that the degree of trust the patient has in the health care professional was one of the factors
most consistently related to medication adherence.”); Leonard L. Berry et al., Patients’

2017] N O N C O M P L I A N T P A T I E N T S 157

patients with low levels of trust frequently delay or forego appropriate
preventive or follow-up care, such as routine check-ups, cancer
screenings, and cholesterol screenings.139 These findings hold not only
when patients mistrust their treating physician, but also when patients
mistrust the health care system generally.140 Patients rejected for
noncompliance therefore may develop a general mistrust of providers
and approach their new physician with wariness, resulting in
diminished treatment adherence and reduced use of needed care. 141

When physicians refuse treatment to noncompliant patients, they
undermine these patients’ trust in physicians. With diminished trust,
noncompliant patients may experience poorer health, as they may be
less forthcoming with their physicians and less willing to seek out and
abide by physicians’ guidance. This state of affairs clearly contravenes
the medical profession’s commitment to promote patients’ well-being
and do no harm. Accordingly, the law and standards of professional
conduct should prohibit physicians from undermining patient trust
through the rejection of noncompliant patients.

3. Avoiding the Stigmatization and Shaming of Patients

Being fired by your physician can be deeply distressing and
stigmatizing. Patients in the United Kingdom who were removed from
their general practitioner’s patient list found removal “profoundly

Commitment to Their Primary Physician and Why It Matters, 6 ANNALS FAM. MED. 6–13 (2008)
(finding that patient trust was positively associated with adherence); Kimberly A. Gudzune et
al., Patients Who Feel Judged About Their Weight Have Lower Trust in Their Primary Care
Providers, 97 PATIENT EDUC. COUNSELING 128, 129 (2014) (summarizing prior studies that
found that patients who trust their primary care providers were more committed to the
primary care relationship and more likely to adhere to medical advice).
139 See Wizdom Powell Hammond et al., Masculinity, Medical Mistrust, and Preventive
Health Services Delays Among Community-Dwelling African-American Men, 25 J. GEN.
INTERNAL MED. 1300 (2010) (finding that Black men with higher medical mistrust were
significantly more likely to delay routine check-ups and cholesterol screenings); Maria C.
Katapodi et al., Distrust, Predisposition to Use Health Services and Breast Cancer Screening:
Results from Multicultural Community-Based Survey, 47 INT’L J. NURSING STUD. 975 (2010)
(finding that distrust of the health care system was associated with lower frequency of
mammograms and provider-given clinical breast exams); Thomas A. LaVeist et al., Mistrust of
Health Care Organizations Is Associated with Underutilization of Health Services, 44 HEALTH
SERVICES RES. 2093 (2009) (finding that patient mistrust of health care organizations is
associated not only with lower adherence, but also with lower health services utilization).
140 See sources cited supra note 139; see also Dale et al., supra note 137, at 1311 (finding that
general medical mistrust among African American males predicted lower continuous
medication adherence over time).
141 See LaVeist et al., supra note 139, at 2012 (“[I]t may be that mistrust emanating from
patient experiences in one aspect of the health care system would lead to general mistrust of
health care.”). For example, patients in the United Kingdom who reported that their prior
general practitioner was “untrustworthy” were more skeptical of their new general practitioner.
See Carolyn Tarrant et al., Continuity and Trust in Primary Care: A Qualitative Study Informed
by Game Theory, 8 ANNALS FAM. MED. 440, 442–43 (2010).

158 C A R D O Z O L A W R E V I E W [Vol. 39:127

stigmati[z]ing,” “threatening,” and “an attack on their right to be a
[National Health Service] patient.”142 These patients also feared that
their removal would negatively affect other physicians’ view of them and
result in their receiving care that differed from what others receive.143
Patients rejected for noncompliance may be particularly prone to these
feelings, as they often feel unfairly judged for their poor health
behaviors given the social, environmental, and behavioral health
challenges they face.144

These feelings of stigma can have psychological consequences and
trigger health-harming stress and maladaptive behaviors. Several studies
demonstrate that feelings of stigma cause stress,145 which in turn can
damage immune defenses, vital organs, and physiological systems.146
Stigma also increases individuals’ risk for behavioral health issues and
poor coping behaviors. For example, studies have found that individuals
who experience weight stigma are at higher risk for depression, anxiety,
substance abuse, and suicide.147 These individuals also engage in binge-
eating and disordered eating patterns and have lower motivation for
health-promoting activities such as exercise.148 To the extent
noncompliant patients fired by their physicians are at risk for similar
problems, they will experience even poorer health and higher medical
costs.

Patients who suffer stigma in the clinical setting also have less trust
in their physicians,149 which as noted above can adversely impact

142 Tim Stokes et al., Patients’ Accounts of Being Removed from Their General Practitioner’s
List: Qualitative Study, 326 BRITISH MED. J. 1316 (2003).
143 The authors of these studies reported that

The dominant form of stigma reported by patients was “felt” stigma (a feeling of
shame and fear of discrimination), particularly a fear that their future health care
would be adversely affected by the “spoiling” of their identity as “good” patients.
Participants feared that their notes had been flagged so that they would receive
different treatment from other [general practitioners] . . . .

See id. at 1317.
144 See Jay A. Jacobson, The Effect of Patients’ Noncompliance on Their Surgeons’ Obligations,
87 SURGICAL CLINICS NORTH AM. 937, 944 (2007) (commenting that patients may feel unfairly
blamed for what they do not feel responsible for when their nonadherence results from
challenging circumstances, such as difficult paying for drugs or an addiction relapse). See
generally infra Section IV.A.1 (explaining why noncompliant patients are not necessarily
blameworthy for their nonadherence and unhealthy behaviors).
145 Mark L. Hatzenbuehler, How Does Sexual Minority Stigma “Get Under the Skin”? A
Psychological Mediation Framework, 135 PSYCHOL. BULL. 707 (2009); Bruce G. Link & Jo C.
Phelan, Stigma and Its Public Health Implications, 367 LANCET 528 (2006).
146 See Paula Braveman, Susan Egerter & David R. Williams, The Social Determinants of
Health: Coming of Age, 32 ANN. REV. PUB. HEALTH 381, 385, 388 (2010) (noting the impact of
stress on health).
147 See Rebecca M. Puhl et al., Overcoming Weight Bias in the Management of Patients with
Diabetes and Obesity, 34 CLINICAL DIABETES 44 (2016).
148 See id.
149 See Jacobson, supra note 144, at 944 (explaining that patients who feel unfairly blamed
for their nonadherence “are likely to lose respect and trust in their physician”).

2017] N O N C O M P L I A N T P A T I E N T S 159

physician-patient communications and lower patient-adherence.150 For
example, some obese patients who experience a lack of empathy from
providers or who feel blamed for their weight are “reluctant to discuss
their weight concerns in light of previous negative experiences.”151
Similarly, researchers found that Black patients who previously
experienced racial discrimination in the health care setting had lower
rates of medication adherence and that this resulted in part from their
diminished trust in their physicians.152 Patients who feel stigmatized
also frequently avoid follow-up care, care that is crucial for managing
ongoing medical issues or early detection of emerging health
problems.153

In sum, when physicians fire or otherwise refuse to treat
noncompliant patients, they inflict great harm on these patients. These
patients experience discontinuity in care, a loss of trust in physicians,
and stigma and shame, all of which contribute to poorer health. Laws
and standards of professional conduct that allow physicians to reject
noncompliant patients thus countenance actions that are harmful to
these patients. In other words, current laws and professional standards
subvert the medical profession’s commitment to the norms of
beneficence and nonmaleficence. Rather than uphold current rules that
rest on dubious ethical grounds, lawmakers and professional
associations should prohibit physicians from rejecting patients for
noncompliance. In doing so, they will reinforce physicians’ moral
commitment to their patients’ welfare and best interests.

C. Honoring Patient Autonomy

The paternalism that characterized the past practice of medicine,
where physicians were presumed to know what is best for patients, has
been replaced with respect for patient autonomy.154 No longer are
patients expected to obediently follow their physicians’
recommendations. Rather, current codes of medical ethics and the laws
of informed patient consent grant patients the right to make medical
decisions consistent with their own values and preferences, decisions

150 See supra notes 137–41 and accompanying text.
151 Puhl et al., supra note 147, at 45.
152 See Yendelela L. Cuffee et al., Reported Racial Discrimination, Trust in Physicians, and
Medication Adherence Among Inner-City African Americans with Hypertension, 103 AM. J. PUB.
HEALTH e55 (2013).
153 See Puhl et al., supra note 147, at 46 (noting that one of the long-term effects of
stigmatization in health care is avoidance of follow-up care).
154 See Jacobson, supra note 144, at 939 (“The long tradition of medical paternalism, doctor
knows best, and following doctor’s ‘orders’ is no longer widely accepted by the American public
or supported by laws related to patient decision making.”); Resnick, supra note 1, at 171
(discussing the demise of medical paternalism).

160 C A R D O Z O L A W R E V I E W [Vol. 39:127

that may run counter to physicians’ recommendations.155 Physicians’
conditioning continued treatment on a patient’s adherence to medical
advice, however, threatens to undermine this fundamental patient right.

Some patients’ nonadherence may be involuntary—they desire to
follow their physicians’ recommendations and live healthier lives, but
they face significant obstacles to doing so or simply lack the willpower.
For many patients, however, their noncompliance is a fully autonomous
choice, a conscious refusal to comply with a physician’s advice after
weighing the relevant benefits and costs for themselves.156 For example,
medications may have deleterious side-effects that some patients wish to
avoid, such as drowsiness that interferes with operating a vehicle or
working.157 Other patients may conclude that dietary restrictions, giving
up smoking, or painful treatments make life less enjoyable.158 Although
some may disagree with a patient’s decision to elevate competing goals
or values above health, the principle of self-determination nevertheless
demands that we respect the patient’s choice.159

When physicians fire or threaten to fire their noncompliant
patients, the physicians’ actions jeopardize patients’ autonomy. Patients
who fear termination of the physician-patient relationship may feel they
have little choice but to follow their physicians’ recommendations, even
if those recommendations are inconsistent with the patient’s preferences
and values.160 As explained by one commentator, “[i]f a physician could
respond to a patient’s refusal of some treatment recommendations by
denying treatment altogether, then patients would have little choice in

155 See Jacobson, supra note 144, at 939 (noting that the current American Medical
Association Code of Ethics “reflects the larger changes in American society that have
diminished the unquestioned authority of leaders and professionals and elevated the autonomy
and rights of individuals and classes of persons, including patients”); Resnik, supra note 1, at
171 (“Medical ethics opinions, codes, and articles written in the last three decades have
emphasized patient’s rights, such as the right to refuse treatment, the right to make medical
decisions, the right to be informed, and so on.”).
156 See Sciberras et al., supra note 2, at e61(2) (“Noncompliance may be considered as a form
of autonomy.”).
157 See Resnik, supra note 1, at 179 (explaining why some patients may have good reasons
for their medication nonadherence).
158 Cf. Orentlicher, supra note 12, at 1581 (listing examples of when patients may “trade
length of life for quality of life”).
159 See Mike W. Martin, Responsibility for Health and Blaming Victims, 22 J. MED. HUMAN.
95, 102 (2001) (“If we disagree with [the patient’s] judgment, we should nevertheless avoid the
self-righteous assumption that there is only one reasonable view of what self-respect requires.”).
160 Orentlicher, supra note 12, at 1581–82. The possibility of physicians engaging in coercive
behavior toward their noncompliant patients is not merely academic conjecture, as some
physicians subject to performance incentives have reported cajoling their patients into
compliance, “scolding them” or acting as the “nagging parent.” See Hibbard et al., supra note
10, at 488 (quoting comments from physicians describing how quality improvement incentives
changed the nature of the physician-patient relationship); McDonald & Roland, supra note 44,
at 123 (stating that U.S. physicians subject to performance incentives tied to patient outcomes
“appeared to increase pressure to cajole and persuade their patients to secure their
compliance”). Firing or threatening to fire noncompliant patients simply takes such behavior to
the next level.

2017] N O N C O M P L I A N T P A T I E N T S 161

their care. They could accept all of their physician’s treatment regimen
or none of it.”

The practice of physicians refusing to treat noncompliant patients
thereby threatens to render patients’ rights to self-determination
meaningless. To guard against the United States returning to an era of
medical paternalism, lawmakers and professional associations should
prohibit physicians from dismissing noncompliant patients or rejecting
prospective patients perceived as noncompliant.

D. Reducing Disparities in Health

Numerous studies have found that the burdens of poor health—
illness, premature death, disability—disproportionately affect vulnerable
populations.161 Racial minorities and individuals with lower levels of
family income and education have higher rates of mortality and
morbidity.162 The prevalence of chronic conditions such as asthma,
diabetes, and hypertension also varies by race and ethnicity, educational
attainment, and family income.163 The federal government has identified
the elimination of disparities in health164 as a national priority, reflecting
a desire to achieve health equity among all Americans.165

161 See CTRS. FOR DISEASE CONTROL & PREVENTION, MORBIDITY AND MORTALITY WEEKLY
REPORT 3 (2013), https://www.cdc.gov/mmwr/preview/ind2013_su.html#
HealthDisparities2013 [hereinafter CDC Health Disparities and Inequalities Report 2013].
162 See id. at 15 (reporting association between mortality and morbidity rates and income
and education levels); The Henry J. Kaiser Family Foundation, State Health Facts: Minority
Health, KAISER FAMILY FOUNDATION, http://kff.org/state-category/minority-health (last visited
Sept. 12, 2017) (reporting higher deaths in 2014 for Blacks as compared to Whites, and higher
rates of cancer (2013), heart disease (2014), asthma (2015), and diabetes (2013) for Blacks as
compared to Whites).
163 See CDC Health Disparities and Inequalities Report 2013, supra note 161, at 93, 100, 146.
According to the CDC, asthma prevalence was 7.9% among Whites, 10.5% among Blacks,
10.8% among American Indians/Alaska Natives, 5.0% among Asians, 14.4% among multi-
race/other-race persons, 15.9% among Puerto Ricans, and 5.4% among Mexicans. See id. at 93.
The prevalence rate for diabetes among adults in 2010 was 4.5 percentage points higher among
Blacks and 4.7% higher among Hispanics than non-Hispanic Whites; 5.8% higher among those
who did not graduate high school than among those with a college degree; and 4.6% higher
among those considered poor as compared to those with high income. See id. at 100. The
prevalence of hypertension among adults from 2007–2010 was 41.3% for Blacks as compared to
28.6% for non-Hispanic Whites, 36.9% for those with less than higher school as compared to
28% for those with a college degree, and 32.8% for those with incomes below the federal
poverty line as compared to 27.6% those with incomes above 500% of the federal poverty line.
See id. at 146.
164 The Centers for Disease Control and Prevention defines disparities in health as observed
differences in health outcomes or health determinants between populations. See id. at 3.
165 See Disparities: Healthy People 2020, OFF. DISEASE PREVENTION & HEALTH PROMOTION,
https://www.healthypeople.gov/2020/about/foundation-health-measures/Disparities (last
updated Aug. 5, 2017) (“During the past 2 decades, one of Healthy People’s overarching goals
has focused on disparities . . . . In Healthy People 2020, that goal was expanded even further: to
achieve health equity, eliminate disparities, and improve the health of all groups.”). “Healthy
People 2020 defines health equity as the attainment of the highest level of health for all people.

162 C A R D O Z O L A W R E V I E W [Vol. 39:127

The emerging trend of physicians rejecting patients for
noncompliance, however, threatens to worsen disparities in health. As
explained below, members of vulnerable populations disproportionately
face social and environmental circumstances that hinder their adopting
healthier lifestyles and adhering to their physicians’ medical advice.
Lower rates of health literacy among vulnerable populations also
negatively impact these patients’ compliance with medical advice.
Consequently, members of vulnerable groups have higher rates of
noncompliance and therefore are more likely to be rejected by
physicians seeking to avoid noncompliant patients. Moreover,
physicians’ tendency to subconsciously apply stereotypes to patients
with certain socio-demographics both negatively influences physicians’
assessments of these patients’ levels of compliance and contributes to
actual differences in patient compliance. Eliminating disparities in
health therefore necessitates a prohibition against physicians refusing
treat to patients for reasons of noncompliance.

1. The Disproportionate Impact of Social and Environmental
Factors

As explained in Section II.A, a range of social and environmental
factors adversely impact individuals’ health behaviors, including their
adherence to medical advice. These factors disproportionately impact
members of vulnerable groups. For example, lower average family
income and lack of paid sick leave contribute to individuals from
vulnerable groups delaying or foregoing needed care.166 Lower income
families also have difficulty affording healthier foods that support
improved weight management and general health, as they generally are
more expensive than less healthy foods.167 In addition, lower-income
communities, as well as rural and predominately minority communities,
often have few if any healthier food retailers168 and existing food outlets

Achieving health equity requires valuing everyone equally with focused and ongoing societal
efforts to address avoidable inequalities, historical and contemporary injustices, and the
elimination of health and health care disparities.” See id.
166 See supra notes 72, 74 and accompanying text (explaining the association between
income and sick leave and individuals’ health). Racial and ethnic minorities are more likely
than Whites to have annual incomes below the federal poverty level. See CDC Health
Disparities and Inequalities Report 2013, supra note 161, at 16 (reporting that in 2011, 16.4% of
non-Hispanic Blacks and 16% of Hispanics had incomes below the federal poverty level, as
compared to 12.4% of non-Hispanic Whites). Only 46% of Hispanic workers have paid sick
days, as compared to 63% of White workers, and only one in three workers earning less than
$15,000 annually have paid sick days while nearly nine in ten workers earning above $65,000
have paid sick days. See INST. FOR WOMEN’S POLICY RESEARCH, PAID SICK DAYS ACCESS AND
USAGE RATES VARY BY RACE/ETHNICITY, OCCUPATION, AND EARNINGS 2, 5 (2016), https://
iwpr.org/wp-content/uploads/wpallimport/files/iwpr-export/publications/B356 .
167 See supra note 76 and accompanying text.
168 See CDC Health Disparities and Inequalities Report 2013, supra note 161, at 20–23.

2017] N O N C O M P L I A N T P A T I E N T S 163

in these communities typically charge higher prices.169 Financial
problems, safety concerns, racism, and other challenges that
disproportionately affect certain groups can lead to chronic stress, with
this stress then adversely impacting adherence to medical advice and
prescribed medications.170

Given these disparities in the social determinants of health, not
surprisingly numerous studies have found that vulnerable populations
have greater noncompliance than other groups. For example, lower-
income adults have greater difficulty controlling their asthma.171 Among
patients with hypertension, racial and ethnic minorities, individuals
with lower levels of family income, and individuals with lower
educational attainment have poorer blood pressure control.172 Studies
also have found that racial and ethnic minorities are less likely than
non-Hispanic Whites to adhere to prescribed medications, 173 and that
those with lower levels of family income and educational attainment

Specifically, the CDC reported that individuals residing in rural census tracts were
approximately four times as likely to lack access to a healthier food retailer as those residing in
urban tracts. See id. at 22. In addition, individuals residing in census tracts with less than 64%
non-Hispanic Whites were half as likely to lack access to a healthier food retailer as tracts with a
higher percentage of non-Hispanic tracts, while persons in census tracts with average income of
below $27,269 were 1.2 times as likely to lack access to a healthier food retailer than census
tracts with higher income. See id. Finally, individuals residing in census tracts where fewer than
27% of the population had a college education were significantly more likely to lack access to a
healthier food retailer (33.3% versus 25.8%). See id. at 22–23.
169 See Renee E. Walker et al., Disparities and Access to Healthy Food in the United States: A
Review of Food Deserts Literature, 16 HEALTH & PLACE 876, 880 (2010) (listing studies that
found that residents living in areas without a supermarket pay more for their food).
170 See infra notes 223–24 and accompanying text. Cf. Megan C. Roberts et al., Racial/Ethnic
and Socioeconomic Disparities in Endocrine Therapy Adherence in Breast Cancer: A Systematic
Review, 105 AM J. PUB. HEALTH e4, e4 (2015) (stating that individuals who have high levels of
social stressors, such as minority women, have “competing social and economic demands [that]
may take priority over medication adherence, leading to suboptimal medication use”).
171 See CDC Health Disparities and Inequalities Report 2013, supra note 161, at 94
(reporting that among adults surveyed between 2006–2010, those with incomes below 100% of
the federal poverty level were 53.9% likely to report having an asthma attack in the past year,
while those with incomes above 450% of the federal poverty level were 48.9% likely to have
reported having had an attack).
172 See id. at 146. Among adults surveyed from 2007–2010, 42.5%, 34.4%, and 30.3% of
Blacks, Hispanics, and Mexican American adults respectively successfully controlled their
blood pressure, as compared to 52.6% of non-Hispanic Whites. Only 41.8% of those with less
than high school controlled their blood pressure, as compared to 52.6% of those with a college
degree, and 46.2% of those with incomes below the federal poverty line did so as compared to
51.4% of those with incomes at or above 500% of the federal poverty line. See id.
173 See Jessica Forsyth et al., Perceived Discrimination and Medication Adherence in Black
Hypertensive Patients: The Role of Stress and Depression, 76 PSYCHOSOMATIC MED. 229, 229
(2014) (noting research finding that Blacks are significantly less likely than Whites to adhere to
antihypertensive medication); Walid F. Gellad et al., Race/Ethnicity and Nonadherence to
Prescription Medications Among Seniors: Results from a National Study, 22 J. GEN. MED. 1572,
1574 (2007) (finding racial disparities in medication adherence among seniors, with 45.3% of
Blacks and 48.8% of Hispanics reporting nonadherence as compared to 41.1% of Whites);
Connie M. Trinacty et al., Racial Differences in Long-Term Adherence to Oral Antidiabetic Drug
Therapy: A Longitudinal Cohort Study, 9 BMC HEALTH SERVICES RES. 24 (2009) (finding that
Black diabetics are have lower medication adherence than White diabetics).

164 C A R D O Z O L A W R E V I E W [Vol. 39:127

also may have lower medication adherence.174
Disparities in social and environmental conditions also may

explain disparities in lifestyle behaviors. Vulnerable populations face
more challenging living and working conditions, which in turn can
cause chronic stress. Stressed individuals frequently engage in harmful
behaviors—such as smoking—in an effort to self-soothe,175 may lack the
physical and emotional energy to exercise, and may have greater
difficulty sleeping.176 Studies of lifestyle behaviors in fact do find
disparities among various groups. Smoking, for example, is much more
prevalent among those with lower levels of educational attainment and
family incomes.177 Government statistics also show that racial and
ethnic minorities and individuals with lower levels of educational
attainment and family income are less likely to meet recommended
guidelines for aerobic physical activity, muscle-strengthening,178 and
hours of sleep each night.179

174 See Jin et al., supra note 3, at 275 (summarizing studies that found an association between
educational level and non-compliance, but noting that some studies did not find an
association); Sharon J. Rolnick et al., Patient Characteristics Associated with Medication
Adherence, 11 CLINICAL MED. & RES. 54 (2013) (finding that medication adherence among
patients with depression, hypertension, diabetes, asthma, chronic obstructive pulmonary
disease, multiple sclerosis, cancer, or osteoporosis was lower among those living in areas with
lower income and educational rates).
175 See infra note 223.
176 See Torbjörn Ǻkerstedt, Psychosocial Stress and Impaired Sleep, 32 SCANDINAVIAN J.
WORK ENV’T & HEALTH 493 (2006) (cross-sectional studies show that stress is associated with
shortened sleep, fragmentation, and possibly a reduction in sleep stages 3 and 4); Linda M.
Delahanty et al., Psychological Predictors of Physical Activity in the Diabetes Prevention
Program, 106 J. AM. DIETETIC ASS’N 698 (2006) (finding that among diabetic patients
participating in a diabetes prevention program, those with lower levels of perceived stress had
higher physical activity levels); J. Firth et al., Motivating Factors and Barriers Toward Exercise
in Severe Mental Illness: A Systematic Review and Meta-Analysis, 46 PSYCHOL. MED. 2869
(2016) (reporting that various studies found that low mood and stress were the most prevalent
barrier to exercise (61% of patients)).
177 In a 2009–2010 survey, 34.6% of those who did not graduate from high school self-
identified as smokers as compared with 13.2% of those with college degrees. See CDC Health
Disparities and Inequalities Report 2013, supra note 161, at 82. In addition, adults with family
incomes below the poverty level are more than twice as likely to smoke than those in the
highest family income group (29.2% versus 13.9%). See CHARLOTTE A. SCHOENBORN ET AL.,
U.S. DEP’T HEALTH & HUMAN SERVS. & CTRS. FOR DISEASE CONTROL & PREVENTION, HEALTH
BEHAVIOR ADULTS: UNITED STATES, 2008–2010, 24 (2013), https://www.cdc.gov/nchs/data/
series/sr_10/sr10_257 .
178 See SCHOENBORN ET AL., supra note 177, at 44–46. In 2008-2010, 50.1% of non-Hispanic
White adults met the 2008 national guidelines for aerobic physical activity and 21.6% met both
the aerobic and muscle-strengthening guidelines. In contrast, only 37% of Blacks and 35.9% of
Hispanic adults met the aerobic guidelines and only 16.7% of Blacks and 12.9% of Hispanics
met both guidelines. Adults with a graduate-level degree were more than twice as likely as those
with less than a high school diploma to have met the guidelines for aerobic physical activity
(63.6% versus 28.9%), and almost three times as likely to have met the muscle-strengthening
guidelines (35.1% versus 11.9%). Finally, adults with incomes above 400% of the federal poverty
level were nearly twice as likely to have met the aerobic guidelines than those with incomes
below the poverty level (57.8% versus 32.4%), and about twice as likely to have met the muscle-
strengthening guidelines (31.4% versus 15.1%). See id. at 44–46.
179 According to the Centers for Disease Control, insufficient sleep increases the risk of a

2017] N O N C O M P L I A N T P A T I E N T S 165

Disparities in health literacy rates also may lead to lower rates of
compliance among certain vulnerable groups. Racial and ethnic
minorities, those with lower levels of educational attainment, and those
living in poverty have lower rates of health literacy.180 Consequently,
patients from these groups often have greater difficulty understanding
and following the health information given to them by physicians and
others.181 This in turn contributes to lower rates of adherence and
poorer self-care maintenance182 among members of these groups as
compared to other patients.183

To summarize, vulnerable populations’ disparate exposure to social
and environmental conditions and lower health literacy contributes to
disparities in compliance and unhealthy lifestyle behaviors. Given these
higher rates of noncompliance, patients from vulnerable populations are
more likely to be fired or rejected by physicians seeking to avoid
noncompliant patients.

2. Physicians’ Implicit Bias

Physicians’ implicit biases also may result in their rejecting patients
from certain groups more frequently than other patients. To simplify
the processing of complex information, people subconsciously apply

range of health conditions, including diabetes, high blood pressure, cardiovascular disease,
obesity, depression, cognitive dysfunction, and injury. See SCHOENBORN ET AL., supra note 177,
at 68. Whereas 72.9% of White adults met the federal government’s Healthy People 2020
objective of adequate sleep (7-8 hours/night), only 64.3% of Black adults did so. 68.8% of adults
with a bachelor’s degree, 72.3% of those with an advanced degree, and 74.1% of those with
family incomes at or above 400% of the poverty level had adequate sleep, as compared to 54.2%
of adults with a GED and 68.6% of those with family incomes below the poverty level. See id. at
69–70.
180 See KUTNER ET AL., supra note 82, at v, 9–14 (reporting lower rates of health literacy
among Black, Hispanic, American Indian/Alaska Native, and multiracial adults; that average
health literacy increases with each higher level of educational attainment; and that adults living
below the poverty level had lower average health literacy than adults living above the poverty
threshold); RIMA RUDD ET AL., EDUC. TESTING SERV., LITERACY AND HEALTH IN AMERICA 3–4
(2004), https://www.ets.org/media/research/pdf/picheath (reporting that health literacy is
strongly related to educational attainment, that White adults have significant higher health
literacy than Black, Hispanic and other adults, and that working adults with additional assets
have higher health literacy).
181 See supra notes 83–84 and accompanying text; see also Chen, supra note 80, at 447
(“Inadequate health literacy can cause difficulties in understanding and following directions
given within the healthcare system.”).
182 For additional discussion of the association between health literacy and adherence and
self-care, see generally Jin et al., supra note 3, at 277 (2008) (summarizing the results of
empirical studies examining the association between health literacy and compliance); Osborn et
al., supra note 80, at 268–69 (summarizing research finding an association between health
literacy and medication adherence); see also supra note 90.
183 See RUDD ET AL., supra note 180, at 43 (stating that disparities in health literacy “may
well increase already existing disparities in health care”); Osborn et al., supra note 80, at 270–71
(concluding that racial and ethnic disparities in health literacy explains in part racial and ethnic
disparities in medication adherence).

166 C A R D O Z O L A W R E V I E W [Vol. 39:127

generalized beliefs about a group to its individual members.184 As
explained by Michelle Van Ryn and Jane Burke, this strategy “can lead
to stereotype usage: the generation of a widely held image of a group of
people through which specific individuals are perceived.”185 The
automatic activation of subconscious stereotypes biases the person’s
judgment about the group’s individual members, often in ways that
conflict with the person’s explicit beliefs.186 For example, a person who
sincerely holds the explicit belief that men and women are equally
competent nevertheless may subconsciously apply negative stereotypes
of women’s competence, resulting in the person judging a female job
applicant as less qualified than a male applicant with a similar resume.187

Numerous studies show that physicians are not immune to
subconscious stereotype usage, and that their implicit biases impact
their professional judgment. Studies directly measuring subconscious
bias188 find that the vast majority of health care professionals have low to
moderate levels of implicit racial and ethnic bias even though most
physicians report no such explicit bias.189 Although less studied than

184 See Michelle Van Ryn & Jane Burke, The Effect of Patient Race and Socio-Economic Status
on Physicians’ Perceptions of Patients, 50 SOC. SCI. & MED. 813, 814 (2000) (“In order to make
the social world more manageable, people often make judgements [sic] about categories or
groups of people and generalize these judgements [sic] to all individuals mentally assigned to
that category or group.”).
185 Id. (citations omitted).
186 See Elizabeth N. Chapman et al., Physicians and Implicit Bias: How Doctors May
Unwittingly Perpetuate Health Care Disparities, 28 J. GEN. MED. 1504, 1505 (2013) (explaining
that implicit bias from the automatic activation of racial, ethnic, gender, and age stereotypes
influences judgment of and behavior toward individuals from the stereotyped groups, and that
this unconscious implicit bias can starkly differ from explicit beliefs).
187 See Corinne A. Moss-Racusin et al., Science Faculty’s Subtle Gender Biases Favor Male
Students, 109 PROCEEDINGS NAT’L ACAD. SCI. 16474 (2012) (reporting the results of a study
where participants rated the résumés of male applicants for a laboratory manager position as
significantly more competent and hirable than identical female applicants).
188 Researchers measure subconscious bias using the computer-based Implicit Association
Test (IAT). This test measures subconscious bias as follows:

The IAT measures the time it takes subjects to match representatives of social groups
(e.g., age, gender, and race) to particular attributes (e.g., good, bad, cooperative, and
stubborn). The IAT operationalizes unconscious bias by hypothesizing that subjects
will match a group representative to an attribute more quickly if they connect these
factors in their minds, regardless of their awareness of this connection. For instance,
the more strongly subjects associate pictures of white persons with good concepts and
pictures of black persons with bad concepts, the more quickly they will match them,
and vice versa. The computerized IAT measures the aggregate time required for these
matching tasks under two conditions (pairings). A difference in average matching
speed for opposite pairings (e.g., black+bad/white+good vs black+good/ white+bad)
determines the IAT score.

Alexander R. Green et al., Implicit Bias Among Physicians and its Prediction of Thrombolysis
Decisions for Black and White Patients, 22 J. GEN. INTERNAL MED. 1231, 1231 (2007).
189 See William J. Hall et al., Implicit Racial/Ethnic Bias Among Health Care Professionals
and Its Influence on Health Care Outcomes: A Systematic Review, 105 AM. J. PUB. HEALTH e60,
e60 (2015) (summarizing studies applying the Implicit Association Test to health care
professionals, all of which but one found evidence of implicit racial and ethnic bias).

2017] N O N C O M P L I A N T P A T I E N T S 167

racial and ethnic bias, evidence suggests that physicians also are not
immune to implicit gender bias,190 and at least one study has found that
heterosexual health care providers have implicit preferences for
heterosexual patients over lesbian and gay patients.191

Studies also demonstrate that physicians’ implicit biases lead them
to unintentionally discriminate against patients from certain socio-
demographic groups despite their genuine desire to provide equitable
care.192 In one study physicians with moderate to high implicit racial
bias scores were less likely to recommend thrombolysis to Black patients
than to White patients.193 Similarly, studies that compare treatment
recommendations for patients identical in all respects except for social
category (e.g., race, gender) also find disparate treatment
recommendations. For example, physicians asked to evaluate
hypothetical patient vignettes were more likely to recommend male
patients for arthroplasty than female patients, while White patients seen
in emergency departments are more likely to receive pain management
interventions than Black and Hispanic patients.194

In addition to impacting treatment recommendations, physicians’
implicit bias may cause them to unfairly judge patients from certain
groups as less compliant than other patients. Several studies have
documented that implicit racial, ethnic, and socioeconomic bias
negatively affect physicians’ views of their minority and lower income
patients’ intelligence, cooperativeness, likelihood of being non-adherent,
and proclivity to engage in risky behavior.195 For example, in a study
that presented physicians with vignettes depicting HIV-positive
patients, physicians predicted that Black men would be less likely than
White men to adhere to prescribed medication regimens.196 Another

190 See Chapman et al., supra note 186, at 1508 (summarizing studies of implicit gender
bias).
191 See Janice A. Sabin et al., Health Care Providers’ Implicit and Explicit Attitudes Toward
Lesbian Women and Gay Men, 105 AM. J. PUB. HEALTH 1831 (2015).
192 See DAYNA BOWEN MATTHEW, JUST MEDICINE: A CURE FOR RACIAL INEQUALITY IN
AMERICAN HEALTH CARE 39 (2015) (“Physicians’ implicit biases lead to unintentional and in
some cases, even unconscious discrimination. The resulting biased behavior may directly
contradict the physician’s sincerely held, explicit beliefs and intentions to provide excellent care
to all patients regardless of their race or ethnicity.”); Hall et al., supra note 189, at e61, e71
(“[E]ven if [providers’] explicit attitudes demonstrate a desire to provide equitable care, health
care providers may unintentionally interact with patients of color less effectively than with
White patients, which may contribute to health disparities.”).
193 See Hall et al., supra note 189, at e72.
194 See Chapman et al., supra note 186, at 1507 (summarizing studies examining implicit
bias).
195 See generally MATTHEW, supra note 192, at 57, 79–91 (explaining how providers’ implicit
beliefs can negatively impact their views of racial and ethnic minority patients and discussing
studies demonstrating this dynamic); Hall et al., supra note 189, at e66–70 (summarizing
studies that found an association between patients’ race and physicians’ implicit and explicit
judgments regarding medical compliance).
196 See Laura M. Bogart et al., Factors Influencing Physicians’ Judgments of Adherence and
Treatment Decisions for Patients with HIV Disease, 21 DECISION PSYCHOL. & RISK PERCEPTIONS

168 C A R D O Z O L A W R E V I E W [Vol. 39:127

study found that physicians perceived Black coronary patients as having
a higher risk of both noncompliance with cardiac rehabilitation and
substance abuse relative to White patients, and perceived coronary
patients with lower socioeconomic status as less likely to comply with
cardiac rehabilitation and more likely to be irrational and lacking self-
control.197 These findings suggest a real risk that physicians will
attribute higher levels of noncompliance to patients with certain socio-
demographics as compared to other patients, regardless of whether the
facts support such judgments. This in turn may lead to physicians firing
patients from certain socio-demographic groups more frequently than
other patients.

Implicit bias also impacts physicians’ verbal and nonverbal
communication in ways that impact patient compliance. Numerous
studies document that physicians’ implicit racial biases influence the
quality of the physician-patient interaction. Differences include “the
length of time doctors spend with minority patients as compared to
Whites; the level of verbal exchange and shared decision-making in
which they engage; their body language; verbal tone, and eye contact;
and their willingness to credit and respond to information provided.”198
Similar communication differences arise in physicians’ interactions with
Hispanic patients.199 Physicians’ implicit racial and ethnic bias also may
impact the content of physician-patient communications, with minority
patients receiving less information about their health status and health
care options than White patients.200

These differences in communication style adversely impact
patients’ judgments about their physicians. For example, studies have
found that physicians’ scores on a test measuring implicit racial bias
were negatively correlated with Black patients’ perceptions of the quality
of the physician-patient communication, a physician’s contextual
knowledge of the patient, patient-centeredness, and physician warmth

28 (2001).
197 See Van Ryn & Burke, supra note 184.
198 MATTHEW, supra note 192, at 108.
199 See id. at 109–10 (discussing research comparing physicians’ interviewing skills with
Anglo-American and Spanish-American patients).
200 As Dayna Matthew explains:

Beyond experiencing different styles of communication, minority patients also
routinely receive less information from their providers about their health and health
choices than white patients. These disparities in the content of the messages that
physicians provide to minorities also appear to be related to unconsciously biased
perceptions of the racial and ethnic groups to which these patients belong, rather
than to individualized judgments about the knowledge and circumstances that
dictate what is appropriate to share with each patient. Some physicians filter the
information they communicate based on their biased expectation that minority
patients cannot afford to pay for excellent health care or will not understand complex
treatment regimens.

Id. at 111.

2017] N O N C O M P L I A N T P A T I E N T S 169

and friendliness.201 Higher physician implicit racial bias scores also were
negatively correlated with whether Black patients felt respected by the
providers, liked the provider, and would recommend their provider to
others.202 Importantly, minority patients’ perceptions of bias can result
in their being less engaged in the physician-patient encounter and
disclosing less information about their health,203 resulting in a “viciously
reciprocal cycle of miscommunication between doctors and
patients . . . .”204

Moreover, the evidence suggests that minority patients respond to
physicians’ implicit bias by reducing both their compliance with medical
advice and future use of medical services.205 Positive physician-patient
interactions promote better patient understanding of medical advice
and greater patient trust, which in turn encourages higher levels of
patient compliance.206 Conversely, patients less satisfied with their
physician-patient interaction are less compliant. Not surprisingly, then,
studies show that physicians’ disparate communication styles are
associated with poorer adherence among racial and ethnic minorities
relative to White patients.207 For example, one study found that
although Black and White patients had similar levels of trust prior to
their initial visits for medical care, their post-visit levels of trust diverged
as a result of disparate physician communication styles; furthermore,
these differences were associated with lower patient adherence among
Black patients relative to White patients.208 Patients who are mistrustful
of or who anticipate prejudicial attitudes from health care providers also
are more likely to avoid future encounters with the health care system.209
This dynamic reveals itself in studies showing that minority patients’

201 See Hall et al., supra note 189, at e72 (summarizing results from studies measuring
association between physicians IAT scores and various domains of patient care).
202 See id.
203 See MATTHEW, supra note 192, at 141 (“[C]onsistent with studies showing [generally]
that when patients perceive that their physicians like, care about, and are interested in them,
they are more likely to volunteer information and actively participate in the clinical encounter,”
studies involving African American patients show that “these patients respond to their white
providers by asking fewer questions, seeking less clarification of information provided by
physicians, and exhibiting a less positive emotional tone in their visits with physicians”); Van
Ryn & Burke, supra note 184, at 823 (“When patients perceive that physicians like them, care
about them and are interested in them as a person, they are more likely to volunteer
information and be more active in the encounter.”).
204 MATTHEW, supra note 192, at 4.
205 See generally id. at 146 (summarizing studies showing an association between physicians’
implicit racial and ethnic bias and minority patients’ poorer adherence and health care
outcomes); Hall et al., supra note 189, at e71–72 (same).
206 See MATTHEW, supra note 192, at 117 (“[When] providers more often create a successful
clinical encounter, and indeed a collaborative partnership, that encourages these patients to
trust their provider and comply with recommendations and instructions.”).
207 See id. at 117–18 (describing studies finding as association between physician-patient
communication styles and disparate adherence rates among White and minority patients).
208 See id. at 118 (summarizing a study conducted by Dr. Howard Gordon of Black and
White lung cancer patients).
209 See supra notes 139–41 and accompanying text.

170 C A R D O Z O L A W R E V I E W [Vol. 39:127

perceived levels of physician bias are associated with fewer follow-up
visits and reduced use of medical services.210

In sum, the research suggests that implicit bias not only distorts
physicians’ perceptions of a patient’s level of compliance based on the
patient’s socio-demographics, but also contributes to actual differences
in the level of compliance among socio-demographic groups.
Consequently, patients from vulnerable populations are more likely to
be fired or rejected for noncompliance by physicians seeking to avoid
noncompliant patients. This in turn reinforces, if not increases, existing
disparities in health between populations. The United States therefore
cannot achieve greater health equity if laws and standards of
professional conduct permit physicians to reject patients for
noncompliance. Lawmakers and professional associations accordingly
should prohibit physicians from doing so.

In sum, current legal rules and ethical standards that permit
physicians to reject noncompliant patients frustrates fundamental
values and policy goals—preserving life and health, protecting patients’
trust in medical professionals, promoting equality, and respecting
human dignity. Society therefore should prohibit physicians from
rejecting noncompliant patients when a physician can offer the
noncompliant patient medically appropriate treatment. Specifically,
Congress and/or state legislatures should adopt laws that protect
noncompliant patients from discrimination in the health care context.
Professional organizations such as the AMA and ACP also should revise
their codes of professional conduct to impose on physicians an
obligation to care for noncompliant patients, with state medical
licensing boards doing the same.

To clarify, physicians’ obligation to treat noncompliant patients
would not be absolute. Physicians should not provide medically
inappropriate care to a noncompliant patient, even if demanded by the
patient. Indeed, if a patient’s noncompliance compromises her ability to
benefit from a particular treatment, providing such care would be
unethical. Moreover, if the physician cannot offer the physician any
medically appropriate treatment, the physician would have no
obligation to enter into or continue the physician-patient
relationship.211 For example, if a surgeon reasonably determines that a

210 See MATTHEW, supra note 192, at 118 (describing the findings of the Gordon study);
Hammond et al., supra note 139, at 1306 (finding that medical mistrust rooted in expectations
of racially biased treatment was associated with Black men delaying routine check-ups and
cholesterol screening). See generally Chapman et al., supra note 186, at 1507 (stating that
minority patients’ negative perceptions of the physician-patient interaction could reduce
patients return for follow-up care).
211 See Orentlicher, supra note 12, at 1582 (“The obligation to treat noncompliant patients
should not be an absolute one,” and “does not imply that patients can demand irrational
medical care . . . . Similarly, when a patient persistently rejects a physician’s proposals for
therapy and the physician has nothing left to offer the patient, the physician is not obligated to
continue the patient-physician relationship.”).

2017] N O N C O M P L I A N T P A T I E N T S 171

noncompliant patient is not a good candidate for surgery, the surgeon
would have no duty to treat the patient. Finally, physicians remain free
to refuse treatment to noncompliant patients for other legitimate
reasons, such as nonpayment or belligerent behavior on the patient’s
part.

IV. JUSTIFICATIONS FOR THE STATUS QUO

Defenders of legal and ethical standards that afford physicians
discretion in deciding whether to treat noncompliant patients offer
several justifications for the current rules: (1) physicians are justified in
terminating the physician-patient relationship because noncompliant
patients should be held accountable for failing to exercise self-care; (2)
physicians should not be forced to assume financial liability for poor
patient outcomes that stem from patients’ lifestyle choices and
nonadherence; (3) requiring physicians to treat noncompliant patients
unduly compromises physicians’ personal autonomy and freedom of
association rights; and (4) firing noncompliant patients serves the
patients’ best interest by motivating the patient to improve their
treatment adherence or, alternatively, allowing the patient to find
another physician better suited to treat the patient. This Part evaluates
these rationales. While not without merit, ultimately these rationales do
not justify a status quo that undermines patients’ health and
compromises fundamental principles of fairness, equality, and
beneficence.

A. The Personal Responsibility Rationale

Many in society hold negative views of those who smoke, are
overweight, unfit, or otherwise engage in unhealthy behaviors.212 This
disdain stems from the underlying assumption that unhealthy behaviors
are the product of faulty character traits such as laziness, weakness, or
ignorance.213 Even some physicians describe their noncompliant
patients in judgmental terms, labelling them as lazy or lacking self-

212 See NAT’L INSTS. OF HEALTH, CLINICAL GUIDELINES ON THE IDENTIFICATION,
EVALUATION, AND TREATMENT OF OVERWEIGHT AND OBESITY IN ADULTS: THE EVIDENCE
REPORT 20 (1998); Bryan Gibson, Psychological Aspects of Smoker-Nonsmoker Interactions:
Implications for Public Policy, 49 AM. PSYCHOLOGIST 1081, 1081 (1994); Lydia Saad, One in
Four Americans Have Less Respect for Smokers, GALLUP (Aug. 5, 2011), http://www.gallup.com/
poll/148850/one-four-americans-less-respect-smokers.aspx.
213 See Gibson, supra note 212 (summarizing studies finding that nonsmokers perceive
smokers less favorably on an array of personal characteristics, including self-discipline,
morality, intelligence, and insecurity); Sophie Lewis et al., How Do Obese Individuals Perceive
and Respond to the Different Types of Obesity Stigma That They Encounter in Their Daily Lives?
A Qualitative Study, 73 SOC. SCI. & MED. 1349, 1350 (2011).

172 C A R D O Z O L A W R E V I E W [Vol. 39:127

control and willpower.214 Some commentators have argued that those
who fail to care for themselves should be held accountable for the
consequences of their unhealthy choices.215 Denying treatment to
noncompliant patients is consistent with this personal responsibility
ethos.216

Two distinct principles underlie the personal responsibility
rationale: First, a patient’s noncompliance justifies ascribing to the
patient a judgment of wrongdoing, which Gerald Dworkin refers to as a
finding of “culpability” or “establish[ing] fault and assign[ing]
blame.”217 Second, termination of the physician-patient relationship
“ought to flow from this first judgment,” which Dworkin refers to as
“liability” or “demands on the person to do something or on others to
act toward him in certain ways.”218 This Section assesses both the
fairness of judging noncompliant patients as morally culpable and
whether physicians’ rejection of noncompliant patients is an appropriate
form of liability. It concludes that a patient’s moral responsibility for her
noncompliance often is open to doubt given the significant impact of
social, environmental, and mental health factors on individuals’
behavior. Moreover, physicians are not the appropriate judges for
determining an individual patient’s culpability for their noncompliance.
Finally, even if noncompliant patients fairly can be characterized as
morally culpable, penalizing them with termination of the physician-
patient relationship goes against the compassion society generally
affords the sick and the health profession’s norms of beneficence and
nonmaleficence.

214 See Resnik, supra note 1, at 170 (“[T]he doctor may view non-adherent patients as a
moral problem. This is the main reason why doctors often describe non-adherent patients in
morally loaded terms, such as ‘lazy,’ ‘difficult’, ‘defiant’, “non-compliant’, or ‘crazy.’”); Puhl et
al., supra note 147, at 44 (describing the negative biases of primary care providers, “including
attitudes that patients with obesity are lazy, lack self-control and willpower, [and are]
personally to blame for their weight . . . .”).
215 See Amy Darby, The Individual, Health Hazardous Lifestyles, Disease and Liability, 2
DEPAUL J. HEALTH CARE L. 787, 788, 792–98 (1999) (highlighting various proposals for holding
individuals responsible for their unhealthy behaviors); Meredith Minkler, Personal
Responsibility for Health? A Review of the Arguments and the Evidence at Century’s End, 26
HEALTH EDUC. & BEHAV. 121, 124–26 (1999) (summarizing the arguments for personal
responsibility for health).
216 See Orentlicher, supra note 12, at 1581 (“Denials of treatment are ethically permissible in
response to patient noncompliance on account of the principle that individuals must assume
responsibility for their actions.”). Orentlicher nevertheless concludes that competing ethical
considerations favor physicians refraining from firing their noncompliant patients. See id.
217 Gerald Dworkin, Voluntary Health Risks and Public Policy: Taking Risks, Assessing
Responsibility, 11 HASTINGS CTR. REP. 26, 28 (1981).
218 Id. Mike Martin similarly speaks of judgment blame, which he defines as “the simple
ascription of wrongdoing to a person who is morally accountable,” liability blame, which
involves “assigning liabilities (costs, penalties, punishment) for harmful consequences,” and
censure blame, which are acts of public criticism, including shunning. Martin, supra note 159,
at 96.

2017] N O N C O M P L I A N T P A T I E N T S 173

1. Noncompliant Patients’ Moral Culpability

Health experts increasingly embrace what Lindsay Wiley calls the
ecological model of health.219 The ecological model “places supposedly
private, individual choices and risks into their social context and
emphasizes structural explanations for health behaviors and
outcomes.”220 It therefore requires that judgments regarding
noncompliant patients’ moral culpability consider the broader social
context that influences individual behavior.

As described in Section III.A, social, environmental, and behavioral
health factors play a significant role in shaping individuals’ behaviors. In
particular, they influence patients’ compliance with their physicians’
recommendations by supporting or constraining an individual’s
capacity to access and pay for recommended medical care, adopt healthy
behaviors, and adhere to physicians’ medical advice. Financial
considerations, transportation and childcare challenges, and lack of sick
leave impact whether individuals obtain timely follow-up care or adhere
to medication regimens.221 Adverse economic and environmental
factors contribute to poor diet and insufficient physical activity.222
Financial problems, unstable housing, and safety concerns cause stress,
and stress in turn can lead individuals to engage in harmful behaviors
such as smoking and substance abuse in an effort to self-soothe.223
Moreover, stressed individuals often lack the physical or emotional
energy or time to carefully follow prescribed medication regimes,
monitor their symptoms, or otherwise self-manage their disease.224
Studies also link patient compliance to health literacy225 and behavioral
health issues, such as depression and anxiety.226 Characterizing
unhealthy behaviors as largely personal failures ignores the significant
influence these factors have on individuals’ conduct.

In highlighting the impact of social, environmental, and mental

219 See Lindsay F. Wiley, Health Law As Social Justice, 24 CORNELL J.L. & PUB. POL’Y 47, 96–
98 (2014) (describing the shift among public health experts away from the behavioral model of
health, which emphasizes individual behavior choices, to the ecological model of health).
220 Id. at 97.
221 See supra notes 73–75 and accompanying text.
222 See supra notes 76–77 and accompanying text.
223 See Sue A. Kaplan et al., The Perception of Stress and Its Impact on Health in Poor
Communities, 38 J. COMMUNITY HEALTH 142, 142, 146–47 (2013) (reporting the results of
focus groups examining how low income individuals perceive stress and its relationship to
health); Fred C. Pampel et al, Socioeconomic Disparities in Health Behaviors, 36 ANN. REV. SOC.
349, 353 (2010) (explaining that those deprived economically and living in disadvantaged
neighborhoods face a variety of chronic stress, and that this may lead to smoking and other
unhealthy behaviors in an attempt to cope or self-medicate).
224 Cf. Chesanow, supra note 5, at 2 (“After a divorce, job loss, or any traumatic event,
depression may set in; taking medication as directed may then be the last thing on the patient’s
mind.”).
225 See sources cited supra note 80.
226 See sources cited supra notes 78–79.

174 C A R D O Z O L A W R E V I E W [Vol. 39:127

health factors on individual’s health behavior, I do not mean to suggest
that all patient noncompliance is involuntary or outside individual
control. Clearly individuals exercise some degree of free will when
choosing a course of action contrary to their physicians’ advice.227
Indeed, some individuals who face challenging social and
environmental circumstances manage to quit unhealthy lifestyles and
adhere to recommended treatment regimes.228 But construing patient
noncompliance as solely a matter of individual choice and personal
responsibility ignores the fact that health behaviors result from a multi-
dimensional interplay of both personal and external factors.

Judgments regarding the moral culpability of noncompliant
patients therefore cannot be made without giving consideration to the
broader social context shaping individuals’ health behaviors. Doing so
raises fundamental questions regarding the extent of noncompliant
patients’ culpability. Can society fairly cast blame on a diabetic patient
for her noncompliance when she has limited financial means and
resides in a food desert with few opportunities for safe exercise? Can
society fairly cast blame on the COPD patient who continues smoking
when he struggles with depression and stressful life events? Can society
fairly cast blame on the surgical patient who fails to follow postoperative
instructions and obtain follow-up care when she has low health literacy
and cannot delay her return to work? As these cases illustrate, when we
view a patient’s noncompliance within the broader social context, for
many patients society rightly softens its moral judgments of their
culpability, even eliminating culpability for patients facing
insurmountable obstacles to healthy behaviors. Accordingly, in many
(perhaps most) cases of patient noncompliance, ascribing wrongdoing
to the patient is unwarranted.229

Moreover, even if some noncompliant patients fairly may be

227 See Minkler, supra note 215, at 126 (“Few . . . would argue that individuals lack any
responsibility for health-related decisions and actions . . . .”).
228 See id. at 130 (“[M]any individuals, despite highly adverse environmental circumstances
and constraints, do manage to quit smoking, dramatically change their diet and exercise
patterns, and in other ways act effectively to improve their health.”).
229 See generally Dworkin, supra note 217, at 31 (arguing that holding individuals
accountable for their unhealthy behaviors on the grounds that they caused their poor health
becomes dubious given “the mixed character of the voluntariness of many behaviors” and
“[t]he difficulty in determining the relative causal role of voluntary vs. nonvoluntary factors in
the genesis of illness”); Martin, supra note 159, at 102–05 (noting that individual responsibility
for illness is open to question given the obstacles many individuals face, and that the
“enormous complexity of causal relationships” for poor health favors compassion rather than
blame); Minkler, supra note 215, at 126 (criticizing the emphasis on personal responsibility
because it ignores “the social context in which the individual decision making and health-
related action take place,” particularly in the case of the poor); Wicclair, supra note 12, at 313
(rejecting the argument that firing noncompliant patients is justified on the grounds of holding
patients accountable for their irresponsible behavior and choices, arguing that there is
“considerable controversy about the extent to which patients are responsible for illnesses”
associated with unhealthy behavior, and that the individual’s responsibility is therefore “open
to question”).

2017] N O N C O M P L I A N T P A T I E N T S 175

judged as morally culpable, physicians are not the appropriate decision
makers for this determination. Many physicians do not fully understand
the complex ways in which social, environmental, and mental health
factors impact their patients’ health-related behaviors.230 Furthermore,
determinations regarding a noncompliant patient’s culpability
ultimately involve value judgments, and there is no reason to think
physicians are uniquely qualified to make these judgments. In fact,
physicians may be uniquely unqualified to make these judgments, as
self-interest may bias them against their noncompliant patients.
Specifically, because firing the noncompliant patient allows physicians
to perform better under performance incentive programs, physicians
may be subconsciously motivated to deem their noncompliant patients
morally culpable and thus deserving of dismissal.231

In sum, the task of assessing a noncompliant patient’s culpability is
both complex and morally ambiguous. Physicians do not possess any
special expertise to make this determination, and casting them as the
moral judges of patients’ noncompliance risks introducing unfair bias
into the process. In practice, then, many noncompliant patients may be
unfairly designated as wrongdoers. This in turn raises fundamental
questions about whether the personal responsibility rationale can justify
physicians’ rejection of their noncompliant patients.

2. Holding Noncompliant Patients Accountable

Even if physicians could fairly judge a noncompliant patient’s
moral culpability, a second question remains—is it just to hold patients
accountable for their noncompliance through the denial of care? I argue
below that the answer to this question is no. Punishing the
noncompliant patient by refusing to enter into or continue the
physician-patient relationship runs counter to both the health
profession’s commitment to benevolence and the compassion society
shows the sick. Accordingly, even if penalizing noncompliant patients in

230 Cf. Nancy P. Chin et al., Social Determinants of (Un)Healthy Behaviors, 13 EDUC. FOR
HEALTH 317, 321 (2000) (noting that a medical education that focuses on the biopsychosocial
model of health produces health providers with insufficient awareness of the social causes of
diseases); Melissa D. Klein et al., Training in Social Determinants of Health in Primary Care:
Does It Change Resident Behavior?, 11 ACAD. PEDIATRICS 387, 387–88 (2011) (advocating for
training primary care physicians to identify and address the social determinants of health, as
many physician residents may not know how to do so and may lack an understanding of these
issues).
231 As I have explained elsewhere, cognitive psychologists have found that when people have
a vested interest in a decision’s outcome, they have an unconscious tendency to form
judgments that suit their desired ends or goals. See generally Jessica Mantel, The Myth of the
Independent Physician: Implications for Health Law, Policy and Ethics, 64 CASE W. RES. L. REV.
455, 498 (2013). Physicians’ judgments regarding their noncompliant patients therefore may be
subconsciously affected by physicians’ financial incentives under various performance incentive
programs. See id. at 504–05.

176 C A R D O Z O L A W R E V I E W [Vol. 39:127

some form would be justified, terminating the physician-patient
relationship is not an appropriate means of doing so.

Commentators have forcefully called for society to show
compassion for the sick “regardless of the cause of the sickness or the
culpability of the afflicted person.”232 This norm of compassion,
however, is more than aspirational rhetoric; it finds expression in the
insurance reforms adopted under the Affordable Care Act (ACA). Prior
to enactment of the ACA,233 private insurers employed a range of
practices that penalized individuals at high-risk of poor health,
including those leading unhealthy lifestyles. Under experience rating
practices, insurers charged higher premiums to those considered at
high-risk, taking into account factors such as whether the individual was
overweight, used tobacco, or engaged in other unhealthy behaviors.234
Insurers also considered an individual’s past and current medical
history, including chronic conditions or other illnesses resulting in part
from poor health habits. 235 For individuals considered extremely high-
risk, insurers would reject their application for insurance altogether or
refuse to re-insure them at the end of a plan year.236 Finally, many
insurers also excluded from coverage pre-existing health conditions,
meaning they would not pay for any future medical expenses associated
with previously diagnosed illnesses that resulted from unhealthy
behaviors.237 Without access to affordable, comprehensive health
insurance, many individuals with poor health behaviors found
themselves unable to access needed care.

The ACA eliminated these practices. Insurers no longer can deny
enrollment or re-enrollment to individuals on the basis of their health238
or deny coverage of pre-existing conditions.239 Nor can they charge
higher-risk individuals steeper premiums (a prohibition known as

232 DANIEL CALLAHAN, FALSE HOPES: OVERCOMING THE OBSTACLES TO A SUSTAINABLE,
AFFORDABLE MEDICINE 199 (1998).
233 Patient Protection and Affordable Care Act, Pub. L. No. 111-148, 124 Stat. 119 (2010).
234 See THE HENRY J. KAISER FAMILY FOUND., HEALTH INSURANCE MARKET REFORMS: RATE
RESTRICTIONS 1 (June 2012), https://kaiserfamilyfoundation.files.wordpress.com/2013/01/
8328 (describing health status rating).
235 See Gary Claxton et al., Pre-existing Conditions and Medical Underwriting in the
Individual Insurance Market Prior to the ACA, HENRY J. KAISER FAM. FOUND. (Dec. 12, 2016),
http://kff.org/health-reform/issue-brief/pre-existing-conditions-and-medical-underwriting-in-
the-individual-insurance-market-prior-to-the-aca (stating that prior to enactment of the ACA,
most individual market insurance policies included general pre-existing condition exclusion
provisions).
236 See id. (explaining medical underwriting); BERNADETTE FERNANDEZ, CONG. RESEARCH
SERV., RL 32237, HEALTH INSURANCE: A PRIMER (2004) (same).
237 See Claxton et al., supra note 235 (explaining exclusion riders, or the practice of insurers
excluding from coverage medical conditions disclosed by an applicant).
238 Patient Protection and Affordable Care Act, Pub. L. No. 111-148, § 2705, 124 Stat. 156
(2010) (codified at 42 U.S.C. § 300gg-4 (2012)).
239 See Pub. L. No. 111-148, § 2704, 124 Stat. 154 (2010) (codified at 42 U.S.C. § 300gg-3
(2012)).

2017] N O N C O M P L I A N T P A T I E N T S 177

“community rating”),240 with the exception of a surcharge on tobacco
users of up to fifty percent.241 Insurers also cannot impose higher cost-
sharing on high-risk individuals.242 Together, these reforms ensure that
health insurance, and in turn medical care, are generally available to all
individuals regardless of lifestyle choices.

Although Republicans in Congress favor repeal and replacement of
the ACA, Republican leaders have voiced support for prohibiting
insurers from denying enrollment or re-enrollment to individuals on
the basis of their health or denying coverage of pre-existing
conditions.243 Moreover, most of the leading conservative health reform
proposals would continue the ACA’s community rating requirement for
those individuals who maintain continuous health care coverage.244

While these insurance reforms specifically address the market for
private health insurance, they more broadly signal society’s reluctance to
punish those with poor health habits by denying them access to health
care (with the glaring exception of higher premiums for tobacco users).
In other words, with the ACA, the United States has rejected an attitude
of blame toward those who fail to exercise self-care in favor of
compassion. Perhaps the public’s compassion reflects an
acknowledgment of and willingness to forgive human imperfection, or a
belief that “[c]ompassion rather than contempt” is more appropriate
when the ill “are the primary victims of their own weakness.”245
Allowing physicians to punish their noncompliant patients by denying
them care, however, violates this norm of compassion for the sick.

Allowing physicians to punish their noncompliant patients also
violates the long-standing professional norms of beneficence and
nonmaleficence. As described above, the beneficence norm requires that
the physician dedicate herself to her patients’ welfare and best interests,
while the corollary norm of nonmaleficence obligates the physician to
refrain from any action that would unnecessarily harm a patient.246 As
argued in Section III.B, firing or otherwise avoiding the noncompliant
patient “hardly demonstrates the[se] commitments.”247

240 See Pub. L. No. 111-148, § 2701, 124 Stat. 155 (2010) (codified at 42 U.S.C. § 300gg
(2012)).
241 See id.
242 See Pub. L. No. 111-148, § 1302, 124 Stat. 165 (2010) (codified at 42 U.S.C.
§ 18022(b)(4)(B) (2012)).
243 See Alex Ruoff, Rand Paul Plan Would OK Denying Coverage for Pre-Existing Conditions,
BLOOMBERG BNA (Jan. 10, 2017), https://www.bna.com/rand-paul-plan-n73014449588; Gene
B. Sperling & Michael Shapiro, How the Senate’s Health-Care Bill Would Cause Financial Ruin
for People With Preexisting Conditions, ATLANTIC (June 23, 2017), https://
www.theatlantic.com/business/archive/2017/06/ahca-senate-bill-preexisting-conditions/
531375.
244 See, e.g., American Health Care Act of 2017, H.R. 1628, 115th Cong. (2017); Better Care
Reconciliation Act, S. 270, 115th Cong. (2017).
245 Martin, supra note 159, at 105.
246 See supra notes 116–17 and accompanying text.
247 Wicclair, supra note 12, at 310. As summarized by one commentator, physicians who fire

178 C A R D O Z O L A W R E V I E W [Vol. 39:127

In sum, imposing liability on the noncompliant patient through
termination of the physician-patient relationship contravenes the
compassion and benevolence generally afforded the sick. When coupled
with concerns that noncompliant patients may be unfairly judged as
wrongdoers, physicians’ rejection of noncompliant patients can only be
described as unjust and excessive. For these reasons, the personal
responsibility rationale does not provide a defensible justification for
current laws and ethical standards that afford physicians discretion on
whether to treat noncompliant patients.

B. The Fairness and Autonomy Rationales

As explained in Part I, physicians will find limited success under
performance incentive programs if they simply improve the clinical care
they give patients; rather, their patients also must adopt healthier
lifestyles and adhere to physicians’ treatment recommendations.
Physicians, however, contend that they can do little to influence their
patients’ health-related behavior, and that holding physicians
accountable for what they cannot control thus unfairly punishes
them.248 As explained by one physician,

When I hear my colleagues talk, it’s the same sort of thing—I’m
being punished for things I can’t control. Patient behavior. You can’t
make people come in [for doctor appointments]. You can’t make
them eat healthy, stop smoking, take their medication. But you can
be punished as a physician if your numbers don’t look good.249

In other words, physicians object to the imposition of financial
liability for an outcome they did not cause and, in their view, are
powerless to change. Avoiding noncompliant patients gives physicians
an escape from this perceived injustice. Consequently, a rule infringing
on physicians’ ability to do so may be criticized as unjust.

Relatedly, physicians may argue that rules that obligate them to
treat noncompliant patients compromise their individual autonomy.250

noncompliant patients “add to the plight of patients who already experience feelings of
confusion, shame and guilt that work against recovery from acute illness and management of
chronic illness.” Martin, supra note 159, at 111.
248 See Hibbard et al., supra note 10, at 490 (reporting that in a 2013 survey of primary care
physicians subject to pay-for-performance, “patients unwillingness to change their behavior”
was listed by respondents as the top obstacle to improving their quality metrics, and more than
a third said that what they found most frustrating about the compensation model was “the fact
that patients’ lifestyle behaviors influenced their salaries”); McDonald & Roland, supra note 44,
at 124 (reporting that California physicians subject to pay-for-performance “believed they were
being held accountable for things beyond their control”).
249 Hibbard et al., supra note 10, at 491; see also Page, supra note 3, at 4 (“[P]atients should
be penalized, not the doctors.”).
250 Cf. Jacobson, supra note 144, at 937–38 (stating that none of his medical students
believed they had an obligation to serve any particular patient or patient population, that they

2017] N O N C O M P L I A N T P A T I E N T S 179

Specifically, they may argue that such a rule violates a time-honored
American norm—the freedom to associate with persons of one’s own
choosing. As previously discussed, current legal and ethical standards
governing the physician-patient relationship reflect this principle,
granting physicians as sellers of medical care the right to reject a
prospective patient or fire an existing patient for virtually any reason.251

This Section evaluates this fairness and autonomy rationales and
concludes that they do not justify legal and ethical rules that permit
physicians to fire their noncompliant patients. Importantly, the fairness
and autonomy rationales ignore the fiduciary nature of the physician-
patient relationship. Like other fiduciaries, physicians have an ethical
obligation to compromise their personal autonomy and financial
interests in favor of the noncompliant patient’s welfare. In addition,
imposing on physicians a duty to treat noncompliant patients would be
consistent with laws that have imposed similar duties on common
carriers, public utilities, and other industries serving important needs.

1. Physicians’ Ethical Obligations Limit Their Autonomy

The traditional freedom of contract found in the marketplace for
goods and services “is predicated on a paradigm of arms-length
bargaining.”252 The relationship between a physician and patient,
however, is not an arms-length relationship but a fiduciary one of
special trust and confidence.253 As such, physicians, like lawyers, are
subject to higher standards of behavior and responsibility than
individuals engaged in ordinary commercial transactions.

The fiduciary norms that govern both the medical and legal
professions find expression in the standards of professional conduct that
govern ongoing relations with patients/clients. Yet only the legal
profession has adopted ethical rules that constrain their members’
ability to unilaterally terminate the client relationship.254 The medical
profession should follow the legal profession’s lead and adopt ethical
rules that limit physicians’ ability to fire their noncompliant patients.

As fiduciaries, physicians generally must give primacy to their
patients’ welfare and best interests.255 This commitment requires
physicians to place their patients’ welfare above their own, including

“resented the idea that others expected or asserted” that they should have such a duty, and that
physicians “[would] not welcome the idea of being told what to do or being obliged to do it”
given that they “value their freedom, autonomy and authority”).
251 See supra Part II.
252 Timothy S. Hall, Third-Party Payor Conflicts of Interest in Managed Care: A Proposal for
Regulation Based on the Model Rules of Professional Conduct, 29 SETON HALL L. REV. 95, 139
(1998).
253 See id.; Rodwin, supra note 115.
254 See infra notes 265–66 and accompanying text.
255 See supra notes 115–16 and accompanying text (discussing the beneficence principle).

180 C A R D O Z O L A W R E V I E W [Vol. 39:127

above their personal financial interests. As stated in the AMA’s Code of
Medical Ethics, “[t]he primary objective of the medical profession is to
render service to humanity; reward of financial gain is a subordinate
consideration. Under no circumstances may physicians place their own
financial interests above the welfare of their patients.”256 Consequently,
once the physician accepts an individual as a patient, any conflict
between the physician’s economic interests and the patient’s welfare
must be resolved in favor of the latter.257

Nevertheless, current ethical standards that obligate physicians to
place their patients’ welfare above their own interests apply only for as
long as the physician agrees to care for the patient.258 As explained in
Part II, physicians are free to terminate the physician-patient
relationship for virtually any reason, including when doing so serves the
physician’s financial interest. In other words, with respect to
termination of the physician-patient relationship, the beneficence norm
yields to physicians’ interests in protecting their personal autonomy and
financial interests. This professional tenet stands in stark contrast to the
ethical rules governing the attorney-client relationship that restrict
when lawyers may withdraw their representation.

The American Bar Association (ABA) Model Rules of Professional
Conduct, which have been adopted by all fifty states and the District of
Columbia (with modifications),259 set forth model ethical rules for
lawyers. Rule 1.16 of the ABA Model Rules of Professional Conduct
regulates the termination of the attorney-client relationship.260 Rule
1.16(b)(1) generally prohibits lawyers from withdrawing from
representing a client if doing so cannot be accomplished “without
material adverse effect on the interests of the client.”261 In focusing on
“the impact the withdrawal will have on the client’s interests, and not

256 CODE OF MEDICAL ETHICS, supra note 62, at 188.
257 See Nathan A. Bostick et al., AM. MED. ASS’N COUNCIL ON ETHICAL & JUDICIAL AFFAIRS,
Report of the Council on Ethical and Judicial Affairs: Physician Pay-for-Performance Programs, 3
IND. HEALTH L. REV. 429, 437 (“Practicing physicians who participate in [pay-for-performance]
programs while providing medical services to patient should maintain primary responsibility to
their patients and provide competent medical care, regardless of financial incentives . . . .”);
Hall, supra note 252, at 130 (“[M]edical ethics recognizes that a physician’s first duty is to her
patient, and that the physician’s personal or financial interest must be subordinated to the
interest of the patient.”); Nancy J. Moore, What Doctors Can Learn from Lawyers About
Conflicts of Interest, 81 B.U. L. REV. 445, 450 (2001) (stating that when conflicts arise between
physicians’ financial interest and the interests of their patients, “it is the ethical duty of the
physicians . . . to resist temptation”).
258 See Thomas H. Boyd, Cost Containment and the Physician’s Fiduciary Duty to the
Patient, 39 DEPAUL L. REV. 131, 137 (1989).
259 See State Adoption of the ABA Model Rules of Professional Conduct, AM. BAR ASS’N,
http://www.americanbar.org/groups/professional_responsibility/publications/model_rules_of_
professional_conduct/alpha_list_state_adopting_model_rules.html (last visited Aug. 7, 2017).
260 See CTR. FOR PROF’L RESPONSIBILITY, AM. BAR ASS’N, ANNOTATED MODEL RULES OF
PROFESSIONAL CONDUCT (8th ed. 2015).
261 MODEL CODE OF PROF’L CONDUCT r. 1.16(b)(1) (AM. BAR ASS’N 2015).

2017] N O N C O M P L I A N T P A T I E N T S 181

the attorney’s reasons for withdrawing,”262 this rule recognizes the
lawyer’s fiduciary obligation to place the client’s interests over the
lawyer’s personal interests. Lawyers who violate Rule 1.16 may be
subject to disciplinary action in the state where licensed.263

Given the similarities between the legal and medical professions,
the ethical rules governing lawyers’ withdrawals from representation are
instructive for the medical profession. The success of both the attorney-
client and physician-patient relationship depends on the client/patient’s
willingness to trust and confide in their lawyer/physician.264 The ethical
rules governing both professions aim to preserve this trust by imposing
fiduciary obligations on its members.265 Yet only the legal profession has
recognized that preserving trust in the profession necessitates a general
prohibition on lawyers withdrawing from representation when doing so
materially harms a client’s interest, particularly when the matter is in
ongoing litigation. Physicians similarly should be prohibited from firing
their noncompliant patients, as such action can undermine patient trust
in the profession,266 compromise patients’ autonomy,267 and harm
patients’ well-being by causing discontinuity in care and feelings of
stigma and shame.268

Physicians may counter that the ethical rules governing lawyers
permit lawyers to withdraw their representation for good cause, even if
detrimental to the client, and that this principle similarly supports
physicians firing their noncompliant patients for good cause. The ABA
Model Rules of Professional Conduct permit the lawyer to withdraw
from representation when the client has engaged in culpable conduct,
such as fraud or insisting on an immoral or unethical course of action.269

262 In re Petition for Distribution of Attorney’s Fees, 870 N.W.2d 755 (Minn. 2015).
263 See MODEL CODE OF PROF’L CONDUCT preamble (2015) (explaining that a lawyer’s
violation of mandatory rules of professional conduct may be the basis of a professional
disciplinary action by a state).
264 See Hall, supra note 252, at 134 (“[T]he physician-patient relationship and the lawyer-
client relationship are both grounded on similar principles of trust and similar ethical duties.”);
Moore, supra note 257, at 447 (“Both doctors and lawyers believe that trust is essential to the
success of the professional relationship.”).
265 See Edward D. Re, The Profession of the Law, 5 J. CIV. RTS. & ECON. DEV. 109, 111 (2000)
(explaining that the elements that comprise the privileges and responsibilities of the legal
profession include “that the client’s trust presupposes that the practitioner’s self-interest is
overbalanced by devotion to serving both the client’s interest as well as the public good”); supra
notes 256–57, 260–62 and accompanying text.
266 See supra Section III.B.2.
267 See supra Section III.C.
268 See supra Sections III.B.1 and III.B.3
269 See MODEL CODE OF PROF’L CONDUCT r. 1.16(b)(2)–(7) (AM. BAR. ASS’N 2015)
(withdrawing representation permitted if “[t]he client persists in a course of action involving
the lawyer’s services that the lawyer reasonably believes is criminal or fraudulent; the client has
used the lawyer’s services to perpetrate a crime or fraud; the client insists upon taking action
that the lawyer considers repugnant or with which the lawyer has a fundamental disagreement;
the client fails substantially to fulfill an obligation to the lawyer regarding the lawyer’s services
and has been given reasonable warning that the lawyer will withdraw unless the obligation is

182 C A R D O Z O L A W R E V I E W [Vol. 39:127

Physicians may argue that they too have good cause to fire their
noncompliant patients in light of these patients’ failure to adhere to
medical advice and adopt healthy behaviors.270 This argument echoes
the personal responsibility rationale discussed in Section IV.A. As
explained above, however, ascribing moral culpability to noncompliant
patients is seldom warranted given the obstacles to healthy behaviors
faced by many patients. Accordingly, a patient’s noncompliance should
not be considered good cause that would justify a physician’s
termination of the physician-patient relationship.

Physicians also may note that Rule 1.16(b)(6) the ABA’s Model
Rules of Professional Conduct permit a lawyer to withdraw if continued
representation of a client “will result in an unreasonable financial
burden on the lawyer.”271 Physicians may argue that they too should be
allowed to terminate the physician-patient relationship when continued
treatment imposes an unreasonable financial hardship, and that the
financial consequences of treating noncompliant patients satisfy this
condition. In applying Rule 1.16(b)(6), however, courts set a high bar
for unreasonableness and are unwilling to approve a lawyer’s request to
withdraw from representation simply because representation will prove
unprofitable.272 As explained by one court,

An attorney has certain obligations and duties to a client once
representation is undertaken. These obligations do not evaporate
because the case becomes more complicated or the work more
arduous or the retainer not as profitable as first contemplated or
imagined. Attorneys must never lose sight of the fact that the
profession is a branch of the administration of justice and not a mere
money-getting trade.273

Consequently, courts only approve lawyers’ requests to withdraw
when the lawyer is unlikely to recover fees274 or any recovery would be

fulfilled; the representation . . . has been rendered unreasonably difficult by the client, or other
good cause for withdrawal exists.”).
270 Cf. Resnik, supra note 1, at 170, 176–77 (explaining that a patient makes a variety of
implicit or explicit promises to the doctor to adhere to their treatment plan or modify poor
health habits, and that when they fail to do so “the patient is not upholding his or her end of the
bargain”).
271 MODEL CODE OF PROF’L CONDUCT r. 1.16(b)(6) (AM. BAR ASS’N 2015).
272 See Eric W. Macaux, Limiting Representation in the Age of Private Law: Exploring the
Ethics of Limited-Forum Retainer Agreements, 19 GEO. J. LEGAL ETHICS 795 (2006) (explaining
that jurisdictions that have applied the financial hardship justification for withdrawing
representation “have set the reasonableness bar high”); Sylvia Stevens, When a Client
Repudiates a Settlement: What Can You Do?, 68 OR. ST. BAR BULL. 9 (2008) (stating that courts
do not look favorable on withdrawal for reasons of financial burden unless the burden is
“unanticipated and significant,” and that “[c]ourts are loathe to allow a lawyer to withdraw
from a case merely because it appears less profitable than originally anticipated”).
273 Haines v. Liggett Grp., 814 F. Supp. 414, 424 (D.N.J. 1993) (citations omitted).
274 See, e.g., City of Joliet v. Mid-City Nat’l Bank of Chi., 998 F. Supp. 2d 689 (N.D. Ill., 2014)
(withdrawal permitted when client unable to pay outstanding fee of $5 million); In re Franke,
55 A.3d 713 (Md. Ct. Spec. App. 2012) (withdrawal permitted when client had not paid

2017] N O N C O M P L I A N T P A T I E N T S 183

“substantially less” than the value of the lawyer’s work.275 The medical
profession similarly should reinforce physicians’ fiduciary obligation to
give primacy to the patient’s welfare over profits.276 Although caring for
noncompliant patients may reduce physicians’ profits under
performance incentive programs, this falls far short of imposing a
significant financial hardship on physicians that would justify dismissal
of the noncompliant patient.

2. Policy Justifications for Limiting Physicians’ Right to Refuse
Treatment to Noncompliant Patients

As argued above, a rule that prohibits physicians from firing their
noncompliant patients reinforces a physician’s fiduciary commitment to
place a patients’ welfare above her own interests. Yet physicians’
fiduciary obligations are triggered only upon entering into the
physician-patient relationship; a physician owes no fiduciary duty to
individuals who are not her patients.277 Arguably, then, physicians
should remain free to reject any prospective patient the physician
believes will be noncompliant. Compelling policy interests, however,
justify restrictions on physicians’ ability to reject both current and
prospective patients.

Although laws regulating the marketplace for goods and services
generally treat sellers and buyers as free agents with broad discretion to
decide with whom they will do business, freedom of association in the
marketplace is not absolute. The law regularly overrides autonomy in
the marketplace in favor of competing values such as equality, human
dignity, and preserving life and health. As described below, imposing on
physicians a duty to treat noncompliant patients would be consistent
with this legal tradition.

Regulators frequently curb the economic independence of actors
that serve important public needs. This so-called public utility
regulation traces back to common law doctrines that imposed on

outstanding fees of $120,000); In re Withdrawal of Attorney, 594 N.W.2d 514 (Mich. Ct. App.
1999) (law firm permitted to withdraw when plaintiff had “virtually no likelihood of recovering
an award of money damages or attorney fees from defendant” and case would require an
additional 1000 or more hours or work).
275 Smith v. R.J. Reynolds Tobacco Co., 630 A.2d 820 (N.J. Super. Ct. App. Div. 1993)
(stating that trial courts should allow lawyers to withdraw if they submitted expert proofs that
any recovery would be substantially less than the value of the work required).
276 Cf. Bostick et al., supra note 257, at 431, 437 (stating that physicians who participate in
performance incentives programs such as pay-for-performance must maintain primary
responsibility to their patients regardless of financial incentives and therefore should “avoid
selectively treating healthier patients for the purpose of bolstering their individual or group
performance outcomes”).
277 See Boyd, supra note 258, at 137 (explaining that physicians’ fiduciary obligations,
including the duty to put a patient’s interests above the physician’s interests, “attach at the time
the physician undertakes to treat the patient”).

184 C A R D O Z O L A W R E V I E W [Vol. 39:127

common carriers and innkeepers an obligation to be reasonable in their
dealings with the public, including a duty to serve all-comers on a
nondiscriminatory basis.278 These principles continue today through
comprehensive regulation of a wide array of industries affected with a
public interest and the power to exploit consumers,279 including the
transportation, communication, banking, and insurance industries.280
Regulation of these industries has been defended on the grounds that
“they wield . . . undue market power over goods or services upon which
the public had grown dependent.”281 Medicine too has acquired these
hallmarks.

The twentieth century witnessed tremendous advances in
medicine, with the practice of medicine evolving into a sophisticated,
highly professional endeavor. As Nicholas Bagley has argued, with these
changes the practice of medicine acquired the same attributes that have
justified public utility regulation of other industries: “the power to abuse
its control over a necessity.”282 Specifically, health care has become
indispensable to individuals’ well-being as medicine has come to play an
ever larger role in relieving suffering and preventing premature death.
Patients, however, cannot self-prescribe or administer medical tests or
treatments, but instead must rely on the physicians and other health
providers who both control access to health care and possess the
necessary knowledge and expertise.283 These features—health care as an

278 See Nicholas Bagley, Medicine As a Public Calling, 114 MICH. L. REV. 57, 58–59, 72 (2015)
(discussing the common law duties of innkeepers and common carriers, including the duty that
all must be served on a nondiscriminatory basis); Shepherd, supra note 135, at 1090 (stating
that under the common carrier doctrine, innkeepers and carriers “had an obligation not to turn
away customers without reasonable justification”). Courts also required that “adequate facilities
must be provided, reasonable rates must be charged, and no discrimination must be made.”
BRUCE WYMAN, THE SPECIAL LAW GOVERNING PUBLIC SERVICE CORPORATIONS AND ALL
OTHERS ENGAGED IN PUBLIC EMPLOYMENT xi (1911).
279 See Bagley, supra note 278, at 59 (explaining that the public utility concept recognized
that the law must constrain the behavior of “[a]ny industry that served an important human
need and had the market power to exploit consumers”); Shepherd, supra note 135, at 1090
(“Today, the concept behind common carrier obligations is retained, less through common law
application, and more through the comprehensive regulation of certain industries affected with
a public interest . . . .”).
280 See Bagley, supra note 278, at 73.
281 Id. at 71.
282 Id. at 84.
283 See Shepherd, supra note 135, at 1089–90 (arguing that one justification for requiring
physicians not to discriminate on unfair grounds is society, through the licensing process, has
granted physicians a self-regulated monopoly over access to health care). Contrary to public
perceptions, public utility regulation was not limited to natural monopolies. As Professor
Bagley explains:

[A] business need not be monopolistic in a strict sense. An extraordinary range of
market features—the costs of shopping around, bargaining inequalities,
informational disadvantage, rampant fraud, collusive pricing, emergency conditions,
and more—could all frustrate competition and so give rise to “virtual” or “practical”
monopolies that would warrant state intervention. By no means was the regulation of

2017] N O N C O M P L I A N T P A T I E N T S 185

important human need and providers’ control over access—leave
patients vulnerable to exploitation and consumer disadvantage.284 This
vulnerability in turn justifies state intervention that protects patients.

Health law includes several examples where courts and regulators
have recognized patients’ vulnerability and adopted laws that promote
patients’ access to health care on a reasonable and nondiscriminatory
basis. As described in Part II, the common law doctrine of patient
abandonment promotes patients’ access to continuous care by imposing
constraints on physicians’ autonomy. Specifically, the doctrine prohibits
physicians from terminating the physician-patient relationship for those
patients needing continuing medical attention without first providing
adequate notice.285 In adopting the patient abandonment doctrine,
courts not only emphasize the fiduciary nature of the physician’s duties,
but also the public nature of the physicians’ calling.286

Several federal laws also bear the trademarks of public utility
regulation. Enacted in 1946, the Hill-Burton Act requires all hospitals
and other facilities that received federal subsidies for construction and
modernization to make their services available on a nondiscriminatory
basis.287 EMTALA similarly imposes on hospitals and its affiliated
physicians a duty to serve certain patients. Aimed at preventing “patient
dumping”—the practice of hospitals refusing to treat or transferring
indigent, uninsured patients requiring emergency care—EMTALA
requires hospital emergency rooms that participate in Medicare (i.e.,
most hospitals) to stabilize any person with an emergency condition or
in active labor.288 Title VI of the Civil Rights Act of 1965 prohibits any
health care provider receiving federal financial assistance, including
Medicare and Medicaid payments, from discriminating against patients
on the basis of race, color, or national origin,289 while the Age
Discrimination Act of 1975 and Title IX of the Education Amendments
of 1972 prohibit discrimination on the basis of age and sex

public callings confined to the natural monopolies that today are thought to be its
proper targets.

Bagley, supra note 278, at 77–78.
284 See id. at 84–85 (discussing the potential for consumer exploitation in the health care
market).
285 See supra text accompanying note 59.
286 See Norton v. Hamilton, 89 S.E.2d 809, 812 (Ga. Ct. App. 1955) (“[P]ublic considerations
which are inseparable from the nature and exercise of [the physician’s] calling” justify imposing
a duty to provide continuing care); Ballou v. Prescott, 64 Me. 305, 313 (1874) (stating that
courts “impose specific duties in connection with and growing out of special undertakings,”
especially when as in the case of the physician-patient relationship the relationship is “public in
nature”).
287 See James F. Blumstein, Court Action, Agency Reaction: The Hill-Burton Act As a Case
Study, 69 IOWA L. REV. 1227, 1228–29 (1984).
288 42 U.S.C. § 1395dd(a)–(b) (2012).
289 Civil Rights Act of 1964, Pub. L. No. 88-352, 78 Stat. 241, 252–53 (codified as amended at
42 U.S.C. §§ 2000d–d-4 (2012)).

186 C A R D O Z O L A W R E V I E W [Vol. 39:127

respectively.290 The American Disabilities Act of 1990 affords similar
protections to persons with disabilities, explicitly including hospitals
and physicians’ offices within its scope.291 Section 1557 of the ACA
echoes these earlier laws, prohibiting discrimination in the health care
setting on the basis of race, color, national origin, sex, and age.292

Although these laws do not impose on the health care industry a
fully developed service obligation to treat all patients, they nevertheless
reflect the public utility tradition of curbing the economic independence
of actors serving an important human need.293 Specifically, they protect
the vulnerability of certain patients in the marketplace for health care
who might otherwise be denied equal access to care. Prohibiting
physicians from rejecting noncompliant patients represents a logical
extension of these protections.294 Nevertheless, the question remains
whether compelling policy interests justify doing so.

For the reasons discussed in Part III, I believe important policy
objectives outweigh physicians’ autonomy interests. Laws and rules of
professional conduct that allow physicians to avoid noncompliant
patients frustrate an important objective of performance incentive
programs: motivating physicians and their affiliated organizations to
improve their patients’ health behaviors. Physicians’ contention that
they can do little to influence their patients’ behaviors, and that holding
them financially accountable for their patients’ noncompliance
therefore unfairly punishes them, is overstated. Providers who
implement the activities described in Section III.A.1 can make real
inroads in increasing patient compliance and, ultimately, in improving
patients’ health. Providers can improve their physician-patient
communications, adopt intensive patient education interventions, and
implement programs that address the social, environmental, and
behavioral health obstacles to patient compliance.295 Allowing
physicians to fire their noncompliant patients would significantly
weaken physicians’ incentives to perform these actions.

Rejection of noncompliant patients frustrates other important
policy goals. As explained previously, patients fired for noncompliance

290 20 U.S.C. § 6101 (age); 20 U.S.C. § 1681 (sex).
291 42 U.S.C. § 12181 (2012).
292 Patient Protection and Affordable Care Act, Pub. L. No. 111-148, § 1557, 124 Stat. 119
(2010) (codified at 42 U.S.C. § 18116 (2012)).
293 See Bagley, supra note 278, at 94 (“[T]aken as a whole, the network of laws regulating
access to medical services reflects the impulse that private actors serving important needs owe a
legal duty to serve the public.”).
294 See Shepherd, supra note 135, at 1090–91 (arguing that requiring “physicians to treat all
those who seek their care (unless there is a good reason to refuse) . . . is consonant with the
centuries old theory of ‘common carrier” obligations” and more recent “of certain industries
affected with a public interest . . . ”). See generally Bagley, supra note 278 (arguing that that the
regulation of medicine, including regulating the duties of hospitals and physicians to fairly
serve the public, is consistent with public utility regulation).
295 See supra notes 91–111 and accompanying text.

2017] N O N C O M P L I A N T P A T I E N T S 187

experience discontinuity in care, and this in turn leads to poorer health
among noncompliant patients and higher medical spending for those
who insure these individuals. 296 Rejection of noncompliant patients also
lowers these patients’ trust in physicians and can leave patients feeling
stigmatized and ashamed.297 Importantly, allowing physicians to refuse
treatment to noncompliant patients undermines patient autonomy, as
patients fearful of being rejected by their physician may feel pressured
into agreeing to treatment recommendations that go against their
personal preferences.298 Finally, the emerging trend of physicians
avoiding noncompliant patients will exacerbate existing disparities in
health.299

So while physicians’ fairness and autonomy interests should not be
dismissed lightly, in the case of noncompliant patients they are
outweighed by other fundamental values—preserving life and health,
promoting equality, and respecting human dignity. Regulators and
professional associations therefore should prohibit physicians from
rejecting noncompliant patients despite such rule’s infringement on
physicians’ autonomy.

C. The Beneficence Rationale

Finally, physicians may argue that professional norms of
beneficence obligate them to reject their noncompliant patients. The
beneficence principle demands that physicians act in a patient’s best
interest.300 In some circumstances a physician may conclude that her
rejection of the noncompliant patient does in fact serve the patient’s best
interests. Specifically, the physician may believe that firing (or
threatening to fire) the patient will spur the patient to improve her
treatment adherence and make better lifestyle choices. I refer to this
justification as the paternalistic rationale. Alternatively, a physician may
determine that she no longer can provide effective care to a
noncompliant patient due to a breakdown in the physician-patient
relationship, and that another physician would be better suited to treat
the patient. I refer to this justification as the alternative physician
rationale. This Section discusses these two beneficence rationales and
argues that both are problematic.

296 See supra Section III.B.1.
297 See supra Sections III.B.2 and III.B.3.
298 See supra Section III.C.
299 See supra Section III.D.
300 See Martin, supra note 159 (defining beneficence as “[t]he obligation to act in patients’
best interests at all times”).

188 C A R D O Z O L A W R E V I E W [Vol. 39:127

1. The Paternalistic Rationale

When persistent counseling fails to change a patient’s behavior, the
physician may view actual or threatened termination of the physician-
patient relationship as a “last resort” to induce compliance with medical
advice.301 In other words, a physician may believe that her rejection of
the patient serves as the means for promoting a desired outcome—
improvement in the patient’s self-care.302 The physician’s actions
therefore are consistent with the beneficence principle, grounded in a
paternalistic impulse that can be summed up as “tough love.”

It is doubtful, however, that this paternalistic justification for
physicians’ rejection of noncompliant patients could be defended
empirically.303 The assumption that physicians threatening patients with
dismissal will successfully motivate patients to change their health-
related behaviors conflicts with practical experience. Despite wide-
spread awareness of the significant health risks associated with smoking,
poor diets, and lack of exercise, many people’s sincere attempts to
change these behaviors are met with failure. The suggestion that
pending rejection by one’s physicians is the carrot (or stick) that
prompts lasting behavior changes simply stretches credulity, particularly
for those patients facing social, environmental, and behavioral health
barriers to healthier behaviors.

2. The Alternative Physician Rationale

A physician also may argue that the beneficence norm supports
dismissal of a noncompliant patient when the relationship has
deteriorated to the point that the physician can no longer provide
competent care. A physician frustrated with her noncompliant patient
may “become emotionally exhausted” or develop “strong negative
emotions” toward the patient.304 These feelings can compromise the
patient’s care, as they may impede effective physician-patient
communication, bias the physician’s decision-making, or diminish
patient trust if the patient senses the physician’s negative attitude.305

301 Wicclair, supra note 12, at 314.
302 See id. at 314 (“Health-based firing is a means to promote desirable changes in [the
patient’s] choices and behavior.”).
303 See id. (“This justification is based on a questionable empirically assumption. It is
doubtful that available evidence supports a claim that either threatening to fire or actually firing
patients is an effective last resort rather than counter-productive act of frustration or
desperation.”).
304 See Tim Stokes et al., Ending the Doctor-Patient Relationship in General Practice: A
Proposed Model, 21 FAM. PRAC. 507, 510 (2004).
305 See Wicclair, supra note 12, at 316 (discussing how a breakdown in the physician-patient
relationship “can impede communication, decision-making, and mutual respect and trust”).

2017] N O N C O M P L I A N T P A T I E N T S 189

Relatedly, the physician who is unable to secure the patient’s compliance
may conclude that another physician will have better success in doing so
given differences in the physicians’ temperament or treatment style.306
For example, another physician may be more effective at
communicating relevant medical information or have a more
encouraging bedside manner that supports patients changing their
health behaviors.307 In such cases, severing the physician-patient
relationship arguably promotes the patient’s best interest by prompting
the patient to find a new physician better able to meet her needs.
Nevertheless, terminating these patients is the wrong mechanism to
achieve this end.

If the law and standards of professional ethics permit physicians to
reject noncompliant patients for the reasons described above, physicians
may invoke the alternative physician rationale to mask more self-
interested motivations. Caring for difficult, noncompliant patients can
be “hard,” “thankless,” and “frustrating.”308 Firing noncompliant
patients allows physicians to escape these negative feelings. Ethically,
however, physicians must strive to put aside their bias against a patient
in favor of compassion and fidelity.309 As explained by one
commentator, physicians must “care for their patients without letting
moral judgment or personal bias cloud their compassion for the
suffering human being in front of them.”310 Accordingly, a physician
should not be permitted to place her own desire to avoid the challenges
inherent in caring for noncompliant patients above her ethical
responsibilities to do so.

Rejecting noncompliant patients can also serve physicians’
financial and reputational interests under performance incentive
programs, as described in Part I. Physicians primarily concerned about
their success under performance incentive programs may use the

306 See id. (stating that termination may be appropriate if “the physician has good reason to
believe that another doctor would be able establish a more effective relationship that will more
successfully promote the patient’s health”).
307 Cf. Martin, supra note 159, at 111 (“Good clinical judgment usually finds a suitable
combination of sympathy and firmness. It remains focused on providing support in the form of
education, reassurance, helpful suggestions, and encouragement to change bad habits . . . .”).
Another physician also may be better suited to care for a patient because they possess more
relevant clinical expertise, such as an ob-gyn with expertise in caring for pregnant women who
are obese, smoke, or are otherwise high-risk. When the treating physician does not have the
clinical expertise to provide the noncompliant patient medically appropriate care, transferring
the patient’s care to a physician with the relevant skill set would be appropriate. Indeed, a
physician may be exposing herself to medical liability if she continues to provide care to a
patient whose clinical needs extend beyond the physician’s expertise.
308 Ramy Sedhorn, Taking Our Oath Seriously: Compassion for Patients, 18 AMA J. ETHICS
69, 70 (2016).
309 See Michael Hawking, Courage and Compassion: Virtues for So-Called “Difficult”
Patients, 19 AMA J. ETHICS 357 (2017) (calling for physicians to treat their patients with
compassion).
310 Micah Johnson, Do Physicians Have an Ethical Duty to Repair Relationships with So-
Called “Difficult” Patients?, 19 AMA J. ETHICS 323, 326 (2017).

190 C A R D O Z O L A W R E V I E W [Vol. 39:127

“patient’s best interest” argument as a benevolent justification for
terminating their noncompliant patients. In particular, physicians may
exaggerate the extent to which there has been a breakdown in the
physician-patient relationship or inflate the alternative physician’s
competence. Simultaneously they may downplay the detrimental impact
that dismissal can have on noncompliant patients.311

Among physicians with a genuine fidelity to their patients’ interests
over the physician’s own financial interests, they nevertheless may be
subconsciously motivated to conclude that they cannot adequately care
for a noncompliant patient. Psychologists studying cognitive motivation
have found that individuals have a subconscious tendency to form
initial judgments that promote their own self-interest.312 More
conscious deliberations then perform the secondary role of rationalizing
the self-serving conclusion.313 Accordingly, physicians subject to
performance incentive programs may, consistent with their self-interest,
be subconsciously biased to conclude that dismissing the noncompliant
patient is “the right thing to do.”314 Specifically, a physician may be
biased toward finding an irrevocable breakdown in her relationship with
a noncompliant patient or concluding that another physician will
provide better care to the patient.

In raising these concerns, I do not mean to characterize the
alternative physician rationale as illegitimate. When a physician’s
negative feelings toward a patient threaten the quality of care provided,
continuing the physician-patient relationship may prove harmful to the
patient. Moreover, if another physician truly could provide more

311 See generally supra Section IV.B.
312 See Dan M. Kahan, Foreword: Neutral Principles, Motivated Cognition, and Some
Problems for Constitutional Law, 125 HARV. L. REV. 1, 19 (2011) (describing “the unconscious
tendency of individuals to process information in a manner that suits some end or goal”). For
example, studies have found that individuals have faster reaction times when generating and
endorsing memories and beliefs consistent with conclusions that promote an individual’s self-
interest or desired ends. See Ziva Kunda, The Case for Motivated Reasoning, 108 PSYCH. BULL.
480, 483–85 (1990) (summarizing studies on biased memory search).
313 See Milton C. Regan, Moral Institutions and Organizational Culture, 51 ST. LOUIS U. L.J.
941, 959–60 (2007) (“[W]e typically engage in moral reasoning after our judgments have been
formed, and . . . we engage in that exercise in order to justify, rather than arrive at, those
judgments.”). See generally DANIEL KAHNEMAN, THINKING, FAST AND SLOW 105 (2011)
(explaining that deliberative processes merely endorse individuals’ initial impressions by
providing justifications for them). This does not mean deliberative reasoning cannot override
our initial impressions—it can—but doing so requires mobilizing substantial mental focus,
something individuals do infrequently, particularly when their mental capacity is otherwise
taxed by the complexity of the situation or performing other tasks. See Don A. Moore & George
Loewenstein, Self-Interest, Automaticity, and the Psychology of Conflict of Interest, 17 SOC. JUST.
RES. 189, 193 (2004) (stating that although “[c]ontrolled processes can override automatic
processes,” studies have found “that when mental capacity is constrained because people are
under cognitive load, it is harder for them to engage in reflection and correction of automatic
judgments”). See generally KAHNEMAN, supra note 313, at 81 (describing the “laziness” of
System 2 deliberative cognitive processes).
314 See Mantel, supra note 231, at 498–505 (explaining how physicians’ self-interest
subconsciously biases physicians to make clinical decisions consistent with their self-interest).

2017] N O N C O M P L I A N T P A T I E N T S 191

effective care the patient’s interests may be better served by switching
physicians. Rather, my point is that physicians may be too quick in
concluding that termination serves the patient’s best interests. When a
physician genuinely believes that insurmountable obstacles threaten the
physician-patient relationship or that another physician would be better
suited for the patient, she should discuss her concerns with the patient.
However, if the patient insists on continuing the physician-patient
relationship, the physician should not be permitted to fire the patient.
This approach would allow physicians to honor the beneficence
principle without undermining the wide-ranging policies and values
served by a prohibition against physicians rejecting noncompliant
patients.

V. SUPPORTING PROVIDERS’ EFFORTS TO IMPROVE PATIENT
COMPLIANCE

Parts III and IV argued that legal and ethical standards should
prohibit physicians from rejecting noncompliant patients. Adopting
rules that bar physicians from discriminating against noncompliant
patients, however, is only one side of the equation. Changing patients’
health-related behaviors is no small task. As described above, complex
social, environmental, and behavioral health factors impact individuals’
capacity to adhere to physicians’ medical advice.315 Identifying and
addressing these underlying causes of nonadherence requires time, skill,
and organizational resources that many physicians currently lack.
Counseling patients on overcoming their personal barriers to
compliance can involve many hours,316 yet declining payment rates have
caused physicians to reduce the amount of time they spend with each
patient.317 In addition, many physicians lack the skills to communicate
effectively with patients about barriers to adherence or lack knowledge
about strategies for improving patients’ health behaviors.318 And while

315 See supra Section III.A.1.
316 For example, clinical studies of one effective counseling technique, known as
motivational interviewing, found that the technique typically involves one to five hours of
counseling. See Bender, supra note 6, at 4 (describing the time commitment involved for
motivational counseling to bring about behavior changes).
317 See Page, supra note 3, at 4 (explaining that lack of time is a problem for physicians
treating nonadherent patients, as declining payment rates push physicians to keep patients
moving in order to maintain their income by increasing their volume). See generally Bender,
supra note 6, at 4 (noting that the time needed to provide effective counseling to noncompliant
patients is “not feasible in most clinical settings”); Chesanow, supra note 5, at 4–5 (stating that
busy physicians do not have the time to “sit down with a patient and tease out his or her unique
personal barriers to compliance,” and “[t]he more patients you are forced to see to pay the bills,
the less time you have to explore and address patient barriers to compliance”).
318 See Hibbard et al., supra note 10, at 493 (stating that the findings from the authors’
physician survey and interviews shows that physicians “may need help with skill development
and help in learning about evidence-based strategies for supporting patient self-management”);

192 C A R D O Z O L A W R E V I E W [Vol. 39:127

accountable care organizations and other large health care providers
may have the personnel and other resources to support interventions
that address barriers to compliance,319 small and medium–sized
physician practices often do not.320 A prohibition against physicians
rejecting noncompliant patients therefore should be balanced with
policies and infrastructure that support physicians and their affiliated
organizations’ efforts to improve patient compliance.

While a detailed discussion of how policymakers can support
efforts to improve patient compliance is beyond the scope of this
Article, I highlight here some matters for policymakers’ consideration.
First, government programs and private payors should ensure that their
payment policies do not hinder providers’ efforts to improve patient
compliance. Time spent by physicians or other health professionals
counseling noncompliant patients must be adequately compensated. For
example, Medicare covers medication management therapy—
comprehensive services provided by pharmacists that include
medication review, care coordination, and follow-up services.321 Payors
also should adequately compensate providers for intensive interventions
designed to improve patients’ health-related behaviors, such as the
diabetes self-management and smoking cessation initiatives described in
Section III.A. Any increase in payors costs for these services potentially
will be offset by a decrease in costly emergency care, hospitalizations, or
other treatments,322 although more research is needed to identify the
most cost-effective interventions.323

Mantel, supra note 113, at 268–69 (noting that many health professionals lack the skills to
assess and develop strategies for addressing the social determinants of health); Page, supra note
3, at 8 (noting that a consultant who works with health providers and plans on improving
patient communication believed that “most physicians have ineffective skills in identifying
patients who will potentially ignore treatment recommendations, and working with them to
become more motivated to engage in their treatment”).
319 See generally Mantel, supra note 113, at 269–71 (explaining that larger health care
organizations may have the capacity to address the social determinants that adversely impact
health).
320 See Steiner, supra note 110, at 583–84 (explaining that improvements in adherence may
best be accomplished by organization-based interventions that require substantial time, an
investment in information systems, staff training, and adoption of a medical home model,
something small practices may not be able to do on their own). See generally Mantel, supra note
113, at 269–71 (stating that small and medium-sized physician practices may be poorly
equipped to address the social determinants of health on their own).
321 Medication Therapy Management, CENTERS FOR MEDICARE & MEDICAID SERVICES,
https://www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/
MTM.html (last updated Aug. 1, 2017, 12:38 PM); CTRS. FOR MEDICARE & MEDICAID SERVS.,
2015 MEDICARE PART D MEDICATION THERAPY MANAGEMENT (MTM) PROGRAMS: FACT
SHEET SUMMARY OF 2015 MTM PROGRAMS 25–26 (Aug. 21, 2015), https://www.cms.gov/
Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/Downloads/CY2015-
MTM-Fact-Sheet .
322 See sources cited supra note 35 and accompanying text.
323 See Leah L. Zullig et al., A Renewed Medication Adherence Alliance Call to Action:
Harnessing Momentum to Address Medication Nonadherence in the United States, 10 PATIENT
PREFERENCE & ADHERENCE 1189 (2016) (“[I]t is widely believed that improving medication

2017] N O N C O M P L I A N T P A T I E N T S 193

Second, mindful of physicians’ perception that they can do little to
improve patients’ health-related behaviors,324 policymakers should fund
research and training on how providers can best address patients’
noncompliance. Physicians and their affiliated organizations would
benefit greatly from research on the most effective and efficient ways to
improve patients’ adherence and health-related behaviors. The National
Institutes of Health (NIH) provides some support for research on
interventions designed to improve adherence325 and health-related
behaviors,326 but additional research is needed.327 In addition to funding
additional research in this domain, government agencies could develop
training programs that support providers in improving their patients’
adherence, as they have done in the areas of patient safety, 328 meaningful
use of electronic health records,329 health care compliance,330 and public

adherence will result in improved individual- and population-level health and reduced health
care spending,” but that more research is needed on the efficacy of various interventions); see
also sources cited supra note 36 and accompanying text.
324 See supra notes 248–49 and accompanying text.
325 See PA-17-060 Oral Anticancer Agents: Utilization, Adherence, and Health Care Delivery,
NAT’L INST. HEALTH, https://grants.nih.gov/grants/guide/pa-files/PA-17-060.html (last visited
Sept. 12, 2017); PA-14-334 Advancing Interventions to Improve Medication Adherence, NAT’L
INST. HEALTH, https://grants.nih.gov/grants/guide/pa-files/PA-14-334.html (last visited Sept.
12, 2017); Department of Health and Human Services, NAT’L INST. HEALTH, https://
grants.nih.gov/grants/guide/pa-files/PA-14-181.html (last visited Sept. 12, 2017). The
Adherence Network is a working group with representatives from various NIH Institutes,
Centers, and Offices working to “provide leadership and vision for adherence research at NIH.”
National Institutes of Health Adherence Network, FACEBOOK, https://www.facebook.com/pg/
NIHAdherenceNetwork/about/?ref=page_internal (last visited Sept. 12, 2017). The National
Institutes of Health Adherence Network also sponsors a distinguished speaker series on
adherence issues. Id.
326 See, e.g., PAR-17-087, Tobacco Use and HIV in Low and Middle Income Countries, NAT’L
INST. HEALTH, https://grants.nih.gov/grants/guide/pa-files/PAR-17-087.html (last visited Sept.
12, 2017); PA-14-114, Behavioral Interventions to Address Multiple Chronic Health Conditions
in Primary Care, NAT’L INST. HEALTH, https://grants.nih.gov/grants/guide/pa-files/PA-14-
114.html (last visited Sept. 12, 2017); PA-14-15, Self-Management for Health in Chronic
Conditions, NAT’L INST. HEALTH, https://grants.nih.gov/grants/guide/pa-files/PA-14-345.html
(last visited Sept. 12, 2017); PA-14-315, Testing Interventions for Health-Enhancing Physical
Activity, NAT’L INST. HEALTH, https://grants.nih.gov/grants/guide/pa-files/PAR-14-315.html
(last visited Sept. 12, 2017).
327 See Hayden B. Bosworth et al., Medication Adherence: A Call for Action, 162 AM. HEART
J. 412 (2011) (reporting that participants in a think tank meeting on the current status of
medication adherence concluded that as “compared with the many thousands of trials for the
efficacy of individual drugs,” research on effective medication adherence interventions is more
limited, and that “prospective research is required to fully understand the effects of various
interventions on adherence”).
328 See Patient Safety Network Training Catalog, AGENCY FOR HEALTHCARE RES. & QUALITY,
https://psnet.ahrq.gov/pset?qs=compliance (last visited Sept. 12, 2017) (listing patient safety
training programs offered by the Agency for Healthcare Research and Quality).
329 See National Learning Consortium, HEALTHIT.GOV, https://www.healthit.gov/providers-
professionals/national-learning-consortium (last updated July 11, 2014) (listing resources,
training, and tools developed by HHS that support providers working towards the
implementation, adoption, and meaningful use of electronic health systems).
330 See HEAT Provider Compliance Training Videos, OFF. INSPECTOR GEN.: U.S. DEP’T
HEALTH & HUMAN SERVICES, https://oig.hhs.gov/newsroom/video/2011/heat_modules.asp (last

194 C A R D O Z O L A W R E V I E W [Vol. 39:127

health.331 Medical schools and other graduate programs for health care
professionals also should incorporate into their curriculums training on
interventions that enhance patient adherence,332 including better
communication between physicians and patients.333

Third, policymakers can help physician practices acquire the
organizational resources necessary for improving patient adherence and
health-related behaviors. For example, efforts to enhance patient
compliance are most effective when delivered by an interdisciplinary
team of health professionals, social workers, and other professionals.334
To help smaller physician practices transition to this model of care,
some states are funding interdisciplinary community health teams that
support multiple physician practices, thereby relieving small practices
from having to each establish and fund their own interdisciplinary
team.335 Alternatively, a few states are providing start-up funding to
physician practices transitioning to the interdisciplinary team model.336
Other states should follow suit and provide similar support to smaller
physician practices.

CONCLUSION

Caring for patients who fail to follow medical advice or adopt
healthy behaviors has long frustrated physicians. Now, caring for these
patients also may harm a physician’s reputation and reduce her income.

visited Sept. 12, 2017); MLN Web-Based Training, CENTERS FOR MEDICARE & MEDICAID
SERVICES, https://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/
MLNProducts/WebBasedTraining.html (last updated Apr. 13, 2017) (listing provider
compliance courses offered by the Centers for Medicare & Medicaid Services).
331 See CDC Learning Connection, CENTERS FOR DISEASE CONTROL & PREVENTION, https://
www.cdc.gov/learning (last updated Aug. 1, 2017).
332 See Berben et al., supra note 90, at 644 (“Health care curricula need to be revised to
include competencies in adherence-enhancing interventions.”); Piette et al., supra note 101, at
632 (stating that their research findings, showing an association between patients’ self-care and
providers’ communication processes, provides “further support for including provider training
in communication as a legitimate and important component of medical education”).
333 See generally Section III.A (discussing how poor physician-patient communications
contributes to poor patient adherence, particularly among patients with low health literacy).
334 See Berry et al., supra note 138, at 11 (stating that coaching and other interventions that
encourage patients to make positive lifestyle changes “are most effective when physicians are
part of a larger team of allied health professionals”).
335 See Mary Takach & Jason Buxbaum, Care Management for Medicaid Enrollees Through
Community Health Teams, COMMONWEALTH FUND 9 (MAY 2013), http://
www.commonwealthfund.org/~/media/files/publications/fund-report/2013/may/1690_takach_
care_mgmt_medicaid_enrollees_community_hlt_teams_520 (describing the functions of
community health teams). States supporting community health teams include Alabama, Maine,
Minnesota, New York, North Carolina, Oklahoma, and Vermont.
336 Start-up funds can be used to hire new staff (e.g., case managers) and pay for other
structural changes. See Kelly Devers et al., Innovative Medicaid Initiatives to Improve Service
Delivery and Quality of Care: A Look at Five State Initiatives, HENRY J. KAISER FAM. FOUND. 8
(Sept. 2011), https://kaiserfamilyfoundation.files.wordpress.com/2013/01/8224 .

2017] N O N C O M P L I A N T P A T I E N T S 195

Because noncompliant patients are likely to get sicker and consume
more health care, the physicians who treat these patients may have
poorer grades on publicly available report cards and lower
reimbursements under payment models that link providers’ payments
to patient outcomes. Physicians can escape this predicament, however,
by firing or otherwise refusing to treat noncompliant patients. Current
legal and ethical standards support physicians doing so, as they grant
physicians broad discretion in deciding whom to treat.

As detailed in this Article, denying treatment to patients for
reasons of noncompliance raises significant policy and ethical concerns.
When physicians fire or refuse to treat noncompliant patients, their
actions are detrimental to these patients’ health and well-being,
undermine patients’ trust in the medical profession, risk compromising
patient autonomy, and increase disparities in health. These harmful
consequences clearly run counter to the medical profession’s professed
fidelity to patients’ best interests and their ethical obligation to put a
patient’s welfare above the physician’s own financial interests. Nor can
the status quo be defended by invoking the ethos of personal
responsibility given the influence social, environmental, and behavioral
health factors have on individuals’ conduct. The laws and standards of
professional conduct that govern the physician-patient relationship
therefore should be revised to prohibit physicians from refusing to treat
noncompliant patients.

• Table of Contents
• Introduction
• I. Avoiding the Noncompliant Patient

A. Physicians’ Incentives Under Value-Based Purchasing
B. Reputational Concerns

• II. Physicians’ Legal and Ethical Obligations
• III. Justifications for Prohibiting Physicians from Discriminating Against Noncompliant Patients

A. Reinforcing the Policy Goals Behind Performance Incentive Programs
1. Lowering Barriers to Patient Compliance and Healthy Behaviors
2. Preserving Incentives for Physicians to Lower Barriers to Compliance and Healthy Behaviors
B. Honoring Professional Norms of Beneficence and Nonmaleficence
1. Eliminating the Adverse Consequences from Discontinuity in Care
2. Protecting Patients’ Trust in Physicians
3. Avoiding the Stigmatization and Shaming of Patients
C. Honoring Patient Autonomy
D. Reducing Disparities in Health
1. The Disproportionate Impact of Social and Environmental Factors
2. Physicians’ Implicit Bias

• IV. Justifications for the Status Quo

A. The Personal Responsibility Rationale
1. Noncompliant Patients’ Moral Culpability
2. Holding Noncompliant Patients Accountable
B. The Fairness and Autonomy Rationales
1. Physicians’ Ethical Obligations Limit Their Autonomy
2. Policy Justifications for Limiting Physicians’ Right to Refuse Treatment to Noncompliant Patients
C. The Beneficence Rationale
1. The Paternalistic Rationale
2. The Alternative Physician Rationale

• V. Supporting Providers’ Efforts to Improve Patient Compliance
• Conclusion