discussion
read the two PowerPoints and answer the question.
describe the ARRA-HITECH programs and how they support EHRs.
Chapter3: Computer and Network Architectures
Robert Hoyt MD
William Hersh MD
After viewing the presentation, viewers should be able to:
Describe the basic components of computing
Discuss the Internet and World Wide Web
Understand the importance of networks in the field of medicine
Compare and contrast wired and wireless local area networks (LANs)
Describe the newest wireless broadband networks and their significance
Learning Objectives
Types:
Microcomputer
Minicomputer
Personal computer (PC) (Desktop)
Supercomputer: mainframe enterprise level
Laptop
Server: robust in terms of speed and memory
Mobile
smartphone
tablet
Computers
Computers recognize the binary digit (bit): 0 and 1
Bits represent everything in a computer: text, images, etc.
8 bits is a byte which can have 256 combinations
Text is represented by 7 bit alphanumeric combinations (ASCII). For more text characters Unicode is used. For example, 16 bit UTF means 16 bit
Images are represented by pixels. Each black/white pixel is 1 bit. Images are “memory hogs”, compared to text
Computers and Data
1000 bytes = 1 kilobyte (or 1K)
1000 kilobytes = 1 megabyte (or 1 meg)
1000 megabytes = 1 gigabyte (or 1 gig)
1000 gigabytes = 1 terabyte
1000 terabytes = 1 petabyte
1000 petabytes = 1 exabyte
1000 exabytes = 1 zettabyte
1000 zettabytes = 1 yottabyte = 1024 bytes
Kilobyte to Zettabyte
Central processing unit (CPU): where the computer chip resides and is related to data processing speed
Memory: random access memory (RAM) is the working memory that is lost when the computer is turned off.
Auxiliary storage: active storage is where data is stored that is needed continuously. Archival storage can refer to tape, hard disks, optical drives or cloud
Input and Output devices: keyboards, mouse, monitors, etc.
Computer Hardware
Software programs instructs the hardware what to do
Application Programming Interfaces (APIs) are a common way now for computers to react with other programs or apps
Operating systems provide a standardized way for software programs to perform tasks
Computer Software
C, C++, C# – used in many modern applications
MUMPS – used in many early medical applications, now renamed to M and with many modern enhancements
BASIC – common in early days of PCs
Python – originated as “scripting” languages for Unix and Web, but achieving larger-scale use, e.g., in machine learning applications
Java – attempt to create standard language for Web applications
JavaScript – scripting language for Web browsers
R – emerging important language for statistics and data analytics
Programming Languages for Computing
Source code is free and shareable. Examples are Linux, Apache, Mozilla, OpenOffice, etc.
Resources for open source software in healthcare:
https://sourceforge.net/directory/business-enterprise/enterprise/medhealth
https://en.wikipedia.org/wiki/List_of_open-source_health_software
Online Registry of Biomedical Informatics Tools (ORBIT) – https://orbit.nlm.nih.gov/
Open Source Software
Spreadsheets are considered flat files or one dimensional; unlike relational databases which link tables together for more robust storage and queries
A database management system (DBMS) is a program that manages the database
Most common language to query the database is structured query language (SQL)
There is a trend towards non-relational database systems, like NoSQL for large “big data”
Database Systems
Note the primary key (Employee ID = 1) in the first table is linked to the second table (Salary)
SQL Database Schema
The Internet is the largest and arguably most important large scale international network
The Internet is a global “network-of-networks” using the Telecommunications Protocol/Internet Protocol stack (TCP/IP) as the communications standard. The TCP/IP stack allows for layering of different standards and technologies based on the participants in an exchange and the payload being exchanged
The Internet began in the late 1960 as a network known as Advanced Research Projects Agency Network (ARPANET) capable of tying together universities and research organizations securely
The Internet and World Wide Web
The World Wide Web (WWW) operates on top of the Internet and was created by Tim Berners-Lee in 1989
The WWW introduced the web browser, a software program that allows for connection to web servers over the Internet using Hypertext Transfer Protocol (HTTP)
Web pages are written using Hypertext Markup Language (HTML), an implementation of a markup language, or method for defining formatting of text in a document, which has become synonymous with the Web
The Internet and World Wide Web
Achieving interoperability on the Internet depends on global use of standards
Standards exist for the exchange of data, such as HTTP; the format of data, such as HTML, and the transport of data, such as TCP/IP
In a TCP/IP network, each device (host) must have an Internet Protocol (IP) address. IP addresses can be distributed amongst different tiers of lower layer networks, or “sub-networks”
The Internet and World Wide Web
In order for addressing to function properly in the presence of a sub-network, the machine must both have an IP address and a routing prefix or “subnet mask” (example: IP address of 192.168.10.1 and subnet mask of 255.254.254.0) in order for it to be considered properly addressable by other network nodes
Two versions of IP addressing exists today, IP version 4 (IPv4) which has been around for more than 40 years
IP version 6 (IPv6) is being used to phase out IPv4 before the complete depletion of assignable addresses brings the growth of the Internet to a complete halt
The Internet and World Wide Web
It is useful to think of the Internet as comprised of two main components, protocols and hardware
Each machine addressable on a network is known as a node
Computers connect to the internet through an Internet Service Provider (ISP) such as Bell South or AT&T
The Internet and World Wide Web
An electronic request for an IP address is sent via the network link provided by the one’s ISP to a DNS server
The DNS server then matches the requested domain name and responds with an IP address
The browser is now capable of sending an HTTP GET request (again routed through one’s ISP provided link) to the IP address returned from the DNS request
The browser can now render and display the document defined by the HTML response on the user’s screen
The Internet and World Wide Web
A message must be sent using small packets of information that can arrive via different routes, useful when there is web congestion, and are reassembled back at one’s computer
All traffic sent using TCP/IP (such as phone calls over the internet (VoIP) and email) are sent using packets
A router is a node which directs the packets on the Internet. The role of ordering these packets and making sure that they make it to their intended recipient in the proper structure is one of the jobs of TCP/IP
The Internet and World Wide Web
The Uniform Resource Locator (URL) is a specified address to a specific resource (e.g. http://www.google.com); the first part of the URL is the protocol identifier and the rest is the resource name
The protocol identifier and the resource name are separated by a colon and two forward slashes. As an example, http://uwf.edu/uwfmain/about describes HTTP as the protocol,“uwf.edu” as the server to which the HTTP request will be made, and “/uwfmain/about” as the path to the resource being requested. The most common domains end in .com, .edu, .org, .net, .mil, .gov and .int
The Internet and World Wide Web
How the Internet Works
Prior to the advent of the Internet, disparate businesses and health care entities were not able to easily exchange data; instead data resided on a local PC or server
Now, much of what we do is “is in the cloud”
Web services are task specific applications which are deployed in a platform independent manner via a series of transactions to and from other web-aware applications/services over a network (such as the Internet)
Web services can be broken down into two categories: RESTful services and SOA services
Web Services
Representational State Transfer (or RESTful) services are lightweight services which use existing Internet infrastructure and World Wide Web (WWW) concepts as their backbone
REST is an architecture, not a standard with endless possibilities as to how REST can be applied to act as a service bus
Communication with a RESTful service is a relatively quick process and can utilize any existing content standard for packaging its messaging. Most commonly, a RESTful service will use XML or JavaScript Object Notation (JSON) for this content delivery
RESTful Services
Operation Types (GET, DELETE, POST, PUT). These HTTP methods can provide four different points of access to a RESTful service. If a URI or URL identifies an object, the HTTP operation type defines an accessor method to that object (e.g. GET a list, POST an update, PUT a new record, DELETE a purged record)
MIME Type (Multipurpose Internet Mail Extensions) is a means of communicating the content type used within a message transferred over the Internet. Typically, in a RESTful service, this would be XML or JSON, but it could be any other type
RESTful Services
RESTful Operations
SOA is based on SOAP, a protocol standard for interacting with web services
These services require a set of standards for content and a service oriented architecture (SOA) stack, a collection of services
The most common standards used in web services transactions are HTTP, as the internet protocol, with XML as the delivery language
Service Oriented Architecture (SOA) Web Services
SOAP (Simple Object Access Protocol): a communication protocol between applications. It is a XML-based platform neutral format for the invocation and response of web services functions over a network
WSDL (Web Services Description Language): a XML document used to describe and locate web services. A WSDL can inform a calling application as to the functionality available from a given service, as well as the structure and types of function arguments and responses
UDDI (Universal Description, Discovery and Integration): a directory for storing information about web services, described by WSDL. UDDI utilizes the SOAP protocol for providing access to WSDL documents necessary for interacting with services indexed
Service Oriented Architecture (SOA) Web Services
UDDI provides the service registry (like the yellow pages) and WSDL describes the services and the requirements for their interaction
A service requester seeking a web service would search a service directory for a function that meets a need using a structured language
Once the service provider is located, a SOAP message can be sent back and forth between the service requester and service provider (next slide)
SOA Web Services
SOA Web Services
A conceptual model was created in 1984 to help understand network architectures
It divides computer-to-computer communication into seven layers known as the OSI Stack
The Open Systems Interconnection (OSI) Model
Application. This is the layer where applications access network services. Examples: software for database access, email and file transfer and the Internet protocols FTP, HTTP and SMTP
Presentation. This layer translates (formats) the data for the application layer for the network. Examples: data encryption and compression
Session. This layer establishes, maintains and terminates “sessions” between computers
The OSI Model Layers
Transport. This layers deals with error recognition and recovery and handles message size issues and can send receipt acknowledgments. The Internet protocol related to this is TCP
Network. This layer is involved with message control, switching and routing. It translates logical addresses into physical addresses
Data link. This layer packages data from the physical layer into frames (special packets) and is responsible for error free from transfer from one computer (node) to another
The OSI Model Layers
Physical. This layer deals with the unstructured raw data stream from the other layers. Specifically, it encodes data and decides whether the bits will be sent via a digital or analog mode and decides if the bits will be transmitted as electrical or optical signals. This layer is involved with communication with devices. Examples: USB, Bluetooth and RS-232
The OSI Model Layers
A network is a group of computers that are linked together in order to share information
A network can share patient information as well as provide internet access for multiple users
Networks can be small, connecting just several computers in a clinician’s office or very large, connecting computers in an entire organization in multiple locations
There are several ways to access the internet: dial-up modem, wireless fidelity (WiFi), a Digital Subscription Line (DSL), 3G/4G telecommunication, cable modem or T1 lines
Networks
Transmission method Theoretical max speed Typical speed range
Dial-up modem 56 Kbps 56 Kbps
DSL 6 Mbps 1.5-8Mbps downlink/128 Kbps uplink
Cable modem 30 Mbps 3-15 Mbps downlink/1-3 Mbps uplink
Wired Ethernet (Cat 5) 1000 Mbps 100 Mbps
Fiber optic cable 100 Gbps 2.5-40 Gbps
T-1 line 1.5 Mbps 1-1.5 Mbps
Wireless 802.11g
Wireless 802.11n 54 Mbps
300 Mbps 1-20 Mbps
40-115 Mbps
WiMax 70 Mbps 54-70 Mbps
LTE 60 Mbps 8-12 Mbps
Bluetooth 24 Mbps 1-24 Mbps
3G 2.4 Mbps 144-384 kbps
4G 100 Mbps 10-70 Mbps
Satellite 10 Mbps 10 Mbps
Data Transfer Rates
Bandwidth is the size of the pipe to transmit packets (a formatted data unit carried by a packet mode computer network)
Packet loss is an issue because packets may rarely fail to reach their destination. The IP Transmission Control Protocol (TCP) makes sure a packet reaches its destination or re-sends it. The User Datagram Protocol (UDP) does not guarantee delivery and is used with, for example, live streaming video
End-to-end delay is the latency or delay in receiving a packet. With fiber optics the latency is minimal
Jitter is the random variation in packet delay and reflects Internet spikes in activity
Data Transfer Affected By:
Networks are named based on connection method, as well as configuration or size. As an example, a network can be connected by fiber optic cable, Ethernet or wireless. Networks can also be described by different configurations or topologies. They can be connected to a common backbone or bus, in a star configuration using a central hub or a ring configuration
Network Types
A PAN is a close proximity network designed to link phones, computers, PDAs, etc. The most common technology to create a wireless personal area network or WPAN is Bluetooth (BT)
BT is designed to wirelessly connect an assortment of devices at a maximum distance of about 300 feet with the most recent Bluetooth devices (version 5.0). It does have the advantages of not requiring much power and connecting automatically. It operates in the 2.4 MHz frequency range
The most common application of Bluetooth today is as a wireless headset to connect to a mobile phone but wearable technology (e.g. fitness devices) is a close second
Personal Area Networks (PANs)
Bluetooth can be used to wirelessly connect computers to keyboards, mice, printers, PDAs and smartphones
Security must be enabled due to the fact that even though the transmission range is short, hackers have taken advantage of this common frequency
Bluetooth Smart or Low Energy 5.0 requires less power, has better battery life and is less expensive. The frequency is the same but the range is less (about 50 meters)
Personal Area Networks (PANs)
PANs can also use other standards: Infrared to connect devices using the IrDA standard, ZigBee networks, Wireless USB and a body area networks (BANs)
A wireless body area network (WBAN) is also known as a body sensor network which is gaining importance in healthcare
Another wireless sensor network protocol known as ANT™ is available for ultra-low power applications. The proprietary network operates on the 2.4 GHz ISM band
Personal Area Networks (PANs)
LANs generally refer to linked computers in an office, hospital, home or close proximity situation
A typical network consists of nodes (computers, printers, etc.), a connecting technology (wired or wireless) and specialized equipment such as hubs, routers and switches
LANs can be wired or wireless
Local Area Networks (LANs)
To connect several computers in a home or office scenario, a hub or a network switch is needed
Routers direct messages between networks and the Internet; whereas, switches connect computers to one another and prevent delay
Unlike Hubs that share bandwidth, switches operate at full bandwidth. Switches are like traffic cops that direct simultaneous messages in the right direction
Wired LANs
To handle larger enterprise demands Gigabit Ethernet LANs are available that are based on copper or fiber optics. Cat5e or Cat6 cables are necessary
Greater bandwidth is necessary for many hospital systems that now have multiple IT systems, an electronic medical record and picture archiving and communication systems (PACS)
Typical LAN shown on next slide
Wired LANs
Typical Wired LAN Schema
Phone lines can connect a computer to the internet by using a dial-up modem. The downside is that the connection is relatively slow
Digital subscription lines (DSL) also use standard phone lines that have additional capacity (bandwidth) and are much faster network connection than dial up
DSL also has the advantage over modems of being able to access the internet and use the telephone at the same time
Wired LANs
Power lines are another option using standard power outlets to create a network. PowerPacket® is inexpensive to install and claims data transfer speeds of 14 Mbps
Ethernet is a network protocol and most networks are connected by fiber or twisted-pair/copper wire connections. Ethernet networks are faster, less expensive and more secure than wireless networks. The most common Ethernet cable is category 5 (Cat 5, 5e) unshielded twisted pair (UTP)
Wired LANs
Wireless networks are based on the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard and operate in the 900 MHz, 2.4 GHz and 5 GHz frequencies
Wireless networks have become much cheaper and easier to install so many offices and hospitals have opted to go wireless
This allows laptop/tablet PCs and smartphones in exam and patient rooms to be connected to the local network or internet without the limitations of hardwiring but it does require a wireless router and access points
Wireless LANs
802.11ac is the newest standard that can operate at speeds up to 900 Mbps with a frequency of 2.4 and 5 GHz
The network router is usually connected to the Internet by an Ethernet cable to DSL or a cable modem. Security must be established using an encryption scheme such as WiFi Protected Access II (WAP2) encryption
Each device on a network has a unique address (MAC) and routers can have security lists which only allow known devices or MACSs into the network
Wireless LANs
Wireless LAN (WLAN)
GANs are networks that connect other networks and have an unlimited geographic area. The problem with broadband technology is that it is expensive and the problem with WiFi is that it may result in spotty coverage
These shortcomings created an initiative known as Worldwide Interoperability for Microwave Access (WiMax), using the IEEE 802.16 standard
This 4G network is about 10 times faster than 3G and has greater capacity which is equally important
Global Area Networks (GANs)
The goal is to be faster than standard WiFi and reach greater distances, such that it might replace broadband services and permit widespread wireless access to the internet by PCs or phones
The introduction of one 4G network (WiMax) was so slow that major carriers adopted Long Term Evolution (LTE) WiFi
Operating in the 700 MHz range it has maximum download rates of 100 Mbps and upload rates of 50 Mpbs
There will be a slow roll out of 5 G in 2018
Global Area Networks (GANs)
In this case the home computer is the client and is attached to the network at work by communicating with a VPN server associated with that network
The Internet can serve as the means of connection with VPN working over both wired and wireless LANs
“Tunneling protocols” encrypt data by the sender and decrypt it at the receiver’s end via a secure tunnel. In addition, the sender’s and receiver’s network addresses can be encrypted
A physician at home may use a VPN to access the work EHR
Virtual Private Networks (VPNs)
Virtual Private Networks (VPNs)
Informaticians need to understand computer and network technologies
Clinicians and IT staff must have an excellent understanding of office networks and the related required security
Wireless networks have become attractive due to faster speeds and lower costs
Wireless broadband has arrived to provide faster Internet access for all
Conclusions
Chapter 4: Electronic Health Records
Robert Hoyt MD
Vishnu Mohan MD
After reading this chapter the reader should be able to:
State the definition and history of electronic health records (EHRs)
Describe the limitations of paper-based health records
Identify the benefits of electronic health records
List the key components of an electronic health record
Describe the ARRA-HITECH programs to support EHRs
Describe the benefits and challenges of computerized order entry and clinical decision support systems
State the obstacles to purchasing, adopting and implementing an electronic health record
Enumerate the steps to adopt and implement an EHR
Learning Objectives
2
There is no topic in health informatics as important, yet controversial, as the electronic health record (EHR)
In spite of fledgling EHRs being around for the past 35-40 years they are still controversial in the eyes of many
Due to the federal government reimbursement programs for EHR use by physicians and hospitals, EHRs are now part of the healthcare landscape
Some of the famous early EHRs are listed on the next slide
Introduction
The Problem Oriented Medical Information System (PROMIS)
American Rheumatism Association Medical Information System (ARAMIS)
Regenstrief Medical Record System (RMRS)
Summary Time Oriented Record (STOR)
Health Evaluation Through Logical Processing (HELP)
Computer Stored Ambulatory Record (COSTAR)
De-Centralized Hospital Computer Program (DHCP)—forerunner of VistA (Veterans Health Administration)
Early EHRs
Electronic Health Record: “An electronic record of health-related information on an individual that conforms to nationally recognized interoperability standards and that can be created, managed and consulted by authorized clinicians and staff across more than one healthcare organization”
While the “experts” can debate the difference between EHR and EMR, we will not and will stick with EHR throughout the textbook and slides
Definition
Paper records are severely limited: less legible, more difficult to retrieve, store and share and unstructured data. Also, electronic records less likely to be missing and available 24/7 from multiple locations. Paper records do not permit clinical decision support
Need for improved efficiency and productivity: clinicians are more productive if charts are available and retrieval of results is faster. EHR access from home while on call helps productivity
Quality of care and patient safety: the factors already described in last two bullets plus clinical decision support, quality reports and secure messaging as part of an EHR
Why do we need EHRs?
Public expectations: EHRs may increase patient satisfaction through faster results, messaging, patient portals, electronic patient education, e-prescribing and online scheduling
Governmental expectations: federal government considers EHR to be transformational and hence why they support reimbursement for use
Why do we need EHRs?
7
Financial savings: EHRs may save money by eliminating transcription and improving coding. Decreased file room storage and faster chart pulls and info retrieval may result in cost savings
Technological advances: computers are much faster, the Internet is more prevalent, wireless and mobile technologies are ubiquitous; all supporting EHRs
Need for aggregated data: healthcare data must be electronic to be shared, stored and analyzed. Research depends on large study populations and data sets which EHRs can provide
Why do we need EHRs?
Need for integrated data: electronic data permits integration with health information organizations, data analytics, public health reporting, artificial intelligence and genomic information
EHR as a transformational tool: select organizations such as the VA and Kaiser Permanente made huge investments in EHRs to standardize care and transform delivery and analysis of healthcare
Why do we need EHRs?
Need for coordinated care: with an aging population with multiple physicians and medications, care coordination is important. Sharing electronically has great potential, but barriers exist as we point out in the chapter on health information exchange
Why do we need EHRs?
Electronic Health Record Key Components
Clinical decision support
Secure messaging
Computerized physician order entry
Practice management
Manage care module
Referral management
Results retrieval
Prior encounter retrieval
Patient reminders
Electronic encounter notes
Multiple input methods
Access via mobile technology
Remote access from home
Electronic prescribing
Integration with images
Integration with physician and patient education
Public health reporting
Quality reports
Problem summary lists
Electronic Health Record Key Components
Ability to scan in data
Evaluation and management help
Ability to graph and track results
Ability to create patient lists
Ability to create registries
Preventive medicine tracking
Privacy/security compliance
Robust backup systems
Ability to generate summaries of care (CCD)
Support for client server or application service provider (ASP) modes
CPOE is an EHR feature that processes orders for medications, lab tests, imaging, consults and other diagnostic tests. It is not the same as electronic prescribing
CPOE has many potential benefits (next slide)
CPOE has the potential to reduce medical errors but the literature is mixed. Most early studies came from a select number of academic institutions with home grown EHRs and large IT departments
Computerized Physician Order Entry (CPOE)
Potential Benefits of CPOE
Koppel et al
Overcomes the issue of illegibility
Fewer errors associated with ordering drugs with similar names,
More easily integrated with decision support systems than paper,
Easily linked to drug-drug interaction warning
More likely to identify the prescribing physician,
Able to link to adverse drug event (ADE) reporting systems
Able to avoid medication errors like trailing zeroes
Creates data that is available for analysis
Can point out treatment and drugs of choice
Can reduce under and over-prescribing,
Prescriptions reach the pharmacy quicker
One study suggested cost savings from reduced length of stay, compared to paper based orders
Some studies have shown improved standardized care with EHRs, but this is not universal
CPOE is difficult to implement in hospitals because it disrupts workflow and slows physicians down. They often don’t realize, however, that CPOE benefits others on the team, such as nurses and pharmacists
CPOE
With CPOE you can embed a variety of tools to assist in decision making. Traditionally, this meant medication alerts and patient reminders. In reality, any software that assists decisions is a CDSS:
Knowledge support: programs embedded into the EHR that educate clinicians or patients
Calculators: part of the EHR
Flow charts and graphs: to look at lab or vital sign trends over time
Clinical Decision Support Systems (CDSSs)
CDSS (continued)
Order sets: inpatient clinical practice guidelines for specific scenarios (e.g. pneumonia), standardizing care
Reminders: remind clinician or patient about pending tests, etc.
Differential diagnosis: software exists that helps clinicians analyze symptoms and signs, to arrive at a diagnosis
Lab and Imaging decision support: what tests are indicated and at what costs?
Public health alerts: primarily infectious disease alerts for new outbreaks, e.g. MERS virus
Clinical Decision Support Systems (CDSSs)
Currently, the vast majoring of eRx occurs as part of an EHR and not a standalone program
69% of office-based prescriptions are now electronic
93% of community pharmacies are connected to the Surescripts network
The next slide lists the potential advantages of eRx over paper-based prescriptions
Electronic Prescribing (eRx)
eRx Potential Benefits
Legible and complete prescriptions
Abbreviations+ unclear decimal points are avoided
The wait to pick up scripts shorter
Fewer duplicated prescriptions
Better compliance with fewer drugs not filled or picked up
Potential to reduce workload for pharmacists
Timely notification of drug alerts and updates
Ability to check formulary status and copays
Can interface with practice and drug management software
The process is secure and HIPAA compliant
Associated with CDDSs
Digital records improve data analysis of prescribing habits
Batch refills can save time
Better use of generic or preferred drugs
Details about drug allergies
Drug-drug interaction alerts
Formulary alerts to tell you drug is either not recommended or not reimbursed
Alerts can exist to ask about pregnancy, kidney or liver function and safety in the elderly
Dosing alerts can arise based on age or size of patient
eRx Clinical Decision Support
Alert fatigue: too many alerts result in deletions, some justified, others not. Hot topic and area of much future research
Prescribing errors still occur with eRx but they are different; wrong drug or wrong dose
There are still issues at the pharmacist’s end but these should improve over time
Still not clear how many adverse drug events are prevented with eRx; perhaps too soon to know
eRx Challenges
Chronic disease: track e.g. diabetes
Research registries: high volume allows research questions to be answered
Safety registries: issues reported to e.g. FDA
Public health registries: immunizations, cancer and biosurveillance
Quality: data could be stored in registry and later forwarded to e.g. CMS
EHR Registries
Prior to EHR adoption, most medical practices used an electronic PMS. Now most are part of their EHR
PMSs are essential to run any practice: for billing, dealing with insurance companies, evaluating physician performance and practice trends
Typical office workflow is shown in next slide
Practice Management Systems (PMSs)
EHR Adoption
The US has been behind many other “developed” countries up until the HITECH ACT that included reimbursement for EHRs
Ambulatory EHR adoption (2015): roughly 86% have EHRs, but some are much more advanced than others. Larger practices adopt at a higher rate due largely to stronger finances
Inpatient (hospital) EHR adoption (2015): perhaps as many as 96% of US hospitals have EHRs and most are participating in the Meaningful Use program. Smaller urban and rural hospitals lag
Just because you own an EHR doesn’t mean you are maximizing the features and benefits (next slide)
Very Few Practices Have Reached Stage 7 Sophistication (HIMSS data second quarter 2017
Financial: in spite of government reimbursement, some practices will gain and some will lose money. What will the long term annual costs be after reimbursement ends? Will some stop using EHRs?
Physician resistance: complying with meaningful use has been onerous and may not result in any immediate and direct benefit to clinicians and patients
Loss of productivity: there is almost always initial loss of productivity and if the practice doesn’t change workflow habits there will be a long term losses as well
EHR Challenges
Workflow changes: everyone must adapt to doing business differently but some seek strange workarounds
Reduced physician-patient interaction: without careful forethought and planning, there will be less eye contact and interaction with patients
Usability issues: some EHRs are not user friendly and require too many mouse clicks or illogical steps, impeding workflow
Integration with other systems: practices may need to build expensive interfaces to communicate with HIOs, practice management systems, etc.
EHR and Meaningful Use Challenges
Lack of interoperability: EHRs are not capable of communicating with each other without additional technology, thus an impediment to data sharing
Privacy concerns: hacking into EHRs could result in loss of privacy for thousands, rather than a single paper chart
Legal: It is not known if EHRs will increase or decrease malpractice over the long haul
Inadequate proof of benefit: in spite on many published studies, there is not adequate proof that EHRs improve quality of care
EHR and Meaningful Use Challenges
Patient safety and unintended consequences: not only are studies suggesting improved patient safety mixed, there is evidence that new medical errors may occur (at least in the short term) with EHR use. “E-iatrogenesis” means medical errors due to technology
Situation worsened by alert fatigue, frequent software upgrades, usability issues, stress to meet meaningful use objectives
Several sentinel failures of major EHRs in large healthcare systems have highlighted EHR vulnerability
EHR and Meaningful Use Challenges
The US federal government (along with the IOM) has opined that EHRs are an important part of healthcare reform
A program for reimbursement for EHR use by clinicians and hospitals under Medicare and Medicaid (HITECH Act) was established in 2009
Clinicians had to: (1) be eligible, (2) register for reimbursement, (3) use a certified EHR, (4) demonstrate and prove Meaningful Use, and (5) receive reimbursement.
As of December 2017, $24.8 billion was spent by Medicare and $12.54 by Medicaid on EHR reimbursement to clinicians
HITECH ACT and EHR Reimbursement
Medicare defines EPs as doctors of medicine or osteopathy, doctors of dental surgery or dental medicine, doctors of podiatric medicine, doctors of optometry and chiropractors
Medicaid defines EPs as physicians, nurse practitioners, certified nurse midwives, dentists and physician assistants (physician assistants must provide services in a federally qualified health center or rural health clinic that is led by a physician assistant). Medicaid physicians must have at least 30% Medicaid volume (20% for pediatricians)
Eligible Professionals (EPs)
The goals of MU are the same as the national goals for HIT: (a) improve quality, safety, efficiency and reduce health disparities; (b) engage patients and families; (c) improve care coordination; (d) ensure adequate privacy and security of personal health information; (e) improve population and public health
EHRs must be certified by several organizations as capable of meeting meaningful use objectives
Meaningful Use Goals
Users must meet required core measures and multiple menu measures (textbook for more details). Quality measures are a major part of meaningful use
There are penalties for hospitals or EPs that don’t comply with Medicare Meaningful Use
Meaningful Use
Low cost that includes 3 month free trial
Fully featured and compliant with Meaningful Use
Available as a client or web based (ASP) model
Appeals to small practices, particularly primary care
Small EHR Example
Amazing Charts
Medium priced for medium sized practices of multiple specialty types
More clinician and patient features to include mobile and a health information exchange (HIE) solution
Medium EHR Example
eClinicalWorks
Intended for very large practices such as Kaiser-Permanente
Includes every aspect of Meaningful Use and numerous innovations such as a comprehensive patient portal and several mobile solutions
Large EHR Example
Epic
Develop an office strategy: why are you considering EHRs? Is your entire staff onboard? Don’t do it just for reimbursement. Plan, plan, plan
Do Research: take advantage of courses, books, articles, EHR survey results, regional extension centers, HIT consultants, etc.
List features: be sure to include inputting methods, backup, warrantees, mobile presence, etc.
Analyze and re-engineer workflow: consider all processes likely to change when you transition from paper to electronic
Implementing an EHR Steps
Use project management tools: these will improve your organization for tasks
Choose client versus ASP model: the web based model will be easier with less of the need for in house IT support
Practice management system needs: should you purchase a combination or build an interface?
Survey your hardware and network needs: will you need more bandwidth? Wireless? How many computer stations and will they require upgrades?
Implementing an EHR Steps
Develop a vendor strategy: create request for proposals (RFPs) for vendors to outline all of your needs, to include price, maintenance, etc. Obtain commitments in writing.
Select a vendor: develop a contract and have it reviewed by legal
Develop a paper to EHR conversion strategy: it is likely you will initially run a dual paper and electronic practice. Textbook discusses this in more detail
Implementing an EHR Steps
Training: you can’t train to much and be sure to discuss the details with your vendor early on
Implementation: decide whether you will phase in implementation or have a “go live” date. Be prepared to decreased productivity for several months and a new glitches along the way
Implementing an EHR Steps
EHRs are felt to be critical for US healthcare reform
Paper based health records are severely limited
EHR reimbursement has greatly increased US adoption
In spite of many potential benefits of EHRs, multiple challenges are associated with adoption
Planning, training and strategizing about EHRs is more important than the actual EHR brand purchased
Conclusions