Electronic Health Record and Meaningful Use Challenges
Many of the same barriers to HIT adoption discussed in Chapter 1 also pertain to EHR adoption and successful attainment of meaningful use.
Although there are models that suggest significant savings after the implementation of ambulatory EHRs, the reality is that it is expensive. Multiple surveys report lack of funding as the number one barrier to EHR adoption.106 In a 2005 study published in Health Affairs, initial EHR costs averaged $44,000 (range $14-$63,000) per FTE (full time equivalent) and ongoing annual costs of $8,500 per FTE. These costs included the purchase of new hardware, etc. Financial benefits averaged about $33,000 per FTE provider per year. Importantly, more than half of the benefit derived was from improved coding.107 This is not a surprise given the fact that studies have shown that physicians often under-code for fear of punishment or lack of understanding what it takes to code to a certain level.108 A 2008 survey reported about one-third of physicians paid between $500-$3,000 per clinician, one-third paid between $3,001-$6,000 and about one-third paid more than $6,000.109 A 2011 study reported on the financial and nonfinancial costs of implementing a commercial EHR in a healthcare network in Texas. They calculated that implementation for a five physician primary care practice would be about $162,000 with $85, 500 in maintenance expenses in the first year. They also estimated that the average end-user would require 134 hours to train and prepare for implementation.110 Another study reported on 5-year return on investment from 49 practices that were part of the Massachusetts eHealth Collaborative, before and after EHR implementation. The study was prior to CMS reimbursement under the HITECH Act but was similar in that the eHealth Collaborative paid for the majority of costs related to purchase and implementation. They found only 27 percent of practices would achieve a positive five year return and that a majority would experience a loss. The average projected loss over five years was $43,473 per physician. There were striking differences between the winners and losers of EHR adoption.111 It is important to consider that integration with other disparate systems such as practice management systems can be very expensive and hard to factor into a cost-benefit analysis. The web-based application service provider (ASP) option is less expensive in the short term and perhaps in the long term, when one factors in the expenses to maintain and upgrade an office client-server network. According to many studies EHR adoption was far higher in large physician practices that could afford the initial high cost.112
Prior to EHR reimbursement lack of support by medical staff was consistently the second most commonly perceived obstacle to adoption.113 Physicians have to be shown a new technology makes money, saves time or is good for their patients. None of these can be proven for certain for every practice. Although physicians should not expect to go paperless from the beginning, at some point it can no longer be optional. It seems clear that CPOE does take longer than written orders but offers multiple advantages over paper as pointed out previously.114 Implementation will not fix old work flow issues and will not work if several physicians in a group are opposed to going electronic. It is now known that some practices have opted to change or discontinue their use of an EHR. A 2007 survey demonstrated fewer than 20% of respondents had uninstalled their EHR in an effort to step down to a less expensive alternative and 8% had returned to paper.115 According to a 2013 Deloitte survey of US physicians 63% of physicians were satisfied with their EHR (48% were somewhat satisfied and 15% were very satisfied).116 Physicians may resist some aspects of the EHR reimbursement program. For example, the American Academy of Family Physicians analyzed CMS meaningful use data and determined that 21% fewer family physicians attested for meaningful use in 2012, compared to 2011. Rates for the specialties were about the same. They theorized that physicians had to attest for 12 months of meaningful use which is onerous and they may have missed the attestation period.117 EHRs are not the only important issue for most physicians. They face increases in overhead while reimbursement wanes, along with ICD-10, HIPAA 5010, new healthcare reform and Red Flag rules, just to mention several looming challenges.
Loss of productivity
It is likely physicians will have to work at reduced capacity for several months with gradual improvement depending on training, aptitude, etc. This is a period when physician champions can help maintain morale and momentum with a positive attitude. According to one systematic review CPOE used on central station desktops for CPOE was not time efficient; the weighted average relative time difference across these studies reported an increase in documentation time of 238.4%.114 Loss of productivity is, in part, due to the change in workflow discussed in the next section.
Work flow changes
Everyone in the office will have to change the way they route information compared to the old paper system. If planning was well done in advance everyone should know how work flow will change. As an example, many offices place the patients chart in the exam room door to indicate that the patient is ready to be seen. How will that be accomplished with an electronic system? Initially, one will have to maintain a dual system of paper and electronic records. Work flow analysis will also determine where computer terminals will be placed in an office or hospital setting for easy access.
Reduced physician-patient interaction
Clinicians will have to maintain eye contact as often as possible and learn to incorporate the EHR into the average patient visit. Use of a movable monitor or tablet PC may help diminish the time the clinician spends not looking at the patient. There are several studies that report computers (EHR access) implemented in the office exam room do not detract from the physician-patient relationship. Some believe that the overall effect of exam room technology depends on the skill of the physician integrating the technology appropriately with the patient.118-120 Because CPOE and inpatient documentation entry takes longer to complete (on average), compared to the paper process there is a concern that attendings or housestaff will be forced to spend more time documenting on the computer and less time with the patient. A study reported in 2013 showed that interns spent only 12% of their time in direct patient-related care, but 40% on the computer.121 A second report in 2013 reported that emergency room physicians spent 28% of their time in direct patient care but 43% of time with data entry. On average, the total number of mouse clicks for a 10 hour shift approached 4,000.122 These findings further strain the already negative perception of many patients that they don’t have enough face time with their physician.
Usability has been defined as the “effectiveness, efficiency and satisfaction with which specific users can achieve a specific set of tasks in a particular environment.”123 Is the software well organized and intuitive such that the user can find what they are looking for with a minimal number of mouse clicks? This is more complicated than what one would expect because there are multiple sub-specialties with unique needs, as well as multiple clinicians who are used to working in a set sequence. Based on several surveys included in this chapter, usability does not necessarily correlate with the amount of money paid for the software. HIMSS now has an EHR usability task force and it is predicted that eventually all certified EHRs will need to pass usability testing.124 At this time CCHIT is the only certification body that includes usability testing, but for 2014 limited usability testing will be a requirement.125 In early 2013 the American Medical Informatics Association (AMIA) published their recommendations to improve EHR usability. They recommended further research and new policy recommendations as well as recommendations to vendors and end-users of EHR systems.126 An article by DesRoches et al. published in 2013 looked at the achievement of meaningful use and the ability to manage patient populations as of early 2012. Ease of use for panel management was subjectively measured and was listed as “easy” by only 43.8% for the ability to generate a list of patients by laboratory results and as high as 75.7% for the ability to provide patients with an after visit summary.127
Integration with other systems
Hopefully, integration with other systems like practice management software was already solved prior to implementation. Users should be prepared to pay significantly for programmers to integrate a new EHR with an old legacy system. An average cost is about $3-$15,000 per interface.128 Most office and hospitals have multiple old legacy systems that do not talk to each other. Systems are often purchased from different vendors and written in different programming languages. If either the EHR or practice management system’s software is upgraded, then interfaces need to be checked and possibly changed. It is now popular to purchase an EHR already integrated with practice management, billing and scheduling software programs.
Quality reporting issues
EHRs have the potential to generate a variety of data necessary for compliance with meaningful use objectives, to include quality reports. Quality reports have been tied to physician reimbursement in several situations. New York City considered basing a physician’s pay on evidence of high quality, but obstacles remain. In early 2013, two reports from Weill-Cornell Medical College in New York City highlighted issues with quality measure reporting. In one study the accuracy of reporting was low, compared to manual chart review. In another study that examined quality reporting in the Primary Care Information Project in New York it was noted that within the first two years of using an EHR there was no improvement in overall quality, even with high levels of technical assistance.129-130
Lack of interoperability standards
Data standards and medical vocabularies are necessary for interoperability. The initial standards have been proposed by ONC and will be covered in more detail in another chapter. Reimbursement for Meaningful Use will mandate that EHRs demonstrate the ability to exchange information. Although numerous standards have already been accepted (separate chapter) they will likely need to be updated and new standards added based on use cases. Furthermore, computers are based on data and not information, as discussed in the chapter on healthcare data, information and knowledge.
The HITECH Act of 2009 introduced a new certification process for EHRs sponsored by ONC, in addition to CCHIT certification. This new certification ensures that EHRs will be able to support Meaningful Use and that they also will be HIPAA compliant. ONC certification includes requirements on database encryption, encryption of transmitted data, authentication, data integrity, audit logs, automatic log off, emergency access, access control and accounting of HIPAA releases of information. The HITECH Act also strengthened the prior HIPAA requirements as they relate to EHRs, particularly in the areas of enforcement of HIPAA and notification of breaches. Both civil and criminal penalties for Business Associates (as well as covered entities) were introduced. Civil penalties in their harshest form can range up to 1.5 million dollars. If a data breach of PHI (protected health information) occurs, all affected individuals must be notified. If more than 500 individuals are affected, HHS must be notified as well. Sale of PHI is prohibited.131 Users of EHRs must:
- Use HIPAA compliant technology
- Provide physical and software security of data systems
- Provide physical and software security of their network(s) including mobile and remote computing
- Provide access control with defined user roles, passwords and user authentication and auditing
- Monitor and manage user behavior
- Have written security policies and procedures
- Have an effective disaster recovery plan132
EHRs pose new potential privacy and security threats for patient data, but with proper technology as well as proper health entity and user behavior, these risks can be mitigated. On the bright side, EHRs offer new safeguards unavailable in the paper record world, like audit trails, user authentication, and back-up copies of records. Further details are available in the chapter on privacy and security.
A 2010 Health Affairs article estimated that malpractice costs in the US are around $55 billion dollars annually (in 2008 dollars) or 2.4% of what the US spends on health care.133 Will EHRs increase or decrease that number? Unfortunately the answer isn’t in yet. Most studies suggesting lower malpractice claims after EHR implementation are not designed to prove cause and effect and may not be generalizable to other practices or regions.134 Arguments can be made for either outcome. On one hand, by increasing the quality of care, theoretically EHRs should reduce malpractice risk. Yet that assumes that quality and malpractice are related in a linear fashion, which may well not be the case. On the other hand, EHRs that are poorly designed, or that contain bugs, could promote inadvertent errors. This risk points to a need for monitoring and corrective action related to EHR-generated errors. The Office of the National Coordinator (ONC) for Health IT understands that a system of monitoring and corrective action for EHR-related errors needs to be implemented. ONC outlined its plans for this in a December 2010 statement.135 As a first step, one can currently report EHR-generated errors to AHRQ-recognized Patient Safety Organizations like PDR Secure.136 Two important areas of potential risks and benefits include documentation of clinical findings and clinical decision support. One might expect that the more comprehensive documentation produced by EHRs will improve a physician’s defense against malpractice. It certainly may. However the automated way that EHRs carry information forward from one note to the next can also promote errors and potential liability, if a piece of data is recorded incorrectly from the start, yet never corrected.137 Guidance on proper coding with EHRs is beginning to appear.138 E-discovery laws now allow electronically stored data related to patient records to be considered discoverable for the purpose of malpractice, so the metadata and audit trails that supplement EHR documentation can be used both to defend and to impeach a physician in a malpractice case.139 Will that be a net benefit or liability for physicians? Decision support alerts and guidelines embedded into EHRs could potentially provide a defense against malpractice claims if their advice is followed. But what if alerts or guidelines are overridden? There may be very appropriate reasons to do so, but will physicians be expected to document the reason for each and every alert they override? Will they run the risk of being penalized if they don’t? Improved access to information provided by health information exchanges (HIEs) should improve the coordination of care, the quality of medical information that is available, and thus the quality of medical decision making. But, will clinicians have a tendency to overlook key nuggets of clinical information simply because they are overwhelmed by the volume of information they receive? Will ready access to outside information on a patient make a physician more liable if he or she doesn’t always actively search for every piece of potentially relevant information? In addition, user errors can arise as users climb a steep learning curve to become proficient with EHRs. Care needs to be taken particularly during the implementation of an EHR to guard against user error. Finally, as EHRs become the standard of care, will practicing without an EHR become a medicolegal liability? At this point in time it is still undetermined whether EHRs will significantly impact the incidence and expense of malpractice in a positive or a negative way.140
Inadequate proof of benefit
Successful implementation of HIT at a medical center with a long standing history of systemic IT support does not necessarily translate to another healthcare organization with less IT support and infrastructure. A systematic review by Chaudry is often cited as proof of the benefits of HIT, but in his conclusion he states “four benchmark institutions have demonstrated the efficacy of health information technologies in improving quality and efficiency. Whether and how other institutions can achieve similar benefits and at what costs, are unclear.”141 There have been five recent articles in the medical literature that failed to demonstrate a significant impact of EHRs on medical quality in the US and in Europe.142-146 A more positive study was published in 2011 of more than 25,000 diabetics in 46 practices that showed achievement of diabetic care was significantly better for practices with EHRs, compared to paper-based practices. They measured intermediate outcomes and not actual patient outcomes, so the impact on morbidity or mortality is not known.147 Following the publication of the fifth edition of this textbook, three other articles related to diabetic care and electronic health records were published. All three studies were observational in nature and measured intermediate outcomes such as hemoglobin A1c levels. Only one study showed significant benefit and that was experienced by Kaiser-Permanente, an advanced integrated delivery network.148-150 A systematic review published in 2012 that looked at the economics of HIT and medication management could find little evidence that CPOE or CDSS were cost effective. Importantly, they noted that the quality of the literature was heterogonous and of poor quality.151 Another systematic review evaluated the impact of point-of-care computer reminders, as part of CPOE/CDSS on physician behavior and found a very small positive effect. Specifically, the review found that the reminders improved adherence to care by a median of only 4.2%.152 There has also been a hope and perception that having prior test results readily available in the EHR would reduce testing duplication. In a large retrospective study of before and after EHR implementation, having access to electronic results of lab and imaging results resulted in increased, rather than decreased ordering.153
Patient safety, EHRs and unintended consequences
Patient Safety. Unfortunately, with implement-tation of most technologies new problems and issues arise that were not considered initially. EHRs are no exception to this observation and a variety of unintended consequences have been reported. Weiner coined the term e-iatrogenesis to mean “patient harm caused at least in part by the application of health information technology.”154 Several studies have shown increased errors as a result of implementing CPOE.41,43,72,155-157 Campbell et al. outlined nine examples of unintended consequences related to CPOE implementation:
- “More work for clinicians
- Unfavorable workflow changes
- Never ending demands for system changes
- Conflicts between electronic and paper-based systems
- Unfavorable changes in communication patterns and practices
- Negative user emotions
- Generation of new kinds of errors
- Unexpected and unintended changes in institutional power structure
- Overdependence on technology”158
Alert fatigue is another common unintended consequence related to CPOE, discussed in more detail in the chapter on patient safety. The US federal government is keenly aware of the unintended consequences associated with HIT and EHRs after reports by the Joint Commission and the Institute of Medicine.159-160 Furthermore, the Pennsylvania Patient Safety Authority published a report on errors related to use of default values in 2013. They reported that wrong-time, wrong-dose, inappropriate auto-stops and wrong-route errors were often related to default values that should have been changed.161 In response to concerns AHRQ released the monograph Guide to Reducing Unintended Consequences of Electronic Health Records in 2011. This Guide discusses unanticipated and undesirable consequences of EHR implementation.162 In mid-2013, ONC released the report HIT Patient Safety and Surveillance Plan. The plan will make EHR error reporting easier, to include allowing the EHR to generate the report to patient safety organizations (PSOs). More details are discussed in the patient safety chapter.163 Reliability. In spite of successful EHR implementations, several dramatic failures were seen in 2013, with EHR shutdowns from 1 to 10 days.164-165 Healthcare organizations must develop backup plans to include temporarily relying on paper-based processes until the EHR is re-established. With better training or re-design some of the technology-related errors are likely to be overcome. More research is needed to obtain a balanced opinion of the impact of EHRs on quality of care, patient safety and productivity. Furthermore, there is a need to study the impact on all healthcare workers and not just physicians.
Over 112,000 health care professionals use Practice Fusion for their EHR
See for yourself why our EHR was ranked #1 for ease of use.