ELECTRONIC HEALTH RECORDS (EHRs) have been in widespread use for more than a decade, with more than 80% of physicians currently reporting EHR adoption (Centers for Disease Control and Prevention, 2016; Healthcare IT News, n.d.; HealthIT.gov, n.d.). Heralded as a promising technology that would change how care is delivered and how we manage populations with chronic medical conditions, enhance safety, and make care more patient-oriented, EHRs not only have, in fact, changed care for the better in some ways but also have fallen short of expectations in other ways and at the same time became a major source of frustration for clinicians and other end users.

From physician blogs (GomerBlog, n.d.; Schnabel & GomerBlog, 2015) to rap parodies (ZDogg, n.d.), where doctors are seen smashing computer keyboards and screens, to town hall meetings ("Mass. Doctors Join to Vent Frustrations," n.d.), physicians have been trying to vent their growing frustrations with EHRs. They cite their lack of usability and functionality for primary care, poor design, and onerous meaningful use requirements.

Most recently, an article in the Annals of Internal Medicine documented negative effects on time allocation and productivity (Sinsky et al., 2016). The article noted that

for every hour physicians provide direct clinical face time to patients, nearly 2 additional hours is spent on EHR and desk work within the clinic day. Outside office hours, physicians spend another 1 to 2 hours of personal time each night doing additional computer and other clerical work.^{(p. 1)}

Other studies have linked clinician dissatisfaction, stress, and burnout to EHR use (Babbott et al., 2014; Linzer et al., 2016).

Our special issue attempts to shed additional light on EHR issues affecting end users including physicians, nurses, nurse practitioners (NPs), physician assistants, and staff in the ambulatory setting. We wanted to bring forward articles that illustrate how EHRs have impacted clinicians and patients with an eye toward solutions.

While stand-alone expert diagnostic and decision support systems have existed for decades, the EHR adoption has become widespread in the last decade since the passage of Meaningful Use incentives in the Recovery Act. EHRs were first promoted as a solution to medical errors by the Institute of Medicine (IOM), now the National Academy of Medicine, in a landmark 1991 report, titled "The Computer-Based Patient Record: An Essential Technology for Health Care" (IOM, 1997). The idea was that EHRs would improve quality and enhance safety and help transform the health system.

As compared with paper charts, EHRs do allow enhanced access to information for both clinicians through easier access to laboratory results and notes and patients via the patient portals. Electronic prescribing has been shown to reduce prescribing errors and improve patient safety (Guthrie et al., 2016; Kirkendall et al., 2016). Systematic literature reviews of EHRs found evidence for improvements in efficiency, effectiveness, patient safety, patient satisfaction, and improved care process outcomes but mixed results in provider satisfaction (Buntin et al., 2011). Clinical decision support and reminders have been shown to help practitioners deliver safer and higher-quality care to patients and the community via improvements in process outcomes (Implementation Science, in press; Romano & Stafford, 2011). From the population health perspective, software tools such as registries and data warehouses have provided opportunities to assess population-based outcomes such as screening and immunization rates, diabetes care, and chronic care management interventions (Menachemi & Collum, 2011).

But along with these benefits, there are also unintended consequences of computerized order entry (Ash et al., 2007) and health information technology, most of which are also problems with EHR use in general. Extra work, effects on workflow, and patient communication have had particularly negative impacts on the job satisfaction and mood of primary care clinicians. In addition to the extra time for documentation as noted by Sinsky et al. (2016), EHRs are felt to take away from doctor-patient interaction and impair the doctor-patient relationship (Gardner & Levinson, 2016).

The articles published in our special issue add to the knowledge base surrounding some of these issues. Bobadilla et al. (in this issue) describe how a thoughtful workflow redesign at the time of EHR implementation using LEAN principles and physician extenders could lead to better clinician workflow and improved outcomes for deep vein thrombosis prophylaxis. Similarly, a study on the use of scribes (Imdieke et al., in this issue) details this emerging work-around and its effects on encounter efficiency and physician satisfaction. Bunce et al. (in this issue) report on feelings of clinicians undergoing quality measurement in a health center setting, highlighting how these quality measures fail to capture their hard work and the socioeconomic challenges that their patients face. While many providers feel that the EHR impedes productivity, Frogner et al. (in this issue) found increased productivity for physicians on EHRs as compared with those on paper, but a reduced productivity for nurses, NPs, and other staff. Finally, recognizing that physicians are not the only users of EHRs, Jones et al. (in this issue) review EHR adoption by NPs and describe the factors associated with increased EHR use among NPs.

EHR integration is very complex, encompassing issues related not only to technology but also to design, end-user satisfaction (from clinicians to patients to data managers), clinical workflow, vendor-related issues, policy implementation, finances, and interoperability (see editorial by Zuckerman, in this issue). Like so many other complicated health-related issues, adoption and optimization will require work and research on multiple fronts. Research is needed to explore, among other things, EHR customization for primary care, alert fatigue, quality, and interoperability improvement.

Critically, research should seek to improve EHR usability, including research by nonacademics and by the designers and vendors, whose products we, as members of health care system, are asked to adopt and embrace. Usability has been defined as the "extent to which a system, product or service can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction..." (International Organization for Standardization, 2010). Vendors and system designers could look at these clinician issues as usability problems and try to fix them with existing usability tools. For example, vendors could measure and improve the time it takes for users to write an EHR note (the efficiency of their use). They could make EHRs easier to learn and measure and minimize their impact on doctor-patient communication patterns. All this could be done with tools that vendors know how to use and (to varying degrees) already use. However, to be valid, this work needs to be done in the real world with real patients, charts, and real-world networks so that the true effectiveness-or lack thereof-of these systems can be understood. Finally, the government and payers need to continue to work toward elimination of the current E&M coding system since, in large part, it has driven EHR use and configuration to maximize note writing for billing purposes at the exclusion of everything else. This has taken a huge toll on clinicians' lives and satisfaction and has contributed to resentment and anti-EHR sentiment, even in those of us who strongly believe in the promise of digital data and their potential to improve quality of care. We look forward to using the next wave of EHR systems-based on and built with consideration of real-life evidence and user experience.

REFERENCES