Beyond the Focus Group: Understanding Physicians’ Barriers to Electronic Medical Records

The Joint Commission Journal on Quality and Patient Safety Information Technology

Beyond the Focus Group: Understanding Physicians’ Barriers to Electronic Medical Records Helen Yan, BS; Rebekah Gardner, MD; Rosa Baier, MPH

C

linicians and policymakers increasingly recognize that electronic medical records (EMRs) can increase adherence to evidence-based practices1 and improve patient safety, for example, by indicating harmful drug-allergy interactions.2 Along with the ability to provide higher-quality care, EMRs have been touted as a cost-effective change for providers and the health care system by reducing redundancy in the system, facilitating work flow and clinical decision making, and decreasing errors.3,4 Effective EMR implementation has the estimated potential of saving the health care system in the United States more than $81 billion annually,5 and state and federal efforts are under way to accelerate adoption.6 Despite the potential benefits, national surveys in the United States demonstrated that only 17% of office-based physicians had a basic EMR system in 2008,7 and only 12% of hospitals had such a system in 2009.8 The slow adoption rate has been attributed to a number of causes, including high financial costs and uncertain payoffs, loss of productivity, concerns with available technology,9 and physician resistance to changes in the culture of care and work flow.10 Although addressing physicians’ perceived barriers is vital to increasing EMR adoption and use, much of the previous research has relied on small focus groups of primary care physicians (PCPs) and may have limited generalizability.1,2,11 Using a mandatory statewide survey of Rhode Island physicians, we sought to examine physicians’ barriers to EMR adoption and to understand how physician characteristics, such as age, practice setting, and specialty, are associated with the perception of these barriers. We hypothesized that physicians with EMRs would differ from those without EMRs, both in the types of barriers identified and their magnitude.

Methods DATA SOURCES Data came from the Rhode Island Department of Health’s 2009 Physician Health Information Technology (HIT) survey. The 184

April 2012

Article-at-a-Glance Background: Although electronic medical records (EMRs)

have potential to improve quality of care, physician adoption remains low. Rhode Island physicians’ perceptions of barriers to EMRs and the association between these barriers and physician characteristics were examined. It was hypothesized that physicians with and without EMRs would differ in the types and magnitude of barriers identified. Methods: Data were drawn from the Rhode Island Department of Health’s mandatory 2009 Physician Health Information Technology (HIT) survey of physicians licensed and in active practice in Rhode Island or an adjacent state. Some1,888 (58.1% of the target population of 3,248 physicians) responded. Respondents, who were invited to provide open-ended comments, were asked to consider 11 issues as barriers to EMR use: Access to technical support, lack of computer skills, availability of a computer in the appropriate location, impact of a computer on doctor–patient interaction, lack of interoperability, privacy or security concerns, start-up financial costs, ongoing financial costs, technical limitations of systems, training and productivity impact, and lack of uniform industry standards. Results: Respondents with EMRs consistently perceived significantly fewer barriers than those without them (p < .0001). For example, 78.9% of physicians without EMRs viewed start-up financial costs as a major barrier versus only 45.8% of physicians with EMRs. Conclusions: An understanding of physicians’ reluctance to use EMRs is critical for developing adoption strategies. Policies to increase EMR adoption should be tailored to different physician groups to achieve maximum effectiveness. Further research into the differences between current EMR users’ and nonusers’ perceptions of barriers may help elucidate how to facilitate subsequent adoption.

Volume 38 Number 4

Copyright 2012 © The Joint Commission

The Joint Commission Journal on Quality and Patient Safety survey is administered annually to all physicians licensed and in active practice in Rhode Island, a small northeastern state with an estimated population of an estimated 1,051,302 (2011),12 and includes physician demographics, information about EMR adoption, and perceptions of barriers to EMR adoptions. We linked the survey with licensure data to create an analytic dataset that included physician characteristics, such as age.

Technical Barriers ■ Lack of computer skills ■ Access to technical support ■ Availability of a computer in the appropriate location Respondents were asked to rate each issue (“not a barrier,” “a minor barrier,” or “a major barrier’) and were also invited to provided free-text comments.

SURVEY ADMINISTRATION The Department of Health and its public reporting contractor, Healthcentric Advisors (Providence), first piloted this survey in 2008 and have administered it every year since 2009, the baseline year. The 2009 survey used licensure data to identify physicians with active medical licenses who were providing direct patient care and were located in Rhode Island, Connecticut, or Massachusetts. In January 2009 all eligible physicians received a letter with a link to the electronic survey, as well as an e-mail notice if they had an available e-mail address. The e-mail subset also received two reminder e-mails. The survey closed in February 2009. The Department of Health calculates five measures of HIT adoption and disseminates results in aggregate and on its public reporting website. A public-use data set is also available, enabling stakeholders to use these data to inform strategies to accelerate EMR adoption, such as incentive payments or state or federal meaningful use.6

OUTCOME MEASURES: BARRIERS TO EMR USE Survey respondents were asked to indicate the extent to which they considered 11 issues as barriers to EMR use. Potential barriers topics were identified through the literature and through stakeholder comments during survey development and testing. The 11 barriers, grouped by theme, were as follows: Financial and Time ■ Start-up financial costs ■ Ongoing financial costs ■ Training and productivity impact Data Exchange ■ Lack of interoperability (that is, inability of different systems to communicate and share information) ■ Lack of uniform industry standards ■ Technical limitations of systems Interpersonal ■ Privacy or security concerns ■ Impact of a computer on doctor–patient interaction

April 2012

INDEPENDENT VARIABLES We looked at the effect of current EMR use, age, practice setting, practice size, number of hours spent in patient care, and specialty on physicians’ perception of barriers. We measured EMR use by the survey questions, “Does your main practice have EMR components?” and, if no, “Do ANY of your practice settings have EMR components?” EMR components were defined as “integrated electronic clinical information systems that track patient health data.” We defined EMR use as a positive response to either question. We used licensure date of birth to calculate four age categories. Physicians’ self-reported practice setting was categorized in terms of office or a hospital, size (< 5 clinicians, 5–10 clinicians, or > 10 clinicians), and patient care hours (< 20 hours or > 20 hours). We grouped specialties into nine categories: primary care, medical subspecialties, surgery and surgical subspecialties, anesthesiology, emergency medicine, obstetrics/gynecology, psychiatry, radiology, and other. Primary care included family medicine, pediatrics, geriatrics, and internal medicine without further subspecialization. Medical subspecialties included internal medicine subspecialties such as cardiology, as well as dermatology, neurology, and ophthalmology. “Other” specialties included pain management, pathology, radiation oncology, and occupational medicine.

DATA ANALYSIS We analyzed the data using STATA v10 (College Station, Texas). First, we conducted bivariate analyses between EMR use and each independent variable to assess demographic characteristics. We used a Pearson chi-square p < .2 as a cutoff for inclusion of independent variables in our regression model. Second, we conducted bivariate analysis between EMR use and each barrier. Finally, we examined the association between the 11 barriers and EMR use using logistic regression modeling, controlling for age, practice setting, practice size, care hours, and specialty. For the regression model and overall ranking of barriers, we collapsed “major barrier” and “minor barrier” responses into one

Volume 38 Number 4

Copyright 2012 © The Joint Commission

185

The Joint Commission Journal on Quality and Patient Safety Table 1. Characteristics of Survey Respondents (N = 1,888)*

Age, years

< 40

Does the Physician Use Electronic Medical Records? Frequency (% of Column Total) No Yes Total 74 (12.1%) 303 (23.8%) 377 (20.0%)

(N = 1,886)

40–49

150 (24.5%)

415 (32.5%)

565 (30.0%)

p < .0001

50–59

210 (34.4%)

385 (30.2%)

595 (31.5%)

≥ 60

177 (29.0%)

172 (13.5%)

349 (18.5%)

Practice Setting

Hospital

117 (19.2%)

587 (46.0%)

704 (37.3%)

(N = 1,887)

Office

493 (80.8%)

690 (54.0%)

1,183 (62.7%)

p < .0001 Practice Size

< 5 clinicians

344 (57.5%)

381 (30.0%)

725 (38.9%)

(N = 1,866)

5–10 clinicians

151 (25.3%)

316 (24.9%)

467 (25.0%)

p < .0001

> 10 clinicians

103 (17.2%)

571 (45.0%)

674 (36.1%)

Care Hours/Week

≤ 20

117 (19.6%)

215 (17.0%)

332 (17.9%)

(N = 1,859)

> 20

480 (80.4%)

1,047 (83.0%)

1,527 (82.1%)

Physician Specialty

Primary care†

206 (34.0%)

463 (36.4%)

669 (35.6%)

(N = 1,879)

Medical subspecialties‡

176 (29.0%)

236 (18.5%)

412(21.9%)

p = .178

p < .0001

Surgery and surgical subspecialties

71 (11.7%)

139 (10.9%)

210 (11.2%)

Anesthesiology

10 (1.7%)

50 (3.9%)

60 (3.2%)

Emergency medicine

8 (1.3%)

74 (5.8%)

82 (4.4%)

Obstetrics/gynecology

32 (5.3%)

65 (5.1%)

97 (5.2%)

Psychiatry

68 (11.2%)

110 (8.6%)

178 (9.5%)

Radiology

9 (1.5%)

55 (4.3%)

64 (3.4%)

Other§

26 (4.3%)

81 (6.4%)

107 (5.7%)

* Source: Authors’ analysis of data from the Rhode Island Health Information Technology survey and the Rhode Island Department of Health licensure database. †

“Primary care” includes the specialties of family medicine, pediatrics, and internal medicine without further subspecialization, with the exception of geriatrics,

which was included in primary care. ‡

“Medical subspecialties” includes internal medicine subspecialties such as cardiology, as well as specialties such as dermatology, neurology, and ophthalmology.

§

“Other” includes specialties not captured in preceding categories, such as pain management, pathology, radiation oncology, and occupational medicine.

category. We calculated the adjusted odds ratios (AORs) to describe the size of the effects, using a likelihood ratio chi-square p < .05 to determine statistical significance. We also reviewed and categorized physicians’ free-text comments.

Results CHARACTERISTICS OF SURVEY RESPONDENTS AND EMR USE Of 3,248 physicians in the target population, 1,888 (58.1%) responded. Younger physicians were more likely to use EMRs than older physicians. Although physicians 60 years of age or older comprised 18.5% of the sample, they accounted for almost a third of physicians who did not use EMRs (Table 1, above). Differences in EMR use also existed at the practice level. Hospitalbased physicians, who comprised only 37.3% of the sample, 186

April 2012

accounted for almost half of all physicians with EMRs. Physicians in large practices (at least 10 clinicians) were more likely to report using EMRs than physicians in small practices. Overall, PCPs comprised the largest percentage of all specialties (35.6%). The percentage (36.4%) of PCPs using EMRs was comparable to the percentage (34.0%) of PCPs not using EMRs. Physicians specializing in surgery and surgical subspecialties or obstetrics/gynecology also had similar percentages of use and nonuse of EMRs. However, physicians specializing in medical subspecialties or psychiatry were less likely to use EMRs, while physicians specializing in anesthesiology, emergency medicine, radiology, or “other” specialties were more likely to use EMRs.

BARRIERS As shown in Figure 1 (page 187), respondents perceived

Volume 38 Number 4

Copyright 2012 © The Joint Commission

The Joint Commission Journal on Quality and Patient Safety Overall Ranking of Physicians’ Perceived Barriers to Electronic Medical Record (EMR) Adoption, from Most Frequently Identified to Least

Figure 1. Respondents perceived EMRs’ training and productivity impact as the largest barrier (77.5%), and lack of computer skills was perceived as a barrier by the fewest number of respondents (34.4%).

EMRs’ training and productivity impact as the largest barrier (77.5%). The technical limitations of an EMR system (76.3%) and the ongoing financial costs (76.2%) were the second and third most popular barriers, respectively. Lack of computer skills was perceived as a barrier by the fewest number of respondents, with 34.4% of physicians viewing it as an obstacle. Respondents with EMRs consistently perceived significantly fewer barriers than those without them (p < .0001; Figure 2, page 188). This difference was particularly apparent among the financial and time barriers: 78.9% of physicians without EMRs viewed start-up financial costs as a major barrier versus only 45.8% of physicians with EMRs. Almost half of nonusers viewed training and productivity as a major barrier, compared with only 28.5% of users. The differences persisted among the other barriers. More than half of nonusers perceived lack of uniform industry standards as a major barrier, whereas only 36.3% of users did. Nearly two fifths of nonusers perceived impact on doctorpatient interaction as a major barrier versus only 17.8% of users.

April 2012

Some 4.6% of EMR users viewed lack of computer skills as a major barrier versus 12.2% of nonusers.

PHYSICIAN CHARACTERISTICS AND BARRIERS Physician characteristics associated with perceiving more barriers to EMR adoption include not having an EMR, being older, working in an office, and specializing in emergency medicine (Appendix 1, available in online article). Practice size and number of hours per week a physician spends in direct patient care were not independently associated with how a physician perceives barriers. The finding that physicians without EMRs were more likely to identify barriers than physicians with them persisted after regression analysis controlling for the other characteristics, supporting our hypothesis that EMR adopters and nonadopters would differ in both the types of barriers identified and in their magnitude. This was true for all barriers and was most pronounced with regard to financial costs: Physicians without

Volume 38 Number 4

Copyright 2012 © The Joint Commission

187

The Joint Commission Journal on Quality and Patient Safety Physicians’ Perceived Barriers to Electronic Medical Record (EMR) Adoption, by EMR Use/Nonuse and Barrier Theme a. Financial and Time Barriers

b. Data Exchange Barriers

c. Interpersonal Barriers

d. Technical Barriers

Figure 2. Data are shown for the 11 barriers, grouped by themes: (a) financial and time, (b) data exchange, (c) interpersonal, and (d) technical. Respondents with EMRs consistently perceived significantly fewer barriers than those without them (p < .0001). This difference was particularly apparent among the financial and time barriers.

EMRs were more than five times as likely to perceive start-up (AOR: 5.26, 95% confidence interval [CI]: 3.70–7.69) and ongoing financial costs (AOR: 5.26, 95% CI: 3.70–7.69) as barriers. This reinforces the unadjusted differences, as presented in Figure 2. Compared with PCPs, surgeons and physicians in medical subspecialties did not, for the most part, significantly differ in their perception of barriers. Emergency medicine physicians, however, were overall more likely to identify barriers than PCPs; they were three times more likely to perceive the impact on doc188

April 2012

tor–patient interaction as well as the availability of a computer in the appropriate location as obstacles. Anesthesiologists and radiologists were both less likely to identify impact on doctor– patient interaction as barriers. Psychiatrists were less likely to identify technical barriers and interpersonal barriers, with one exception: They were 1.5 times more likely to perceive privacy or security concerns as a barrier (Appendix 1).

FREE-TEXT COMMENTS Approximately one in four respondents (n = 470, 24.8%)

Volume 38 Number 4

Copyright 2012 © The Joint Commission

The Joint Commission Journal on Quality and Patient Safety provided free-text comments, which reiterated the substantial financial costs and productivity loss associated with implementing an EMR system. They also underscored the perception that physicians would be unlikely to directly benefit from EMR implementation, citing an increase in the work of documentation, as well as difficulties in using a technology that is often not userfriendly. However, a number of physicians brought up the comprehensiveness and ease of using the EMR system in US Veterans Health Administration medical centers.

Discussion Understanding physicians’ reluctance to use EMRs is critical for developing targeted strategies to effectively accelerate adoption. We found that physicians across a wide spectrum of specialties and practice settings perceived substantial barriers to EMR implementation, and, significantly, noted differences in the type and magnitude of barriers among physicians with and without EMRs. We also found wide variation in perceptions of barriers among different physician specialties. These differences persisted after controlling for other physician and practice characteristics captured in our data. Overall, the largest barriers to EMR use among the respondents were related to finances, time, and technical limitations of available EMR systems, which is consistent with previous qualitative studies.1,9,11,13 Physicians without EMRs consistently perceived more EMR barriers and a greater impact of these barriers than physicians with EMRs, supporting other studies.14–16 This may be because physicians who perceive greater barriers are less likely to adopt the technology or the converse: That after physicians have adopted EMRs, their perception of barriers may differ (for example, they may have found the barriers to be more manageable than they originally believed or to not represent impediments). There may also be a psychological bias to view decisions more favorably after one makes them.17 The fact that physicians without EMRs perceived more barriers to implementation, in terms of greater number and severity, suggests that physicians may underor overestimate barriers before EMR implementation. A pre-post implementation evaluation could help determine which perceptions reflect true barriers and enable policy makers to tailor messaging and strategies to correct any misperceptions and address actual barriers. Similarly, analyses by market (for example, different types of hospitals or outpatient clinics) could help assess the extent to which market characteristics, such as EMR maturity, relate to perceptions of barriers. Start-up and ongoing financial costs were perceived as much greater barriers to physicians who had not adopted EMRs, consistent with previous research.14–16,18 Up-front costs to install an

April 2012

EMR may range from $16,000 to $36,000 per physician,19 a significant investment that does not lead to immediate returns, which makes many physicians reluctant to adopt. However, a survey of 4,203 ambulatory care physicians in Florida revealed that physicians with EMRs or who were planning to implement within one year (“imminent adopters”) were significantly less likely to view up-front costs or inadequate return on investment as barriers,14 reinforcing our finding that perceptions between EMR adopters and nonadopters differ. We did not attempt to determine whether the physicians with EMRs had greater financial resources or whether they, as stated, were acting according to a psychological bias to view their decisions favorably, after they had made them.17 In any case, payer incentive payments may help to decrease financial costs and mitigate these important barriers. Compared with physicians younger than 40 years of age, older physicians were more likely to perceive barriers to EMR adoption across several domains; most significantly, inadequate computer skills. Computer proficiency is vital to efficient use of EMRs, and younger physicians might be faster typists and more technology-savvy. Younger physicians are also more likely to have been exposed to EMRs during residency. In addition, physicians who are older and likely have established practices might be more resistant to the change in work flow associated with implementation of a new system. They might also have to incur greater costs in converting paper charts if they have larger patient panels. Finally, older physicians are closer to retirement and therefore less likely to see the return on their investment. Office-based physicians perceived greater barriers overall, across most domains, than hospital-based physicians. Whereas office-based physicians, for the most part, control their office revenue and have critical decision-making roles in purchasing the technology, hospital-based physicians might have less influence over decisions to implement EMRs. Financing for an EMR in an office-based practice comes directly out of the physician’s pocket, and the training and productivity impact directly affects the physician and his or her staff. In a hospital, this impact is spread across multiple departments and larger groups of physicians. In addition, unlike their office-based colleagues, many hospital-based physicians do not perceive an impact from an EMR on the doctor–patient interaction, perhaps because their documentation often occurs after an encounter in the patient’s room rather than as the clinical interaction is taking place. There is one exception to the trend of office-based physicians’ reporting of more substantial barriers than hospital-based physicians: Office-based physicians were less likely to identify the availability of a computer in the appropriate location as a barrier, most

Volume 38 Number 4

Copyright 2012 © The Joint Commission

189

The Joint Commission Journal on Quality and Patient Safety likely because offices that implement EMRs install computers in each exam room, whereas hospitals may place only a small number of computers in each wing. Differences among physician specialties persisted after controlling for the physician and practice characteristics captured in our data. Emergency medicine physicians were three times more likely than PCPs to perceive the impact on doctor–patient interaction as a barrier. In addition, they were more likely to identify the availability of a computer in the appropriate location as a barrier. The hectic environment of the emergency department might magnify productivity and technical barriers; an increasing emphasis on improving patient flow and a length of stay measured in minutes may make the prospect of any additional time spent in using an EMR even less acceptable. In contrast, anesthesiologists and radiologists were less likely than other physicians to identify obstacles to EMR adoption and were particularly less likely to perceive the impact of a computer on doctor-patient interaction as a barrier. This finding may reflect the nature of their practice, which typically has less intensive patient contact. Psychiatrists were just as likely as or less likely than PCPs to perceive barriers, except in the area of privacy or security concerns. Mental health records often have more stringent legal protections than standard health records, which may make psychiatrists more wary of the privacy and security of such records in an EMR system. In addition, it is likely that physicians’ perceptions of EMR functions differ by specialty. For example, hospitalists might be more concerned about the ease of entering notes and retrieving longitudinal data, whereas surgeons might be more focused on entering orders. Differences in EMR system use likely color perceptions of EMR barriers.

LIMITATIONS Although this study is notable for its large sample size, wide cross-section of practicing physicians, and systematic collection of perceived barriers to EMR implementation, there are several limitations. First, the data are self-reported and identifiable. Respondents might be more likely to have EMRs or to have a stronger opinion than nonrespondents on EMR barriers. The fact that physicians were told that a lack of survey response would be reported as nonuse of an EMR may have decreased the likelihood of including physicians without EMRs in the sample. In addition, the knowledge that their name and other identifiers are captured might render them more likely to self-report “socially desirable” responses to barrier and other questions, even though their responses would not be publicly reported at the physician level. Second, we classified physicians as EMR adopters or nonadopters on the basis of self-report (and self-per190

April 2012

ception) of whether or not they had “EMR components,” which were defined as “integrated electronic clinical information systems that track patient health data.” Rhode Island’s public report also calculates a more restrictive measure of EMR adoption, which limits EMRs to systems with certification and specific functionalities, as reported for national studies.7,8 Because the primary objective of this analysis was to compare EMR adopters with nonadopters, we chose to use the broad measure that directly reflects physicians’ self-report. Third, the survey was administered electronically, which required access to a computer, possibly increasing the likelihood that the physician responding also had an EMR. In addition, EMR users had a larger incentive to respond to the survey because the two largest health insurers in Rhode Island based their EMR incentive payments on physicians’ survey responses, although physicians qualifying for these incentives were told by the insurers that they would be subject to possible audit. Fourth, the survey was administered in a single state, which may limit its generalizability. Finally, the survey provides cross-sectional data, taken at one point in time, which prevents the evaluation of a causal relationship between physician characteristics and the perception of barriers to EMR use. However, the survey is administered annually, which may present an opportunity to link physician records longitudinally and gain a better understanding of the relationship between perceptions to barriers and EMR adoption among physicians without EMRs who later implement them.

Conclusions The findings indicate that EMR adopters and nonadopters differed markedly in their perceptions of barriers to EMR implementation. Additional research is needed to elucidate a causal relationship between the perception of barriers and the subsequent adoption of an EMR: for example, in a longitudinal prepost EMR adoption survey, evaluate the effectiveness of financial incentives and social marketing strategies to encourage physicians to employ EMRs, and assess the relationship between various submarkets (for example, different types of hospitals or outpatient clinics) and perceptions of barriers. Conceptual frameworks for evaluating barriers should help standardize future research,20,21 and there are numerous federal initiatives to address known barriers; for example, through meaningful use payments,6 training curricula development,22 and efforts to address industry standards.23 In addition, physicians’ perceptions of EMR barriers vary by specialty and practice settings, among other characteristics, emphasizing the importance of tailoring EMR adoption policies to different physician groups in order to achieve maximum effectiveness. For example, EMR adoption

Volume 38 Number 4

Copyright 2012 © The Joint Commission

The Joint Commission Journal on Quality and Patient Safety strategies in the emergency department could emphasize adequate numbers of computers, ready technical support, and minimal disruption to the doctor–patient interaction, insofar as emergency medicine physicians view these issues as particularly strong barriers to EMR implementation. Although initiatives to subsidize costs and improve data exchange are important, policy makers will need to make physicians feel that investing in an EMR will not only benefit their patients but themselves, as well. J The annual Physician HIT Survey is funded by the State of Rhode Island’s Healthcare Quality Reporting Program, a legislatively mandated quality reporting program. The survey is jointly administered by the Rhode Island Department of Health and its public reporting contractor, Healthcentric Advisors. HIT adoption is publicly reported annually, and analytic files are available upon request by contacting the corresponding author.

Helen Yan, BS, formerly a Student, Brown University, Providence, Rhode Island, is a Student, Boston University School of Law, Boston. Rebekah Gardner, MD, is Senior Medical Scientist, Healthcentric Advisors, Providence; and Assistant Professor of Medicine, Warren Alpert Medical School, Brown University, Providence. Rosa Baier, MPH, is Senior Scientist at Healthcentric Advisors; and Teaching Associate, Warren Alpert Medical School. Please address correspondence to Rosa Baier at [email protected].

Online-Only Content

8

See the online version of this article for

Appendix 1. Association Between Physicians’ Perceptions of Electronic Medical Record (EMR) Barriers and Various Physician and Practice Characteristics: A Logistic Regression Analysis

References 1. Terry AL, et al. Implementing electronic health records: Key factors in primary care. Can Fam Physician. 2008;54(5):730–736. 2. Holden RJ. Physicians’ beliefs about using EMR and CPOE: In pursuit of a contextualized understanding of health IT use behavior. Int J Med Inform. 2010;79(2):71–80. 3. Wang SJ, et al. A cost-benefit analysis of electronic medical records in primary care. Am J Med. 2003 Apr 1;114(5):397–403. 4. Bates DW, Kuperman GJ. The role of health information technology in quality and safety. In From Front Office to Front Line: Essential Issues for Health Care Leaders, 2nd ed. Oak Brook, IL: Joint Commission Resources, 2011, 87–108.

April 2012

5. Hillestad R, et al. Can electronic medical record systems transform health care? Potential health benefits, savings, and costs. Health Aff (Millwood). 2005;24(5):1103–1116. 6. Office of the National Coordinator for Health Information Technology, US Department of Health & Human Services. Electronic Health Records and Meaningful Use. (Updated Feb 9, 2011.) Accessed May 18, 2011. http://healthit.hhs.gov/portal/server.pt?open=512&objID=2996&mode=2. 7. DesRoches CM, et al. Electronic health records in ambulatory care—A national survey of physicians. N Engl J Med. 2008 Jul 3;359(1):50–60. 8. Jha AK, et al. Use of electronic health records in U.S. hospitals. N Engl J Med. 2009 Apr 16;360(16):1628–1638. 9. Miller RH, Sim I. Physicians’ uses of electronic medical records: Barriers and solutions. Health Aff (Millwood). 2004;23(2):116–126. 10. Harvard School of Public Health, Harvard Public Health Review. Electronic Health Records. Hand L. Fall 2008. Accessed Feb 23, 2012. http://www.hsph.harvard.edu/news/hphr/fall-2008/fall08ehealth.html. 11. Informing Science Insitute, Issues in Informing Science and Information Technology. Resistance to Electronic Medical Records (EMRs): A Barrier to Improved Quality of Care. Meinert D. 2005. Accessed Feb 23, 2012. http://informingscience.org/proceedings/InSITE2005/I41f100Mein.pdf. 12. U.S. Census Bureau. State & County QuickFacts: Rhode Island. (Updated Jan 17, 2012.) Accessed Feb 22, 2012. http://quickfacts.census.gov/qfd /states/44000.html. 13. Burt CW, Sisk JE. Which physicians and practices are using electronic medical records? Health Aff (Millwood). 2005;24(5):1334–1343. 14. Menachemi N. Barriers to ambulatory EHR: Who are ‘imminent adopters’ and how do they differ from other physicians? Inform Prim Care. 2006; 14(2):101–108. 15. Wisconsin Department of Health Service. 2008 Wisconsin Ambulatory Health Information Technology Survey. Mar 31, 2009. Accessed Feb 23, 2011. http://www.dhs.wisconsin.gov/ehealth/EHR/2008EHRsurveyfinal.pdf. 16. Kemper AR, Uren RL, Clark SJ. Adoption of electronic health records in primary care pediatric practices. Pediatrics. 2006;118(1):e20–24. 17. Mather M, Johnson MK. Choice-supportive source monitoring: Do our decisions seem better to us as we age? Psychol Aging. 2000;15(4):596–606. 18. Gans D, et al. Medical groups’ adoption of electronic health records and information systems. Health Aff (Millwood). 2005;24(5):1323–1333. 19. Simon SR, et al. Physicians and electronic health records: A statewide survey. Arch Intern Med. 2007 Mar 12;167(5):507–512. 20. Boonstra A, Broekhuis M. Barriers to the acceptance of electronic medical records by physicians from systematic review to taxonomy and interventions. BMC Health Serv Res. 2006 Aug 6;10:231. 21. Vishwanath A, Scamurra SD. Barriers to the adoption of electronic health records: Using concept mapping to develop a comprehensive empirical model. Health Informatics J. 2007;13(2):119–134. 22. Office of the National Coordinator for Health Information Technology, U.S Department of Health & Human Services. Curriculum Development Centers Program. (Updated Jan 19, 2010.) Accessed Feb 23, 2012. http://healthit.hhs.gov/portal/server.pt/community/curriculum_development _centers_program/1418/home /16935. 23. Office of the National Coordinator for Health Information Technology, US Department of Health & Human Services. State Health Policy Consortium. (Updated Aug 13, 2010.) Accessed Feb 23, 2012. http://healthit.hhs.gov/portal/server.pt/community/healthit_hhs_gov__state _health_policy_consortium/3035.

Volume 38 Number 4

Copyright 2012 © The Joint Commission

191

The Joint Commission Journal on Quality and Patient Safety Online-Only Content

8

Appendix 1. Association Between Physicians’ Perceptions of Electronic Medical Record (EMR) Barriers and Various Physician and Practice Characteristics: A Logistic Regression Analysis*

* The associations are presented as odds ratios (95% confidence intervals). The table cells with lighter highlighting are statistically significant at p < 0.05; cells with darker highlighting are statistically significant at p < 0.01. Ob/gyn, obstetrics and gynecology. †

Reference group.

AP1

April 2012

Volume 38 Number 4

Copyright 2012 © The Joint Commission