Health Information Technology for Ambulatory Care in Health Systems

Health systems are important in driving electronic health record adoption in ambulatory clinics, although the uptake of key functionalities varies across systems.


Objectives: The adoption and use of health information technology (IT) by health systems in ambulatory care can be an important driver of care quality. We examine recent trends in health IT adoption by health system—affiliated ambulatory clinics in the context of the federal government’s Meaningful Use and Promoting Interoperability programs.

Study Design: We analyzed a national sample of 17,861 ambulatory clinics affiliated with 1711 health systems, using longitudinal data (2014-2016) from the HIMSS Analytics annual surveys.

Methods: We used descriptive analyses and linear probability models to examine the adoption of electronic health records (EHRs), as well as 16 specific functionalities, at the clinic level and the system level. We compared the differential trends of adoption by various characteristics of health systems.

Results: We find that the adoption of an EHR certified by the Office of the National Coordinator for Health IT (ONC) increased from 73% to 91%. However, in 2016, only 38% of clinics reported having all 16 health IT functionalities included in this study. Small health systems lag behind large systems in ambulatory health IT adoption. Patient-facing functionalities were less likely to be adopted than those oriented toward physicians. Health information exchange capabilities are still low among ambulatory clinics, pointing to the importance of the ONC’s recent Promoting Interoperability initiative.

Conclusions: The relatively low uptake of health IT functionalities important to care improvement suggests substantial opportunities for further improving adoption of ambulatory health IT even among the current EHR users.

Am J Manag Care. 2020;26(1):32-38.

Takeaway Points

  • The adoption of an electronic health record (EHR) certified by the Office of the National Coordinator for Health Information Technology (ONC) among ambulatory clinics increased from 73% to 91% between 2014 and 2016.
  • However, in 2016, only 38% of clinics reported having all 16 health information technology (IT) functionalities included in this study.
  • Small health systems significantly lag behind large systems in adoption.
  • Patient-facing functionalities were less likely to be adopted than functionalities oriented toward physicians.
  • Health information exchange capabilities are still low, pointing to the importance of the ONC’s continuing focus on interoperability.
  • Our findings suggest substantial opportunities for further improving adoption of ambulatory health IT even among the current EHR users.

Advancing the adoption and use of health information technology (IT) has been an important focus of the federal government’s efforts to improve quality of care during the past decade. Two major pieces of legislation promoted health IT adoption: the Health Information Technology for Economic and Clinical Health (HITECH) Act,1 enacted as part of the American Recovery and Reinvestment Act of 2009, and the Patient Protection and Affordable Care Act2 in 2010. The multistage incentive programs under HITECH, commonly known as Meaningful Use, provided more than $30 billion to about a half-million healthcare providers and were essential to the initial adoption of and sustained use of health IT.3 Nevertheless, the penetration of health IT has proceeded at different paces for hospitals and providers in ambulatory settings. For example, by 2013, more than 59% of hospitals had adopted at least a basic electronic health record (EHR), whereas the proportion of clinics using an EHR was estimated to be about 48%.4 As the adoption and use of health IT grow, evidence supporting the positive impacts of health IT has also been accumulating, although not all health IT implementations have produced effective results.5 Furthermore, there is evidence of unintended consequences, such as physician burnout.6 Although most empirical studies linking IT to quality of care using large national samples were conducted in hospital settings,7,8 a recent study found that higher rates of ambulatory EHR adoption at the county level were associated with reductions in hospitalization between 2003 and 2012.9

Along with policies to promote health IT adoption, significant and rapid changes have been occurring in the organizational landscape of healthcare, with health systems assuming an increasingly important role in delivering care.10 A typical health system, often formed through consolidation with the expectation of better clinical integration and higher-value care, can include multiple hospitals, numerous ambulatory clinics, and even an insurance plan. Some studies have found positive associations between care quality and increased integration of care delivery (ie, various healthcare services being delivered with coordination among different providers) within health systems,11,12 although whether the formation of health systems actually leads to quality improvement is still an open question.13

However, little is known about how the adoption and use of health IT varies across health systems. Different health systems may vary in their capabilities of pooling resources to overcome cost, implementation, or usage barriers to health IT, and they may also employ different strategies to adopt and implement IT.14 In this study, we examine the recent trends in the adoption of health IT by health system—affiliated ambulatory clinics and make 2 important contributions to this topic. First, despite the large existing literature on health IT adoption in hospitals, relatively few studies have examined adoption in the ambulatory setting at the national level. Most previous research on ambulatory health IT either examined only basic EHR adoption without measuring adoption of specific functionalities15 or studied selected functionalities using cross-sectional data.16 One recent study examined use of specific functionalities among a cohort of 566 small primary care clinics.17 Our study extends this work by analyzing the adoption of 16 health IT functionalities with a longitudinal sample of ambulatory care clinics between 2014 and 2016. Second, our study focuses on clinics that are affiliated with (ie, owned, leased, or managed by) health systems to provide an understanding of ambulatory health IT adoption from a delivery system perspective. Specifically, we identify patterns of adoptions at both the ambulatory clinic level and the health system level and examine how adoption varies by clinic and health system characteristics. Our analysis provides an up-to-date understanding of the advancement of adoption of health IT in the ambulatory setting and the extent to which a core set of functionalities are being adopted to support patient care.


Data and Sample

We used 2014-2016 data from the HIMSS annual surveys of ambulatory clinics, provided by HIMSS Analytics LOGIC Market Intelligence Platform. The HIMSS surveys collected detailed information on the adoption and use of health IT for more than 75% of ambulatory care clinics affiliated with health systems.18 HIMSS defines a health system as an organization that owns at least 1 hospital.14 Although data from hospital surveys conducted by HIMSS have been widely used in previous health IT research,19 the HIMSS ambulatory survey data have been used in only a relatively small number of studies and government reports.20 Despite its limited use to date, the HIMSS ambulatory health IT database is an important source of information because of the longitudinal survey design and the large sampling frame.21

For our analysis, we included only ambulatory clinics that responded to the survey in all 3 years. They accounted for 92% of all clinics surveyed in 2014. Because we intended our analysis to be focused on ambulatory clinics in which providers directly interact with patients in delivering primary and/or specialty care, as these care sites were likely to benefit from the commonly adopted health IT functionalities,22 we excluded certain types of facilities (eg, urgent care, podiatry, and wellness centers). Finally, we excluded 6% of the clinics because they did not report the adoption statuses of all the functionalities examined in this study. Our final study sample contained 17,861 clinics and 1711 health systems (Table 1).

Health IT Functionalities

The National Academy of Medicine identified 8 core areas of health IT that are instrumental to the improvement of quality, safety, and efficiency in care delivery: health information and data, results management, order entry and management, decision support, electronic communication and connectivity, patient support, administrative processes, and reporting and population health.23 In our study, we measured the adoption of an ambulatory EHR system certified by the Office of the National Coordinator for Health IT (ONC), as well as the adoption of 16 specific health IT functionalities, which represent 7 of the 8 core areas defined by the National Academy of Medicine (Table 2) and were included by HIMSS in its annual survey of ambulatory care clinics. These specific functionalities are also at the center of the 3 stages of Meaningful Use programs (eg, clinical decision support) as well as their recent extension, Promoting Interoperability (eg, capabilities of exchanging health information).3,24 Functionalities in the area of reporting and population health were not covered in the HIMSS ambulatory surveys during our study period.

Characteristics of Ambulatory Care Clinics and Health Systems

We included ambulatory care clinic and health system characteristics in our analyses. For ambulatory clinics, we examined type (primary care vs specialty), location (urban vs rural), number of physicians, and ownership (owned vs leased or managed) by the parent health system. At the health system level, we included characteristics that capture both the inpatient and the outpatient aspects of care provided by each system: being a single-hospital system, number of licensed hospital beds, number of ambulatory clinics within each system, and profit status. We included number of licensed beds in our analysis because hospital(s) may influence a health system’s overall IT strategy and thus have implications on how ambulatory care clinics within the system adopt health IT. The size of a health system can potentially be quantified by various measures, such as the total number of physicians, the total annual revenue, and the total number of outpatient visits. We used the number of ambulatory clinics within a system as a proxy to measure system size for 2 reasons. First, the number of clinics is highly correlated with other measures of size available in our data. Second, the clinics are often the units for health IT implementation, and hence the number of clinics might be directly impactful in terms of cost and efficiency in adopting IT.25


We first descriptively examined the ambulatory clinics’ trends of adopting EHR and specific IT functionalities and tested whether the changes in adoption rates were statistically significant. To analyze the adoption trends at the health system level, we calculated adoption rates of EHR and health IT functionalities by aggregating the adoption statuses of all ambulatory clinics within a system, weighted by the number of physicians in each clinic. We then stratified health systems into 5 groups based on the number of ambulatory clinics within each system (1, 2, 3-10, 11-30, or >30) and computed the average adoption rates by stratum.

Linear probability (regression) models were used to examine whether ambulatory health IT adoption was associated with specific clinic characteristics (location, clinic type, ownership status, number of physicians in the clinic) and health system characteristics (being a single-hospital system, profit status, number of ambulatory clinics in the system, number of total licensed hospital beds in the system). The outcomes were the adoption statuses of individual IT functionalities as well as an aggregate indicator for whether a clinic adopted all 16 functionalities. Dummy variables for each period were included in all models, and standard errors were clustered at the health system level to adjust for potential intrasystem correlations.


Table 1 summarizes the characteristics of the study sample by year. The majority of the ambulatory care clinics in our sample were primary care providers (64%), were located in urban areas (82%), were owned by health systems (98%), and had 5 or fewer physicians (74%). The health systems included in the study mostly were single-hospital systems (74%) and nonprofit (95%), had 10 or fewer ambulatory care clinics (66%), and had 150 or fewer licensed hospital beds (52%). Although all sample characteristics were stable during the study period, the total number of health systems decreased from 1711 to 1601, likely because of consolidations.

The trends in the adoption of health IT functionalities are presented in Table 2. From 2014 to 2016, the adoption of all 16 functionalities increased significantly at the clinic level, by similar magnitudes. In 2016, 91% of the clinics in our sample had a certified EHR, an increase of 18 percentage points from 2014. Among the examined functionalities, clinical data repository had consistently the highest average adoption rate (61% in 2014 and 68% in 2016), whereas patient health record was the least adopted (38% in 2014 and 48% in 2016). In 2014, only 28% of clinics had adopted all 16 functionalities examined in this study, increasing to 38% in 2016. Across individual functionalities, the proportion of nonadopters was still large in 2016, between 32% and 52%. Patient-facing functionalities (eg, patient health record, patient portal) were less adopted than physician-oriented functionalities (eg, clinician charting, computerized provider order entry [CPOE]), and the differences persisted over time.

At the health system level, the adoption patterns varied greatly depending on the size of the system and the specific functionality, as shown in the Figure (complete results included in eAppendix Table 1 and eAppendix Figure [eAppendix available at]). Although there was an increasing trend in the adoption of all functionalities across all subgroups, the large systems (with >30 ambulatory clinics) showed not only a much higher level of adoption but also significantly faster increases over time compared with small systems (with ≤10 clinics). In 2016, the average adoption of CPOE among the large systems (69%) was more than double the rate among the small systems (28%) and also had a greater increase from 2014 (9 vs 5 percentage points). Similar patterns were seen in the adoption of other functionalities such as clinical decision support and e-prescribing. Only a modest proportion of small systems (13%-22% in 2016) had the capability to exchange clinical health information between clinics and hospitals, whereas 57% of the large systems were able to do so. Also, most patient-facing functionalities were present in only a fraction of small systems (8%-18% in 2016), and their adoption rates in large systems were around 50%, although these rates were still significantly lower than those of other functionalities. Finally, only about 5% of the small systems had adopted all 16 functionalities by 2016, whereas the proportion was much higher among the large systems (38%).

As shown in Table 3, several clinic and system characteristics were significant correlates of health IT adoption (P ≤.05). At the clinic level, being a primary care clinic and having a larger number of physicians were associated with an increased likelihood of adoption. At the system level, being nonprofit and having more ambulatory clinics in the system were associated with increased adoption levels of ambulatory health IT. In particular, nonprofit systems were, on average, between 20 and 35 percentage points more likely to adopt various functionalities. Systems with more than 30 ambulatory clinics were between 35 and 45 percentage points more likely to adopt compared with systems with only 1 clinic. Across functionalities, clinics with more than 20 physicians were associated with an increase of 20 percentage points in probability of adoption relative to solo physician clinics. Multispecialty care clinics were, on average, about 5 percentage points less likely to adopt the functionalities than primary care clinics. Being in urban areas, being a single-hospital system, and having more licensed hospital beds were not significantly associated with adoption. The significant and positive coefficients of the time indicators were consistent with descriptive trends observed earlier.


The principal findings from this study show that most ambulatory clinics had an ONC-certified EHR in 2016. However, specific functionalities designed to improve care quality vary in their rate of adoption, and considerable gaps remain in the full adoption of any specific functionality as well as in the adoption of all functionalities. Although the proportion of ambulatory clinics having all 16 functionalities increased by 10 percentage points over the 3-year period, almost two-thirds of clinics still did not have the complete set. This is consistent with results of other recent studies16,17 and may partly explain the mixed findings on the impact of the EHR adoption found in previous research.7,26 Our study findings suggest substantial opportunities for further improving ambulatory health IT among current EHR users.

In particular, patient-facing functionalities such as patient portals and patient health records were about 10 to 20 percentage points less likely to be adopted than physician-based functions such as clinical data repositories and clinician charting. Patient health records and patient portals are considered core elements of health IT that can potentially improve patient-centered care through better engagement and patient experience.27 Because ambulatory providers are often the principal source of care for most patients, the use of these functionalities in such settings is especially important, and hence their lower adoption rate is of concern. Another area of concern is health information exchange (HIE), which is viewed as a crucial tool for improving coordination of care and efficiency.28,29 Consistent with findings from other studies,17 our results show that by 2016, close to half of the clinics were still not able to exchange clinical information with hospitals or other clinics.

Equally important, we find substantial disparities in the trends of adoption across health systems of different sizes. Compared with large systems, smaller health systems had a 2-, 3-, or even 4-fold lower rate of adopting important functionalities such as CPOE, clinical decision support, and HIE. Moreover, the gap in adoption rates is increasing, with large systems adopting new functionalities faster than small systems. For example, whereas the majority of the large systems had HIE capabilities in 2016, about 80% of systems with 10 or fewer clinics were still without HIE. As health IT is de facto becoming an indispensable component of care delivery, the gaps in adoption may translate into differences in quality performance.17,28 The observed disparity in adoption may be due to economy of scale as well as differential resource barriers30 (eg, up-front cost, staff training, prior IT experiences) faced by systems of different sizes.

The results from the regression analysis indicated that a number of clinic and system characteristics were important predictors of ambulatory health IT adoption, especially the number of clinics within a system and the number of physicians within a clinic, consistent with the descriptive patterns. We also found that, possibly because of their inclination to reduce short-term investment cost, for-profit health systems were much less likely (17-20 percentage points less) to adopt ambulatory IT.


Our study has several limitations. First, we used self-reported survey data, and previous research has demonstrated that health IT surveys in general may be subject to measurement errors.31 However, in terms of major functionalities, results of a recent study showed reasonable agreement between HIMSS ambulatory data and other sources of information.32 Second, the information in the data did not fully capture all potential attributes of health systems that may be related to adoption of health IT. For example, our analysis did not include organizational factors such as governance structures and managerial processes, which can be critical to the implementation of technology.33 Analyzing such factors may provide important insights for better understanding the barriers to the adoption and use of health IT. Third, the responses in the HIMSS data on health IT functionalities are mostly binary (yes/no) and do not contain more granular information about capability and use. Nevertheless, the HIMSS survey does include a broad set of functionalities with some specificity (eg, clinical decision support used for basic medication screening). Finally, because our data did not include independent ambulatory clinics or physician groups unaffiliated with health systems, our findings are not representative of the entire universe of ambulatory providers. Based on the much lower adoption rates among small health systems, we expect independent providers outside health systems to fare even worse.34


Small health systems lagged in their adoption of ambulatory health IT. Patient-facing functionalities were less adopted than physician-oriented functionalities. Such gaps in adoption may lead to differences in performances across health systems. The fact that health IT capabilities are low among ambulatory clinics supports the ONC’s continuing focus on interoperability. As the healthcare market continues to consolidate, the number of small systems may decrease, but they probably will still play important roles in care delivery, highlighting the ongoing need to develop strategies to support health IT adoption by smaller providers of care. Additionally, efforts to promote the adoption of HIE and patient-facing functions such as patient portals are needed.


The authors would like to thank Donald Miller and Tracy Johnson at The Pennsylvania State University for their research assistance.Author Affiliations: The Pennsylvania State University (YS, AA-R, DS), University Park, PA; RAND Corporation, Boston, MA (RSR, SHF), and Santa Monica, CA (CLD); Southern California Evidence-based Practice Center, RAND Corporation (PS), Santa Monica, CA.

Source of Funding: This work was supported through the RAND Center of Excellence on Health System Performance, which is funded through a cooperative agreement (1U19HS024067-01) between the RAND Corporation and the Agency for Healthcare Research and Quality.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (YS, RSR, SHF, PS); acquisition of data (YS, DS, CLD); analysis and interpretation of data (YS, AA-R, RSR, SHF, CLD); drafting of the manuscript (YS, AA-R, CLD); critical revision of the manuscript for important intellectual content (YS, AA-R, RSR, SHF, PS, DS, CLD); statistical analysis (YS); obtaining funding (RSR, DS, CLD); administrative, technical, or logistic support (CLD); and supervision (YS, PS, CLD).

Address Correspondence to: Yunfeng Shi, PhD, The Pennsylvania State University, 504E Ford Bldg, University Park, PA 16802. Email:

1. Blumenthal D. Launching HITECH. N Engl J Med. 2010;362(5):382-385. doi: 10.1056/NEJMp0912825.

2. Reisman M. The Affordable Care Act, five years later: policies, progress, and politics. P T. 2015;40(9):575-600.

3. CMS, HHS. Medicare and Medicaid programs; electronic health record incentive program—stage 3 and modifications to Meaningful Use in 2015 through 2017: final rules with comment period. Fed Regist. 2015;80(200):62761-62955.

4. Henry J, Pylypchuk Y, Searcy T, Patel V. Adoption of electronic health record systems among U.S. non-federal acute care hospitals: 2008-2015. Office of the National Coordinator for Health Information Technology website. Published May 2016. Accessed June 12, 2019.

5. Jones SS, Rudin RS, Perry T, Shekelle PG. Health information technology: an updated systematic review with a focus on meaningful use. Ann Intern Med. 2014;160(1):48-54. doi: 10.7326/M13-1531.

6. Downing NL, Bates DW, Longhurst CA. Physician burnout in the electronic health record era: are we ignoring the real cause? Ann Intern Med. 2018;169(1):50-51. doi: 10.7326/M18-0139.

7. Spetz J, Burgess JF, Phibbs CS. The effect of health information technology implementation in Veterans Health Administration hospitals on patient outcomes. Healthc (Amst). 2014;2(1):40-47. doi: 10.1016/j.hjdsi.2013.12.009.

8. McCullough JS, Parente ST, Town R. Health information technology and patient outcomes: the role of information and labor coordination. RAND J Econ. 2016;47(1):207-236. doi: 10.1111/1756-2171.12124.

9. Miller A. Health information technology in ambulatory care settings: effects on quality and disparities — final report. Agency for Healthcare Research and Quality website. Published 2015. Accessed May 5, 2019.

10. Hwang W, Chang J, Laclair M, Paz H. Effects of integrated delivery system on cost and quality. Am J Manag Care. 2013;19(5):e175-e184.

11. Newhouse RP, Mills ME, Johantgen M, Pronovost PJ. Is there a relationship between service integration and differentiation and patient outcomes? Int J Integr Care. 2003;3:e15. doi: 10.5334/ijic.91.

12. Weeks WB, Gottlieb DJ, Nyweide DE, et al. Higher health care quality and bigger savings found at large multispecialty medical groups [erratum in Health Aff (Millwood). 2010;29(6):1275. doi: 10.1377/hlthaff.2010.0519]. Health Aff (Millwood). 2010;29(5):991-997. doi: 10.1377/hlthaff.2009.0388.

13. Koch TG, Wendling BW, Wilson NE. The effects of physician and hospital integration on Medicare beneficiaries’ health outcomes. Federal Trade Commission website. Published July 2018. Accessed August 3, 2019.

14. Hüsers J, Hübner U, Esdar M, et al. Innovative power of health care organisations affects IT adoption: a bi-national health IT benchmark comparing Austria and Germany. J Med Syst. 2017;41(2):33. doi: 10.1007/s10916-016-0671-6.

15. Xierali IM, Hsiao CJ, Puffer JC, et al. The rise of electronic health record adoption among family physicians. Ann Fam Med. 2013;11(1):14-19. doi: 10.1370/afm.1461.

16. DesRoches C. Progress and challenges in electronic health record adoption: findings from a national survey of physicians. Ann Intern Med. 2015;162(5):396. doi: 10.7326/L15-5060.

17. Rittenhouse DR, Ramsay PP, Casalino LP, McClellan S, Kandel ZK, Shortell SM. Increased health information technology adoption and use among small primary care physician practices over time: a national cohort study. Ann Fam Med. 2017;15(1):56-62. doi: 10.1370/afm.1992.

18. Rumball-Smith J, Shekelle P, Damberg CL. Electronic health record “super-users” and “under-users” in ambulatory care practices. Am J Manag Care. 2018;24(1):26-31.

19. Miller AR, Tucker CE. Can health care information technology save babies? J Polit Econ. 2011;119(2):289-324. doi: 10.1086/660083.

20. The state of health information technology in California. California Health Care Foundation website. Published November 2011. Accessed July 20, 2019.

21. Gresenz CR, Laughery SP, Miller A, Tucker CE. Health IT and ambulatory care quality. RAND Corporation website. Published 2015. Accessed August 11, 2019.

22. Higgins TC, Crosson J, Peikes D, McNellis R, Genevro J, Meyers D. Using health information technology to support quality improvement in primary care. Agency for Healthcare Research and Quality website. Published March 2015. Accessed July 25, 2019.

23. Institute of Medicine Committee on Data Standards for Patient Safety. Key Capabilities of an Electronic Health Record System: Letter Report. Washington, DC: National Academies Press; 2003.

24. Connecting health and care for the nation: a shared nationwide interoperability roadmap. Office of the National Coordinator for Health Information Technology website. Published 2015. Accessed July 25, 2019.

25. Fleming NS, Culler SD, McCorkle R, Becker ER, Ballard DJ. The financial and nonfinancial costs of implementing electronic health records in primary care practices. Health Aff (Millwood). 2011;30(3):481-489. doi: 10.1377/hlthaff.2010.0768.

26. Holroyd-Leduc JM, Lorenzetti D, Straus SE, Sykes L, Quan H. The impact of the electronic medical record on structure, process, and outcomes within primary care: a systematic review of the evidence. J Am Med Inform Assoc. 2011;18(6):732-737. doi: 10.1136/amiajnl-2010-000019.

27. Walker DM, Sieck CJ, Menser T, Huerta TR, Scheck McAlearney A. Information technology to support patient engagement: where do we stand and where can we go? J Am Med Inform Assoc. 2017;24(6):1088-1094. doi: 10.1093/jamia/ocx043.

28. Esmaeilzadeh P, Sambasivan M. Patients’ support for health information exchange: a literature review and classification of key factors. BMC Med Inform Decis Mak. 2017;17(1):33. doi: 10.1186/s12911-017-0436-2.

29. Menachemi N, Rahurkar S, Harle CA, Vest JR. The benefits of health information exchange: an updated systematic review. J Am Med Inform Assoc. 2018;25(9):1259-1265. doi: 10.1093/jamia/ocy035.

30. Palabindala V, Pamarthy A, Jonnalagadda NR. Adoption of electronic health records and barriers. J Community Hosp Intern Med Perspect. 2016;6(5):32643. doi: 10.3402/jchimp.v6.32643.

31. Diana ML, Kazley AS, Menachemi N. An assessment of Health Care Information and Management Systems Society and Leapfrog data on computerized provider order entry. Health Serv Res. 2011;46(5):1575-1591. doi: 10.1111/j.1475-6773.2011.01259.x.

32. Rudin RS, Shi Y, Fischer SH, et al. Level of agreement on health information adoption and use in survey data: a mixed-methods analysis of ambulatory clinics in 1 US state. JAMIA Open. 2019;2(2):231-237. doi: 10.1093/jamiaopen/ooz004.

33. Avgar AC, Litwin AS, Pronovost PJ. Drivers and barriers in health IT adoption: a proposed framework. Appl Clin Inform. 2012;3(4):488-500. doi: 10.4338/ACI-2012-07-R-0029.

34. Felt-Lisk S, Johnson L, Fleming C, Shapiro R, Natzke B. Toward understanding EHR use in small physician practices. Health Care Financ Rev. 2010;31(1):11-22.