Physicians complain about the time costs and other effects of electronic medical records (EMRs).1–3 In a small survey,4 family practice physicians reported an EMR-associated loss of 48 minutes of free time per clinic day (P < .05). We collaborated with the American College of Physicians (ACP) to revise the instrument from this study and surveyed the ACP’s national sample of internists to determine the extent of this problem.
Methods |
The ACP invites 1% of its members, including internal medicine attending physicians and trainees (resident and fellows), into its research panel,5 narrows the candidates by random sampling to ensure balance, and then adds nonmember internists. On December 12, 2012, the ACP mailed a 19-question survey to its panelists (900 ACP member and 102 nonmember internists at that time) who provided ambulatory care, and left it in the field for 10 days. The survey (Q11-Q12) focused on free time to get a sense of the EMR’s overall effect on internist’s time budget, including nonclinic time (Table 1). The National Institutes of Health’s Office of Human Research Subjects Protections declared the protocol “not human subjects research.” We used R statistical package for all statistical analyses, the Wilcoxon 2-sample test and the χ2 test to assess univariate associations, and the Wilcoxon 1-sample test to assess time change within responses.
Table 1.
Survey Instrument–Selecteda Questions and Responses
| Survey Questions (n = 411), Responsesb | No. (%) |
|---|---|
| Q1. Which of the following best describes your ambulatory care practice? c | |
| Independent solo practice | 35 (8.5) |
| Independent 2-physician practice | 15 (3.6) |
| Independent group of 3-5 physicians | 24 (5.8) |
| Independent group of 6-20 physicians | 33 (8.0) |
| Independent group of 21-50 physicians | 14 (3.4) |
| Independent group of ≥51 physicians | 43 (10.5) |
| HMO | 8 (1.9) |
| Hospital-based practice | 71 (17.3) |
| University- or academic medical center–based practice | 122 (29.7) |
| Community health center–based practice | 31 (7.5) |
| VA hospitals | 8 (1.9) |
| Other (7 write-ins) | 7 (1.7) |
| Do not provide care for any ambulatory patients (Terminate survey if selected.)c | 0 |
| Q2. Does your practice currently use an EMR that includes orders, visit notes, and other clinical information? | |
| Yes | 411 (100) |
| No (Terminate survey if selected.)c | 0 |
| Q4. Who is the vendor of your practice’s current EMR/EHR system? If you have to use different EMRs at different practice sites, choose all that apply. (Respondents chose from list of 15 EMR system vendors and included 46 write-ins.) | |
| Q5. What clinical functions do you regularly perform on the EMR computer? (Choose all that apply.) | |
| Prescription writing | 383 (93.2) |
| Test ordering | 351 (85.4) |
| Visit note writing | 381 (92.7) |
| Responding to EMR inbox messages and requests (eg, approving refill requests; reviewing and reacting to test results messages; responding to patient and miscellaneous clinical and administrative requests) | 344 (83.7) |
| Q7. How long have you had any EMR/EHR system in this practice? | |
| <1 year | 72 (17.5) |
| 1 to <2 y | 66 (16.1) |
| 2 to <3 y | 67 (16.3) |
| 3 to <5 y | 73 (17.8) |
| ≥5 y | 133 (32.4) |
| Q11. What impact, if any, has using the EMR/EHR had on your free time? | |
| I have much more free time (eg, I get home much earlier and/or have much less patient chart– and practice-related work at home) (Ask Q12) | 9 (2.2) |
| I have somewhat more free time (eg, I get home somewhat earlier and/or have somewhat less patient chart– and practice-related work at home.) (Ask Q12) | 52 (12.7) |
| It’s had no impact on my free time. (Assume zero minutes; skip to Q14.) | 106 (25.8) |
| I have somewhat less free time (eg, I get home somewhat later and/or have somewhat more patient chart– and practice-related work at home). (Skip to Q13) | 153 (37.2) |
| I have much less free time (eg, I get home much later and/or have much more patient chart– and practice-related work at home). (Skip to Q13.) | 91 (22.1) |
| Q12. On average how many minutes of free time have you gained per clinic day? (n = 61), mean (SD) | 39.0 (21.4) |
| Q13. On average, how many minutes of free time have you lost per clinic day? (n = 244), mean (SD) | −77.5 (40.9) |
| Q15. Which tasks, if any, take more of your overall time to perform using the EMR than they did before you used the computer? (Consider both the time to complete each task as well as any increase/decrease in the number of tasks. Choose all that apply.) (n = 410) d | |
| Nothing takes longer | 42 (10.2) |
| Managing calls and messages, refill requests, and new test results (With the EMR, these are usually in-box functions.) | 129 (31.5) |
| Writing visit notes | 262 (63.9) |
| Ordering and scheduling tests | 166 (40.5) |
| Finding and reviewing medical record data | 143 (34.9) |
| Writing prescriptions | 101 (24.6) |
| Reading the clinical notes that other clinicians enter into your office EMR | 132 (32.2) |
| Other (28 write-ins [which included 7 unique responses]) | 28 (6.8) |
Abbreviations: EHR, electronic health records; EMR, electronic medical records; HMO, health maintenance organization.
These are the 8 questions related to the key outcome variables, labeled with their original question numbers.
For each variable, n = 411 and data are presented as number (percentage) except where stated otherwise.
We excluded the 6 respondents who provided no ambulatory care (Q1), the 62 who did not use an EMR (Q2), and the respondents who did not answer Q2 from the totals in this table.
One responder failed to answer Q15.
Results |
Of 845 invitees, 485 opened the e-mail (a 62.5% contact rate). We removed 69 who reported no EMR use or no ambulatory practice (a 53.6% response rate [416 of 776]), and 5 who did not answer the main outcome question, to yield 411 respondents for analysis. While most (3 of 4) were attending physicians, trainees were more likely to respond. Except for the proportion of ACP members (P < .001), respondents and nonrespondents did not differ overall across 8 measured characteristics (Table 2).
Table 2.
Comparison of Respondents to Nonrespondents
| Variablea | No. (%) | P Value for Test of Difference | ||
|---|---|---|---|---|
| Nonrespondents | Responders | Total | ||
| No. (%) | 360 (46.4) | 416 (53.6) | 776 (100) | |
| Age (total n = 692)b | ||||
| Mean (SD) | 47.6 (14.2) | 45.9 (13.1) | 46.6 (13.6) | .22 |
| Median (IQR) | 47 (36.75–58) | 46 (34–57) | 47 (35–57) | NA |
| Sex (total n = 662)b | .63 | |||
| Male | 200 (69.7) | 269 (71.7) | 469 (70.8) | NA |
| Female | 87 (30.3) | 106 (28.3) | 193 (29.2) | NA |
| Specialty | .99 | |||
| General internist | 212 (58.9) | 244 (58.7) | 456 (58.8) | .99 |
| Subspecialist internist | 122 (33.9) | 141 (33.9) | 263 (33.9) | .99 |
| Other | 26 (7.2) | 31 (7.5) | 57 (7.3) | .99 |
| Main professional activity | .14 | |||
| Trainee (residency/fellowship) | 77 (21.4) | 109 (26.2) | 186 (24) | .14 |
| Attending (primary activity = practice) | 256 (71.1) | 268 (64.4) | 524 (67.5) | .06 |
| Attending (primary activity = not practice)c | 27 (7.5) | 39 (9.4) | 66 (8.5) | .42 |
| Primary activity = research | 10 (2.8) | 7 (1.7) | 17 (2.2) | .43 |
| Primary activity = teaching | 6 (1.7) | 16 (3.8) | 22 (2.8) | .11 |
| Primary activity = administration | 9 (2.5) | 13 (3.1) | 22 (2.8) | .76 |
| Primary activity = other activity | 2 (0.6) | 3 (0.7) | 5 (0.6) | .99 |
| Time in patient care, % | .92 | |||
| <25 | 14 (3.9) | 13 (3.1) | 27 (3.5) | .70 |
| 25–49 | 16 (4.4) | 21 (5) | 37 (4.8) | .82 |
| 50–74 | 52 (14.4) | 58 (13.9) | 110 (14.2) | .92 |
| ≥75 | 278 (77.2) | 324 (77.9) | 602 (77.6) | .89 |
| Location of patient care | .09 | |||
| All outpatient | 85 (23.6) | 104 (25) | 189 (24.4) | .72 |
| Primarily outpatient with some inpatient | 163 (45.3) | 157 (37.7) | 320 (41.2) | .04 |
| Primarily inpatient with some outpatient | 112 (31.1) | 155 (37.3) | 267 (34.4) | .09 |
| ACP member status | <.001 | |||
| Member | 305 (84.7) | 388 (93.3) | 693 (89.3) | NA |
| Not a member | 55 (15.3) | 28 (6.7) | 83 (10.7) | NA |
| Associate flagd (total n = 693)b | .28 | |||
| Not an associate | 229 (75.1) | 276 (71.1) | 505 (72.9) | NA |
| Associate | 76 (24.9) | 112 (28.9) | 188 (27.1) | NA |
Abbreviations: ACP, American College of Physicians; IQR, interquartile range; NA, not applicable.
Data are presented as number (percentage) based on the numbers listed in the column header unless otherwise specified.
The 83 non-ACP members did not have information on age, sex, and associate status account for most of the missing. The numbers missing for each factor for nonrespondents and respondents, respectively, were 56 and 28 for age, 73 and 41 for sex, and 55 and 28 for associate flag.
Represents the total of the subcategories that follow.
Trainees are called associates in the ACP database.
The respondents used 61 distinct EMR systems (Q4) and came from a broad range of practice types (Q1). Nine EMRs were used by 20 or more respondents, accounted for 324 (78.8%) of all users, and users of each of these lost free time (P < .05 for all). Of these 9, the Veterans Affairs’ Computer Patient Record System (CPRS) was associated with the least free time loss (−20 minutes) (P = .04). Respondents were experienced EMR users. Most (70.6%) used all of the EMR functions listed in Q4, and 82.5% had EMRs in their practice for more than a year (Q7).
Among all respondents, 89.8% reported that at least 1 data management function was slower post-EMR adoption, and 63.9% reported that note writing took longer. Surprisingly, a third (33.9%) reported that it took longer to find and review medical record data with the EMR than without, and a similar proportion, 32.2%, that it was slower to read other clinicians’ notes.
The mean time loss for attending physicians was −48 minutes per clinic day (P < .001), or 4 hours per 5-day clinic week. The mean loss for trainees was −18 minutes per day, less than that of attending physicians (P < .001). For the 59.4% of all respondents who did lose time, the mean loss was −78 minutes per clinic day, or 6.5 hours per 5-day clinic week.
Discussion |
The loss of free time that our respondents reported was large and pervasive and could decrease access or increase costs of care. Policy makers should consider these time costs in future EMR mandates. Ambulatory practices may benefit from approaches used by high-performing practices6—the use of scribes, standing orders, talking instead of e-mail—to recapture time lost on EMR. We can only speculate as to whether better computer skills, shorter (half-day) clinic assignments with proportionately less exposure to EMR time costs, or other factors account for the trainees’ smaller per-day time loss.
Funding/Support:
This research was supported in part by the Intramural Research Program of the National Institutes of Health and National Library of Medicine (NIH-NLM), and the American College of Physicians (ACP).
Role of the Sponsor:
The NIH-NLM and the ACP had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer:
The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the ACP, NIH-NLM, or the US Department of Health and Human Services.
Footnotes
Conflict of Interest Disclosures: None reported.
References
- 1.HIMSS13 Blog. EHR satisfaction diminishing. American EHR blog. 2013. http://www.americanehr.com/blog/2013/03/himss13-ehr-satisfaction-diminishing/. Accessed April 18, 2014. [Google Scholar]
- 2.Vogel L. Cut-and-paste clinical notes confuse care, say US internists. CMAJ. 2013;185(18):826. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Weir CR, Hammond KW, Embi PJ, Efthimiadis EN, Thielke SM, Hedeen AN. An exploration of the impact of computerized patient documentation on clinical collaboration. Int J Med Inform. 2011;80(8):e62–e71. [DOI] [PubMed] [Google Scholar]
- 4.McDonald CJ, McDonald MH. Electronic medical records and preserving primary care physicians’ time: comment on “electronic health record-based messages to primary care providers.” Arch Intern Med. 2012;172(3):285–287. [DOI] [PubMed] [Google Scholar]
- 5.Internal Medicine Insider. American College of Physicians, 2014. https://iminsider.org/faq.php?SES=6dbcf1ab3b92904c7a1cbec86d049d14. Accessed February 19, 2014.
- 6.Sinsky CA, Willard-Grace R, Schutzbank AM, Sinsky TA, Margolius D, Bodenheimer T. In search of joy in practice: a report of 23 high-functioning primary care practices. Ann Fam Med. 2013;11(3):272–278. [DOI] [PMC free article] [PubMed] [Google Scholar]