[Skip to Content]
[Skip to Content Landing]
Figure 1.
Mean Electronic Health Record (EHR) Time per Patient Encounter vs Median Half-Day Clinic Volume by Mean Billing Level
Mean Electronic Health Record (EHR) Time per Patient Encounter vs Median Half-Day Clinic Volume by Mean Billing Level

Mean EHR use time per encounter was 10.8 minutes (dotted horizontal line) (range, 5.8-28.6 minutes). Linear regression fits and 95% CIs are displayed. Low billing level was defined as level 1-2 evaluation and management office visits, preoperative and postoperative encounters, brief or intermediate cosmetic evaluations, vision examinations, or special procedures. Medium billing level was defined as level 3 office visits, comprehensive cosmetic evaluations, intermediate or established comprehensive eye examinations, or refractive surgery consultations. High billing level was defined as level 4-5 office visits, eye examinations that included treatment, or new comprehensive eye examinations.

Figure 2.
Mean Electronic Health Record (EHR) Time per Patient Encounter vs Clinic Volume by Mean Billing Level
Mean Electronic Health Record (EHR) Time per Patient Encounter vs Clinic Volume by Mean Billing Level

A, Low billing level, defined as level 1-2 evaluation and management office visits, preoperative and postoperative encounters, brief or intermediate cosmetic evaluations, vision examinations, or special procedures. B, Medium billing level, defined as level 3 office visits, comprehensive cosmetic evaluations, intermediate or established comprehensive eye examinations, or refractive surgery consultations. C, High billing level, defined as level 4-5 office visits, eye examinations that included treatment, or new comprehensive eye examinations. Linear regression fits and 95% CIs are displayed.

Table 1.  
Characteristics of 27 Stable Faculty Ophthalmologists and Their Encounters Using the EHR System in 2014a
Characteristics of 27 Stable Faculty Ophthalmologists and Their Encounters Using the EHR System in 2014a
Table 2.  
Time Requirements by Ophthalmologists During Patient Encounters for Activities During Examinationa
Time Requirements by Ophthalmologists During Patient Encounters for Activities During Examinationa
Table 3.  
Time Requirements for EHR Use per Encounter and per Median Half-Day Clinica
Time Requirements for EHR Use per Encounter and per Median Half-Day Clinica
1.
Committee on Improving the Patient Record, Division of Health Care Services, Institute of Medicine.  The Computer-Based Patient Record: An Essential Technology for Health Care. Revised edition. Washington, DC: National Academies Press; 1997.
2.
Blumenthal  D.  Implementation of the federal health information technology initiative.  N Engl J Med. 2011;365(25):2426-2431.PubMedGoogle ScholarCrossref
3.
Jamoom  EW, Yang  N, Hing  E.  Adoption of certified electronic health record systems and electronic information sharing in physician offices: United States, 2013 and 2014.  NCHS Data Brief. 2016; (236):1-8.PubMedGoogle Scholar
4.
Henry  J, Pylpchuck  Y, Searcy  T, Patel  V.  Adoption of Electronic Health Record Systems Among US Non-Federal Acute Care Hospitals: 2008-2015. Washington, DC: Office of the National Coordinator for Health Information Technology; 2016. ONC Data Brief, No. 35.
5.
Verdon  DR. (Slideshow) Medical Economics EHR survey probes physician angst about adoption, use of technology. http://medicaleconomics.modernmedicine.com/medical-economics/content/tags/ehr/slideshow-medical-economics-ehr-survey-probes-physician-angst-abo. Published February 10, 2014. Accessed May 17, 2016.
6.
Carayon  P, Wetterneck  TB, Alyousef  B,  et al.  Impact of electronic health record technology on the work and workflow of physicians in the intensive care unit.  Int J Med Inform. 2015;84(8):578-594.PubMedGoogle ScholarCrossref
7.
Wormer  BA, Colavita  PD, Yokeley  WT,  et al.  Impact of implementing an electronic health record on surgical resident work flow, duty hours, and operative experience.  Am Surg. 2015;81(2):172-177.PubMedGoogle Scholar
8.
Chan  P, Thyparampil  PJ, Chiang  MF.  Accuracy and speed of electronic health record versus paper-based ophthalmic documentation strategies.  Am J Ophthalmol. 2013;156(1):165-172.e2.PubMedGoogle ScholarCrossref
9.
Poissant  L, Pereira  J, Tamblyn  R, Kawasumi  Y.  The impact of electronic health records on time efficiency of physicians and nurses: a systematic review.  J Am Med Inform Assoc. 2005;12(5):505-516.PubMedGoogle ScholarCrossref
10.
VanDenKerkhof  EG, Goldstein  DH, Lane  J, Rimmer  MJ, Van Dijk  JP.  Using a personal digital assistant enhances gathering of patient data on an acute pain management service: a pilot study.  Can J Anaesth. 2003;50(4):368-375.PubMedGoogle ScholarCrossref
11.
Asan  O, D Smith  P, Montague  E.  More screen time, less face time—implications for EHR design.  J Eval Clin Pract. 2014;20(6):896-901.PubMedGoogle ScholarCrossref
12.
Toll  E.  A piece of my mind: the cost of technology.  JAMA. 2012;307(23):2497-2498. PubMedGoogle ScholarCrossref
13.
Blumenthal  D, Collins  SR.  Health care coverage under the Affordable Care Act—a progress report.  N Engl J Med. 2014;371(3):275-281.PubMedGoogle ScholarCrossref
14.
Litvak  E, Bisognano  M.  More patients, less payment: increasing hospital efficiency in the aftermath of health reform.  Health Aff (Millwood). 2011;30(1):76-80.PubMedGoogle ScholarCrossref
15.
Koppel  R, Lehmann  CU.  Implications of an emerging EHR monoculture for hospitals and healthcare systems.  J Am Med Inform Assoc. 2015;22(2):465-471.PubMedGoogle Scholar
16.
R Core Team.  R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2014.
17.
Bates  D, Maechler  M, Bolker  B, Walker  S.  Fitting linear mixed-effects models using lme4.  J Stat Softw. 2015;67:1-48. doi:10.18637/jss.v067.i01Google ScholarCrossref
18.
Hothorn  T, Bretz  F, Westfall  P.  Simultaneous inference in general parametric models.  Biom J. 2008;50(3):346-363.PubMedGoogle ScholarCrossref
19.
Pandit  RR, Boland  MV.  The impact of an electronic health record transition on a glaucoma subspecialty practice.  Ophthalmology. 2013;120(4):753-760.PubMedGoogle ScholarCrossref
20.
Lo  HG, Newmark  LP, Yoon  C,  et al.  Electronic health records in specialty care: a time-motion study.  J Am Med Inform Assoc. 2007;14(5):609-615.PubMedGoogle ScholarCrossref
21.
Linzer  M, Manwell  LB, Williams  ES,  et al; MEMO (Minimizing Error, Maximizing Outcome) Investigators.  Working conditions in primary care: physician reactions and care quality.  Ann Intern Med. 2009;151(1):28-36, W6-W9. PubMedGoogle ScholarCrossref
22.
Shaw  MK, Davis  SA, Fleischer  AB, Feldman  SR.  The duration of office visits in the United States, 1993 to 2010.  Am J Manag Care. 2014;20(10):820-826.PubMedGoogle Scholar
23.
Centers for Disease Control and Prevention.  The State of Aging and Health in America 2013. Atlanta, GA: Centers for Disease Control & Prevention, US Dept of Health and Human Services; 2013.
24.
Long  C, Tsay  EL, Jacobo  SA, Popat  R, Singh  K, Chang  RT.  Factors associated with patient Press Ganey satisfaction scores for ophthalmology patients.  Ophthalmology. 2016;123(2):242-247.PubMedGoogle ScholarCrossref
25.
Blumenthal  D, Causino  N, Chang  YC,  et al.  The duration of ambulatory visits to physicians.  J Fam Pract. 1999;48(4):264-271.PubMedGoogle Scholar
26.
Tsiga  E, Panagopoulou  E, Sevdalis  N, Montgomery  A, Benos  A.  The influence of time pressure on adherence to guidelines in primary care: an experimental study.  BMJ Open. 2013;3(4).PubMedGoogle Scholar
27.
Kumarapeli  P, de Lusignan  S.  Using the computer in the clinical consultation; setting the stage, reviewing, recording, and taking actions: multi-channel video study.  J Am Med Inform Assoc. 2013;20(e1):e67-e75.PubMedGoogle ScholarCrossref
28.
Tai-Seale  M, Olson  CW, Li  J,  et al.  Electronic health record logs indicate that physicians split time evenly between seeing patients and desktop medicine.  Health Aff (Millwood). 2017;36(4):655-662.PubMedGoogle ScholarCrossref
29.
Linzer  M, Poplau  S, Babbott  S,  et al.  Worklife and wellness in academic general internal medicine: results from a national survey.  J Gen Intern Med. 2016;31(9):1004-1010.PubMedGoogle ScholarCrossref
30.
Makam  AN, Lanham  HJ, Batchelor  K,  et al.  Use and satisfaction with key functions of a common commercial electronic health record: a survey of primary care providers.  BMC Med Inform Decis Mak. 2013;13:86.PubMedGoogle ScholarCrossref
31.
Sanders  DS, Read-Brown  S, Tu  DC,  et al.  Impact of an electronic health record operating room management system in ophthalmology on documentation time, surgical volume, and staffing.  JAMA Ophthalmol. 2014;132(5):586-592. PubMedGoogle ScholarCrossref
32.
Verghese  A.  Culture shock—patient as icon, icon as patient.  N Engl J Med. 2008;359(26):2748-2751.PubMedGoogle ScholarCrossref
33.
Alkureishi  MA, Lee  WW, Lyons  M,  et al.  Impact of electronic medical record use on the patient-doctor relationship and communication: a systematic review.  J Gen Intern Med. 2016;31(5):548-560.PubMedGoogle ScholarCrossref
34.
Hribar  MR, Read-Brown  S, Reznick  L,  et al.  Secondary use of EHR timestamp data: validation and application for workflow optimization.  AMIA Annu Symp Proc. 2015;2015:1909-1917. PubMedGoogle Scholar
35.
Raghupathi  W, Raghupathi  V.  Big data analytics in healthcare: promise and potential.  Health Inf Sci Syst. 2014;2:3. doi:10.1186/2047-2501-2-3PubMedGoogle ScholarCrossref
36.
Hoover  R.  Benefits of using an electronic health record.  Nursing. 2016;46(7):21-22.PubMedGoogle ScholarCrossref
37.
Patel  VL, Arocha  JF, Kaufman  DR.  A primer on aspects of cognition for medical informatics.  J Am Med Inform Assoc. 2001;8(4):324-343.PubMedGoogle ScholarCrossref
38.
Thornton  JD, Schold  JD, Venkateshaiah  L, Lander  B.  Prevalence of copied information by attendings and residents in critical care progress notes.  Crit Care Med. 2013;41(2):382-388.PubMedGoogle ScholarCrossref
39.
Zheng  K, Haftel  HM, Hirschl  RB, O’Reilly  M, Hanauer  DA.  Quantifying the impact of health IT implementations on clinical workflow: a new methodological perspective.  J Am Med Inform Assoc. 2010;17(4):454-461.PubMedGoogle ScholarCrossref
40.
Zhang  J, Chen  Y, Ashfaq  S,  et al.  Strategizing EHR use to achieve patient-centered care in exam rooms: a qualitative study on primary care providers.  J Am Med Inform Assoc. 2016;23(1):137-143.PubMedGoogle ScholarCrossref
41.
Shanafelt  TD, Hasan  O, Dyrbye  LN,  et al.  Changes in burnout and satisfaction with work-life balance in physicians and the general US working population between 2011 and 2014.  Mayo Clin Proc. 2015;90(12):1600-1613.PubMedGoogle ScholarCrossref
42.
Shanafelt  TD, Dyrbye  LN, Sinsky  C,  et al.  Relationship between clerical burden and characteristics of the electronic environment with physician burnout and professional satisfaction.  Mayo Clin Proc. 2016;91(7):836-848.PubMedGoogle ScholarCrossref
43.
Babbott  S, Manwell  LB, Brown  R,  et al.  Electronic medical records and physician stress in primary care: results from the MEMO Study.  J Am Med Inform Assoc. 2014;21(e1):e100-e106.PubMedGoogle ScholarCrossref
44.
Centers for Medicare & Medicaid Services, Dept of Health and Human Services. Evaluation and management services. https://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/MLNProducts/Downloads/eval-mgmt-serv-guide-ICN006764.pdf. Updated August 2017. Accessed September 12, 2017.
45.
Martin  SA, Sinsky  CA.  The map is not the territory: medical records and 21st century practice.  Lancet. 2016;388(10055):2053-2056.PubMedGoogle ScholarCrossref
46.
Zulman  DM, Shah  NH, Verghese  A.  Evolutionary pressures on the electronic health record: caring for complexity.  JAMA. 2016;316(9):923-924.PubMedGoogle ScholarCrossref
47.
Horsky  J, Ramelson  HZ.  Development of a cognitive framework of patient record summary review in the formative phase of user-centered design.  J Biomed Inform. 2016;64:147-157.PubMedGoogle ScholarCrossref
48.
Zahabi  M, Kaber  DB, Swangnetr  M.  Usability and safety in electronic medical records interface design: a review of recent literature and guideline formulation.  Hum Factors. 2015;57(5):805-834.PubMedGoogle ScholarCrossref
49.
Middleton  B, Bloomrosen  M, Dente  MA,  et al; American Medical Informatics Association.  Enhancing patient safety and quality of care by improving the usability of electronic health record systems: recommendations from AMIA.  J Am Med Inform Assoc. 2013;20(e1):e2-e8.PubMedGoogle ScholarCrossref
50.
American Medical Association. Improving care: priorities to improve electronic health record usability: executive summary. https://www.aace.com/files/ehr-priorities.pdf. Accessed Mar 23, 2017.
Original Investigation
November 2017

Time Requirements for Electronic Health Record Use in an Academic Ophthalmology Center

Author Affiliations
  • 1Department of Ophthalmology, Casey Eye Institute, Oregon Health & Science University, Portland
  • 2Department of Medical Informatics and Clinical Epidemiology, Oregon Health & Science University, Portland
JAMA Ophthalmol. 2017;135(11):1250-1257. doi:10.1001/jamaophthalmol.2017.4187
Key Points

Question  What are the time requirements for ophthalmologists’ use of electronic health records?

Findings  In this single-center cohort study of 27 ophthalmologists, mean total ophthalmologist examination time was 11.2 minutes per patient, of which 27% was spent on electronic health record use, 42% on conversation, and 31% on patient examination. Mean total ophthalmologist time spent using the electronic health record was 10.8 minutes per encounter, translating to 3.7 hours per day using the electronic health record (2.1 hours during patient examinations, and 1.6 hours outside the clinic session).

Meaning  Although simultaneous electronic health record and conversation or examination time were not determined, this study suggests that electronic health record use requires substantial time by ophthalmologists, with variability in electronic health record use patterns.

Abstract

Importance  Electronic health record (EHR) systems have transformed the practice of medicine. However, physicians have raised concerns that EHR time requirements have negatively affected their productivity. Meanwhile, evolving approaches toward physician reimbursement will require additional documentation to measure quality and cost of care. To date, little quantitative analysis has rigorously studied these topics.

Objective  To examine ophthalmologist time requirements for EHR use.

Design, Setting, and Participants  A single-center cohort study was conducted between September 1, 2013, and December 31, 2016, among 27 stable departmental ophthalmologists (defined as attending ophthalmologists who worked at the study institution for ≥6 months before and after the study period). Ophthalmologists who did not have a standard clinical practice or who did not use the EHR were excluded.

Exposures  Time stamps from the medical record and EHR audit log were analyzed to measure the length of time required by ophthalmologists for EHR use. Ophthalmologists underwent manual time-motion observation to measure the length of time spent directly with patients on the following 3 activities: EHR use, conversation, and examination.

Main Outcomes and Measures  The study outcomes were time spent by ophthalmologists directly with patients on EHR use, conversation, and examination as well as total time required by ophthalmologists for EHR use.

Results  Among the 27 ophthalmologists in this study (10 women and 17 men; mean [SD] age, 47.3 [10.7] years [median, 44; range, 34-73 years]) the mean (SD) total ophthalmologist examination time was 11.2 (6.3) minutes per patient, of which 3.0 (1.8) minutes (27% of the examination time) were spent on EHR use, 4.7 (4.2) minutes (42%) on conversation, and 3.5 (2.3) minutes (31%) on examination. Mean (SD) total ophthalmologist time spent using the EHR was 10.8 (5.0) minutes per encounter (range, 5.8-28.6 minutes). The typical ophthalmologist spent 3.7 hours using the EHR for a full day of clinic: 2.1 hours during examinations and 1.6 hours outside the clinic session. Linear mixed effects models showed a positive association between EHR use and billing level and a negative association between EHR use per encounter and clinic volume. Each additional encounter per clinic was associated with a decrease of 1.7 minutes (95% CI, -4.3 to 1.0) of EHR use time per encounter for ophthalmologists with high mean billing levels (adjusted R2 = 0.42; P = .01).

Conclusions and Relevance  Ophthalmologists have limited time with patients during office visits, and EHR use requires a substantial portion of that time. There is variability in EHR use patterns among ophthalmologists.

Introduction

Electronic health record (EHR) systems have transformed modern medical care.1 The transition from paper-based records to EHRs within hospitals and ambulatory practices has been aggressively promoted by federal initiatives.2 Among all office-based clinicians in the United States, 74% had a certified EHR in 2014.3 Adoption within nonfederal acute care hospitals is almost universal, with more than 95% possessing a certified EHR.4

However, physicians have raised concerns that EHR implementation has negatively affected their productivity.5 Published studies in the surgical, intensive care, and ophthalmology settings have reported mixed outcomes with respect to physician documentation times in the EHR: some report longer times, while others report shorter times.6-10 Additional studies have raised concerns that EHR use may negatively affect the physician-patient relationship.11,12

Meanwhile, these problems are exacerbated by the growing complexity and documentation associated with medical care. Physicians are pressured to see more patients in less time because of challenges involving accessibility and cost of care.13,14 Through new programs, such as the Medicare Access and CHIP (Children’s Health Insurance Program) Reauthorization Act of 2015, physician reimbursement is based increasingly on value-based purchasing models, which require additional documentation to measure quality and cost of care. These factors are increasing the documentation and time burden for physicians.

To our knowledge, few studies have rigorously examined the association of EHRs with ophthalmologists’ time requirements. The purpose of our study was to address this gap in knowledge by examining the time requirements of EHR use for ophthalmologists in an outpatient setting using time-motion and data analytic methods.

Methods
Study Institution

Oregon Health & Science University (OHSU) is an academic medical center in Portland, Oregon. The department of ophthalmology at OHSU Casey Eye Institute includes 57 attending faculty ophthalmologists who perform more than 115 000 annual outpatient examinations. In 2014, five of the top 10 physicians at OHSU with the highest annual outpatient clinical volume were ophthalmologists. This department provides primary eye care and serves as a tertiary referral center in the Pacific Northwest and nationally. In 2006, OHSU implemented an institution-wide EHR system (EpicCare; Epic Systems). This vendor is a market share leader among large hospitals.15 All ambulatory practice management, documentation, order entry, medication prescribing, operating room management, and billing tasks are performed using components of this EHR. This study was approved by the OHSU Institutional Review Board, including waiver of consent for retrospective data analytics study of outpatient encounters.

Study Design

This study involved the following 2 parts: a time-motion study and a data analytics study. Through manual observations, the time-motion study examined aspects of time requirements for EHR use compared with other activities during patient encounters. The data analytics study used EHR time stamps to perform larger-scale analysis of time requirements for EHR use during and after encounters.

Time-Motion Study

Five ophthalmologists from different subspecialties (comprehensive, cornea, glaucoma, pediatrics, and retina) were observed for a minimum of 5 half-day clinic sessions between September 1, 2013, and December 31, 2016. Observers used mobile devices and software (Numbers; Apple Inc) to record time-motion data of ophthalmologists starting and ending patient examinations and switching between 3 predefined activities with patients: EHR use, conversation, and examination. When EHR use or examination occurred concomitantly with conversation, the time was attributed to the nonconversation activity. Observers were trained by two of us (S.R.-B. and M.R.H.), performed parallel observation with us, and compared their observations with our observations to ensure that data collected were consistent. Time-motion data were processed to determine the duration and distribution of time spent in examination rooms with patients. The following observed encounters were excluded: encounters with incomplete timing data, additional encounters for the same patient (only the first encounter was included), and encounters that included EHR scribes (who were not widely used within the department at the time of the study).

Data Analytics Study: EHR Time Stamps

Because manual time-motion observation is cumbersome, we developed methods using EHR time stamps to broaden our analysis to all OHSU ophthalmologists. For our study, we identified “stable faculty ophthalmologists” as attending physicians who worked at OHSU for at least 6 months before and after the study period. This definition minimized bias from physicians with growing or shrinking practices. Ophthalmologists with fewer than 200 patient encounters in 2014 or who did not use the EHR were excluded. Physician demographic characteristics (sex, age, years in practice, and ophthalmic subspecialty) were gathered using publicly available data.

We collected time stamp data for patient encounters by querying audit logs of EHR system use along with data about patient encounters. For each ophthalmologist, we counted the minutes with time-stamped events in audit log records for every encounter, both during (between patient check-in and check-out) and after the patient encounter until it was completed. We validated this approach by comparing EHR time stamp data findings against reference standard data from manual time-motion observations during 363 patient encounters.

For analysis of factors affecting EHR time requirements from January 1, 2014, to December 31, 2014, we identified the median half-day clinic volume for ophthalmologists and the primary billing code for each encounter, which was used as a proxy for complexity. Based on input from the ophthalmology departmental billing manager, we categorized the billing level as low (level 1-2 evaluation and management office visits, short preoperative and postoperative encounters, brief or intermediate cosmetic evaluations, vision examinations, and special procedures), medium (level 3 office visits, comprehensive cosmetic evaluations, intermediate or established comprehensive eye examinations, and refractive surgery consultations), or high (level 4-5 office visits, eye examinations that include treatment, and new comprehensive eye examinations).

Statistical Analysis

We analyzed factors that affected ophthalmologists’ EHR use for outpatient encounters using 2 different models in R.16 First, analyzing all encounters, we used the lme4 function17 to perform a linear mixed effects analysis of total EHR interaction, including various fixed effects with the random effects of the ophthalmologist and the patient. We used a log normal transformation of EHR interaction time in the model. P values were calculated using the glht R function in the multcomp library.18 Second, aggregating data for ophthalmologists, we created a linear model with interactions using characteristics of each ophthalmologist along with his or her typical encounter characteristics. P < .05 (2-sided) was considered significant.

Results
Study Physicians

Based on our study criteria, 27 stable ophthalmologists were identified (and are defined) (Table 1) at OHSU in 2014. A total of 46 519 outpatient encounters were completed in 2014 by these ophthalmologists. Overall, 10 ophthalmologists (37.0%) were female, and the mean (SD) age was 47.3 (10.7) years (median, 44; range, 34-73 years). Ten subspecialties were represented, with the largest numbers in glaucoma and retina (4 ophthalmologists each). In 2014 the mean annual volume per ophthalmologist was 1872 (range, 235-4833) patient encounters.

Time-Motion: Manually Observed Time Requirements

Five ophthalmologists were observed for 32 half-day clinics and 444 patient encounters. Eighty-one (18.2%) observed encounters were excluded (23 [5.2%] because of additional visits and 36 [8.1%] because of missing data), leaving 363 encounters for analysis. Lengths of manually observed ophthalmologist-patient encounters ranged from 2 to 60 minutes. Table 2 summarizes the mean ophthalmologist time requirements per encounter for each of the in-examination activities (EHR use, conversation, and examination). Overall, ophthalmologists spent a mean (SD) of 11.2 (6.3) minutes per encounter (3.0 [1.8] minutes [27%] for EHR use, 4.7 [4.2] minutes [42%] for conversation, and 3.5 [2.3] minutes [31%] for examination).

Data Analytics: Validation of EHR Time Stamp Data

For each of the 363 observed encounters, we used time stamp methods to estimate EHR time use. The mean time spent using the EHR based on manually observed data was 3.0 minutes compared with 4.3 minutes for the same encounters based on EHR time stamp data. The mean difference using these 2 methods was 1.3 minutes (range, −5.7 to 9.5 minutes). Overall, time use measurements by EHR time stamps were within 3 minutes of the reference standard manually observed time-motion data in 278 of 363 encounters (76.6%).

EHR Use Analysis

We created the following 2 models of the data: one focusing on encounters and another on ophthalmologists. For both models, we considered various effects, but the only significant ones were encounter billing level and clinic volume.

Model 1 was a mixed linear model of all encounters, with random effects (encounter’s ophthalmologist and patient) combined with interacting fixed effects (encounter’s session volume and billing level). This model showed that increasing the billing level of an encounter was associated with an increase in EHR use in that encounter, but that the outcome was mitigated by clinic volume. For example, at the median clinic volume (14 encounters), EHR use time increased by 26% at medium billing levels compared with low billing levels, by 64% at high billing levels compared with low billing levels, and by 30% at high billing levels compared with medium billing levels (P < .001 for all comparisons). However, increases in clinic volume were associated with decreases in EHR use and were affected by billing level. For low–billing level encounters, clinic volume was associated with a minor influence of clinic volume on EHR time per encounter (0.01% per additional encounter; P < .001). Clinic volume for high–billing level encounters was associated with a greater influence (−0.64% per additional encounter; P < .001), while medium level encounters were associated with the greatest influence (−1.10% per additional encounter; P < .001).

Model 2 used data aggregated by ophthalmologist. Figure 1 displays the mean EHR use time per encounter for each ophthalmologist as a function of clinic volume. The mean (SD) total time spent by ophthalmologists for EHR use was 10.8 (5.0) minutes per encounter (range, 5.8-28.6 minutes per encounter). As shown in Figure 1 and Table 3, increases in ophthalmologists’ median half-day clinic volume were associated with decreases in EHR use time per encounter. Furthermore, Figure 2 displays the influence of mean billing level on mean EHR use time per encounter for each ophthalmologist. Overall, ophthalmologists with the highest half-day clinic volumes had the lowest mean billing levels, while ophthalmologists with high mean billing levels had the lowest half-day clinic volumes and their clinic volume was associated with the biggest influence on their EHR use. The influence of volume on EHR time ranged from near zero for ophthalmologists with lower mean billing levels (Figure 2A) to −1.7 minutes (95% CI, −4.3 to 1.0) per additional encounter for ophthalmologists with higher mean billing levels (Figure 2B) (adjusted R2 = 0.42; P = .01 for high mean billing level).

Extrapolating our findings from half-day clinics, study ophthalmologists spent a mean 3.7 hours using the EHR for a full-day clinic session: 2.1 hours during encounters, and 1.6 hours outside the clinic session. High-volume ophthalmologists spent the most time using the EHR outside of clinic sessions, with a mean 2.3 hours of EHR use per full-day clinic.

Discussion

The key findings from the manual time-motion observation and data analysis of EHR time stamps were that ophthalmologists have limited time with patients, time requirements associated with EHR use are significant, and ophthalmologists differ in their patterns of EHR use.

The first key finding is that ophthalmologists have limited time to spend with patients. Manual observation showed that ophthalmologists spent a mean of 11.2 minutes per encounter (Table 2). Although physicians report feeling pressured for time, studies show that the length of patient visits has increased in all primary care specialties except ophthalmology and psychiatry.19-22 Physicians face growing time pressures as care of chronic medical issues becomes more complex23 and as increasing emphasis is placed on reporting patient satisfaction and quality of care metrics.24 During patient visits, physicians are expected to address acute and chronic issues, provide preventive care, and form a relationship with the patient.25 These time pressures may have negative consequences for patients, as research suggests that physician performance suffers under time pressure.26

The second key finding is that EHR use requires significant time by ophthalmologists. In this study, EHR use accounted for 27% of ophthalmologist time during encounters, representing a mean of 10.8 minutes per encounter and 1.9 hours per half-day clinic session (Table 3). Although there were statistically significant differences between the time for EHR use and times for conversation and examination (Table 2), it is not clear if this is a clinically relevant difference. The magnitude and proportion of time spent using the EHR are consistent with previous research.7,19,27 The time requirements for a full day of clinic (2.1 hours during examinations and 1.6 hours outside the clinic session) are consistent with a similar study outside ophthalmology28 and represent a significant time burden. Finally, there were 2 ophthalmologists who appeared to be outliers for the time spent per encounter (Figure 1). Removing these 2 ophthalmologists did not result in a significant change in mean (SD) EHR use time (9.6 [2.7] minutes). We included all ophthalmologists in the analysis to represent the inherent variability in EHR use time among different clinicians. Overall, these quantitative study findings support physicians’ reports of using EHRs during and after work hours for excessive lengths of time.29,30

The purpose of this study was to analyze the amount of time clinicians spent using the EHR, not to determine its positive or negative outcomes. Although studies have shown that the nature of EHR documentation is different from paper documentation,31 the overall association of EHRs with quality of care warrants further study. That said, adoption of EHRs has altered the nature of the physician-patient relationship.32,33 Although EHRs have enormous potential to improve quality of care,34-36 they also create clerical and cognitive burdens, bloated records with copied and pasted text, and frequent interruptions that threaten the potentially beneficial outcomes.37-39 Physicians have expressed concern about pressures from the competing demands of patient care and EHR requirements.40 Furthermore, 57% of physicians display 1 or more signs of burnout and 45% report dissatisfaction with work-life balance.41 Several surveys have shown a correlation between EHR use and physician burnout due to EHR-related stress and significant EHR time spent outside traditional work hours.29,42,43 Future research should investigate ways to lessen these burdens.

The final key study finding is that there is significant variation among ophthalmologists regarding time requirements and EHR use patterns (Figures 1 and 2). In general, higher clinic volumes are associated with less EHR time per encounter and more total EHR time (Table 3). Billing level, a reflection of the complexity of the patient encounter and documentation, also affects the variability of EHR use among ophthalmologists.44 Encounters billed at a higher level were associated with longer EHR use times, but this outcome was mitigated by clinical volume. Ophthalmologists with medium and high mean billing levels had lower clinic volumes and their EHR use was the most affected by changes in clinic volume. In addition, there was greater variability in EHR use per patient encounter among ophthalmologists within the medium and high mean billing levels. Ophthalmologists with low mean billing levels saw the most patients, displayed a wide variety of clinic volumes, and were not affected by the influence of clinic volume on EHR use. Future studies should focus on understanding the differences in EHR use among similar physicians, not only for identifying habits that improve efficiency but also for ensuring that shortcuts (eg, copying and pasting text) are not detrimental to patient care.

All ophthalmologists in this study spent more than 5.8 minutes on average using the EHR per patient encounter, suggesting that there is a minimum amount of EHR time needed per encounter. Ophthalmologists differed in when they used the EHR: among study physicians, 8 of 27 (30%) completed more than half of their EHR use after the encounter, and high-volume ophthalmologists spent the most time using the EHR after the clinic session. It was not possible to discern why these ophthalmologists used the EHR after the clinic session and whether those reasons consisted of examination time constraints, work flexibility, comfort level with the EHR, preservation of the physician-patient relationship, or other factors. Future studies will further investigate physician work patterns, including which parts of the patient encounter documentation and communication are completed during vs after the examination.

In summary, these findings highlight the importance of developing EHRs to meet the needs of patients and physicians and to develop appropriate training programs to improve the quality and efficiency of care.45,46 Addressing EHR usability through user-centered design methods will be an important step toward improving EHR workflow and data accessibility.47 Professional groups have long advocated for improved EHRs designed in collaboration between clinicians and vendors, yet the efficiency of current EHRs continues to be a challenge.40,48-50 We hope that this study will inspire clinicians and system designers to collaborate on improving EHR systems.

Limitations

Several study limitations should be noted. This study was limited to the manual observation of 5 ophthalmologists and the EHR time stamp analysis of 27 ophthalmologists at 1 institution and the findings may not be generalizable to other specialties, clinical settings, or EHRs. However, the consistency between EHR time stamp data and manual observation findings and between the quantitative results of this study and previously published qualitative findings suggests that these methods are generalizable. Although we believe ophthalmology is a representative domain for other fields because it is a fast-paced specialty with both medical and surgical components, additional studies are needed. This study did not investigate the association between EHR use and satisfaction. Further research is needed to determine if the time requirements for EHR are correlated with positive or negative feelings about EHRs. Time-saving practices such as copying and pasting text were not included in the analysis. We imagine that the use and effect of these practices would vary by clinician and believe that future studies will be important to address the association of such practices with EHR use time. For simplicity, we used only the primary office visit billing code as a proxy for encounter complexity. Future studies are needed to investigate the influence of additional procedures on encounter complexity. At the study institution, ancillary staff and trainees perform and record intake examinations (history, medications, and preliminary examination findings) and the clinician completes the examination. Although this is a fairly standard workflow in ophthalmology, our findings may not apply to alternative workflows without this initial examination. This study inferred EHR use from time stamps with a coarse (1-minute) granularity that may have led to the overestimation or underestimation of ophthalmologist EHR use. Although time-motion studies reveal more precise information from a smaller number of encounters, we found that it was impractical given the large number of physicians involved. In practice, ophthalmologists often talk with patients while simultaneously examining them or using the EHR. We recorded this multitasked time as the nonconversation activity, which may have caused a bias of decreased conversation time. Without recording multitasked time in this way, a time-motion study would have been impractical.

Conclusions

Overall, these study findings support anecdotal reports that EHRs create significant time burdens for ophthalmologists both during and after clinics. Ophthalmologists varied with regard to their time requirements and EHR use, suggesting that there may be approaches to address these challenges through training and experience. These findings have important implications for all practicing physicians regarding clinical efficiency, patient and physician satisfaction, design of improved EHR systems, and quality of care.

Back to top
Article Information

Corresponding Author: Michael F. Chiang, MD, Department of Ophthalmology, Oregon Health & Science University, 3375 SW Terwilliger Blvd, Portland, OR 97239 (chiangm@ohsu.edu).

Accepted for Publication: August 26, 2017.

Published Online: October 12, 2017. doi:10.1001/jamaophthalmol.2017.4187

Author Contributions: Dr Chiang had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Read-Brown, Hribar, Parikh, Yackel, Chiang.

Acquisition, analysis, or interpretation of data: Read-Brown, Hribar, Reznick, Lombardi, Chamberlain, Bailey, Wallace, Chiang.

Drafting of the manuscript: Read-Brown, Hribar, Bailey.

Critical revision of the manuscript for important intellectual content: Read-Brown, Hribar, Reznick, Lombardi, Parikh, Chamberlain, Wallace, Yackel, Chiang.

Statistical analysis: Read-Brown, Hribar, Wallace, Chiang.

Obtained funding: Read-Brown, Hribar, Chiang.

Administrative, technical, or material support: Read-Brown, Parikh, Chamberlain, Yackel, Chiang.

Study supervision: Read-Brown, Hribar, Chiang.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Chiang reported serving as an unpaid member of the Scientific Advisory Board for Clarity Medical Systems and as a consultant for Novartis. No other disclosures were reported.

Funding/Support: This work was supported by grants T15 LM007088, K99 LM12238, and P30 EY010572 from the National Institutes of Health and by unrestricted departmental funding from Research to Prevent Blindness.

Role of the Funder/Sponsor: The funding sources 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.

References
1.
Committee on Improving the Patient Record, Division of Health Care Services, Institute of Medicine.  The Computer-Based Patient Record: An Essential Technology for Health Care. Revised edition. Washington, DC: National Academies Press; 1997.
2.
Blumenthal  D.  Implementation of the federal health information technology initiative.  N Engl J Med. 2011;365(25):2426-2431.PubMedGoogle ScholarCrossref
3.
Jamoom  EW, Yang  N, Hing  E.  Adoption of certified electronic health record systems and electronic information sharing in physician offices: United States, 2013 and 2014.  NCHS Data Brief. 2016; (236):1-8.PubMedGoogle Scholar
4.
Henry  J, Pylpchuck  Y, Searcy  T, Patel  V.  Adoption of Electronic Health Record Systems Among US Non-Federal Acute Care Hospitals: 2008-2015. Washington, DC: Office of the National Coordinator for Health Information Technology; 2016. ONC Data Brief, No. 35.
5.
Verdon  DR. (Slideshow) Medical Economics EHR survey probes physician angst about adoption, use of technology. http://medicaleconomics.modernmedicine.com/medical-economics/content/tags/ehr/slideshow-medical-economics-ehr-survey-probes-physician-angst-abo. Published February 10, 2014. Accessed May 17, 2016.
6.
Carayon  P, Wetterneck  TB, Alyousef  B,  et al.  Impact of electronic health record technology on the work and workflow of physicians in the intensive care unit.  Int J Med Inform. 2015;84(8):578-594.PubMedGoogle ScholarCrossref
7.
Wormer  BA, Colavita  PD, Yokeley  WT,  et al.  Impact of implementing an electronic health record on surgical resident work flow, duty hours, and operative experience.  Am Surg. 2015;81(2):172-177.PubMedGoogle Scholar
8.
Chan  P, Thyparampil  PJ, Chiang  MF.  Accuracy and speed of electronic health record versus paper-based ophthalmic documentation strategies.  Am J Ophthalmol. 2013;156(1):165-172.e2.PubMedGoogle ScholarCrossref
9.
Poissant  L, Pereira  J, Tamblyn  R, Kawasumi  Y.  The impact of electronic health records on time efficiency of physicians and nurses: a systematic review.  J Am Med Inform Assoc. 2005;12(5):505-516.PubMedGoogle ScholarCrossref
10.
VanDenKerkhof  EG, Goldstein  DH, Lane  J, Rimmer  MJ, Van Dijk  JP.  Using a personal digital assistant enhances gathering of patient data on an acute pain management service: a pilot study.  Can J Anaesth. 2003;50(4):368-375.PubMedGoogle ScholarCrossref
11.
Asan  O, D Smith  P, Montague  E.  More screen time, less face time—implications for EHR design.  J Eval Clin Pract. 2014;20(6):896-901.PubMedGoogle ScholarCrossref
12.
Toll  E.  A piece of my mind: the cost of technology.  JAMA. 2012;307(23):2497-2498. PubMedGoogle ScholarCrossref
13.
Blumenthal  D, Collins  SR.  Health care coverage under the Affordable Care Act—a progress report.  N Engl J Med. 2014;371(3):275-281.PubMedGoogle ScholarCrossref
14.
Litvak  E, Bisognano  M.  More patients, less payment: increasing hospital efficiency in the aftermath of health reform.  Health Aff (Millwood). 2011;30(1):76-80.PubMedGoogle ScholarCrossref
15.
Koppel  R, Lehmann  CU.  Implications of an emerging EHR monoculture for hospitals and healthcare systems.  J Am Med Inform Assoc. 2015;22(2):465-471.PubMedGoogle Scholar
16.
R Core Team.  R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2014.
17.
Bates  D, Maechler  M, Bolker  B, Walker  S.  Fitting linear mixed-effects models using lme4.  J Stat Softw. 2015;67:1-48. doi:10.18637/jss.v067.i01Google ScholarCrossref
18.
Hothorn  T, Bretz  F, Westfall  P.  Simultaneous inference in general parametric models.  Biom J. 2008;50(3):346-363.PubMedGoogle ScholarCrossref
19.
Pandit  RR, Boland  MV.  The impact of an electronic health record transition on a glaucoma subspecialty practice.  Ophthalmology. 2013;120(4):753-760.PubMedGoogle ScholarCrossref
20.
Lo  HG, Newmark  LP, Yoon  C,  et al.  Electronic health records in specialty care: a time-motion study.  J Am Med Inform Assoc. 2007;14(5):609-615.PubMedGoogle ScholarCrossref
21.
Linzer  M, Manwell  LB, Williams  ES,  et al; MEMO (Minimizing Error, Maximizing Outcome) Investigators.  Working conditions in primary care: physician reactions and care quality.  Ann Intern Med. 2009;151(1):28-36, W6-W9. PubMedGoogle ScholarCrossref
22.
Shaw  MK, Davis  SA, Fleischer  AB, Feldman  SR.  The duration of office visits in the United States, 1993 to 2010.  Am J Manag Care. 2014;20(10):820-826.PubMedGoogle Scholar
23.
Centers for Disease Control and Prevention.  The State of Aging and Health in America 2013. Atlanta, GA: Centers for Disease Control & Prevention, US Dept of Health and Human Services; 2013.
24.
Long  C, Tsay  EL, Jacobo  SA, Popat  R, Singh  K, Chang  RT.  Factors associated with patient Press Ganey satisfaction scores for ophthalmology patients.  Ophthalmology. 2016;123(2):242-247.PubMedGoogle ScholarCrossref
25.
Blumenthal  D, Causino  N, Chang  YC,  et al.  The duration of ambulatory visits to physicians.  J Fam Pract. 1999;48(4):264-271.PubMedGoogle Scholar
26.
Tsiga  E, Panagopoulou  E, Sevdalis  N, Montgomery  A, Benos  A.  The influence of time pressure on adherence to guidelines in primary care: an experimental study.  BMJ Open. 2013;3(4).PubMedGoogle Scholar
27.
Kumarapeli  P, de Lusignan  S.  Using the computer in the clinical consultation; setting the stage, reviewing, recording, and taking actions: multi-channel video study.  J Am Med Inform Assoc. 2013;20(e1):e67-e75.PubMedGoogle ScholarCrossref
28.
Tai-Seale  M, Olson  CW, Li  J,  et al.  Electronic health record logs indicate that physicians split time evenly between seeing patients and desktop medicine.  Health Aff (Millwood). 2017;36(4):655-662.PubMedGoogle ScholarCrossref
29.
Linzer  M, Poplau  S, Babbott  S,  et al.  Worklife and wellness in academic general internal medicine: results from a national survey.  J Gen Intern Med. 2016;31(9):1004-1010.PubMedGoogle ScholarCrossref
30.
Makam  AN, Lanham  HJ, Batchelor  K,  et al.  Use and satisfaction with key functions of a common commercial electronic health record: a survey of primary care providers.  BMC Med Inform Decis Mak. 2013;13:86.PubMedGoogle ScholarCrossref
31.
Sanders  DS, Read-Brown  S, Tu  DC,  et al.  Impact of an electronic health record operating room management system in ophthalmology on documentation time, surgical volume, and staffing.  JAMA Ophthalmol. 2014;132(5):586-592. PubMedGoogle ScholarCrossref
32.
Verghese  A.  Culture shock—patient as icon, icon as patient.  N Engl J Med. 2008;359(26):2748-2751.PubMedGoogle ScholarCrossref
33.
Alkureishi  MA, Lee  WW, Lyons  M,  et al.  Impact of electronic medical record use on the patient-doctor relationship and communication: a systematic review.  J Gen Intern Med. 2016;31(5):548-560.PubMedGoogle ScholarCrossref
34.
Hribar  MR, Read-Brown  S, Reznick  L,  et al.  Secondary use of EHR timestamp data: validation and application for workflow optimization.  AMIA Annu Symp Proc. 2015;2015:1909-1917. PubMedGoogle Scholar
35.
Raghupathi  W, Raghupathi  V.  Big data analytics in healthcare: promise and potential.  Health Inf Sci Syst. 2014;2:3. doi:10.1186/2047-2501-2-3PubMedGoogle ScholarCrossref
36.
Hoover  R.  Benefits of using an electronic health record.  Nursing. 2016;46(7):21-22.PubMedGoogle ScholarCrossref
37.
Patel  VL, Arocha  JF, Kaufman  DR.  A primer on aspects of cognition for medical informatics.  J Am Med Inform Assoc. 2001;8(4):324-343.PubMedGoogle ScholarCrossref
38.
Thornton  JD, Schold  JD, Venkateshaiah  L, Lander  B.  Prevalence of copied information by attendings and residents in critical care progress notes.  Crit Care Med. 2013;41(2):382-388.PubMedGoogle ScholarCrossref
39.
Zheng  K, Haftel  HM, Hirschl  RB, O’Reilly  M, Hanauer  DA.  Quantifying the impact of health IT implementations on clinical workflow: a new methodological perspective.  J Am Med Inform Assoc. 2010;17(4):454-461.PubMedGoogle ScholarCrossref
40.
Zhang  J, Chen  Y, Ashfaq  S,  et al.  Strategizing EHR use to achieve patient-centered care in exam rooms: a qualitative study on primary care providers.  J Am Med Inform Assoc. 2016;23(1):137-143.PubMedGoogle ScholarCrossref
41.
Shanafelt  TD, Hasan  O, Dyrbye  LN,  et al.  Changes in burnout and satisfaction with work-life balance in physicians and the general US working population between 2011 and 2014.  Mayo Clin Proc. 2015;90(12):1600-1613.PubMedGoogle ScholarCrossref
42.
Shanafelt  TD, Dyrbye  LN, Sinsky  C,  et al.  Relationship between clerical burden and characteristics of the electronic environment with physician burnout and professional satisfaction.  Mayo Clin Proc. 2016;91(7):836-848.PubMedGoogle ScholarCrossref
43.
Babbott  S, Manwell  LB, Brown  R,  et al.  Electronic medical records and physician stress in primary care: results from the MEMO Study.  J Am Med Inform Assoc. 2014;21(e1):e100-e106.PubMedGoogle ScholarCrossref
44.
Centers for Medicare & Medicaid Services, Dept of Health and Human Services. Evaluation and management services. https://www.cms.gov/Outreach-and-Education/Medicare-Learning-Network-MLN/MLNProducts/Downloads/eval-mgmt-serv-guide-ICN006764.pdf. Updated August 2017. Accessed September 12, 2017.
45.
Martin  SA, Sinsky  CA.  The map is not the territory: medical records and 21st century practice.  Lancet. 2016;388(10055):2053-2056.PubMedGoogle ScholarCrossref
46.
Zulman  DM, Shah  NH, Verghese  A.  Evolutionary pressures on the electronic health record: caring for complexity.  JAMA. 2016;316(9):923-924.PubMedGoogle ScholarCrossref
47.
Horsky  J, Ramelson  HZ.  Development of a cognitive framework of patient record summary review in the formative phase of user-centered design.  J Biomed Inform. 2016;64:147-157.PubMedGoogle ScholarCrossref
48.
Zahabi  M, Kaber  DB, Swangnetr  M.  Usability and safety in electronic medical records interface design: a review of recent literature and guideline formulation.  Hum Factors. 2015;57(5):805-834.PubMedGoogle ScholarCrossref
49.
Middleton  B, Bloomrosen  M, Dente  MA,  et al; American Medical Informatics Association.  Enhancing patient safety and quality of care by improving the usability of electronic health record systems: recommendations from AMIA.  J Am Med Inform Assoc. 2013;20(e1):e2-e8.PubMedGoogle ScholarCrossref
50.
American Medical Association. Improving care: priorities to improve electronic health record usability: executive summary. https://www.aace.com/files/ehr-priorities.pdf. Accessed Mar 23, 2017.
×