Patient safety issues from information overload in electronic medical records: A systematic review

Abstract

Background and Objective: Electronic health records (EHR) have become ubiquitous in medicine and continue to grow in informational content. Little has been documented regarding patient safety from the resultant information overload. The objective of this literature review is to better understand how information overload in EHR affects patient safety.

Methods: A literature search was performed using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) standards for literature review. PubMed and Web of Science were searched and articles selected that were relevant to EHR information overload based upon keywords. 

Results: The literature search yielded 31 articles meeting criteria for the study. Information overload was found to increase physician cognitive load and error rates in clinical simulations. Overabundance of clinically irrelevant information, poor data display, and excessive alerting were consistently identified as issues that may lead to information overload. An addition, information overload increased the risk of physician burnout due to clerical burden leading to significantly higher rates of medical error. 

Conclusion: Information overload in EHR may result in higher error rates and negatively impact patient safety. Further studies are necessary to define the EHR role in adverse patient safety events and to determine methods to mitigate these errors.

Introduction

Electronic health records continue to increase in usage around the world, with multiple governments requiring implementation, in part due to the improvements in patient safety.¹ The EHR has demonstrated improved patient safety by improving rates of workflows, policies, and practices that promote patient safety when compared to paper charts.² Despite its advantages over paper-based documentation, EHR use has resulted in new physician-related challenges that may lead to increased medical errors.³ Indeed, in 2012, in anticipation of the potential for medical errors, the AMIA Board of Directors convened a task force to produce recommendations on enhancing patient safety by improving the usability of the EHR.⁴ Despite highlighting 14 usability principles to improve the design of EHR’s, the authors only mentioned the concept of minimizing cognitive load, but did not provide any further discussion or solutions for information overload.

A major complaint of physicians is the large amount of required extraneous patient information in each medical chart.⁵ In addition to the expanding written text within the EHR, other datatypes sources such as radiological data, genomic data, and predictive analyses compound the volume of information. Studies analyzing primary care physicians’ EHR usage have shown that they spend up to double the amount of time documenting in the EHR than they do interacting with patients. ⁶ This increased clerical burden on physicians is not only a source of frustration but may compromise patient safety.⁷ An overload of information in a patient’s chart, or “note bloat”, may impair comprehension when reviewing medical records, leading to potential errors in clinical decision making. ⁸ This can be compounded by poorly organized EHR software that is optimized for billing rather than patient care. ⁹ These EHR issues may increase physicians’ cognitive load and leave them more susceptible to making mistakes.¹⁰[Figure 1] A study analyzing data from 2013-2016 in Pennsylvania hospitals identified 1,956 adverse patient safety events blamed on the EHR in that time span with 557 of them being directly attributed to EHR usability.¹¹ 

Overload from EHR’s can also negatively affect physician well-being, as noted in a Finnish study by Vainiomaki et al. where they surveyed 3,781 Finnish physicians and found overload from higher time pressure and lower job control from EHR’s.¹²

The purpose of this literature review is to evaluate the effect of EHR information overload on patient safety. Our hypothesis is that information overload negatively effects patient safety. 

Methods

A systematic review and qualitative analysis were performed to identify factors related to EHR information overload and patient safety using PubMed and Web of Science covering publications from the past 10 years and completed on 8/20/2019.  Selected full text articles were obtained and reviewed for those focused on patient safety implications of EHR information overload. Filters were set for English only and full text availability. References of selected articles were also reviewed and used as an additional source of literature.  All published study types were included.  All data analyses were descriptive.  Institutional review board review was not obtained as the study was limited to published information and did not include any human subjects.

Keywords used included “electronic health record” and “electronic medical record” in conjunction with the terms “information overload”, “cognitive overload”, “note bloat”, “usability”, and “patient safety”. Additional searches with different combinations of these terms included “electronic health record information overload patient safety”, “electronic health record usability patient safety”, “electronic health record cognitive overload patient safety”. Studies were deemed relevant if they: (a) defined the issue of information overload, (b) described how information overload fits into the current model of EHR safety analysis, and/or (c) provided data to demonstrate how information overload and poor EHR usability affect physician comprehension of clinical data. Further literature was sought out to understand the effect EHR has on physician burnout, as well as the link between burnout and medical error. Keywords included “EHR physician burnout” and “physician burnout medical error”.

Articles were screened by one reviewer by first assessing the title and then the abstract for relevance to the topic. If the title included either (a) no keywords or (b) “electronic health record” or “electronic medical record” but no further keywords of interest, the article was excluded. Titles that were focused on healthcare professionals other than physicians were not chosen.

PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) standards for literature review requirements were followed for items applicable to this literature review.

Results

As of 8/20/2019, a total 6,421 records where obtained by searching PubMed and the Web of Science, and adding relevant references from the database articles. 6,031 titles were discarded due to irrelevant topics or duplication. 390 abstracts were reviewed, and 346 were discarded due to lack of original research or not physician-focused. This left 44 full-text articles that were assessed for eligibility, leaving 31 studies that fulfilled the inclusion criteria and used for the study [Figure 2][Table 1]. All of the studies obtained correlated information overload with some level of negative effects on physicians. Based upon the results of these 31 studies, two main topics were consistently raised relating to information overload and patient safety – the resultant cognitive burden and physician burnout. 

The cognitive burden of information overload

Beasley et al.¹³ stated that “information overload occurs when there are too many data, e.g. written, verbal and non-verbal, and physician’s memory, for the clinician to organize, synthesize, draw conclusions from, or act on.” The overload of information can occur from copying & pasting into charts, use of templates, excessive alerts, and adding data that is necessary for billing but effectively useless for clinical care.^13,14 

Ahmed et al.¹⁵ illustrated the effect of cognitive load differences on error rates when using an EHR for clinical decision making. A group of 20 intensive care unit physicians were asked to review patient data in the form of a conventional versus novel streamlined EHR. The novel EHR was specifically designed to only display information that was deemed most salient to these physicians. The National Aeronautics and Space Administration (NASA) task load index (an objective measure of task load from 0-100; higher indicates more work load) was a median of 58 for the conventional EHR versus 38.8 for the novel version. Completion of the task using the conventional EHR took approximately twice as long and was associated with a median four times as many errors per subject as the novel user interface. This was consistent with the hypothesis that increased task load has significant detrimental effects on physicians’ ability to analyze data.

Eye tracker technology in the intensive care unit was studied by Wright et al. to pinpoint what aspects of a chart physicians actually utilize ¹⁶. They found that dynamic data such as vitals and lab values were reviewed most consistently, and that other routine information is unnecessary and hinders usability. 

Koopman et al.⁹ performed a cognitive task analysis with 16 primary care physicians using simulated patient cases to better understand what information they considered most important for medical decision making. A consistent finding among these physicians was that the assessment and plan was reviewed first because it provided the majority of the necessary information in a concise manner. The physicians were frustrated by the review of systems section as it mostly provided redundant information and was another source of clutter. Physicians in the study also identified drivers of note overload: Billing (checklists for each section, especially review of systems), quality improvement measures (e.g., diabetic foot examination), avoiding malpractice, compliance (e.g., documenting informed consent, patient education), and the visit history and physical exam. An earlier study by Clarke et al. found similar results when they interviewed 15 primary care physicians about their information needs, finding the review of systems “superfluous,” and contributing to information overload.¹⁷

Belden et al.¹⁸ expanded on the idea of restructuring the fundamental structure of notes in the EHR to decrease cognitive overload. The traditional “SOAP” (Subjective, Objective, Assessment, Plan) note was compared to a newly proposed “APSO” (Assessment, Plan, Subjective, Objective) format with an option to hide other extraneous information. A simulated case with 16 physicians demonstrated that simply changing the format of the note without changing any of the actual data had a positive effect. The APSO note performed better in regard to usability, and the physicians strongly endorsed this style as more practical.

Information overload can also be mitigated through educating physicians to write more efficient notes. Kahn et al.¹⁹ demonstrated that physicians who undergo a training session and use a template write notes that are 25% shorter and take 1.3 hours less time.

A study done by Senathirajah et al. with 11 physicians reviewing the same patient data showed a significant increase in reading efficiency with a user composable interface versus a traditional EHR. 72% of patient data was reviewed more than once in conventional EHR’s compared to 17% in the user composable version. A conclusion offered by these authors was that the poor usability of conventional EHR’s decreases physician comprehension, requiring data to be revisited multiple times until it is fully understood.²⁰ 

However, simply allowing for user composability does not guarantee increased efficiency as illustrated by Ratwani et al.²¹ The usability and safety of Cerner and Epic were assessed by having 4 different groups of 12-15 physicians at different institutions (two groups using Epic, two groups using Cerner) complete basic tasks such as ordering imaging, labs, and medication for fictitious patients. Performance was assessed by tracking error rates, clicks, and task completion time between the four groups. Results showed up to an 8-fold difference in task completion time and clicks between the groups at different sites using the same EHR. Both EHR’s are user composable, but factors such as implementation protocols and physician training varied between the two sites and were hypothesized as reasons for the vast difference in proficiency.

Alert fatigue is another potential source of information overload. In a survey of 2,590 primary care physicians, 69.6% reported receiving more information than they could effectively manage. 29.8% reported incidents where they personally missed test results that delayed patient care.¹⁴ Another study demonstrated that a clinician’s likelihood of accepting best practice reminders dropped markedly with increases in the number of reminders, number of repeated reminders for the same patient, and overall patient complexity.²² A program to decrease alerts of lower importance in the Department of Veteran’s Affairs was developed by Shah et al. in 2018, resulting in a reduction of mean daily notifications per physician from 128 to 116, and a concomitant savings of 1.5 hours of work per week per physician.²³

Khairat et al.²⁴ demonstrated how the burdens of EHR’s affected physicians differently depending on the stage of their careers. Six clinical case simulations were performed by ER residents and attendings, followed by a survey to assess perceived workload and satisfaction for EHR’s. Attending physicians showed significantly higher levels of frustration with the EHR in general compared to residents. Information overload was rated more significant for residents, while attendings found excessive alerting to be a more negative factor.

Physician burnout

Physician frustration can be a result of information overload, with up to half of a work day spent working on an EHR and an additional 1-2 hours at home, according a study by Sinsky et al.⁵ Marked decreases in time spent with patients is reported by physicians to be a large source of dissatisfaction and burnout.²⁴ A 2017 survey of primary care physicians showed that 75% of doctors reporting burnout attributed it to the burden of the EHR as the primary cause.²⁵ A 2019 survey of 282 clinicians from 3 different institutions gave more insight on the specific factors that lead to EHR burnout. The most significant problems associated with EHR‘s included information overload, excessive data entry, and notes geared toward billing rather than patient care.²⁶ Another survey of 1,792 physicians in 2019 revealed that physicians had a 2.8 times the odds of being burned out when they felt there was not enough time in the day for documentation.²⁷ Burnout increases the risk of depression, substance abuse, strained relationships, and suicide among physicians, in addition to a significantly higher incidence of medical errors.^28,29 Tawfik et al.²⁹ reported that physicians with burnout had more than twice the odds of self-reported medical errors, after adjusting for specialty, work hours, fatigue, and work unit safety rating.

Discussion

Patient safety is paramount in all aspects of medical care and any efforts to improve it should be pursued. The EHR’s effect on patient safety is complex. When implemented properly, it can reduce medication errors and provide a potentially safer alternative to paper-based methods.^2,30 However, its use has caused information overload as an unintended consequence.⁴ In addition to the growing text within the written notes, information within the EHR has also expanded from radiology results, laboratory results, alerts, demographics, predictive analyses, and more. The massive amount of data required in each patient’s chart has become potentially obstructive to patient care, and can hinder the physician-patient interaction. Efficiently extracting clinically relevant information from the EHR can be a difficult task for physicians.^15,16,20,21 This increased cognitive load placed on physicians makes them more prone to clinical errors, which puts patient safety at risk.¹⁵ In addition, the stresses of information overload contribute to physician frustration and burnout, which can also lead to an increase in medical errors.²⁹ A worrisome implication of these results in light of the continuous growth of information, is that without any changes, the rate of medical errors will more than likely continue to worsen over time.

In order to minimize the effects of information overload, various solutions have been proposed. One such solution is a customizable EHR to ensure that important data is easier to find.^15,20 Studies that tested this type of software showed significant reductions in error rates and improvements in efficiency. Pickering et al.³¹ introduced a novel user interface called AWARE (Ambient Warning and Response Evaluation) for use in the intensive care unit. The program synthesizes all of the data on a patient to a more readable and concise format, thus allowing physicians to make significantly quicker and safer decisions on patient care. 

However, customizability does not provide the same benefit for all physicians. For example, a physician’s level of training has implications on how best to customize an EHR. Attendings and residents have significant differences in what they find challenging with the EHR²⁴, and thus, their respective interactions with the EHR must be assessed when customizing the EHR to minimize information overload and improve patient safety. In addition, user composability requires proper training techniques with implementation in order to maximize its potential.²¹

Changing the order of the clinical note may also improve usability^18,19. Putting salient information at the beginning of the note may allow physicians to spend less time searching through extraneous information, and thus reduce the cognitive burden. However, notes are not just read, but also written, and retraining physicians to alter the order of notes may lead to an increased burden in and of itself. Software may potentially be designed to reformat a written SOAP note into presentation as an APSO note, but this has not been developed or studied to our knowledge. Thus, changing the traditional order of notes may require further research before implementation.

Alert fatigue can further contribute to information overload. Excessive alerting has been shown to alter decision making and cause physicians to deviate from best practice.²² Interventions aimed at reducing unnecessary alerting have been proven to decrease time spent with the EHR.²³ Any addition of alerts in EHR’s must be taken with great caution due to the increased risk of information overload, and efforts must constantly be made to minimize their usage.

One unintended consequence of information overload is the potential for physician burnout. Information overload results in an increase in tedious clerical work, reduction in physician-patient time, and a hampering of physician efficiency.⁵ This has negatively impacted physician well-being and is a significant cause of physician burnout.^24-28 Physicians with burnout have an increased rate of clinical errors, further increasing the risk of adverse patient safety events.²⁹ Thus, information overload not only had negative consequences for patient care, but it also negatively effects the well-being of physicians.

Conclusion

A review of the literature demonstrates that EHR information overload can have a negative impact on patient safety, in addition to contributing to physician burnout and further increased medical error rates. Customizable EHR, shortened clinical notes, and reduced alerting may be helpful interventions to minimize cognitive load and improve patient safety. EHR information overload may also contribute to physician burnout and negatively affects physician wellness.  Further research to understand the impact of information overload on patient safety is necessary in order to more effectively develop improved EHR’s focused on improving patient safety and reducing the burden on physicians.

Abbreviations

EHR: Electronic Health Record

PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses

AMIA: American Medical Informatics Association

NASA: National Aeronautics and Space Administration

SOAP: Subjective-Objective-Assessment-Plan

APSO: Assessment-Plan-Subjective-Objective

AWARE: Ambient Warning and Response Evaluation

Declarations

Competing interests

There are no conflicts of interest from any of the authors of this manuscript.

Funding

There was no funding for this literature review.

Authors' contributions

SN conducted the literature search and wrote initial drafts of the manuscript. EG provided topic and guidance of the project, as well as wrote many revisions. NL also provided input on the direction of the project.

Acknowledgements

There were no further contributors to this review article.

References

1. Zelmer J, Ronchi E, Hyppönen H, et al. on behalf of the OECD Health ICT Benchmarking Pilot Group. International health IT benchmarking: learning from cross-country comparisons. J Am Med Inform Assoc. 2017;24(2):371–379
2. Tanner C, Gans D, White J, Nath R, Pohl J. Electronic health records and patient safety: Co-occurrence of early EHR implementation with patient safety practices in primary care settings. Appl Clin Inform. 2015;6(1):136-147. Accessed Jun 16, 2018. doi: 10.4338/ACI-2014-11-RA-0099
3. Adler-Milstein J, Everson J, Lee SD. EHR adoption and hospital performance: Time-related effects. Health Serv Res. 2015;50(6):1751-1771. Accessed Jun 25, 2018. doi: 10.1111/1475-6773.12406
4. Middleton B, Bloomrosen M, Dente MA, et al. 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:e2–e8
5. Sinsky C, Colligan L, Li L, et al. Allocation of physician time in ambulatory practice: A time and motion study in 4 specialties. Annals of Internal Medicine. 2016;165(11):753. https://search.proquest.com/docview/1848074823. doi: 10.7326/M16-0961
6. Arndt BG, Beasley JW, Watkinson MD, et al. Tethered to the EHR: Primary care physician workload assessment using EHR event log data and time-motion observations. Ann Fam Med. 2017;15(5):419-426. Accessed Jul 9, 2018. doi: 10.1370/afm.2121
7. Sittig DF, Wright A, Ash J, Singh H. New unintended adverse consequences of electronic health records. Yearb Med Inform. 2016(1):7-12. Accessed Jun 18, 2018. doi: 10.15265/IY-2016-023
8. March CA, Steiger D, Scholl G, Mohan V, Hersh WR, Gold JA. Use of simulation to assess electronic health record safety in the intensive care unit: A pilot study. BMJ open. 2013;3(4):e002549. http://www.ncbi.nlm.nih.gov/pubmed/23578685. doi: 10.1136/bmjopen-2013-002549
9. Koopman RJ, Steege LMB, Moore JL, et al. Physician information needs and electronic health records (EHRs): Time to reengineer the clinic note. J Am Board Fam Med. 2015;28(3):316-323. Accessed Jun 19, 2018. doi: 10.3122/jabfm.2015.03.140244
10. Zulman DM, Shah NH, Verghese A. Evolutionary pressures on the electronic health record: Caring for complexity. JAMA. 2016;316(9):923-924. http://dx.doi.org/10.1001/jama.2016.9538. doi: 10.1001/jama.2016.9538
11. Howe JL, Adams KT, Hettinger AZ, Ratwani RM. Electronic health record usability issues and potential contribution to patient harm. JAMA. 2018;319(12):1276-1278. https://jamanetwork.com/journals/jama/fullarticle/2676098. Accessed Jun 13, 2018. doi: 10.1001/jama.2018.1171
12. Vainiomäki S, Aalto AM, Lääveri T, Hyppönen H. et al. Better Usability and Technical Stability Could Lead to Better Work-Related Well-Being among Physicians. Appl Clin Inform. 2017; Oct;8(4):1057-1067. doi: 10.4338/ACI-2017-06-RA-0094 
13. Beasley JW, Wetterneck TB, Temte J, et al. Information chaos in primary care: Implications for physician performance and patient safety. J Am Board Fam Med. 2011;24(6):745-751. Accessed Jun 30, 2018. doi: 10.3122/jabfm.2011.06.100255
14. Singh H, Spitzmueller C, Petersen NJ, Sawhney MK, Sittig DF. Information overload and missed test results in electronic health Record–Based settings. JAMA Intern Med. 2013;173(8):702-704. https://jamanetwork-com.libproxy.temple.edu/journals/jamainternalmedicine/fullarticle/1657753. Accessed Jun 15, 2018. doi: 10.1001/2013.jamainternmed.61
15. Ahmed A, Chandra S, Herasevich V, Gajic O, Pickering BW. The effect of two different electronic health record user interfaces on intensive care provider task load, errors of cognition, and performance. Crit Care Med. 2011;39(7):1626-1634. Accessed Jun 18, 2018. doi: 10.1097/CCM.0b013e31821858a0
16. Wright MC, Dunbar S, Macpherson BC, et al. Toward designing information display to support critical care. A qualitative contextual evaluation and visioning effort. Appl Clin Inform. 2016;7(4):912-929. Accessed Jun 18, 2018. doi: 10.4338/ACI-2016-03-RA-0033
17. Clarke MA, Steege LM, Moore JL, Koopman RJ, Belden JL, Kim MS. Determining primary care physician information needs to inform ambulatory visit note display. Appl Clin Inform. 2014 Feb 26;5(1):169-90. doi: 10.4338/ACI-2013-08-RA-0064
18. Belden JL, Koopman RJ, Patil SJ, Lowrance NJ, Petroski GF, Smith JB. Dynamic electronic health record note prototype: Seeing more by showing less. J Am Board Fam Med. 2017;30(6):691-700. Accessed Jul 31, 2018. doi: 10.3122/jabfm.2017.06.170028
19. Kahn D, Stewart E, Duncan M, et al. A prescription for note bloat: An effective progress note template. J Hosp Med. 2018;13(6):378-382. Accessed Jun 18, 2018. doi: 10.12788/jhm.2898
20. Senathirajah Y, Kaufman D, Bakken S. User-composable electronic health record improves efficiency of clinician data viewing for patient case appraisal: A mixed-methods study. EGEMS (Wash DC). 2016;4(1):1176. Accessed Jun 18, 2018. doi: 10.13063/2327-9214.1176
21. Ratwani RM, Savage E, Will A, et al. A usability and safety analysis of electronic health records: A multi-center study. J Am Med Inform Assoc. 2018;25(9):1197-1201. Acessed June 15, 2019. doi: 10.1093/jamia/ocy088
22. Ancker JS, Edwards A, Nosal S, Hauser D, Mauer E, Kaushal R. Effects of workload, work complexity, and repeated alerts on alert fatigue in a clinical decision support system. BMC Med Inform Decis Mak. 2017;17(1):36. Accessed Jun 30, 2018. doi: 10.1186/s12911-017-0430-8
23. Shah T, Patel-Teague S, Kroupa L, Meyer AND, Singh H. Impact of a national QI programme on reducing electronic health record notifications to clinicians. BMJ Qual Saf. 2019;28(1):10–14. doi:10.1136/bmjqs-2017-007447
24. Khairat S, Burke G, Archambault H, Schwartz T, Larson J, Ratwani RM. Perceived Burden of EHRs on Physicians at Different Stages of Their Career. Appl Clin Inform. 2018;9(2):336–347. doi:10.1055/s-0038-1648222
25. 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. Accessed Jul 23, 2018. doi: 10.1016/j.mayocp.2016.05.007
26. Robertson SL, Robinson MD, Reid A. Electronic health record effects on work-life balance and burnout within the I3 population collaborative. J Grad Med Educ. 2017;9(4):479-484. Accessed Aug 12, 2018. doi: 10.4300/JGME-D-16-00123.1
27. Kroth PJ, Morioka-Douglas N, Veres S, et al. Association of Electronic Health Record Design and Use Factors With Clinician Stress and Burnout. JAMA Network Open. 2019;2(8). doi:10.1001/jamanetworkopen.2019.9609
28. Gardner RL, Cooper E, Haskell J, et al. Physician stress and burnout: The impact of health information technology. J Am Med Inform Assoc. 2019;26(2):106-114. Accessed June 14, 2019. doi: 10.1093/jamia/ocy145
29. Lacy BE, Chan JL. Physician burnout: The hidden health care crisis. Clin Gastroenterol Hepatol. 2018;16(3):311-317. Accessed Jul 26, 2018. doi: 10.1016/j.cgh.2017.06.043
30. Tawfik DS, Profit J, Morgenthaler TI, et al. Physician burnout, well-being, and work unit safety grades in relationship to reported medical errors. Mayo Clin Proc. 2018. Accessed Jul 19, 2018. doi: 10.1016/j.mayocp.2018.05.014
31. King J, Patel V, Jamoom EW, Furukawa MF. Clinical benefits of electronic health record use: National findings. Health Serv Res. 2014;49(1 Pt 2):392-404. Accessed Jul 30, 2018. doi: 10.1111/1475-6773.12135
32. Pickering BW, Herasevich V, Ahmed A, Gajic O. Novel representation of clinical information in the ICU: Developing user interfaces which reduce information overload. Appl Clin Inform. 2010 Apr 28;1(2):116-31. doi:10.4338/ACI-2009-12-CR-0027

Table 1