Trends in integration of medical practices helped create market-based incentives to develop better communication between the various fragmented pieces of the US health system. Federal legislation in 2008 (MIPPA) and 2009 (ARRA) created special payments to physicians to induce the adoption of electronic medical records (EMRs), also referred to as electronic health records (EHRs) and personal health records (PHRs). In this blog entry, we explore some of the potentially important issues surrounding the role of EMRs in litigation, where the discovery of medical documents and records increasingly involves digitally stored data.[1-4] These issues also pertain to the use of EMRs in research and should in some cases be considered by investigators in designing studies and analyzing EMR data.[5, 6]
The most popular systems include Epic, Allscripts, eClinicalWorks, NextGen, Cerner, and GE Healthcare, although there are many more manufacturers and systems available. There are some variations in different EMR products, but most systems capture a consistent set of core elements.[6-14] Most EMR systems are designed to record and integrate the following domains of information: patient characteristics and demographics; encounter log (office visits; ER; inpatient); allergies; immunizations; medications (history; current); laboratory (history; current); imaging (with links to actual images); diagnoses (ICD-10 codes; past and at encounter); procedures (CPT codes for encounters); notes; correspondence; and insurance and payment information. Larger users, such as large hospitals and health systems (e.g., regional health systems) may customize EMR systems based on: (a) scope of services, (b) types of patients, (c) types of payers, and (d) geographic dispersion of integrated providers.[12, 15-19]
The widespread use of EMR systems by health care facilities (i.e., hospitals, clinics, long-term care facilities, etc.) has improved real-time access to complete patient information and, in turn, improved patient safety and quality of care. EMR systems produce a legible template-driven record, in contrast to paper charts which often vary in consistency in terms of legibility and accuracy. EMRs are generally an improvement over handwritten medical records because they typically rely on standardized language, including the use of drop-down menus, checklists, etc. They also typically link together several data sources (e.g., outpatient clinic, laboratory, radiology, inpatient hospital, pharmacy, etc.), linkages that are very difficult to maintain and update using paper records.
In the litigation and research context, it seems at first glance that EMRs would make things easier and more transparent to all parties involved. And in some cases that may very well be the case. But at the same time EMRs pose some interesting and potentially important challenges. These can be summarized as follows:
- EMRs generally contain “more” information, in terms of volume and type, than paper records. But this added information and functionality brings with it an interesting challenge: what constitutes a “complete” medical record? Is the EMR defined by only its core elements, or does the EMR also include some or all the sources to which it is linked? Longitudinal data are also a potential issue. Paper records are often limited to the time period for which the patient was treated in the medical practice, clinic, or hospital. EMRs have the potential to extend that time period considerably, either by way of facilitating back linking with a patient’s previous EMR records or by linking with external sources that contain more history (e.g., a lab provider or specialized clinic).[20] Again, this can create some uncertainty around what might be considered a full or complete medical record; just because linkages exist, are they relevant?
- While most agree that EMRs provide more information, in some cases they may result is a loss of fidelity, for two reasons. First, EMRs are designed to facilitate easy data entry. The theory is that the easier it is to enter data, the more accurately and more frequently those data will be entered.[21, 22] To this end, EMRs utilize dropdown menus, buttons, and other “user-friendly” interface tools to more efficiently facilitate data entry. One potential limitation of menus, however, is that not all conceivable circumstances can be accounted for or captured. Thus, in some cases it is possible that clinicians entering data will be compelled to choose a value that is not exactly or precisely the same as what they may have written in a paper record. The second type of potential accuracy loss is that which may be the result of “cutting and pasting” from a previous entry in the case of fields that accommodate free-form text.[23] For example, some clinicians may find it easier to copy text entered previously (by themselves or by another clinician on a different shift) if the patient’s condition had not markedly or noticeably changed. Again, with paper records, having to write new text may in some cases encourage the revealing of slight differences in condition, regardless of the clinical importance of those differences.
- EMRs offer a host of audit-type options, which help identify who entered data and when, but depending on the system and how that system may have been modified by the host health care facility, the integrity and reliability of audits may vary substantially. This is not to say that facilities have the ability to alter EMR data without leaving a footprint; they generally do not. However, there may be some normal and reasonable clinical reasons to be able to edit and revise EMR data entries, to ensure record accuracy or to update a field left blank due to lack of data (e.g., results from laboratory tests were incomplete). Most EMR systems (again depending on configuration) will permit such revisions. A potential challenge in litigation and research is, in some cases, the accuracy and transparency of such edits in audit trails and other logs.[2, 3] For example, some facilities may have modified systems to provide easy updating and modifications of some commonly-used fields, whereas other fields may be more controlled and require steps more likely to establish a clear accounting of the “who, when, and why.”[24, 25] Again, there exist many good reasons to update charts; such changes are not per se problematic.
In sum, EMRs offer seemingly limitless potential in improving the flow of information in litigation settings. And they have already been established as the new backbone of patient-level research in medicine and life sciences, especially the area of research commonly referred to as “real-world evidence” (as distinguished from evidence resulting from tightly controlled clinical trials). Overall, it is likely that EMRs have added both accuracy and completeness to data used in litigation and clinical research, despite the limitations we discuss above. However, it is important to understand the nuances and limitations of EMR data. Litigators need to know what to ask for in discovery, and how to evaluate the accuracy and completeness of what they obtain through discovery. They also need to understand what would be considered “normal” in terms of the accuracy and handling of medical records.
Likewise, researchers should approach EMR data the same way that they approach any data and conduct the usual consistency and accuracy checks that would be employed with non-EMR-based data sources, such as claims data. It is common to find discrepancies in claims data, and we should expect to find similar discrepancies in EMR data. In both the litigation and research arenas, discrepancies do not per se suggest a “problem” of some sort; more often, they are simply an artifact of the complexities of collecting medical and clinical details fraught with substantial heterogeneity across patients.
(By John Schneider, PhD and Cara Scheibling) (Avalon Health Economics, April 27, 2020)
References
1) Brooks, R.M., A civil litigator’s guide to discovery obligations in the context of electronic medical records. Health Care Law Mon, 2009. 2009(2): p. 2-8.
2) Dimick, C., E-discovery. Preparing for the coming rise in electronic discovery requests. J ahima, 2007. 78(5): p. 24-9; quiz 33-4.
3) Horn, W.S., Easing e-discovery. The electronic discovery reference model and the information management reference model. J ahima, 2010. 81(1): p. 44-6.
4) McLean, T.R., et al., Electronic medical record metadata: uses and liability. J Am Coll Surg, 2008. 206(3): p. 405-11.
5) Bowman, S., Impact of electronic health record systems on information integrity: quality and safety implications. Perspect Health Inf Manag, 2013. 10: p. 1c.
6) Hayrinen, K., K. Saranto, and P. Nykanen, Definition, structure, content, use and impacts of electronic health records: a review of the research literature. Int J Med Inform, 2008. 77(5): p. 291-304.
7) Evans, R.S., Electronic Health Records: Then, Now, and in the Future. Yearb Med Inform, 2016. Suppl 1: p. S48-61.
8) Huang, M.Z., C.J. Gibson, and A.L. Terry, Measuring Electronic Health Record Use in Primary Care: A Scoping Review. Appl Clin Inform, 2018. 9(1): p. 15-33.
9) Lee, Y.T., et al., Association between Electronic Medical Record System Adoption and Healthcare Information Technology Infrastructure. Healthc Inform Res, 2018. 24(4): p. 327-334.
10) Mills, S., Electronic Health Records and Use of Clinical Decision Support. Crit Care Nurs Clin North Am, 2019. 31(2): p. 125-131.
11) Moreno-Conde, A., et al., Evaluation of clinical information modeling tools. J Am Med Inform Assoc, 2016. 23(6): p. 1127-1135.
12) Sundvall, E., et al., Configuration of Input Forms in EHR Systems Using Spreadsheets, openEHR Archetypes and Templates. Stud Health Technol Inform, 2019. 264: p. 1781-1782.
13) Vuokko, R., et al., Impacts of structuring the electronic health record: Results of a systematic literature review from the perspective of secondary use of patient data. Int J Med Inform, 2017. 97: p. 293-303.
14) Yang, L., et al., A Graphical Representation Model for Electronic Health Records: A Preliminary Study. Stud Health Technol Inform, 2019. 264: p. 1622-1623.
15) Blijleven, V., et al., Workarounds Emerging From Electronic Health Record System Usage: Consequences for Patient Safety, Effectiveness of Care, and Efficiency of Care. JMIR Hum Factors, 2017. 4(4): p. e27.
16) Carrell, D.S., et al., Challenges in adapting existing clinical natural language processing systems to multiple, diverse health care settings. J Am Med Inform Assoc, 2017. 24(5): p. 986-991.
17) Kopanitsa, G., Integration of Hospital Information and Clinical Decision Support Systems to Enable the Reuse of Electronic Health Record Data. Methods Inf Med, 2017. 56(3): p. 238-247.
18) Opoku-Agyeman, W. and N. Menachemi, Are there differences in health information exchange by health system type? Health Care Manage Rev, 2016. 41(4): p. 325-33.
19) Zahabi, M., D.B. Kaber, and M. Swangnetr, Usability and Safety in Electronic Medical Records Interface Design: A Review of Recent Literature and Guideline Formulation. Hum Factors, 2015. 57(5): p. 805-34.
20) Zhao, J., et al., Learning from heterogeneous temporal data in electronic health records. J Biomed Inform, 2017. 65: p. 105-119.
21) Baumann, L.A., J. Baker, and A.G. Elshaug, The impact of electronic health record systems on clinical documentation times: A systematic review. Health Policy, 2018. 122(8): p. 827-836.
22) Clynch, N. and J. Kellett, Medical documentation: part of the solution, or part of the problem? A narrative review of the literature on the time spent on and value of medical documentation. Int J Med Inform, 2015. 84(4): p. 221-8.
23) Weis, J.M. and P.C. Levy, Copy, paste, and cloned notes in electronic health records: prevalence, benefits, risks, and best practice recommendations. Chest, 2014. 145(3): p. 632-8.
24) Jayabalan, M. and T. O’Daniel, Access control and privilege management in electronic health record: a systematic literature review. J Med Syst, 2016. 40(12): p. 261.
25) Saranto, K. and U.M. Kinnunen, Evaluating nursing documentation – research designs and methods: systematic review. J Adv Nurs, 2009. 65(3): p. 464-76.