Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Aug 31;4(1):1203.
doi: 10.13063/2327-9214.1203. eCollection 2016.

A General Framework for Considering Selection Bias in EHR-Based Studies: What Data Are Observed and Why?

Sebastien Haneuse  1 , Michael Daniels  2
Affiliations

A General Framework for Considering Selection Bias in EHR-Based Studies: What Data Are Observed and Why?

Sebastien Haneuse et al. EGEMS (Wash DC). .

Abstract

Electronic health records (EHR) data are increasingly seen as a resource for cost-effective comparative effectiveness research (CER). Since EHR data are collected primarily for clinical and/or billing purposes, their use for CER requires consideration of numerous methodologic challenges including the potential for confounding bias, due to a lack of randomization, and for selection bias, due to missing data. In contrast to the recent literature on confounding bias in EHR-based CER, virtually no attention has been paid to selection bias possibly due to the belief that standard methods for missing data can be readily-applied. Such methods, however, hinge on an overly simplistic view of the available/missing EHR data, so that their application in the EHR setting will often fail to completely control selection bias. Motivated by challenges we face in an on-going EHR-based comparative effectiveness study of choice of antidepressant treatment and long-term weight change, we propose a new general framework for selection bias in EHR-based CER. Crucially, the framework provides structure within which researchers can consider the complex interplay between numerous decisions, made by patients and health care providers, which give rise to health-related information being recorded in the EHR system, as well as the wide variability across EHR systems themselves. This, in turn, provides structure within which: (i) the transparency of assumptions regarding missing data can be enhanced, (ii) factors relevant to each decision can be elicited, and (iii) statistical methods can be better aligned with the complexity of the data.

Keywords: 2014 Group Health Seattle Symposium; Comparative Effectiveness Research (CER); Electronic Health Record (EHR); Methods; Missing Data; Selection Bias.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Summary of Weight-Related Information for 12 Patients in the Group Health EHR–Based Study of Treatment for Depression and Weight Change Gray lines indicate when an encounter occurred; blue dots indicate a weight measurement; and black lines indicate that the patient disenrolled prior to the 24-month mark.
Figure 2.
Figure 2.
Alternative Specifications for Observance of Complete Weight Data at 24 Months Post-treatment Initiation Panel (a) corresponds to the traditional, single mechanism approach to selection bias; panel (b) corresponds to one possible implementation of the proposed framework that acknowledges the complexity of EHR data.
Figure 3.
Figure 3.
Process Flow Representation of the Proposed Framework for Selection Bias in EHR-Based Studies That Can Be Used in Conjunction with Table 2
Figure 4.
Figure 4.
Summary information Regarding Disenrollment and Censoring of Patient Follow-Up During the 24-Month Interval Post-treatment Initiation among the 8,631 patients in the Antidepressants and Weight Change Study Who Had an Observed Weight Measurement at Baseline The right-hand panel shows estimates of the cumulative probability of being disenrolled and being either disenrolled or censored. The left-hand panel shows 617 patients with more than one distinct period of enrollment during follow-up, specifically the distribution of length of the first gap in enrollment.
Figure 5.
Figure 5.
Fitted Sampling Weights Obtained from the Standard Single Missingness Mechanism Framework Compared to Those Obtained from Impementation of the Proposed Framework with Three Sub-mechanisms: Active Enrollment, Initiation of an Encounter, and Recording of a Body Weight Measurement
See this image and copyright information in PMC

References

    1. Schneeweiss S, Avorn J. A review of uses of health care utilization databases for epidemiologic research on therapeutics. Journal of Clinical Epidemiology. 2005 Apr;58(4):323–337. - PubMed
    1. Weiner MG, Embi PJ. Toward reuse of clinical data for research and quality improvement: the end of the beginning? Annals of Internal Medicine. 2009 Sep 1;151(5):359–360. - PubMed
    1. Weiskopf NG, Weng C. Methods and dimensions of electronic health record data quality assessment: enabling reuse for clinical research. Journal of the American Medical Informatics Association. 2013 Jan 1;20(1):144–151. - PMC - PubMed
    1. Gallego B, Dunn AG, Coiera E. Role of electronic health records in comparative effectiveness research. J Comp Eff Res. 2013 Nov;2(6):529–532. - PubMed
    1. Institute of Medicine (US) Committee on Comparative Effectiveness Research Prioritization. Initial national priorities for comparative effectiveness research. Washington, D.C: National Academies Press; 2009.

LinkOut - more resources