|
|
|
|
|
Aman Bhandari, MPH, Health Policy and Management, University of California, Berkeley, 3333 California St. Suite 265, San Francisco, CA 94143, 415-514-1596, aman1@berkeley.edu and Todd Wagner, PhD, VA Palo Alto Health Care System and Stanford, 795 Willow Rd., 152-MPD, Menlo Park, CA 94025.
Objective: Self-report methods are often used for capturing healthcare utilization. The implementation of HIPAA, growth of multi-site trials, and relatively low cost of collecting and cleaning self-report data makes this method attractive. However, the accuracy of such data, especially given its use for estimating healthcare costs, is of paramount concern. Our purpose was to conduct a systematic review of studies that have evaluated the accuracy of self-report utilization data and to provide a conceptual model identifying major issues to consider when collecting, analyzing and reporting self-report utilization data. Design: A comprehensive literature search was conducted using several databases. The following keywords were used to identify relevant articles: interviews, questionnaires, recall, self assessment, self-report, survey design and utilization or healthcare utilization. 32 articles that assessed the accuracy of self-reported healthcare utilization were reviewed. Findings: Questions about accuracy are the major problems with self-reported healthcare utilization data, and under-reporting seems to be the most common error. As recall time increased, under-reporting and over-reporting problems increased, but under-reporting was substantially greater at 12 months. Additionally, the evidence suggests age as the predominant factor impacting accuracy. A number of techniques can be used to increase recall and survey accuracy. Conclusions: Self-report surveys of utilization are being increasingly used and are of variable accuracy. In designing and analyzing self-report surveys, researchers should consider the following factors: (1) respondent cognitive abilities, (2) recall timeframe, (3) utilization type, (4) utilization frequency, (5) questionnaire design (6) data collection mode, and (7) the type of statistical analyses.
Learning Objectives: After reviewing the poster, the participant (learner) in this session will be able to
Keywords: Methodology, Survey
Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.