Researchers have frequently relied on case identification using clinician-based screening as the standard. This study evaluates a self-administered screening questionnaire developed for use in the Veterans Health Study. We compared concordance between elderly patients' reports of selected chronic illnesses and the medical record. The purpose of this study was to determine the feasibility of using a self-report screening questionnaire for case identification in an outcomes study of elderly respondents. Reports of the presence of selected chronic illnesses were compared in a sample of patients (N = 402) receiving outpatient care between 2 data sources, patient self-report and medical record, to determine overall concordance in 5 common chronic conditions (hypertension, adult-onset diabetes mellitus, chronic low-back pain, osteoarthritis of the knee, and chronic lung disease). Discordance between the 2 data sources varied by condition. Differences in reporting were small for diabetes and hypertension, intermediate for chronic lung disease, and larger for osteoarthritis of the knee and chronic low-back pain, where the chart did not identify substantial proportions of cases reported in the questionnaire. Use of patient-reported screening questionnaires, which are self-administered, is a valid, cost-efficient method to identify some chronic illnesses. Using medical records alone may result in underestimation of some symptom-based conditions.
Center for Health Quality, Outcomes, and Economic Research (CHQOER), Edith Nourse Rogers Memorial Veterans Hospital, Bedford, Mass (Drs Miller, Lincoln, and Kazis); the Health Services Department, Boston University School Public Health, Boston, Mass (Drs Miller and Kazis); and the Department of Mathematics, Boston University, Mass (Dr Lee).
Corresponding author: Lewis E. Kazis, ScD, Center for Health Quality, Outcomes, and Economic Research (CHQOER), Edith Nourse Rogers Memorial Veterans Hospital (152), 200 Springs Rd, Bldg 70, Bedford, MA 01730 (e-mail: Lek@bu.edu).
SF-36 is a registered trademark of the Medical Outcomes Trust.
This study was supported by grant SDR 91–006S from the VA Health Services Research and Development Service, Department of Veterans Affairs, Washington, DC.
The authors gratefully acknowledge the contributions of Holly Sims, Janet Krause, William Wagner, Brian Allard, Donna Deveau, Lillian Michalczyk, Diane Tardiff, and James McDonald who participated in the screening phase of the Veterans Health Study. Special thanks to Ann Walsh who coordinated each period of screening. We also acknowledge the very helpful suggestions of Avron Spiro III, PhD, on the earlier drafts of this article.