Cross-sectional comparison of live and interactive voice recognition administration of the SF-12 health status survey.

ABSTRACT

OBJECTIVE: To compare interactive voice recognition (IVR) and live telephone methods for administering the SF-12 health status survey (SF-12). STUDY DESIGN: Patients with low back pain received either IVR or live interviews in a cross-sectional design with partial randomization. The interviews consisted of the SF-12 and some additional questions specific to low back pain. PATIENTS AND METHODS: Complete findings were obtainable from 229 patients. Summary scales were compared by using multivariate analysis of variance with mean comparisons for continuously scored items. Response frequencies for categorically scored items were compared by using the chi-square test. RESULTS: The 2 methods produced similar results on the Physical Component Summary scale but not the Mental Component Summary scale. Compared with patients who had a live telephone interview, the patients using IVR acknowledged significantly greater overall mental interference, greater general emotional concerns, and poorer mood and overall health. CONCLUSIONS: Because IVR eliminates the demand characteristics of responding to a personal interviewer, it may be a desirable way to evaluate sensitive topics. It also may reduce costs of data entry, labor, and measurement error.