Comparison of health estimates between Victorian Population Health Surveys and National Health Surveys

Abstract

Abstract Objective: The computer‐assisted telephone interviewing (CATI) method has grown rapidly in recent years as an epidemiological tool for obtaining data on health issues. However, it is still argued that coverage, methods and lower response rates from CATI surveys compared with face‐to‐face interviewing m ay have an impact on the validity of the health estimates obtained. This paper compares demographic and health estimates from the Victorian Population Health Survey (VPHS) 2001, using CATI, with the National Health Survey (NHS) 2001 that is based on face‐to‐face interviews. Methods: The profiles of the VPHS (2001) and NHS (2001) respondents were compared, specifically demographic characteristics (age, gender, marital status, country of birth, and employment status) and health‐related estimates such as self‐reported health status, prevalence of diabetes and asthma, smoking status, daily intake of fruit, categories of psychological distress and private health insurance status. Results: In both surveys, the demographic characteristics and health‐related estimates of the adult population were remarkably similar as shown by the similarity of weighted estimates and overlapping confidence intervals. Conclusions: The degree to which derived estimates correspond in the two independent surveys lends support to the reliability of the CATI method used for collection of routine population health related data.