Medical consumption of respondents and non-respondents to a mailed health survey

摘要: Non-response bias can distort the results of health surveys. The occurrence selective non-response be assessed when data are available for both respondents and non-respondents. objective this study was to compare medical consumption non-respondents a mailed survey. A survey conducted among approximately 13,500 adults parents 1,500 children aged 5-15 years. net response rate 70.4%. panel set that could matched with all eligible persons. This comprises administrative information on hospitalizati ons, annual care expenditures demographic variables. show associated age, sex, degree urbanization type insurance. After correcting differences in variables, differ utilization several types care. Relatively more users than non-users responded. Response not related severe conditions such as in-patient hospital conclusion from is used measure consumption, will result small overestimation utilization.