Participation in a follow up study of health among unemployed and employed people after a company closedown: drop outs and selection bias.

This article presents an analysis of study loss in a three year longitudinal study of 1816 employees at two Danish shipyards, one of which closed just before the start of the study. The aim of the study was to elucidate the health consequences of a company closedown and the subsequent job loss. The study population responded to an identical questionnaire in 1983, 1984 and 1985. The participation rate in 1983 was 73%. The percentage of drop outs in 1984 and 1985 was 15-20% of the total numbers responding in the first year. We examined this study loss in relation to the characteristics of the respondents, using the first round of data collection as a baseline. In a multivariate analysis, age and seniority at the shipyard were the only variables that were significantly associated with non-participation in 1984 and 1985. It is concluded that the drop outs did not appear to introduce a serious risk of selection bias in 1984 and 1985. A comparison between drop outs and late respondents in 1984 and 1985 did not support the supposition that information about late respondents can be extrapolated to drop outs.