Abstract

ABSTRACT Spontaneous abortions in hospitals were analysed from two sources—membership files of the Union of Rubber and Leather Workers (about 10 000 women) and records of the personnel of a rubber factory (about 1600 women). Two frequencies of spontaneous abortions were calculated for each population analysed: rate (No spontaneous abortions X 100/No pregnancies) and ratio (No spontaneous abortions X 100/No births). The two frequencies were increased for all union members compared with all Finnish women. The frequencies, however, did not appreciably differ when the pregnancies occurred during union membership as compared with the pregnancies before or after membership. The frequency of spontaneous abortions was higher for the short-time union members than for those employed for longer periods, but the increased frequency did not correlate with union membership. The employees of a rubber factory had slightly fewer spontaneous abortions on average than the community population. The women employed in the rubber factory for three to 23 months were found to have appreciably higher frequencies of spontaneous abortions than the women employed for longer periods. The present study showed the feasibility of using cases of spontaneous abortions in hospitals in an occupational study with longitudinal employment data. Women with short periods of employment appeared to have more spontaneous abortions than those with longer periods of employment suggesting the presence of selection mechanisms, perhaps with some analogies to the “healthy worker effect” in occupational mortality studies. The presence of such selection mechanisms deserve serious consideration in occupational reproductive epidemiology.