Perinatal mortality after Chernobyl in former Soviet countries.

DATA AND METHODS: Perinatal mortality data for the FSU countries are available in the Health for All database (WHO). In this study, data from Ukraine, Belarus, the Russian Federation, Moldova, and Estonia are analyzed. The regression model uses a long-term exponential trend with flexible time dependence and two superimposed bell-shaped terms (Model 1). In a second approach, the bell-shaped excess terms are replaced by the inverse of gross domestic product per capita (the GDP term), which serves as a proxy for the possible impact of the socio-economic crisis after the collapse of the Soviet Union. The possible strontium exposure of pregnant women (the strontium term) is added as a second covariate (Model 2). RESULTS: Model 1 fitted the data of all five countries well. The observed increases in PM rates in the 1990s were greater in Belarus than in Ukraine and Russia. Model 2 regressions also fit the data well, except for Ukraine. In Belarus, Russia, and Moldova, the GDP term alone explained the deviation of PM rates from the predicted trend; adding the strontium term did not significantly improve the fit. Only in Ukraine and Estonia was the effect of the strontium term statistically significant. In 1987, increases in PM were found in all countries except Estonia, where PM peaked in 1988. CONCLUSION: The deviation of PM rates in the 1990s from the long-term trend is related to GDP per capita. An effect of the strontium term is detected only in Ukraine and Estonia, where sharp increases in PM were observed in the early 1990s, well before the trough of GDP in the second half of the 1990s.