Does unemployment cause the death rate peak in each business cycle? A multifactor model of death rate change

Abstract

Natural time series and prospective studies are combined to determine the contribution of many causal factors to the business cycle variation of the death rate. The variation of housing and nutrition together accounts for roughly a tenth of the death rate fluctuation. Drug consumption accounts for about one-sixth, with 11 percent of the total variation due to alcohol and 6 percent due to cigarette smoking. Social relationship changes, both as sources of stress and as means of relief, account for the greatest part (72 percent) of the business cycle variation of the death rate.