[Sex-differential mortality and causes of death in a Swedish city 1750-1850].

The secular decline of general mortality in Sweden began about 1810. For men and women in working ages the death rate trend diverged however, leading to increased excess male mortality during the first half of the century. One important issue concerning this process is to analyse how changes in the sex-specific incidence of different causes of death contributed to this divergence in mortality. Unfortunately relevant demographic data (number of deaths from certain causes according to sex and age and related to population at risk) for such an analysis have not (to any larger extent) been compiled on the national level, but is available for practically all Swedish parishes from the middle of the eighteenth century, at least to 1830 (and in many cases even thereafter). This case study investigates the sex-differential mortality generally and for different causes of death and the change in the disease panorama for men and women (15-59 years of age) in the town of Linköping during the "pre-transitional" period 1750-1814 and the following period 1815-1849. The primary sources are the nominal registers of deaths and funerals for the town. The results are compared with data from two rural parishes in the area surrounding Linköping as well as with national data. Problems concerning the interpretation of historical statistics on causes of death are also discussed. According to the information given by the parish registers, marked male excess mortality was the case for the majority of the cause of death categories used for this study. Changes in cause-specific mortality which to a large extent contributed to the widening gap in death rates between adult men and women in Linköping during the first decades of the nineteenth century were increased male excess mortality in lung consumption (lungsot), hygiene related diseases or symptoms (nerv- och rötfeber, rödsot etc.), accidents, stroke (slag) and the vague and ambiguous category wasting disease (tärande sjukdom) and also a considerable reductin in maternal mortality. At least the changes of the sex-specific mortalilty patterns for accidents and maternal mortality seem to have been more general phenomena. It is not possible to make universal generalizations from local studies like this, but it can help in framing questions and generating hypotheses for further studies. Some questions at issue raised by this investigation are, for example, the impact of alcoholic consumption for the increased incidence of accidents among men and for other causes of death such as lung consumption and stroke as well as the role of midwifery for the decrease in maternal mortality.