Unexpected high mortality rates found in a review of district statistics: what action is appropriate?

The regular review of statistical and other data is now a routine public health activity in district health authorities. The interpretation of findings indicative of high or otherwise unexpected rates of mortality or disease may be complicated by small numbers, and also by questions concerning the reliability with which the available data describe the problems. Judgement about whether to undertake further investigation of these problems is subject to uncertainty about the likely effectiveness and efficiency of the investigation, and of any remedial action that would be expected to follow. Examples of these problems are explored in relation to the finding of a high rate of infant and late neonatal mortality in a district population, and a high rate of female suicide. The late neonatal death rate showed a rising trend over a period of four years and was significantly elevated above national and regional rates. The perinatal death rate was not correspondingly low, nor had there been an excess of multiple births over the period. A review of neonatal death certificates did not demonstrate any identifiable problems relating to either the cause or place of death, and the high rate subsequently reverted to the expected level without further intervention. In the case of female suicide, analyses including deaths by misadventure and open verdicts were carried out to allow for possible differences between men and women in the certification of suicide. However, this did not alter the conclusions, and further investigation of the circumstances of death is planned to assess the potential for preventive action in the locality.(ABSTRACT TRUNCATED AT 250 WORDS)