Our results, although indicating smaller effect sizes than previous meta-analyses, do support the moderate efficacy of BIs. Further research is outlined.

When performing a meta analysis, it is often necessary to combine results from several 2 x 2 contingency tables. The Mantel-Haenszel model assumes a common measure of association between the treatment and outcome variables across the tables. A Bayesian method is described for drawing inferences regarding the measure of association, for checking the plausibility of the Mantel-Haenszel model, and for drawing inferences regarding the success rates for the individual studies. While the methodology is readily extendable to random effects models, a fixed effects approach avoids the complex statistical modelling of a mixture distribution which is required for the good application of random effects models.

Parasuicide admissions to the Regional Poisoning Treatment Centre in Edinburgh are reviewed over the seven year period 1968-74. Special emphasis is given to trends in parasuicide rates for Edinburgh City, but social and clinical data are also described. Though the mean annual increase for admissions is 10-6 per cent, recent years have shown a fall in the rates for men and a levelling off for women. There have been increases in the rates for the young, for men in social classes 4 and 5 and for divorced women, and in poisonings with psychotropic drugs and alcohol consumption among women. At the same time it is important to note variables which have not changed: the relative risks by age and sex, repetition rates, the diagnostic picture, poisoning with non-prescribed drugs, and the rank order of municipal ward rates: and variables which have diminished: the rates for divorced men, overcrowding, domestic gas and barbiturate poisoning, and drug misuse. A comprehensive explanation of parasuicide in the contemporary scene would have to explain both the consistencies and the changing trends. The answer to the central question of why parasuicide is changing remains elusive.