The results of a study on certified sickness amongst nurses in a Welsh Health District

There is a general impression given in the Occupational Health literature that certain socioeconomic variables are associated with increasing days lost from sickness absence. This writer does not wish here to take issue with these findings. Instead it is hoped to show that when certified sickness data for staff in a Health District is analysed in the search for associations between days lost and the variables age, sex, length of time in post, marital status and hours worked there is no overall convincing evidence that these variables are associated with more or less days lost by workers in the population from which these samples are taken. Calculations of the power of the study were made to compute the likelihood of the study of detecting a hypothesised degree of association as a "statistically significant" one assuming a critical p. value of 0.05.