Relationship between sickness absence and meteorological factors.

Sickness absence rates are higher in winter than in summer and also the incidence of many diseases is affected by climatic conditions. Both these facts lead one to suspect that there is a genuine causative relationship between the weather and sickness absence. The object of this investigation is to develop a model to illustrate this relationship in order to obtain some idea of its precise form. Any effect of weather on the weekly incidence of sickness absence can be assumed to take one of three forms: (i) a steady long-term cumulative effect whereby the more adverse weather conditions in winter eventually affect the health and absence behaviour of employees; (ii) a more short-term effect whereby deviations of weather conditions about the seasonal norm may influence the sickness absence frequency in the same or subsequent weeks; (iii) the spread of infectious disease, whereby one week's sickness absence frequency is determined to some extent by the amount of infection brought over from previous weeks. Though waves of infection are influenced by weather conditions (e.g., they are more frequent in winter), their occurrence is largely unpredictable and they must be considered separately from (i) and (ii). The regression model described below is designed to assess the importance of the effects of types (ii) and (iii) after the elimination of the broad seasonal trends indicative of effects of type (i). This model is illustrated by sickness absence data from one organization.