Absence from work in relation to length and distribution of shift hours

Walker, J., and de la Mare, Gwynneth (1971). Brit. J. industr. Med., 28, 36-44. Absence from work in relation to length and distribution of shift hours. A long period on night shift or even permanent night work has sometimes been suggested for those on continuous shift work to allow circadian rhythms to adapt. As the weekly hours of work have been reduced there is some evidence that a permanent night shift is practical, and about 12% of all shift workers are on this type of work. However, the case for permanent night shift must be established on grounds of both effectiveness and acceptability. The present studycompares the absence experience, including sickness absence, ofpermanent day workers and permanent night workers matched for age and job in three undertakings which contained a range of working conditions. The question of the relationship between absence from work and total hours worked including overtime has been reopened, and in comparing absence from work according to the type of shift the total hours worked must also be taken into account. The relationship between the average hours when a man was at work and the amount of absence was tested. The men in the three undertakings worked a wide range of voluntary overtime. The results showed that in two undertakings long-term absence, mainly sickness absence, was higher on the night shift than on the day shift; and, in the third, absence was about the same on the two shifts. As the work load was less in the two undertakings with a higher absence on the night shift it was suggested that selective factors were operating. These results may be contrasted to studies which have compared the absence of rotating shift workers and day workers. In all three undertakings there was a tendency for absence to be less among high overtime workers than among those who worked medium or small amounts of overtime, although the trends were not consistent. There was no evidence at all that high overtime and absence from work were positively associated. The implications of these results are discussed.