The incidences of influenza A H1N1 and those of the subgroups whose families got ILI in both cohorts were compared by a Cox regression model and log-rank test. Results: There were 189 and 270 workers who got H1N1 infection during the follow-up period in each cohort. In this period 317 workers in Cohort 1 were asked to wait at home for several days (100% obeyed). The intervention group (Cohort 1) showed a statistically significant lower risk (p for log-rank test=0.033) compared with the control (Cohort 2), and the hazard ratio of the intervention was 0.799 [0.658-0.970] after adjusting for age, sex, BMI and smoking status. The workers who were asked to wait at home showed H1N1 infection more frequently (49 out of 317) compared with the workers whose family got ILI but were not asked to wait and work regularly (77 out of 990,). Conclusions: The waiting on full pay policy in the workplace reduced the overall risk of influenza A H1N1 by about 20% in one flu season in Japan. This kind of non-vaccine measure will be a promising option in workplaces to control the next flu pandemic. (J Occup Health 2011; 53: 287-292)

No abstract