Because poor sleep quality can reduce quality of life and increase prevalence of illness in workers, interventions are becoming increasingly important for businesses. To evaluate how sleep quality is affected by one-on-one behavioral modification when combined with group education, we conducted a randomized, controlled trial among day-shift white-collar employees working for an information-technology service company in Japan. Participants were randomly allocated to groups receiving either sleep hygiene group education (control group), or education combined with individual sleep modification training (one-on-one group). Occupational health professionals carried out both procedures, and sleep quality was assessed using the Pittsburgh Sleep Quality index (PSQi). PSQi scores were obtained before and after the intervention period, and changes in scores were compared across groups after adjustments for age, gender, job title, smoking and drinking habits, body-mass index, and mental health as assessed using k6 scores. The average PSQi score for the control group decreased by 0.8, whereas that of the one-on-one group decreased by 1.8 (difference of 1), resulting in a significantly greater decrease in score for the one-on-one group (95% confidence interval: 0.02 to 2.0). These results show that, compared to sleep hygiene group education alone, the addition of individual behavioral training significantly improved the sleep quality of workers after only three months.

To evaluate the effects of a combined sleep hygiene education and behavioral approach program on sleep quality in workers with insomnia, we conducted a randomized controlled trial at a design engineering unit in Japan. Employees evaluated for insomnia by the Athens Insomnia Scale (≥6 points) were divided into an intervention and control group. The intervention group received a short-term intervention (30 min) program that included sleep hygiene education and behavioral approaches (relaxation training, stimulus control, and sleep restriction) performed by occupational health professionals. We calculated differences in change in Pittsburgh Sleep Quality Index (PSQI) scores between the two groups from baseline to three months after the start of intervention after adjusting for gender, age, job title, job category, average number of hours of overtime during the study period, marital status, smoking habit, average number of days of alcohol consumption per week, exercise habits, K6 score, and baseline PSQI score. Results showed that the average PSQI score decreased by 1.0 in the intervention group but increased by 0.9 in the control group. Additionally, the difference in variation between the two groups was 1.9 (95% confidence interval: 0.6 to 3.4), which was significant. Taken together, these results indicate that the intervention program significantly improved the sleep quality of workers with insomnia.

Relationship between Turnover andPeriodic Health Check-Up Data among Japanese Hospital Nurses: A Three-year Follow-Up Study: Takashi SHIMIZU, et al. University of Occupational and Environmental Health-The present study explored the relationship between turnover and periodic health check-up data among Japanese hospital nurses. The subjects were 379 registered nurses in a Japanese hospital and the duration of our study period was three years. By the proportional hazard model, we investigated the relative risks (RRs) of the turnover rate associated with the following: lifestyles (irregularity of meals, taking breakfast), work environment (working hours, work schedule), and health status (body mass index [BMI], serum total cholesterol [T-cho], taking medication, sleep disturbance). We obtained the data on turnover from October, 1997 to September, 2000. During the observed period we were able to follow up on 363 female nurses (96%) and the turnover total amongst them was 100. We divided the subjects into a younger (21-25-yr-old) and an older (>25-yr-old) group because of the interaction between age and turnover. In the younger group, we found that sleep disturbance influenced turnover significantly. On the other hand, in the older nurses, BMI and T-cho contributed to turnover significantly. Our results imply the possibility of a partial association between turnover and periodic health check-up data among the nurses. (J Occup Health 2005; 47: 327-333)