Citation

Jia H, Lubetkin EI. Public Health Rep. (1974) 2009; 124(5): 692-701.

Affiliation

Department of Biostatistics, Mailman School of Public Heath and School of Nursing, Columbia University, 617 West 168th St., New York, NY 10032, USA. hj2198@columbia.edu

Copyright

(Copyright © 2009, Association of Schools of Public Health)

DOI

unavailable

PMID

19753947

PMCID

PMC2728661

Abstract

OBJECTIVES: Although numerous studies have examined health-related quality of life (HRQOL) longitudinally, little is known about the impact of seasonality on HRQOL. We examined trend and seasonal variations of population HRQOL. METHODS: We used data from the monthly Behavioral Risk Factor Surveillance System (BRFSS). We examined monthly observed mean physically and mentally unhealthy days from January 1993 to December 2006, using the structural time-series model to estimate the trend and seasonality of HRQOL. RESULTS: We found overall worsening physical and mental health during the time period and a significant and regular seasonal pattern in both physical and mental health. The worst physical health was during the winter and the best physical health was during the summer. The mean number of physically unhealthy days in January was 0.63 days higher than in July. The worst mental health occurred during the spring and fall, but the magnitude of the seasonal effect was much smaller. The difference between the best and worst months of mentally unhealthy days was approximately 0.23 days. We found significant differences in unadjusted and season-adjusted unhealthy days in many counties. CONCLUSIONS: Our findings can be used to examine time-varying causal factors and the impact of interventions, such as policies designed to improve population health. Our findings also demonstrated the need for calculating season-adjusted HRQOL scores when examining cross-sectional factors on the population HRQOL measures for continuous surveys or longitudinal data.

Language: en