Background: The question of whether social and psychological factors contribute to cancer etiology has attracted attention for centuries. Although evidence has identified genetic, environmental, lifestyle, and socioeconomic factors as potentially increasing the risk of cancer, the contribution from social and especially psychological factors has been questioned, especially due to lack of prospective studies. The goal of this study was to investigate, in a longitudinal setting, the association between risk of cancer and measures of self-reported social network, self-reported health (physical and mental), and quality of life.
Methods: In 1993, 4488 cancer free individuals aged 31-33 years from The Copenhagen Perinatal Cohort were asked to rate their social network, their physical - and mental health, and quality of life. The study population was followed until end of 2006 in the Danish Hospital Discharge Register to obtain information on registration with a cancer diagnosis. Risk estimates were calculated using Cox proportional hazard regression.
Results: During the follow-up period, 102 individuals were diagnosed with cancer. After adjustment for age, sex, income, lifestyle factors, and other diseases, individuals rating their quality of life to be poor had a hazard ratio of 1.90 (95% CI=1.1-3.4) for cancer compared to individuals with a high quality of life. Individuals rating their quality of life to be good had an age adjusted hazard ratio of 1.31 (95% CI=0.8-2.2). Inserting a time-lad of three years to minimize reverse causation did not change the estimates notably. Self-rated physical and mental health were significantly associated with the risk of cancer, but these estimates became insignificant after adjustment for confounding factors. Social network was not associated with risk of cancer.
Conclusion: In this study, with a relatively strong design, the risk of cancer was almost doubled in individuals rating their quality of life to be poor compared to individuals with the most positive rating of their quality of life. Our results suggest that broad assessment of general well-being as assessed with global quality of life self-rating seems to be a better predictor of cancer risk than more specific information on social network and self-rated health.