The relationship between prior psychiatric diagnosis and brain cancer diagnosis in the U.S. military health system.

ABSTRACT

PURPOSE: Prior research suggested the increased likelihood of brain cancer diagnosis following certain psychiatric diagnoses. This association may result from detection bias or suggest an early sign for brain cancer. This study investigated whether psychiatric illness may be an early manifestation of brain cancer while considering potential effects of detection bias. METHODS: This case-control study used the data from the Department of Defense's Central Cancer Registry and the Military Health System Data Repository. Four cancer-free controls and one negative-outcome control (cancers not associated with psychiatric illness) were matched to each brain cancer case diagnosed from 1998 to 2013 by age, sex, race, and military status. The groups were compared in the likelihood of having a pre-existing psychiatric diagnosis using conditional logistic regression. RESULTS: We found a significant association of psychiatric illnesses with brain cancer (Odds Ratio (OR) = 2.63, 95% confidence interval (CI) = 2.18-3.16) and other cancers (OR = 1.80, 95% CI = 1.49-2.19), compared to non-cancer controls. The association was stronger for psychiatric diagnoses within three months before cancer (brain cancer OR = 26.77, 95% CI = 15.40-46.53; other cancers OR = 4.12, 95% CI = 1.96-8.65). The association with psychiatric disorders within 3 months were higher for small brain tumors (OR = 128.32, 95% CI = 17.28-952.92 compared to non-cancer controls) while the OR was 2.79 for other cancers (95% CI = 0.86-8.99 compared to non-cancer controls). CONCLUSION: Our findings suggest an association between diagnosed psychiatric illnesses and subsequent brain cancer diagnosis, which may not be solely explained by detection bias. Psychiatric illness might be a sign for early detection of brain cancer beyond the potential effects of detection bias.