The trend and pattern of depression prevalence in the US: Data from National Health and Nutrition Examination Survey (NHANES) 2005 to 2016

Background: To assess the trend and pattern of depression prevalence among adult population by demographic characteristics and severity of disease from 2005 to 2016. Methods: We used six cycles of National Health and Nutrition Examination Survey. The Patient Health Questionnaire (PHQ-9) was used to measure of depression with the total score ranges 0-27. Socio-demographic variables including age, gender, marital status, race, education, and poverty income ratio (PIR) were used. The weighted prevalence was calculated for each cycle. Logistic regression was used for assessing time-trends in the prevalence of depression. Results: A total of 31,191 participants aged>20 years were included. The intensity of prevalence trend was different among subgroups of demographic characteristics. Prevalence in females was 1.5-fold more than that in males and increased approximately 2-fold during the study period. The prevalence in "never married" and "other marital status" subgroups was almost 1.5-fold and 2-fold more than that in "married" subgroup, respectively. Prevalence in the never married subgroup increased 2-fold from the first cycle to the latest one. The prevalence of depression was more in the aged<60 years, race group other than white, low education and PIR <= 1 subgroups. But, the intensity of the increasing trend was more in the subgroups of aged >= 60 years, white race and PIR>1. Limitations: the possibility of age-period-cohort (APC) effect that we were unable to assess and control them. Conclusions: In sum, there was an overall increasing trend of depression which was different by demographic characteristics based on subtype of depression.