The association between body mass index (BMI) and depression is complex and controversial. The present study examined the relationship between BMI and new-onset depression during 7 years of follow-up in 20,212 adult women attending Primary Health Care Centres in Navarra, Spain.

The Atención Primaria de Navarra (APNA) study is a dynamic prospective cohort study. A total of 20,212 women aged 18–99 years (mean age: 50.7±18.5 years) without depression at baseline were selected from 2004 to 2011. We estimated the incidence of depression. We used the Kaplan-Meier analysis to predict the survival curve. The risk of depression onset according to different measures of BMI at baseline was assessed using Cox regression analyses.

During the 7 years of follow-up, depression appeared in 8.9% (95% CI 8.5–9.3). The highest rates of depression incidence at follow-up occurred in underweight and obese women (9.8% [95% CI 7.3–12.9] and 10.3% [95% CI 9.5–11.1] respectively). The distribution of depression incidence by weight category was U-shaped. The risk of depression increased over time with an observed Kaplan-Meier estimation of 6.67. After adjusting for age, underweight and obese women at baseline have increased risk of depression onset during the follow-up period compared with normal weight women (HR=1.48, 95% CI=1.09–2.00 and HR=1.14, 95% CI=1.01–1.29 respectively).

In this 7-year prospective study in the APNA women population, depression emerged in 8.9%. Being underweight or obese (not overweight) at baseline is significantly associated with future onset of depression.