Background: Symptoms of depression can be recurrent or limited to one episode. This study discusses the prospective association between psychological health, measured as change in depression symptoms, and the risk of diabetes mellitus in Australian women. Methods: Data obtained from the Mater-University of Queensland Study of Pregnancy. Depression was measured using the Delusions-Symptoms: States Inventory. To examine possible transitions over time, depression was grouped into four categories and assessed at different phases over the 21-year period. Multiple logistic regression models and sensitivity analysis to assess the robustness of our analytical strategy were performed. Results: Three hundred and one women reported diabetes 21 years after the index pregnancy. Almost one-third of the women who reported depression symptoms continued to report these at a subsequent follow-up (FU) phase. About 1 in 20 women who had not reported depression symptoms at the 5-year FU did so at the subsequent 14-year FU. In prospective analyses, we did not find a significant association between diabetes and negative change (not depressed to depressed, at subsequent phase); however, for women with positive history of symptoms of depression and women with persistent symptoms, there was a 1.97-fold (95% confidence interval [CI]: 1.14-3.40) to 2.23-fold (95% CI: 1.09-4.57) greater risk of diabetes. Conclusions: Our study suggests that an increased risk of diabetes is significantly associated with persistent depression symptoms. It highlights the importance of recognizing depression symptoms in terms of women's psychological wellbeing and thus provides a basis for targeting those most at risk.