Type 2 diabetes mellitus and the risk of onset of depression, a meta-analysis

Abstract

Background Due to the aging of the population and the increasing prevalence of overweight and obesity, the global prevalence of diabetes had been steadily rising. The onset of depression among this diabetes population is a public health issue because of the negative impact on diabetes self-care, treatment adherence and increases in healthcare costs.

Three previous meta-analyses have showed that diabetes patients are at an increased risk of developing depression when compared to their non-diabetes counterparts. However, none of these meta-analyses have included primary studies that consisted of primarily Chinese subjects or have considered studies published in Chinese. Thus, there is a knowledge gap for a meta-analysis to study the onset of depression among both Chinese and non-Chinese diabetes subjects.

Methods Studies will be selected based on the pre-specified inclusion and exclusion criteria. Studies that are ambiguous to this study criterion will not be selected. Studies are retrieved from three electronic databases, PUBMED, Cochrane library and Wanfang databases. Keywords of “diabet*” and “depress*” where “*” indicated a truncated term were used to identify articles related to diabetes and depression. Only longitudinal studies are selected. Chinese and English articles that are published on or before 31 July 2013 are considered. Besides data collection, quality assessment are also performed based on a tool developed with reference to the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) statement.

Results 12 longitudinal studies are identified after the quality assessment procedures, with one study being a Chinese study from Taiwan. In this meta-analysis, the pooled unadjusted odds ratio was 1.26 [95% confidence interval: 1.09, 1.45]. The pooled adjusted hazards ratio is 1.26 [95% confidence interval: 1.12, 1.41]. Since both pooled unadjusted odds ratio and pooled adjusted hazards ratio estimates did not include 1, it may imply that the increases in risk of depression onset among diabetes subjects are significant, whether controlled for confounding factors or not.

Conclusion This meta-analysis showed a significant increase in the risk of depression onset among Chinese and non-Chinese diabetes subjects, when compared to their non-diabetes counterparts. There are possible public health implications for depression screening needs for the diabetes population in Hong Kong.