Depression, cognitive dysfunction and other factors associated with 5-year overall mortality in type 2 diabetes mellitus: a pilot prospective observational study

BACKGROUND: Psychological predictors of overall mortality in the Russian population of Type 2 diabetic patients and their impact compared to biological risk factors have not been studied.AIM. To identify clinical, laboratory and psychological factors independently associated with the 5-year overall mortality in Type 2 diabetic patients in the Moscow region.MATERIALS AND METHODS: This open label observational prospective study included 178 consecutive type 2 diabetic patients (women 145, men 33, age range 37 to 82 years, duration of diabetes 0,5 to 30 years). At baseline, in addition to the standard clinical, laboratory and instrumental work-up, all patients were assessed for depression, cognitive dysfunction and diabetes-related quality of life. No study-related intervention was performed; all patients were followed up and treated by their local physicians. After 5 years, we assessed the patients’ vital status (alive or dead). Multiple logistic regression was used to identify baseline patients’ characteristics, which were significantly and independently associated with 5-year overall mortality. Taking into account the exploratory type of multiple regression, the results were considered significant at α<0.1.RESULTS: At 5 years, 150 (84%) patients were alive and 15 (8,4%) were dead; no information could be obtained for the rest 13 (7,3%) patients. The analysis of 165 patients with the verified outcome, independent and significant associations with the death outcome were found for male gender (odds ratio [OR] 6,36 [95%CI 0,91–44.40]; p=0.06), age (OR 2.06 [1.30–3.27]; p<0.002), chronic heart failure (CHF) (OR 2.78 [1.25–6.2]; р=0.012), Hamilton depression scale score (OR 1,18 [1.03–1.34]; р=0.016), cognitive dysfunction score (Roschina scale) (OR 1.20 [1.05–1.35]; р=0.006), and age — body mass index interaction (OR 0,98 [0,97–0,997]; р = 0,013). The predicted probability of death within the next 5 years in men and women was 22,9% and 6,7%, respectively. The highest score of cognitive dysfunction was associated with a 25% predicted probability of death and the lowest, with a 2% probability of death; predicted probabilities of death for the highest and lowest depression scores were 26% and 2%, respectively. The 5-year predicted probability of death in the patients without CHF was 6,7%, with CHF I NYHA functional class, 9,8%, II functional class 13,6%, III functional class 18,2%, and IV functional class 23,5%. All other baseline clinical, laboratory, demographic, psychological and socioeconomic variables were not significantly associated with the 5-year survival rate. The model was not verified on an external cohort.CONCLUSION: Cognitive dysfunction and depression have a significant negative impact on the 5-year mortality rate at much higher degree, than glycemic control, any diabetes-related complications and cardiovascular disorders, excluding CHF. The results obtained highlight the importance of the diagnosis and treatment of depression and cognitive dysfunction in type 2 diabetes mellitus.