Stress Predicts Development of Impaired Glucose Metabolism

Perceived stress and stressful life events predict the development of impaired glucose metabolism (IGM) over 5 years in previously normoglycemic individuals, according to results from the Australian Diabetes, Obesity, and Lifestyle study AusDiab)....

The study by Emily Williams, PhD, Monash University, Melbourne, Australia, showed that perceived stress increased the risk for incident IGM over 5 years by between 1.04 and 1.06, depending on the model used. Using the same models, high levels of stressful life events also increased the risk for incident IGM by between 1.24 and 1.35 in the same longitudinal cohort.

Dr. Williams reported that, "The effect size sounds quite small but for every point increase [in these models], there is a 4% increased risk of developing IGM, so stress is quite a strong risk factor for IGM." "And we think stress management should be incorporated into multiple health behavioral interventions for the most effective prevention and management of diabetes."

AusDiab included 11,247 adults older than 25 years who were randomly selected from 42 areas of Australia. At baseline, a 2-hour, 75-g oral glucose tolerance test was given along with the Perceived Stress Questionnaire and a life events score to measure psychosocial adversity.

At 5 years, more than 6,500 of the original participants returned for follow-up during which another 2-hour oral glucose test was taken and the questionnaires re-administered.

"We used the outcome of a polled analysis of fasting glucose, impaired glucose tolerance, and diabetes to have a larger category of impaired glucose metabolism to try and tap into a wider range of abnormal glucose metabolism, and by measuring perceived stress as well as the experience of stress, we tried to measure both objective and subjective markers of stress. Only subjects who were normoglycemic at baseline were included in the analyses," said Dr. Williams.

At 5-year follow-up, 474 subjects had progressed to IGM. Adjusting for age, sex, and education, logistic regression analyses showed that perceived stress increased the odds of IGM by 1.06. Controlling for the same variables, those reporting high levels of stressful life events were 34% more likely to have developed IGM at 5 years compared with those reporting low levels of stressful life events.

When health behaviors were added to the model, results showed that perceived stress increased the odds of IGM by 1.05. The same model also showed that those reporting high levels of stressful life events had a 35% higher risk of developing IGM compared with those reporting low levels of stressful life events. Adding obesity to the mix attenuated the effect of perceived stress as a risk factor for IGM but not by much, at an odds ratio of 1.04.

Similarly, obesity slightly attenuated the risk for stressful life events contributing to IGM at an odds ratio of 1.26. Lastly, when all variables plus traditional cardiovascular disease (CVD) risk factors were added to the analysis, perceived stress still had the same effect on IGM risk at an odds ratio of 1.04. Again, compared with those who reported low levels of stressful life events, those reporting high levels of stressful life events had a 24% greater chance of developing IGM at 5 years when analyzed in the final model.

Table 1. Perceived Stress as a Risk Factor for Impaired Glucose Metabolism

Controlling for
Odds Ratio

Model 1: Age, sex, education
1.06

Model 1 plus health behaviors (model 2)
1.05

Model 2 plus obesity (model 3)
1.04

Model 3 plus cardiovascular disease risk factors (model 4)
1.04



Table 2. Stressful Life Events as a Risk Factor for Impaired Glucose Metabolism

Controlling for
Odds Ratio

Model 1 low life stress vs high life stress
1.34

Model 2 low life stress vs high life stress
1.35

Model 3 low life stress vs high life stress
1.26

Model 4 low life stress vs high life stress
1.24

Investigators also evaluated how stress affected glycemic control over time among subjects who already had diabetes at baseline. Interestingly, said Dr. Williams, there no relationships between stress and glycemic control was observed in men, but among women with diabetes at baseline, both perceived stress and stressful life events were shown to predict elevated glycosylated hemoglobin at follow-up, after adjustment for other risk factors (P = .024).

Dr. Williams stated that, "All of the evidence in CVD suggests that stress is a key independent risk factor for the development of heart disease, but it hasn't been done in diabetes, and yet they are on the same chronic disease trajectory. "So there is no reason to think stress isn't involved in the development of diabetes too and even more so because diabetes requires so much daily management it's bound to affect a person's experience."