Can mortality over 25 years be predicted in healthy middle-aged women?

Objective It is assumed that psychosocial, psychophysiological and behavioural factors can limit the life expectancy of women who do not suffer from a serious illness. We investigated whether these factors are related to long-term survival. Methods The study group consisted of 300 presumably healthy women aged 30 to 65 years (mean age 57.5 years) randomly selected from the normal population. Their behavioural and psychosocial risk profiles as well as their exercise, smoking, sleep and dietary habits were recorded using standardised procedures. Depressive symptoms were measured by the Pearlin questionnaire. Time and frequency domain measurements were performed using 24-h ECG. At follow-up after 26,1 years (range 25.3-28.3), all-cause mortality was assessed in 299 (99.7%) of the 300 women. In a new exploratory analysis using machine learning algorithms, we compared the effects of psychosocial and behavioural mechanisms with other risk factors. Results We found that 101 (33.9%) of the previously healthy women had died. A sedentary lifestyle or regular intensive exercise (HR 1.62 (95% CI 1.10-2.40; p = 0.016) and HRV without cardiac arrhythmia (HR 0.20, 95%CI 0.81-0.48; p = 0.001) were found to be important predictors in the model. Current smoking, depressive symptoms, SDNN and C-reactive protein were included in the predictive model but were not found to be significant. Social integration, diet and sleep disturbances were not included in the model. Conclusion The data confirmed our hypothesis about the influence of behavioural and psychophysiological risk factors on long-term survival. Targeted preventive measures in various areas identified here could contribute to a prolongation of life.