Midlife Stressors May Increase Risk of Dementia in Women

From AccessMedicine from McGraw-Hill
S. Andrew Josephson, MD

08/27/2010; AccessMedicine from McGraw-Hill © 2010 The McGraw-Hill Companies

Dementia is an important health concern that has become even more prevalent with an aging population. Identification of modifiable risk factors for dementia is a key goal of research given the potential to prevent this potentially devastating and costly disorder. Some lines of evidence point to stress and its relation to cognitive decline as one such risk factor, perhaps through activation of glucocorticoid pathways. A recent study (Johansson et al, 2010) used a unique data set in order to explore whether stress indeed is a risk factor for the development of dementia.

The authors performed a prospective longitudinal study using a cohort of women enrolled in the Prospective Population Study of Women in Gothenburg, which began in 1968 by enrolling 1462 patients born from 1908 to 1930. All nondemented subjects were asked about stress as part of a standard evaluation in 1968, 1974, and 1980.
Participants were asked if they had experienced any period of stress greater than 1 month in relation to work, health, or family situation.
Stress was defined for the subjects as feelings of irritability, tension, nervousness, fear, anxiety, or sleep disturbance. Responses included no stress, one period of stress during the last 1 year or 5 years, multiple periods of stress during the last 5 years, or constant stress during the last 1 or 5 years.
Dementia was diagnosed in these patients using standard criteria and serial neuropsychological examinations including informant interviews. Multiple possible confounders were recorded, including education, marital status, vascular risk factors, and weight.

Frequent/constant stress within 5 years was reported in 20% of the patients in 1968, 23% in 1974, and 15% in 1980. Of the 1415 patients assessed in 1968, 161 (11%) developed dementia over a 35-year follow-up period. A total of 105 dementia cases were thought to result from Alzheimer's disease (73 of whom had coexisting cerebrovascular disease), 40 from pure vascular dementia, and 16 from other causes of dementia. The risk of dementia was increased in patients reporting frequent/constant stress in 1968 [hazard ratio (HR), 1.60; 95% confidence interval (CI), 1.10–2.34], in 1974 (HR, 1.65; 95% CI, 1.12–2.41), and in 1980 (HR, 1.60; 95% CI, 1.01–2.52). Adjustments for multiple confounders did not change the results. Occasional stress was not associated with risk of development of dementia.

Patients with frequent/constant stress were found to have an increased risk of developing Alzheimer's disease, with and without coexisting cerebrovascular disease, but not an increased risk of pure vascular dementia. Associations held true even when examining dementia onset before and after age 70 separately as well as with exclusion of those with onset of dementia before 1992 in an attempt to exclude those whose stress may have been an early manifestation of incipient dementia.

This intriguing study certainly has a number of limitations.
Patients' ability to cope with stress or view a situation as stressful may be variable; this variability may or may not be related to underlying neuropathology.
A single question probably does not robustly assess life stress; however, recall bias is certainly diminished using this methodology of direct questioning at different time points rather than relying on recall of stress many, many years prior.
For the clinician, this study suggests that midlife stress may indeed serve as a risk factor for the development of dementia later in life. Whether mitigation of this stress will help prevent some forms of dementia or delay their onset remains to be seen in future studies.

References