Wellness in a Heartbeat

 

People who are married are less likely to develop dementia than those who have never been married or are widowed, new research suggests.

 

A large meta-analysis showed a 42% higher risk for dementia in lifelong singles and a 20% higher risk in widowed people, compared to those who were married. The association did not apply to divorced individuals.

 

"Our research suggests that the possible protective effect is linked to various lifestyle factors known to accompany marriage, such as living a generally healthier lifestyle and having more social stimulation as a result of living with a spouse or partner," lead author Andrew Sommerlad, MD, a geriatric psychiatrist and Wellcome Trust research fellow at University College London, United Kingdom, told fellow researchers.

 

"Doctors should be aware of this increased risk and have a higher level of suspicion of dementia in unmarried people," he added.

 

The study was published online November 28 in the Journal of Neurology, Neurosurgery and Psychiatry.

 

"Marital status has the potential to affect dementia risk by increasing daily social interaction," which may "improve cognitive reserve," the authors write.

 

On the other hand, bereavement or divorce in people once married "may promote dementia development through stress, which is pathogenic and associated with increased dementia risk." Moreover, the authors note, "being unmarried is associated with adverse health behaviors and a range of poorer health outcomes."

 

Previous studies have pointed to a protective effect of marriage in improving cancer survival. Widowhood is associated with increased disability in older people, and unmarried individuals have been shown to be less healthy than their married counterparts.

 

"We were interested in identifying changes that people may be able to make in their lifestyles to reduce their risk of dementia, and we wondered whether having more social contact may reduce dementia risk," said Dr Sommerlad.

"We thought that marriage might help us to answer this, as married people are likely to have more interaction with people throughout their lifetimes."

 

To further explore this question, the researchers decided to "synthesize" previous research and identified 15 studies that met their inclusion criteria.

 

Studies were required to use prospective or retrospective cohort (n = 9), case-control (n = 2) or cross-sectional (n=4) design. Marital status had to be reported separately from other aspects of social network. At least half of the included study were required to consist of individuals aged 65 years or older at the time of dementia ascertainment or to include stratified data about the individuals in that age range. Participants had to be drawn from a community-dwelling population.

 

The mean age of participants at study inception was 72.8 years (SD, 7.2 years). In the cohort studies, the duration of follow-up before dementia assessment was 3 to 20.9 years (mean, 8.5; SD, 5.5).

 

Of the studies, eight were conducted in European countries, four in Asia, two in the United States, and one in Brazil. Of the total number of people included in the studies (n = 812,047), 29,610 had any form of dementia.

 

Married people accounted for 27.8% to 80.1% of the sample. Widowed, divorced, and lifelong single people accounted for 7.8% to 48.0%, 0% to 16%, and 0% to 32.6% respectively.

 

The main meta-analysis compared widowed, divorced, or lifelong single people to married people in the risk for all-cause dementia. In nine studies, widowed people were found to have a 1.20 relative risk (RR) of dementia (95% confidence interval [CI], 1.02 - 1.41), compared with married people. In seven studies, the RR for divorced vs married people was 0.99 (95% CI, 0.71 - 1.37).

 

In the six studies that examined dementia risk in lifelong single people, the RR was 1.42 (95% CI, 1.07 - 1.90), compared to married people.

 

In studies that focused on Alzheimer's disease, the pooled RR was 1.24 (95% CI, 0.97 - 1.60) in widowed vs married people. There were no differences between married and lifelong single or divorced people. There was no effect of marital status on vascular dementia risk in pooled estimates from the three studies that examined it.

 

No significant differences by sex were found. However, for dementia risk in widowed vs married people, the pooled risk estimates from the three studies that adjusted only for age and sex was higher than the five that adjusted additionally for education or baseline cognitive function.

 

There was no further attenuation of the effect in the three studies that also adjusted for physical health. For lifelong single people, the risk estimate for dementia was unaffected by adjustment for education, although the RR of dementia in single vs married people decreased from 1.45 (95% CI, 0.97 - 2.19) to 1.23 (95% CI, 1.17 - 1.29) in studies that adjusted for physical health.

 

"We find that much of the increased risk in widowed people is attenuated after adjustment for education and that confounding by physical health explains part of the increased risk of dementia in lifelong single people," the authors note.

 

The investigators caution that the findings are limited by the methodology of the studies analyzed. For example, they were unable to investigate the impact of the duration of being widowed or divorced, because the studies did not include this information.

 

"A possible explanation for the link between marital status and dementia is that particular personality characteristics underlie the tendency not to marry, as well as the later risk of dementia," Dr Sommerlad said.

He noted that there were fewer divorced people than widows or singletons in the studies, which might account for the difference between levels of risk in these groups.

 

Additionally, "stress contributes to dementia risk, and previous research has suggested that loss of a spouse is the most stressful event that we might experience."

 

He acknowledged being surprised by the strength of the findings and noted that the higher risk for unmarried people remained, even when physical health was taken into account, "suggesting that the benefit of marriage is due to more than just improving physical health."

 

Commenting on the study Christopher Chen, MD, associate professor, Departments of Pharmacology and Psychological Medicine, Yong Loo Lin School of Medicine, National University of Singapore, said that the study's implications are important because they "suggest possible factors that may help prevent dementia."

 

For example, "if married people have a healthier lifestyle and more social engagement than lifelong single or widowed persons, then measures to improve such factors, regardless of marital status, may be beneficial," said Dr Chen, coauthor of an accompanying editorial.

 

Moreover, "if the stress of widowhood increases the risk of dementia, then reducing these effects may be a helpful preventive measures. ... The mechanisms underlying the association of marriage and reduced dementia risk suggests that potentially modifiable risk factors for dementia exist, and it remains important to obtain evidence from well-designed clinical trials," he added.

 

Until then, "ways of destigmatizing dementia and producing dementia-friendly communities more accepting and embracing of the kinds of disruptions that dementia can produce should progress alongside biomedical and public health programs."

 

Dr Sommerlad emphasized that although dementia can be difficult to diagnosis in people who "attend clinics alone, steps should be taken to ensure thorough assessment in those who may have the condition."