Those who are widowed could have a 20% higher risk, the paper found.
That higher risk for unmarried people remained even after researchers accounted for a person's physical health, said Andrew Sommerlad, a research fellow and psychiatrist at University College London in England, who was lead author of the paper.
Yet the relationship between marriage and dementia risk is not causal: "We don't think that it is marriage itself or wearing a wedding ring which reduces people's risk of dementia," he said.
"Instead, our research suggests that the possible protective effect is linked to various lifestyle factors which are 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," he said.
Globally, about 47 million people have dementia
, and there are nearly 10 million new cases every year, according to the World Health Organization, which ranks dementia as the seventh leading cause of death
Alzheimer's disease may contribute to up to 70% of those cases
, according to the WHO.
Alzheimer's affects an estimated 5.4 million Americans, making it the sixth leading cause of death
among all adults and the fifth leading cause for those 65 and older, according to the US Centers for Disease Control and Prevention.
Link between marriage and dementia
For the paper, the researchers systematically searched databases for previous studies on marital status and dementia. In some of the studies, cohabiting couples were classified as married.
"Not all the studies gave information on this, but it is usual in such studies for non-married cohabiting partners to be classified as married," Sommerlad said. "We would therefore expect cohabiting people be similar to married."
The researchers identified, reviewed and analyzed 15 studies published between the inception of the databases and December 5, 2016.
In total, the studies included 812,047 people. Some were conducted in European countries, Asia and the US, and one was from Brazil.
After analyzing the studies, the researchers found that people who had been single all their lives and those who were widowed were more likely to develop dementia compared with those who were married at the time of the studies, despite their age and sex.
That increased risk appeared to be similar to other known dementia risks, such as having diabetes
or high blood pressure
, Sommerlad said.
The researchers found no evidence that dementia risk in divorced people differed from those who were married, and they could not examine whether the duration of being widowed or divorced had any influence on the findings.
With pooled data from multiple studies, the new paper had adequate power to test the hypothesis that marriage could impact dementia risk, Dr. Bryan Woodruff, a neurologist at Mayo Clinic in Phoenix, wrote in an email.
However, "the findings for lifelong single and divorced participants should be interpreted with some caution since they were a much smaller proportion of the sample studied," said Woodruff, who was not involved in the new paper but has researched widowhood and dementia
The researchers also found that the dementia risk for lifelong single people is possibly reduced in recent years, since "single people born during the first quarter of the 20th century had a 40% higher risk than married people, whereas later studies only found a 24% higher risk," Sommerlad said.
"As being unmarried becomes more of a social norm, it is likely that lifestyle differences between married and unmarried people are lessening," he said.
All in all, "our findings, from large populations, across numerous countries and time periods are the strongest evidence yet that married people are less likely to develop dementia," the researchers wrote in the paper.
Yet the study had some limitations.
"In this type of study, we cannot be certain about the explanations underlying these findings, so further research should examine the contribution of social contact and health behaviors, and consider how long after widowhood is dementia risk higher," Sommerlad said.
In other words, although the study found an association between being married and reduced risk of dementia, the findings can't prove that marriage directly decreases dementia risk.
Importance of staying socially active
"While this study suggests that married or cohabiting people are less likely to develop dementia, that doesn't necessarily mean that getting married -- or a partner -- will reduce a person's risk for dementia," said Keith Fargo, the Alzheimer's Association's director of scientific programs and outreach, who was not involved in the new paper.
Rather, "there are other factors associated with being married or unmarried that may play a role," he said.
Those so-called modifiable risk factors
include diet, physical activity, smoking, sleep and social engagement.
"The evidence continues to accumulate from multiple studies that eating well, exercising regularly, getting enough sleep and pursuing mentally stimulating activity is good for everyone in terms of reducing risk of dementia," Woodruff said.
For instance, engaging in regular exercise that elevates your heart rate, playing games that make you think strategically and volunteering in your community are on the Alzheimer's Association's list of 10 lifestyle habits to reduce risk of cognitive decline
"If social engagement is an important aspect of the protective effect of marriage identified in the study, single individuals should make an effort to avoid social isolation," Woodruff said.
Staying socially active probably relates to the findings in the new paper, Fargo said.
"The advice that the Alzheimer's Association gives, which I do want to point out is the same for single people or married or cohabiting people, is to increase the amount of social activities that they have," he said. "One of the ways that we do that is through marriage or through having a partner that we live with, but it's certainly not the only way that it can be done."
To examine how such modifiable factors could influence dementia risk, Fargo said, the Alzheimer's Association plans to launch a large two-year, $20 million clinical trial called US POINTER
"At the end of that study, we should be able to say some much more strong things about the effect of modifiable risk factors and dementia risk," he said.
It doesn't 'mean that dementia is easily preventable'
Christopher Chen, an associate professor at the National University of Singapore
, and Dr. Vincent Mok, a professor at The Chinese University of Hong Kong
, wrote an editorial commentary
accompanying the new study in the Journal of Neurology, Neurosurgery and Psychiatry on Tuesday.
They say the new paper sheds light on how both marriage and dementia are culturally and socially determined as well as intertwined.
"The institution of marriage is undergoing rapid changes with the acceptance of same-sex marriages and alternatives to marriage such as cohabitation. Moreover, the known effect of race and ethnicity as well as education and income on marriage rates is changing the proportion of the population getting married," Chen and Mok wrote.
"It is also important to acknowledge that dementia and its prodrome has a huge impact on marriage, leading to transition and loss," they wrote. "Hence, although potentially modifiable risk factors for dementia exist, this does not mean that dementia is easily preventable."