Marital status and dementia risk

Friday July 3 2009

“Being married protects you against Alzheimer's in later life,” the Daily Mail reported. The newspaper said that researchers found that people who have a partner in middle age have half the risk of developing dementia as those who live alone. However, getting divorced or becoming widowed in middle age triples the risk of dementia.

It has often been suggested that marital status has an impact on health and quality of life. This study assessed the relationship between marital status in 2,000 middle-aged people (average age 50) and their cognitive impairment about 21 years later. It found that the risk of any cognitive impairment almost doubled if a person was alone in middle age compared to having a partner. Being without a partner in both middle age and later life further increased risk.

Cognitive function was not measured at the start of the study, so it is difficult to prove that marital status is related to cognitive impairment in later life. If the association does exist, the reasons behind it are difficult to establish. The risk of dementia is likely to be governed by a complex interaction of several factors, such as lifetime personal, social and intellectual interactions, health, lifestyle and medical and genetic factors.

Where did the story come from?

The research was carried out by Krister Håkansson from the Department of Psychology at Växjö University, Sweden, and colleagues from other institutions in Sweden and Finland.

The study was funded by Kuopio University Hospital, the Academy of Finland, a grant from the EU, the Swedish Council for Working Life and Social Research, the Finnish Cultural Foundation, the Foundation of Juho Vainio, the Gamla Tjänarinnor Foundation, the Helsingin Sanomain 100-vuotissäätiö and the Gun and Bertil Stohne Foundation.

The study was published in the peer-reviewed British Medical Journal .

What kind of scientific study was this?

This cohort study assessed whether being married in middle age is related to cognitive function in later life. It used participants from a previous study called the cardiovascular risk factors, aging and dementia (CAIDE) study. These people were recruited from four separate population samples in 1972, 1977, 1982 and 1987. The four samples were randomly selected from the population register of the Kuopio and Joensuu regions of eastern Finland and consisted of 30,078 adults aged 30 to 59 (average age 50.4 years). The sample was stratified so that there were at least 250 participants from each sex and from each of three 10-year age intervals. In 1998, 2,000 surviving members of this cohort were randomly selected to be interviewed again, 1,449 of whom (73%) chose to participate. By this time, they were aged 65–70 and the average follow-up time was 20.9 years.

At the beginning of the study and then at follow-up, the researchers assessed the participants’ marital status and categorised them as single, married/cohabiting, divorced or widowed. They combined data on marital status at the two time points to create different categories of marital transition, for example whether participants were married at both times or married then widowed.

At follow-up, cognitive impairment was assessed using the mini-mental state examination (MMSE) for screening. The results of this were then used to decide whether further clinical examination and diagnostic testing (including brain imaging) were needed. Dementia was diagnosed using valid diagnostic criteria and defined as either mild cognitive impairment, Alzheimer’s disease or other forms of dementia. The researchers also used laboratory methods to investigate whether participants were carriers of the apolipoprotein E e4 allele (considered to be a risk factor for Alzheimer’s and vascular dementia). For the 551 selected participants who chose not to take part in the 1998 assessments, information on diagnoses of dementia was obtained from local hospitals and healthcare centres.

In their analyses, the researchers collapsed the middle-age marital status categories into the following: living with a partner/married, single, separated/divorced, or widowed. They combined marital status in middle-age and later life to give the following categories of marital transition: cohabiting with a partner on both occasions, cohabiting with a partner in middle age but not in later life, and living without a partner on both occasions. (The researchers said that too few people were single in middle age but cohabiting in later life to be included in the analyses).

The researchers assessed the relationship between marital status and cognitive impairment in later life, taking into account (adjusting for) other possible risk factors (assessed in middle age) of apolipoprotein E e4 allele status, education and occupation, sex and age, smoking, BMI, blood pressure, cholesterol, physical activity and depression.

What were the results of the study?

In the 1998 assessment, 294 of the 1,449 participants scored 24 or less on the MMSE (the cut-off score, below which the researchers considered there to be evidence of cognitive impairment). Through further assessment and diagnostic testing, 82 were diagnosed with mild cognitive impairment, 48 with Alzheimer’s (or 52 if different diagnostic criteria were used) and nine with other types of dementia. When data from the hospital records of the 551 people who did not participate in the later assessment were included, overall 113 people were considered to have dementia (76 of whom had Alzheimer’s) and 1,887 were considered to be without dementia.

People without a partner (single/separated or widowed) in middle age were twice as likely to demonstrate cognitive impairment in later life compared to those who had a partner. When the researchers looked at the separate categories, they found that people who were widowed at middle age had more than twice the risk of any cognitive impairment, but the risk for those who were single/separated was not significant.

Those who were single/separated or widowed in middle age and were still in the same category at follow-up had almost three times the risk of cognitive impairment compared with married/cohabiting people.

Those who were widowed at both time points had seven times the risk of Alzheimer’s disease compared to married/cohabiting people (odds ratio 7.67, 95% confidence interval 1.6 to 40.0). The presence of the apolipoprotein E e4 allele further increased the risk of Alzheimer’s disease in people who were widowed at both time points.

What interpretations did the researchers draw from these results?

The authors concluded that living with a partner may have a protective effect against the development of cognitive impairment in later life. They say that social and genetic factors may explain the large increase in risk of Alzheimer’s disease for widowed people who are apolipoprotein E e4 carriers. They note that the increased risk among widowed people compared with single people indicates that social and other factors may be involved.

What does the NHS Knowledge Service make of this study?

This research assessed the relationship between marital status in middle age (average age 50) and cognitive impairment about 21 years later in 2,000 Finnish people. This study has strengths in being reportedly one of the first studies to have examined the effect of marital relationships and their long-term influence on dementia. It found that not having a partner in middle age doubled the risk of any cognitive impairment in later life compared to having a partner. Being without a partner in both middle age and later life further increased the risk. Specifically, people who were widowed seemed to have higher risk of cognitive impairment. A few points to note:

  • The study only assessed cognitive impairment at follow-up. It is not clear whether any members of the sample already had cognitive impairment at the start of the study when marital status was assessed. Without taking into account cognitive impairment at the beginning of the study, it is difficult to conclude that marital status is responsible for impairment in later life.
  • The marital status categories used in the study may have meant some relationships were misclassified. Personal relationships are not always easy to group according to a simple convention of married/cohabiting, single/separated or widowed. In addition, such categorisation is unable to take into account all the complexities and detail of the individuals’ situations, such as how long a relationship lasted, whether it was amicable and whether the person had a close, supporting network of family and friends.

The reasons behind the possible associations are not clear. The authors suggest a “brain reserve hypothesis”, the idea that various forms of social and intellectual interaction are protective against dementia. They discuss this theory in some depth. However, it is likely to be a complex relationship between several factors, such as personal, social and intellectual interactions and health, lifestyle, medical and genetic factors, that affect the risk of dementia.

Analysis by Bazian
Edited by NHS Choices