Low Level of Social Integration Tied to Suicide in Women

Pauline Anderson

July 29, 2015

Women with a lot of social ties and who participate regularly in social groups have substantially lower rates of suicide than those who are less socially integrated, even after adjusting for poor mental health, a new study suggests.

In this analysis, the statistically significant association between social integration and suicide was even stronger than that found in a recently published similar study among men by the same authors.

The current study has important implications for suicide prevention, said the authors, led by Alexander Tsai, MD, PhD, Center for Global Health, Massachusetts General Hospital, Boston. Although it is important to ask patients about participation in social relationships to individually target assessment and treatment, a public health approach to suicide prevention may also be beneficial, they write.

The study was published online July 29 in JAMA Psychiatry.

Optimal Prevention

Researchers analyzed data from the US Nurses' Health Study, an ongoing prospective cohort study of women who were aged 30 to 55 years when the study began in 1976. The current analysis included 72,607 women for whom follow-up data were available from 1992 to 2010.

Investigators measured social integration using a seven-item index that was added to the Nurses' Health Study survey in 1992, when study participants were 46 to 71 years old, and was administered again in 1996. The index includes questions about marital status, size of social network, and participation in religious or other social groups. From responses to these questions, individuals were categorized into four groups: I (lowest level of social integration) to IV (highest level).

The social integration index had a mean value of 6.62. Most participants fell into the highest category of social integration.

During 1,209,366 person-years of follow-up, there were 43 suicides. The most common means of suicide was poisoning, followed by firearms and explosives, and strangulation and suffocation.

Even after adjusting for age, employment status, body mass index, physical activity, alcohol intake, smoking status, and history of diabetes, hypertension, and high cholesterol, the hazard of suicide was lowest among individuals in the highest category of social integration (adjusted hazard ratio [AHR], 0.23; 95% confidence interval [CI], 0.09 - 0.59) and the second highest category (AHR, 0/26; 95% CI, 0.09 - 0.74). There was a decreasing trend across the categories (P = .001).

The findings were reinforced when comparing women assessed in both 1992 and 1996. Participants who had the highest level of social integration in both 1992 and 1996 had a reduced hazard of suicide during the subsequent 14 years of follow-up (AHR, 0.15; 95% CI, 0.03 - 0.65); other trajectories of social integration (for example, no change in or decreasing social integration) were less protective. There was a decreasing trend across trajectory categories (P = .03).

Sensitivity analyses that variously looked at social integration as a continuous variable, that excluded persons with a significant health issue, such as cancer or a serious cardiovascular condition, and that removed those with poor mental health status did not substantially alter the findings.

Community-based interventions "may be able to reduce the overall burden of suicide more effectively than intensive efforts focused on 'high risk' individuals," the authors note.

"Clearly, the 'high-risk' and 'population-strategy' approaches each have their advantages and disadvantages, and the optimal prevention strategy likely requires a judicious mix of both," they write.

A weakness of the study was that participants in the Nurses' Health Study are racially, generationally, and socioeconomically homogeneous. As well, the number of suicides was relatively low.

Another drawback was that the social integration index fails to capture relationship quality and other aspects of social integration. And because chronic conditions were less prevalent among women who were socially well integrated, it is possible that social integration is a proxy for good health.

Elegant Perspective

In an accompanying editorial, Eric Caine, MD, Injury Control Research Center for Suicide Prevention, Department of Psychiatry, University of Rochester Medical Center, in New York, notes that the study provides an "elegant" perspective on the relationship between social integration and suicide risk.

Although the study findings make "great sense intuitively" and support what has been described as important protective forces, "it is uncommon to see the effect so clearly demonstrated using a design that is not subject to the biases of post-mortem sampling, retrospective recall, and the need to depend on some type of matching to create a comparison group," Dr Caine writes.

But it is "essential" to recognize the study's limitations, he says. The research included only women who were "remarkably homogeneous" with respect to education, income, and other factors related to socioeconomic status. As well, the "tools" used in 1992 to measure social integration did not utilize computer-based technologies that are available today.

The study also involved relatively few suicides, even among those who were socially isolated. However, the size of the study sample "helps to mitigate" concerns about a study of relatively rare events, said Dr Caine.

He notes that interventions that were introduced years ago to change personal habits ― smoking cessation, healthy diet, and exercise ― in order to reduce cardiovascular disease risks were driven largely at the cultural, social, and political level rather than at the individual level.

Results of the new study "invite further research to explore whether factors or behaviors that reflect longstanding measures of individual social integration predict a person's mindset when he or she is suicidal," he notes.

The social part of the biopsychosocial medical model of suicide has been the weakest link in the paradigm and needs invigorating, he says. "Like heart disease 50 years ago, we do not need to have absolute certainty about the mechanism of action to begin to test and implement essential, broadly targeted preventive interventions."

The authors and Dr Caine report no relevant financial relationships.

JAMA Psychiatry. Published online July 29, 2015. Full text, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: