Bipolar disorder linked to early death

It also comes with a higher risk of dying early. Now, a study puts into perspective just how large that risk is, and how it compares with other factors that can shorten life.

In two different groups, people with bipolar disorder were four to six times more likely as people without the condition to die prematurely, the study finds.

By contrast, people who had ever smoked were about twice as likely to die prematurely than those who had never smoked -- whether or not they had bipolar disorder.

A team from the University of Michigan, home to one of the world's largest longterm studies of people with bipolar disorder, reports their findings in the journal Psychiatry Research.

The stark difference in mortality, and the differences in health and lifestyle that likely contributed to it, should prompt more efforts at preventing early deaths, say the researchers.

"Bipolar disorder has long been seen as a risk factor for mortality, but always through a lens of other common causes of death," said Anastasia Yocum, Ph.D., lead author of the study and data manager of the research program at the Heinz C. Prechter Bipolar Research Program.

"We wanted to look at it by itself in comparison with conditions and lifestyle behaviors that are also linked to higher rates of premature death."

Two large data sources yield similar findings

Yocum and her colleagues, including Prechter Program director Melvin McInnis, M.D., started by looking at deaths and related factors among 1,128 people who had volunteered for the program's longterm study of people with and without bipolar disorder.

They found that all but 2 of the 56 deaths since the study began in 2006 were from the group of 847 people in the study who had bipolar disorder.

With statistical adjustments, their analysis shows that having a diagnosis of bipolar disorder made someone six times more likely to die during a 10-year period than people taking part in the same study who did not have bipolar disorder.

By comparison, study participants who had ever smoked or were over age 60 were more than twice as likely to die in that same time as people who never smoked or were under 60, regardless of bipolar status.

The researchers then turned to another source of data to see if they could find the same effect.

They analyzed years' worth of anonymous patient records from more than 18,000 people who get their primary care through Michigan Medicine, U-M's academic medical center.

Among this group, people with bipolar disorder were four times as likely to die during the study period than those with no record of bipolar disorder.

The team studied records from more than 10,700 people with bipolar disorder and a comparison group of just over 7,800 people without any psychiatric disorder.

The only factor associated with an even higher chance of dying during the study period in this pool of people was high blood pressure.

Those who had hypertension were five times more likely to die than those with normal blood pressure, no matter whether they had bipolar disorder or not.

By contrast, smokers were twice as likely to die as never-smokers in this sample, and those over age 60 were three times as likely to die, both regardless of bipolar status.

"To our major surprise, in both samples we found that having bipolar disorder is far more of a risk for premature death than smoking," said McInnis, a professor of psychiatry at the U-M Medical School.

He hopes the findings will spur more action in the medical and public health communities to address the many factors that contribute to this extra-high risk of death in people with bipolar disorder.

"Over the years there have been all kinds of programs that have been implemented for smoking prevention and cardiovascular disease awareness, but never a campaign on that scale for mental health," he said, noting that about 4% of Americans live with bipolar disorder while about 11.5% of Americans smoke.

Other differences between groups

Yocum and McInnis note that people with bipolar disorder in both groups were much more likely than the people without bipolar disorder to have ever smoked, consistent with past studies.

Nearly half (47%) of the U-M patients with bipolar disorder had a history of smoking, as did 31% of the Prechter participants with bipolar disorder.

By comparison, smoking among those without bipolar disorder stood at 29% of the U-M patients and 8% of the Prechter participants.

People with bipolar disorder in both groups were also much more likely to be female, and female gender was associated with a slightly lower risk of early death.

In the Prechter cohort, people with bipolar disorder were much more likely to have asthma, diabetes, high blood pressure, migraines, fibromyalgia and thyroid conditions than those who had not been diagnosed with bipolar disorder.

Within the group of Prechter study participants who have bipolar disorder, being a smoker and scoring higher over time on a standardized survey of depression symptoms were both associated with a doubled risk of death, compared with participants who had bipolar disorder but didn't smoke or scored lower over time on depression ratings.

Interestingly, the researchers found no association between risk of death and the number of years Prechter participants had been taking medications for mental health symptoms.

There was also no association with scores for anxiety and mania.

Among just people with bipolar disorder in the U-M patient sample, high blood pressure also was associated with a fivefold higher risk of death, while smoking was associated with a nearly twofold risk of death.

Information on depression scores or medication use over time was not available for this group.

A path forward

Both Yocum and McInnis say the findings, combined with studies on the health status, health risk behaviors and specific causes of death for people with bipolar disorder, could inform efforts to improve the health and quality of life for people with the condition.

Past research has shown that people with bipolar disorder are more likely to have metabolic syndrome, which puts them at higher risk for diabetes and cardiovascular conditions because of a combination of factors related to waist size, cholesterol, blood sugar and blood pressure.

Medications for bipolar disorder can contribute to this.

Also important: the secondary effects of bipolar disorder's symptoms.

Lack of activity, poor diet, drug/alcohol overuse and lower education attainment and employment rates also increase overall health risk, while health insurance coverage and access to care may be less consistent.

Educating more teens and adults on how to cope with stress, distress and mood fluctuations, and how to identify and get help for depression symptoms, could be part of increasing early intervention, the researchers say.

Bipolar disorder often begins showing itself as depression, and there are currently no good ways to predict which people will go on to develop bipolar disorder, though a family history of the condition is known to increase risk.

Genetic research at the Prechter Program and elsewhere is studying these contributing factors.

"Bipolar disorder is never going to be listed on the death certificate as the main cause of death, but it can be an immediate or secondary contributor to a death, including in suicides," said Yocum, who notes that cross-sectional studies have found that on average people with bipolar disorder die 8 to 10 years earlier than other people their age.

Similarly, McInnis says, smoking is rarely listed on death certificates, but is well understood to be a major risk factor leading to cancers and cardiovascular emergencies that do get listed as causes of death.

That's why it has received so much attention from agencies and organizations running public health campaigns.

"We need to know more about why people with bipolar have more illnesses and health behaviors that compromise their lives and lifespan and do more as a society to help them live more healthily and have consistent access to care," he said.

The Prechter Program is still accepting people with bipolar disorder, and those without, into its longitudinal study.

In addition to Yocum and McInnis, the study's authors are Emily Friedman and Peisong Han from the U-M School of Public Health and Holli S. Bertram from Psychiatry.

The study was funded by the Heinz C. Prechter Bipolar Research Fund at the University of Michigan Eisenberg Family Depression Center and the Richard Tam Foundation, as well as the National Institute of Mental Health and the National Center for Advancing Translational Sciences, both part of the National Institutes of Health (MH100404, MH106434, TR002240)

The researchers used the Michigan Medicine DataDirect tool to explore anonymous patient data.