Email Icon Facebook icon Twitter Icon GooglePlus Icon Contact

User Top Menu

New Study Looks at Causes of Premature Mortality in People With Schizophrenia

6 Nov 2015

November 7, 2015. People with schizophrenia have a much greater risk of dying prematurely compared to the general population. A new study published in JAMA Psychiatry provides stark details about the causes behind these premature deaths. The study, which was led by Mark Olfson of Columbia University in New York City, analyzed 74,003 deaths in a cohort of over 1.1 million people with schizophrenia living in the United States.

Reinforcing earlier research, the study found that people with schizophrenia lose more than 28 years of life, particularly to cardiovascular disease. But the study also brought new insights, highlighting smoking-related mortality and accidents.

Several previous studies, including one by Dickerson et al. (Dickerson et al., 2014), and a large meta-analysis (see SRF related news report) have examined mortality rates and causes in people with schizophrenia. But this study is unique because of the size of its cohort. "The new study is larger than all of the 40 or so prior studies of this topic combined," Olfson told SRF.

"What I thought was unique about this study and interesting was the ability to really break down a lot of the different causes of death," said Benjamin Druss of Emory University in Atlanta, Georgia, who was not involved with the study. "If you have a million people, you can get a much more granular understanding of what people are dying from."

In a commentary accompanying the study, Shuichi Suetani, Harvey Whiteford, and John McGrath of the Queensland Center for Mental Health Research in Wacol, Australia, describe the findings as "shocking but not surprising" given the strong previous research. They write, "The obvious challenge arising from these findings is how to do more to prevent and treat the risk factors and diseases identified by Olfson and colleagues in individuals with schizophrenia."

A fine-grained analysis

To perform their analysis, Olfson and colleagues cross-referenced national data from the Medicaid claims records of 20- to 64-year-old patients with schizophrenia with national mortality data to look at patterns of mortality between 2001 and 2007. "Most people who have schizophrenia in the United States are in the Medicaid program," said Olfson. "While it's not completely comprehensive, it is, in a broad sense, nationally representative." The researchers compared the rate and age at which people with schizophrenia died from various causes to the general population after adjusting for age, sex, race, and region of the country.

Overall, they found that adults with schizophrenia were more than 3.5 times as likely to die during the follow-up period compared to the general population of adults. Moreover, each deceased person lost, on average, 28.5 years of life. The researchers also took a closer look at each particular cause of death, calculating the standardized mortality ratio (or SMR)—the ratio of actual deaths of people with schizophrenia versus expected deaths in a general population—and the mean years of potential life lost for each one.

The leading cause of death, accounting for about one-third of all natural deaths, was cardiovascular disease. Ischemic heart disease accounted for about one-half of these deaths. The standard mortality ratio for cardiovascular disease was 3.6, showing that people with schizophrenia were much more likely to die from this cause compared to the general population. The mean potential years of life lost from this cause was 26.8 years.

Approximately one in six natural deaths was due to cancer in people with schizophrenia. The overall SMR for cancer was 1.8. Out of the six cancers examined, lung cancer was the cause of the most deaths and had the highest SMR, 2.4. These results may stem from the fact that about two-thirds of people with schizophrenia are smokers, and those who smoke tend to smoke more heavily than people in the general population.

Other prominent natural causes of death were likely also linked to smoking, including chronic obstructive pulmonary disease, which killed people with schizophrenia at a rate of almost 10 times that expected for the general population, and influenza and pneumonia, which had an SMR of 7. "Smoking and smoking-related conditions really came to the fore," Olfson said. "Schizophrenia has been previously linked to increased risk of death from respiratory diseases, but these new numbers provide greater detail than has been available in prior research."

About 13 percent of the deaths had an unnatural cause. Approximately one-quarter of these deaths were due to suicide (SMR of 3.9), and about 58 percent were caused by accidents (SMR of 3.2). About half of the accidental deaths were substance-induced deaths (SMR of 4.8), although it is likely that some of these deaths were actually suicides.

Olfson told SRF that he was surprised by the number of accidental deaths, because motor vehicle accidents account for a substantial portion of accidental deaths in the general population, and most people with schizophrenia do not drive. "But, in fact, you can see that the rates [of accidental deaths in people with schizophrenia] are three times as high," he said. "I hope that greater attention is focused on trying to determine in greater detail what are the sort of situations that place people with schizophrenia at high risk of accidental deaths so preventative interventions can be put into place."

Policy implications

While this study does have some limitations, including a lack of information about the patients' key health risk factors (e.g., substance abuse, smoking status, and body mass index), the possible underreporting of suicides due to stigma and/or misclassified accidental deaths, and the limited age group examined (if elderly patients with schizophrenia had been included, the SMRs may have decreased), the results give a fairly comprehensive picture of the causes of premature death in people with schizophrenia—a picture that largely confirms prior studies.

"This is one of the most consistent findings that you'll find in any literature," said Druss. "Almost every study that's been done in this area has found a mortality gap."

These results also provide further evidence that the gap in premature mortality between people with schizophrenia and the general population is widening. "Our overall finding of over a 3.5-fold increase in standardized mortality ratio is in the ballpark but is larger than many recent studies," Olfson told SRF. "It's consistent with a greater gap opening up."

In their commentary, Suetani, Whiteford, and McGrath suggest that it may be time to move beyond epidemiological research and focus on solutions. "What is needed now is a thorough realignment of policy and funding to support action at the clinical level," they write. "Incremental fixes, such as sporadic government edicts, clinician education about the problem, promotion of practice guidelines, patient education to be more assertive 'consumers,' and implementation of integrated electronic medical records, have not had the necessary effect."

Suetani et al. recommend that "[p]hysical health outcomes need to be measured and individualized targets set for each consumer, with funders and healthcare providers held accountable for achieving these agreed goals." They suggest that interventions target lifestyle risk factors, including smoking status, diet, and physical activity.

Ending the fragmented healthcare that many people with psychiatric illness experience may be the first step toward successfully implementing such integrations. "These sorts of numbers really call out for greater coordination and integration of services," Olfson told SRF. "… whether that means bringing in more psychiatrists and other mental health professionals into general medical care settings or the reverse—bringing primary care physicians and nurse practitioners into the community mental health centers so that patients can have their blood pressure checked and can have their weight followed more carefully, and [be given] recommendations in terms of diet and exercise, and smoking cessation."

Druss told SRF that there are programs currently trying integrative models of healthcare. "It's a major focus of the Medicaid Health Homes waiver program. It's a focus of [the Substance Abuse and Mental Health Services Administration's] home healthcare integration initiative that's providing medical services onsite at community mental health centers."—Summer Allen.


Olfson M, Gerhard T, Huang C, Crystal S, Stroup TS. Premature Mortality Among Adults With Schizophrenia in the United States. JAMA Psychiatry. 2015 Oct 28:1-10. Abstract

Suetani S, Whiteford HA, McGrath JJ. An Urgent Call to Address the Deadly Consequences of Serious Mental Disorders. JAMA Psychiatry. 2015 Oct 28:1-2. Abstract