By Linda Carroll
(Reuters Health) - Muslims in the U.S. are more likely to have attempted suicide than members of other religions, a new study suggests.
Survey responses from a weighted sample of 2,836 Americans revealed that self-identified Muslims were 2.18 times as likely as Protestant respondents to report a lifetime suicide attempt, while Catholics (adjusted odds ratio 1.20), other Christians (aOR 1.18), and Jews (aOR 0.93) were at lower risk. Across all groups, level of religiosity didn't affect the odds of reporting a lifetime suicide attempt, the authors note in JAMA Psychiatry.
"We need to focus our attention on suicide awareness measures in marginalized communities," said the study's lead author, Dr. Rania Awaad, a clinical associate professor and director of the Muslim Mental Health and Islamic Psychology Lab at the Stanford School of Medicine in California. "Because Muslim communities are faith-based, it is important to custom-tailor suicide awareness efforts to their unique needs."
"It is also important to conduct in-depth analysis as to the exact reasons why Muslim communities are severely impacted, and to further study the Islam-inspired protective factors against completing suicide (as evidenced by previous studies) as compared to attempting suicides (as revealed by our study)," Dr. Awaad said in an email.
To explore whether Muslims in the U.S. might have a higher likelihood of having attempted suicide during their lifetimes, the researchers turned to the 2019 Institute for Social Policy and Understanding national community-based survey, which was conducted over landline, cell phone and online by Social Science Research Solutions. Muslim and Jewish participants were oversampled and other religious groups were weighted to provide nationally representative and projectable estimates of the U.S. population 18 years and older.
Along with their religious affiliation, respondents were asked to self-identify their race and ethnicity using the following categories: African American, Arab, Asian/Chinese/Japanese/Indian/Pakistani, Native American/American Indian/Alaska Native, Native Hawaiian/Pacific Islander, mixed, Hispanic, white, or other.
Lifetime history of suicide attempts was assessed with a question adapted from the Columbia Suicide Severity Rating Scale: "Have you ever tried to do anything to try to kill yourself or make yourself not alive anymore?"
The survey's response rate for the prescreened landline and cell phone sample was 22.8%, and 4.1% for the listed-telephone sample. The web panel response rate for Muslim respondents was 6.4% and the web probability panel response rate for general population respondents was 14.3%.
The final sample included 2,376 participants, of which 809 (34%) were Muslim, 1,226 (52%) were men, 1,522 (65%) were white, 801 (34%) were aged 30 to 49 years, 637 (29%) had an annual income more than $100 000, 726 (31%) had a bachelor's degree, 1,132 (48%) reported religiosity as very important, and 1,908 (81%) were born in the U.S.
Looking at the rate of suicide attempts by religious affiliation, the researchers found that 7.9% of Muslim (809), 5.1% of Protestant (314), 6.1% of Catholic (245), and 3.6% of Jewish (415) respondents reported a lifetime suicide attempt. When the researchers adjusted for demographic factors, they found Muslim respondents were 2.18 (P value 0.02) times as likely as Protestant respondents to report a lifetime suicide attempt.
Respondents who identified as atheist/agnostic (aOR 1.87) and those described as "other" (aOR 2.24) were also more likely than Protestants to report a lifetime suicide attempt. Women (aOR 1.35) were more likely than men to report a lifetime attempt as well.
Participants who identified as Jewish, Catholic, other Christian denominations, and atheist/agnostic had no significantly different odds of reporting a suicide attempt in the last year compared to Protestant respondents, the authors note.
When evaluating the study, it's important to remember that "these findings are based on a telephone and web panel based survey with a relatively low response rate, and that the recruitment for Muslim (and Jewish) participants was handled differently to other groups, which can bias results in several ways," said Dr. Paul S. Nestadt, an assistant professor, co-director of the McGlasson Anxiety Disorders Clinic and Attending Psychiatrist, JH Esketamine Clinic in the department of psychiatry and behavioral sciences at the Johns Hopkins School of Medicine in Baltimore.
"However, some of the basic numbers presented beside the primary findings are already in line with what we know about general suicide attempt rates and about attempt ratios in female: male," Dr. Nestadt said in an email. "These sorts of confirmations of known rates and proportions are encouraging and imply good sampling methodology and instill confidence in the novel findings."
"Past international reports have found Muslim suicide attempt rates to be higher as well, though actual suicide death rates to be lower," said Dr. Nestadt, who was not involved in the new research. "Religion has long been thought to be protective against suicide, but this study found no correlation between increasing religious observance and suicide attempts. Note that suicide attempts are a very poor proxy for suicide itself, but they are often used (in research) because they are more common."
In general, Islam is thought to be a protective religion like Orthodox Judaism, said Dr. Alex Dombrovski, an associate professor of psychiatry at the University of Pittsburgh School of Medicine, in Pennsylvania. "The finding they report is thus surprising."
Still, the study has some weaknesses, said Dr. Dombrovski, who was not involved in the research. For one thing, it had a P value less than 0.05, "but for a study like this, you would want to see it at 0.01, or even better, 0.001," he added. Another issue is that the differences between the groups weren't apparent until the researchers adjusted for demographics, Dr. Dombrovski said. "Usually, you want to see an effect with and without adjustment," he added. "Spurious effects may emerge because some variables might be correlated."
SOURCE: https://bit.ly/2V16t47 JAMA Psychiatry, online July 21, 2021.