Self-harm Episode Dramatically Raises 1-Year Suicide Risk

Nancy A. Melville

March 24, 2017

An episode of self-harm dramatically increases the risk for suicide within 1 year, new research shows.

Investigators at Columbia University Medical Center in New York City found that during the 12 months after nonfatal self-harm, the rate of completed suicide was 37 times higher than in a matched-population cohort. They also found that the rate of suicide was even higher if the initial self-harm event involved a violent method, such as the use of a gun, compared to a nonviolent method, and that this risk was greatest within the first 30 days of the first episode.

"Adults who present for treatment following self-harm often require urgent medical attention. However, these patients also provide clinical opportunities for suicide prevention," first author Mark Olfson, MD, MPH, professor of psychiatry at Columbia University Medical Center, told Medscape Medical News.

"They are at exceptionally high short-term risk of suicide, and clinical attention should focus on ensuring their safety, particularly during the critical first few weeks," Dr Olfson added.

The study was published online March 21 in the American Journal of Psychiatry.

Early Vigilance Required

Although previous studies have shown that as many as 1 in 7 adults who commit suicide had been treated for self-harm in the previous year, efforts to identify specific markers of suicide risk in these patients have been impeded by relatively low overall numbers of suicides.

To better determine the risk using a larger population, Dr Olfson and colleagues evaluated data on adults from the national Medicaid database who were diagnosed as having a nonfatal self-harm event and for whom follow-up information for up to 1 year was available.

They found that among 61,297 patients with self-harm events, 19.7% (12,012) were treated for a repeated nonfatal self-harm event within 1 year; 0.16% (99) had suicide death in the first 30 days following the self-harm event; and 0.23% (144) had suicide death in the following 335 days.

Despite the low percentages, the 1-year suicide rate of 439.1 per 100,000 person-years is 37.2 times higher than the rate of 11.8 per 100,000 person years in the general population, the authors report.

The use of a violent method in the initial self-harm event was associated with a significantly higher risk for suicide death compared to nonviolent events (hazard ratio [HR], 7.5; 95% confidence interval [CI] = 5.5 - 10.1), with the risk even higher specifically with firearms (HR = 15.86; 95% CI = 10.7 - 23.4), compared to a reference of poisoning.

However, the increased risk for suicide associated with a violent method vs a nonviolent method was only significant in the first 30 days of the initial self-harm event (HR = 17.5; 95% CI = 11.2 - 27.3), not in the following 335 days.

"Among those who survived self-harm with violent methods, the risk of completing suicide over the following month was approximately 10 times greater than the risk over the subsequent 11 months," the authors note.

Dr Olfson said the concentration of the increased risk only in the first 30 days was a surprise.

"We were anticipating that people who used violent methods would remain at highest risk. This finding underscores the critical importance of working to ensure the safety of patients who use violent self-harm methods during the first few weeks," he said.

Stark Reminder About Firearms

As expected, the use of violent methods in the first self-harm event was much more likely to result in death: two thirds of the completed suicides involved violent methods, and more than 40% of all initial susicide deaths involved firearms.

"These results provide a stark reminder of the urgent need to keep firearms away from individuals who have used them to harm themselves," Dr Olfson said.

Males were twice as likely to have suicide death as females (0.63% vs. 0.28%), and older adults were more than three times as likely to have suicide death as younger adults (0.76% vs. 0.21).

The risk was also higher for white adults (0.45%) than for blacks (0.14%) or Hispanics (0.26%).

These sex differences could be attributable to a tendency of men to use more lethal methods, and suicidal intent also tends to be greater among male than female self-harm patients, the authors note.

They add that among those whose initial self-harm event was fatal, only 40.7% had received outpatient mental health care during the last 6 months of their lives.

The study's findings are consistent with those reported in a 2008 Finnish study published in the American Journal of Epidemiology. That study, which involved patients who had been hospitalized for suicide attempts, reported that the suicide risk was highest during the first week following hospital discharge.

However, that study did not separately assess the risk for male and female patients, the authors say.

The current study is notable for providing insights into the risk during a specific high-risk period, Dr Olfson said.

"No other study has evaluated suicide risk over a period of less than 1 year following self-harm," he said. "Yet it is short-time horizons that are of greatest clinical importance to patients and family members."

The study received support from the National Institutes of Health and the Agency for Healthcare Research and Quality. Dr Olfson has disclosed no relevant financial relationships. Coauthor Shuai Wang, MD, is an employee of Quartet Health. Coauthor Tobias Gerhard, PhD, has received research grant funding from Bristol-Myers Squibb; he serves on an external safety review committee for a Merck study; and he has provided expert consultation to law firms on behalf of Roche and Pfizer.

Am J Psychiatry. Published online March 21, 2017. Abstract

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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.

processing....