Supportive text messages did not decrease suicidal ideation or “risk incidents” among military personnel at risk for suicide in a randomized trial, but there appeared to be significant benefit for some important secondary outcomes including suicide attempts.
No significant reductions were observed in terms of suicide risk incidents — inpatient admission or evacuation associated with suicidality — or current suicide ideation at 12 months among military personnel who were assigned to the text-based intervention compared with standard care alone, reported Amanda Kerbrat, MSW, of the University of Washington, and colleagues.
Called Caring Contact, the intervention involved 11 brief texts sent by mental health professionals — such as “Hope this week is going well for you” and “Hope all’s well and you’re taking good care of yourself” — sent on the first study day, again later that week, and then eight more times over the following year as well as on participants’ birthdays.
However, compared to individuals receiving standard care alone, those receiving Caring Contact did have a lower risk of experiencing suicidal ideation throughout the year (88% vs 80%; OR 0.56, 95% CI 0.33-0.95) and attempting suicide (15% vs 9%; OR 0.52, 95% CI 0.29-0.92), Kerbrat and colleagues wrote in JAMA Psychiatry.
The intervention was based on a caring letters study in the 1970s co-designed by Jerome Motto, who hypothesized that increasing connection would decrease the rates of suicide in a civilian population, and that this could be accomplished by sending a series of letters expressing care and concern, Kerbrat told MedPage Today.
“Those contacts have to be initiated by a concerned person and have to put no demand or expectations on the recipient,” she added. “He believed long term contact like this could persuade a suicidal person to keep on living.”
Although the primary results of this study were null, even small benefits on suicide ideation could be “remarkable and potentially very meaningful,” wrote Murray Stein, MD, MPH, of the University of California, San Diego, and colleagues, in an editorial accompanying the study.
Stein noted this intervention’s failure to meet its primary endpoints could be due to military personnel being “embedded in a rich social milieu that, if nothing else, is hardly isolating,” and that, in fact, the social disconnectedness theory behind Caring Contact might not apply to this population.
“It is reasonable to consider that the failure to see larger effects might at least partly be because of an intervention-environment mismatch,” Stein and colleagues wrote. “This is an important consideration for suicide prevention approaches in general: how to deliver the right intervention at the right time.”
Charles W. Hoge, MD, of the Walter Reed Army Institute of Research, and a noted expert on military suicide who authored an additional editorial, agreed that no single suicide prevention tactic will be effective in every population or situation. He also argued that “unrealistic goals,” such as treating all suicides as preventable, are part of the problem.
Although strategies such as regular depression screenings in primary care, evaluating a patient’s lethal means (particularly since “two-thirds of veteran suicide involve firearms”), and following up after hospitalization “are all responsible clinical steps,” Hoge wrote, packaging them into standardized guidelines “can evolve into overly complex requirements that go well beyond the evidence.”
Kerbrat and colleagues collected data from English-speaking U.S. Army soldiers (54%) and Marines (46%) with histories of suicidality. Participants were randomized and stratified by military installation as well as their number of lifetime suicide attempts.
Midway through the trial, participants reported that Caring Contact responses felt “mechanical,” leading Kerbrat and colleagues to add a second protocol in which they sent an additional text message. For example, when participants responded to a text with “I’m fine, thanks,” their assigned mental health professional would respond with a smiley emoticon or “You’re welcome.”
Of the 658 participants randomized, the majority were white (60.7%) and male (82%), with an average age of 25.2 years. In the year before enrollment, two thirds (66.7%) had experienced a suicide risk incident, and 44.3% had attempted suicide at some point in their lives.
Follow-up data at 12 months were lacking for about 30% of each group — mostly because they refused or could not be located. “Missing data were assumed to be missing at random and addressed via multiple imputation,” the researchers wrote.
Limitations included the second protocol added midway through the trial, a low rate of suicidal behavior across the sample, and the failure to conduct diagnostic interviews. All individuals in this study reported their suicidal thoughts to a military clinician, indicating these results might not be generalizable to individuals who don’t disclose this information, the authors noted.
The authors received funding from the Military Suicide Research Consortium and the National Institute of Child Health and Development of the National Institutes of Health.
One co-author is a co-founder with equity stake in a technology company, Lyssn.io.
This study was financially supported by the Military Suicide Research Consortium and the National Institute of Child Health and Development of the National Institutes of Health.
Stein is a consultant for Actelion, Aptinyx, Bionomics, Dart Neuroscience, Healthcare Management Technologies, Janssen, Neurocrine Biosciences, Oxeia Biopharmaceuticals, Pfizer, and Resilience Therapeutic.
A co-author of Stein’s editorial received support from Sanofi Aventis, co-owns DataStat Inc., and was a consultant for Johnson & Johnson Wellness and Prevention, Sage Pharmaceuticals, Shire, and Takeda.