Study Highlights Effective, Low-Cost Suicide Prevention Strategies

Three interventions for the follow up of patients at risk for suicide who entered a general hospital emergency department (ED) were found to be cost-effective in a National Institute of Mental Health (NIMH) study published September 15 in Psychiatric Services. In previous studies, these interventions were found to reduce suicide risk significantly, but they have not been adopted widely.

One of the interventions studied, sending a series of caring postcards or letters to suicidal patients as a follow-up to their ED visit, was found to be less expensive than usual care. Under usual care, patients discharged from the ED who are identified as at risk for suicide are sent home with a list of names and phone numbers of mental health professionals in the community; thus, it is up to patients to take the initiative to schedule an appointment. The subsequent visit is also part of usual care.

The other two interventions—telephone outreach and cognitive-behavioral therapy (CBT)—cost more than the postcards and usual care, but the costs were still well below the standard benchmark of cost-effectiveness the researchers set.

“The purpose of these interventions is to provide psychosocial support to individuals after ED discharge, help maintain contact, and nudge people toward follow-up treatment,” said lead investigator Michael Schoenbaum, Ph.D., senior advisor for mental health services, epidemiology, and economics in NIMH’s Division of Services and Intervention Research.

For the study, researchers built a computer model of a general hospital ED and used patient data from previously published relevant studies. General hospital EDs are high-risk settings that treat more than 500,000 people each year for self-injuries, according to the Centers for Disease Control and Prevention. The subjects were aged 18 and older and had sought treatment in a general hospital ED.

The study authors estimated that follow-up postcards reduced the rate of suicide attempts or reattempts by 45 percent compared with usual care alone; phone outreach, one to three months after discharge, reduced the rate of suicide attempts or reattempts by 34 percent; and up to nine sessions of CBT reduced the rate of suicide attempts or reattempts by 50 percent, across the one-year study period.

Special software was used to carry out repeated simulations of the chain of events for a roughly year-long period from the time an individual entered the modeled ED. Thousands of simulations showed the range of outcomes possible and the probabilities of each. Researchers computed the mean expected costs and life-years per person during the study period for each of the four interventions (usual care, postcards, phone outreach, and CBT), as well as the incremental costs, life-years, and cost-effectiveness ratio of postcards, phone outreach, and CBT compared with usual care, Schoenbaum said.

Interventions can affect costs in two ways, according to the study: the direct cost of delivery, including the intervention itself and any associated health care use, and by altering the incidence of subsequent suicide attempts and death. For delivery costs, researchers used data on health service use reporting in a corresponding clinical trial, which referred to the relevant CPT codes and assigned costs based on national rates from the 2014 Medicare Physician Fee Schedule. For ED visits and hospitalization, researchers calculated average costs based on an analysis of corresponding events in the Healthcare Cost and Utilization Project database for people discharged alive and for those who died while hospitalized.

Mean costs per suicide-risk patient were $1,962 with usual care, according to the modeling analysis. Compared with usual care, mean costs were 0.07 percent lower with postcards, and 0.05 percent and 0.25 percent higher with phone outreach and CBT, respectively. Each intervention reduced mortality. The estimated mean incremental cost per life-year was $4,300 for phone outreach and $18,800 for CBT—well below a conservative threshold of $50,000 that the researchers set as the amount society is willing to spend per life year saved.

“Based on the science already done and published, and now the costs, we are at the point where we can recommend widespread use of these interventions,” Schoenbaum said. “There is enough clinical evidence supporting the interventions, and we can afford them.”

However, he noted that the success of these interventions is influenced by the extent to which people at risk of suicide are identified in the ED.

“We need to be better at identifying people in the first place,” he said. “We’re missing many of the people who are presenting with a physical problem who may also be suicidal, but they will tell us only if we ask them. Also, the ED [health professionals] may classify some people wrongly because they regard a poisoning or injury as suspicious. They may misidentify a person as a risk who is not (false positive), and some patients may misrepresent being injured just to get shelter and food for the evening.”

Support for this research was provided under professional services contracts with the National Institute of Mental Health. ■