SAN FRANCISCO – Teens who answered “no response” for one or more questions during a standard suicide screening in the ED typically have a risk of suicide nearly on par with those who answered “yes” to at least one question, results of a study showed.
“Risk for suicidality was substantial among both groups,” Tricia Hengehold of the University of Cincinnati reported at the Pediatric Academic Societies meeting. “About half of each group fit the medium-risk category, although the yes group had more teens in the high-risk category comparatively.”
Devonyu/Thinkstock
In their study, she and her coinvestigators compared all teens who answered yes or no response (NR) to any question on the Ask Suicide-Screening Questions (ASQ). A total of 3,386 adolescents, aged 12-17 years, were screened when each presented to the ED with a complaint other than a psychiatric one. The data came from a preexisting study not initially designed for studying nonresponders.
The majority of teens (93%) answered no to all ASQ questions: 5% answered yes to at least one ASQ question, regardless of other answers, and 2% answered no response on at least one of the ASQ questions, but did not answer yes to any of them.
The average age of participants was 14 years among all response groups, but females were overrepresented in the yes and NR groups: 74% of females answered yes to one of the ASQ questions, compared with just 26% of males. Similarly, 79% of females were classified as nonresponders, vs. 21% of males. In the negative screen group, however, females (54%) and males (46%) were much more evenly represented.
Patients who answered yes or NR also were more likely to have Medicaid or Medicare than commercial insurance or a self-pay arrangement. Within the yes group, 56% had public insurance, and 39% had private insurance. In the NR group, 43% had public insurance, and 53% had private insurance.
Any teens who answered yes or NR should have undergone evaluation by a mental health professional, but those answering yes were more likely to get this evaluation than nonresponders. Nearly all (93%) of those answering yes received the evaluation, compared with 79% of nonresponders. Yet half of the nonresponders who were evaluated were recommended for further follow-up, not far behind the 65% recommended from among the yes group.
A clinically significant risk of suicide existed among 93% of those answering yes and 85% of nonresponders, the investigators found. About a third (33%) of the yes group were classified as high risk – suicidal ideation within the past week or no treatment after a previous attempt – while 16% of the nonresponders were.
“The NR group was more often in an earlier stage of change than the yes group, presenting with a greater percentage of teens in the precontemplation category,” Ms. Hengehold said. The precontemplation category, which included 27% of the NR group and 17% of the yes group, referred to teens who did not believe they would benefit from working with a mental health professional.
The contemplation category referred to teens who said they were still thinking about whether to meet with a mental health professional or that they would seek treatment if their suicidality worsened. This category included 9% of yes responders and 14% of nonresponders.
A higher proportion of yes responders (56%) than nonresponders (41%) had reached the preparation stage, which meant they had agreed to set up treatment or had received a referral within a month after their ED visit. A similar percentage of yes responders and nonresponders were currently in treatment, seeing a mental health professional intermittently, or had successfully received therapy.
“The sociodemographic characteristics of teens that endorse a no response are very similar to those of positive endorsers,” she concluded. She noted the potential importance of including an NR option on suicide-screening instruments.
Although nonresponders had clinically significant mental health concerns that indicated a need for further evaluation, these adolescents were less ready to engage in mental health services than those answering yes.
“This may be important when designing suicide risk interventions for each group,” she reported.
The research was funded by the National Institutes of Health. Ms. Hengehold and her associates reported no relevant financial disclosures.