An ED-based RCT of Lethal Means Counseling for Parents of At-Risk Youth

An ED-based Randomized Clinical Trial of Lethal Means Counseling for Parents of Youth at Risk for Suicide

Colorado School of Public Health, Harvard School of Public Health (HSPH), American Foundation for Suicide Prevention

The National Action Alliance to Prevent Suicide recently released a research agenda aimed at significantly reducing suicide over the next decade. Aspirational Goal 12, "Reduce access to lethal means that people use to attempt suicide," calls for identifying effective strategies to reduce a suicidal person's access to firearms and other lethal means. A promising strategy is to counsel patients seen in the emergency department (ED) for a psychiatric emergency to reduce access to firearms (the most lethal suicide method) and medications (the most common method of suicide attempt). To date, however, few studies have evaluated changes in firearm storage practices among those who received lethal means counseling (LMC), and those that have, including a pilot conducted in Colorado by the study team, have lacked control groups. Results from the pilot, which provided emergency department based LMC counseling to parents of suicidal adolescents, found that among gun-owning parents, 33% had unlocked guns at home on the day of the ED visit and none did on follow up. Using the piloted LMC protocol, we will conduct the first randomized, controlled trial (RCT) of the effectiveness of ED-based LMC on firearm and medication storage. The proposed RCT, to be conducted in five Colorado hospitals, will test whether parents of at-risk adolescents who are treated in hospitals that have (vs. have not yet) implemented our LMC protocol are more likely to store household firearms and medications safely. In addition, we will conduct in-depth, qualitative interviews with parents who have received LMC counseling to better understand those factors that affect parents' willingness and ability to make changes to firearm and medication storage. We will also conduct qualitative interviews with clinicians to understand factors affecting clinician engagement in LMC. AIM 1: To assess the effectiveness of an ED-based LMC intervention to improve how parents of pediatric patients (age 10-17) who visit the ED for a mental health emergency store household firearms and medication. AIM 2: To examine how attitudinal and contextual factors shape a) parents' decisions about firearm and medication storage following LMC, and b) clinicians' delivery of LMC messages.

Suicide is the second leading cause of death for U.S. youth 10-17 years of age. Nationally, 51% of all suicides, and 40% of 10-17 year-olds' suicides, are completed with a firearm. In Colorado, which has the 6th highest adolescent suicide rate in the nation, more youth more youth die by suicide than in motor vehicle crashes. Every U.S. study that has addressed the relationship between access to firearms and suicide has found that such access increases risk of suicide, not only for the gun owner, but for all household members. Indeed, the relative risk of suicide associated with household firearms is highest for children and young adults, and elevated further in those households in which guns are stored loaded and unlocked. Some of the largest reductions in national suicide rates have occurred when access to a commonly used and highly lethal suicide method was reduced. Consequently, reducing a suicidal person's access to lethal means of suicide ("means restriction") is a vital component of any effective national strategy for reducing suicide rates. Reducing access to firearms is particularly important given their greater lethality compared with other methods commonly used in US suicides and the very short deliberation period preceding nearly half of suicide attempts. Nonetheless, means restriction interventions remain uncommon. The CDC's recently released Research Priorities for the National Center for Injury Prevention and Control, acknowledges this gap and calls for evaluation of the feasibility, scalability, and economic efficiency of means restriction strategies. A recent study estimated that over 40% of youth suicide victims were seen in the ED in the year preceding their death. EDs have been identified by the National Action Alliance for Suicide Prevention as one of two key arenas for improved services to reduce the proportion of at-risk youth who progress to suicide, yet LMC in emergency departments remains uncommon: ED clinicians often do not offer any lethal means counseling to at-risk patients and their families, and many do not counsel on reducing access to firearms in particular. For example, Grossman and colleagues reported that although 80% of emergency nurses in Illinois had recent experience with suicidal adolescents, only 28% provided LMC to parents. A record review found that psychiatric residents at a psychiatric emergency department assessed firearm access in only 3% of pediatric patients. ED-based lethal means counseling with parents of youth is a promising approach, but no RCTs have been conducted: Previous studies on LMC point to the need for fully powered RCTs in this area. A prospective follow-up study at a hospital ED where staff were instructed to deliver LMC messages to parents of all at-risk youth found that parents exposed to LMC were more likely to reduce the youth's access to lethal means than parents who were not exposed (75% vs. 48% for prescription drugs and 63% vs. 0% for firearms), but the number of gun- owning households (n=15) was very small. Promising evidence also comes from the pilot investigation that members of the study team conducted at Children's Hospital Colorado. Our research team, which includes experts in suicide prevention, LMC, evaluation science, qualitative research, adolescent mental health, and emergency medicine, is well-suited to rigorously evaluate the study aims using a mixed-methods approach. Our five hospital sites will collaborate with study personnel to: a) establish IRB procedures and approvals, b) develop a system (in nearly all sites via the electronic medical record) to flag eligible patients, document provision of intervention services, and provide parent contact information to the study team, c) require the clinicians who will provide LMC to take the online training during their sites' two-week phase-in period, d) host an in-person training with a study investigator at a staff meeting during the phase-in period, and e) implement the new lethal means counseling protocol on their start date. A lead ED clinician at each site will be the point of contact to help manage logistics and access medical record information for the study in accordance with HIPAA regulations. The Colorado-based project coordinator (Brandspigel) will work with hospitals to develop site-specific consent and data transfer processes.

suicide, lethal means, counseling, youth, parents, firearms, medication, suicidal, emergency department

We will use a stepped wedge sample design, a type of cluster randomized crossover design in which different clusters (in our case, hospitals) switch permanently from a control to an intervention state at different time points (steps) (Woertman, 2013). We chose a cluster-randomized design to prevent contamination that would likely occur were providers or patients randomized to receive different interventions within hospitals. All clusters start in the control condition (usual care), with hospitals switching to the intervention at consecutive 3- month time points, where the time of the switch is randomized for every hospital. At the end of the study period, all clusters will have switched from a control to an intervention condition.

During the intervention phase, mental health clinicians at EDs of participating hospitals are trained in lethal means counseling and implement the new protocol uniformly with eligible families.

Clinicians provide applicable families with locking devices and handouts on safe storage of firearms and medications.

Self-reported firearm and medication storage practices following the ED visit, as compared with on the day of the visit. Specifically, we will assess the following among gun-owning parents: 1) removal of firearms from the home; 2) change in firearm storage practices from keeping a firearm unlocked to storing it locked; and 3) change in practice from keeping a firearm loaded to unloaded. In addition, we will assess parental storage of medications (from unlocked to locked).

Qualitative, face-to-face or video semi-structured in-depth interviews with clinicians and parents/caregivers of adolescents eligible for our RCT using a grounded theory approach. The focus of the interviews will be on understanding how the intervention effects behavior change with an emphasis on reducing at-risk adolescents' access to firearms.

Self-reported firearm and medication storage practices several months after the ED visit, as compared with on the day of the visit. Specifically, we will assess the following among gun-owning parents: 1) removal of firearms from the home; 2) change in firearm storage practices from keeping a firearm unlocked to storing it locked; and 3) change in practice from keeping a firearm loaded to unloaded. In addition, we will assess parental storage of medications (from unlocked to locked).

Inclusion Criteria: Age limits pertain to the patient seen at the ED for a psychiatric or substance abuse crisis. It is their parents/guardians who receive the intervention. Parents must speak either English or Spanish and have a working telephone number or email address. Exclusion Criteria: A patient in institutional care would be excluded. Parents/guardians who were not at the ED with their child are excluded.

Miller M, Salhi C, Barber C, Azrael D, Beatriz E, Berrigan J, Brandspigel S, Betz ME, Runyan C. Changes in Firearm and Medication Storage Practices in Homes of Youths at Risk for Suicide: Results of the SAFETY Study, a Clustered, Emergency Department-Based, Multisite, Stepped-Wedge Trial. Ann Emerg Med. 2020 Aug;76(2):194-205. doi: 10.1016/j.annemergmed.2020.02.007. Epub 2020 Apr 16.