Suicide Risk Before Mental Health Treatment Initiation: Implications for Screening and Access to Care

In 2012, the U.S. Surgeon General and the National Action Alliance for Suicide Prevention jointly published an updated National Strategy for Suicide Prevention (first issued in 2001). This statement proposed 13 goals and 60 objectives to reduce suicide in the United States (e.g., improving surveillance systems, reducing access to lethal means, and improving suicide screening across settings) (1). Suicide rates have continued to rise through the first two decades of the 21st century (2), and the COVID-19 pandemic has placed considerable strain on the U.S. health care system. As underscored by the U.S. Surgeon General’s 2021 Call to Action (3), more work needs to be done. (For readers interested in the National Strategy resources, we refer to zerosuicide.edc.org and sprc.org.) This Open Forum builds on recent epidemiological research investigating suicide risk that coincides with initiating mental health treatment to highlight key challenges related to screening and access to suicide prevention, as well as to discuss interventions intended to reduce suicide risk before initiation of mental health treatment.

Mounting research has demonstrated that the greatest risk for suicidal behavior precedes antidepressant initiation (4, 5), pharmacologic treatment for attention-deficit hyperactivity disorder (6), and mood-stabilizing antiseizure drug treatment (7) (measured in various ways across studies, including 30 days, 90 days, or 1 year before treatment initiation). Although we do not know the extent to which these findings would apply to other psychiatric medications (e.g., benzodiazepines), one study suggests that these observations do extend to psychotherapy (4). The studies help elucidate a common clinical phenomenon: individuals seek treatment in times of distress coinciding with increased suicidality risk. It is possible that individuals who initiate mental health treatment are at increased risk for suicidality because of underlying factors that prompted treatment initiation. Importantly, suicidality risk is present prior to clinical engagement. The clinical implication of studies on treatment initiation and suicide risk is sobering—if initiation of medication or psychotherapy serves as the point at which suicide screening and subsequent intervention commence, it may occur too late for some patients. The timing of suicide risk relative to that of treatment initiation highlights considerations for suicide screening and access to care (3).

Screening is crucial to suicide prevention, because it serves as a key point for providers to communicate compassion, triage care, and plan treatment. Determination of the optimal times to screen patients for underlying or emerging suicide risk is complex. Universal screening initiatives have received criticism, and we note that we are not necessarily advocating for universal screening. However, what is clear from the research cited above is that screening should not be delayed until the start of mental health treatment, which raises issues about access to and continuity of care. When a relationship with a provider is established, a patient who has preexisting mental disorders or suicidality clearly requires ongoing monitoring for suicidality. This monitoring may be more feasible in mental health outpatient settings where discussions about suicidality and provider training on suicide intervention are likely to be more prevalent. In instances in which psychotherapy precedes medication initiation or vice versa, collaboration among mental health professionals provides an opportunity to share findings from ongoing suicide risk assessments before the start of additional treatment.

Although suicide screening performed by mental health providers may be preferred (3), nonpsychiatric providers prescribe most of the psychiatric medications dispensed in the United States (8), and the majority of individuals who die by suicide had been in contact with a primary care provider in the year before their death (9). General practitioners prescribing psychiatric medications may have insufficient integration with mental health professionals. In such instances, it may be particularly difficult to determine how to expand suicide screening. These decisions may depend on differences in the sensitivity and specificity of screening tools, as well as on personnel or resource limitations. For example, the ninth item on the Patient Health Questionnaire–9 (PHQ-9) is frequently used to identify suicidality in outpatient settings, but it does not differentiate between thoughts about death and self-injury, nor does it specifically ask about suicidal ideation. These limitations lead to high false-positive rates compared with suicide-specific screens (e.g., the Columbia Suicide Severity Rating Scale [C-SSRS]) (10). On the other hand, false negatives due to a lack of (sensitive) screening continue to propagate, reflecting patients’ beliefs that treatment is not necessary or helpful, stigmatizing beliefs about suicidality, concerns regarding confidentiality, fear of hospitalization, and desire to manage suicidality on one’s own (11). Whether the PHQ-9 serves as a gateway screen is setting specific, but suicide-specific screeners (e.g., the Ask Suicide-Screening Questionnaire) and subsequent assessments (e.g., the full C-SSRS) are necessary to clarify suicide risk. However, single suicide items or suicide questionnaires, although helpful in triaging, are limited in their capacity to develop an understanding of suicide risk and inform clinical decisions. A conceptualization of individual (e.g., psychiatric and medical histories) and contextual factors (e.g., support systems and experiences of discrimination) helps to provide a more robust understanding of factors maintaining suicidality. The process of screening and thorough assessment can be time consuming, likely requires additional training or training in mental health, and may be most effective in the context of a trusting and caring relationship with a provider.

It is crucial that suicide prevention does not end at risk assessment, because these screening tools are only as effective as the actions taken afterward. However, if, as recent data suggest, approximately 3.5% of suicide screens are positive in outpatient settings (12), considerable resources would be required to administer and interpret screens, conduct follow-up assessments, provide brief interventions, alter medication dosage, and connect patients to further care. Additionally, many providers not trained in mental health are uncomfortable discussing suicidality for various reasons, including concern of liability, misconceptions about encouraging suicidality, or lack of self-efficacy if suicidality is endorsed (13). Without changes at the health care–systems level that address workforce capacity and accessibility, the burden on individual providers may be too great. Efforts related to access to care are needed to support, increase, and maintain adequate suicide prevention.

The possibility that suicide risk is highest before treatment initiation raises the question about what to do when individuals are not yet established in or do not have access to care. It is very likely that patients will seek care from a provider when their mental health is already deteriorating. Intervening before treatment initiation is possible when relationships with providers are long standing, but health care in the United States is often intermittent and fragmented.

To identify individuals in need and connect them with care, the health care system needs to increase the number of mental health providers, improve care coordination across providers, and train individuals outside the health care system. First, the shortage of mental health professionals is a chronic issue in the United States and is greatly exacerbated in rural areas (14). If providers are accepting new patients, these patients may be placed on long waitlists. Legislation is unlikely to resolve shortages alone; factors such as public perceptions around the mental health profession, burnout and poor job satisfaction, and low salaries may be reasons people do not enter or remain in the profession. Regardless, as a first step, legislation may help alleviate some of the burden, both for patient access to care and for retention of mental health providers. Examples include expanding loan repayment programs to encourage professionals to work in underserved areas and improving Medicaid reimbursement rates for mental health treatment. Other legislative tactics may include increasing the scope of practice for those trained in mental health in certain states (e.g., psychiatric mental health advanced practice nurses) and broadening telehealth policies to ensure that reimbursement expands the reach of existing providers. Related policies that increase access to high-speed Internet could further address barriers in access to care. Additionally, policies that fund research initiatives examining stepped care models with technology-based interventions would be beneficial to increase care access without recruiting and training a new workforce and to potentially alleviate waitlist issues.

Second, policies targeting care coordination can help facilitate suicide detection before mental health treatment. Integrated behavioral health care models may offer an efficient and practical approach to connecting providers with patients. Various models have been proposed (e.g., accountable care organizations, patient-centered medical homes, and colocated behavioral health in primary care), although research is needed to determine which models are most effective for specific populations. Policies that support these models and research that evaluates them are crucial for adopting and tailoring programs in the future. In addition to care delivery models, care coordination may be facilitated through training of medical providers, such as primary care physicians, on suicide prevention protocols. Engagement with and maintenance of such interventions have been challenging, potentially because of time constraints and competing priorities, again emphasizing the need for coordinated care models.

Third, expanding the catchment area of suicide detection to sites outside traditional health care settings is crucial and includes leveraging organizations that are embedded in communities and have existing relations with community members (e.g., nonprofit organizations, churches, and schools). Such expansion would address not only structural barriers to access to care but also issues related to historical and pervasive racism in health care systems that are harmful to patients of color. Gatekeeper training models (i.e., programs that train individuals in contact with community members to identify suicide risk and refer those at risk to care), however, have mixed support, potentially because of the low base rates of suicide in many communities, leading to skill deterioration (15). Gatekeeper training models may be more effective for populations with elevated suicide risk (e.g., veterans and youths belonging to sexual and gender minority groups), whereas community partnerships with health care groups may benefit from universal suicide prevention programs (e.g., gun safety programs with local firearm retailers and stable housing options in neighborhoods) (15).

Conclusions

Research demonstrates that increased suicide risk precedes the initiation of at least some mental health treatments, highlighting complex clinical considerations to address such risk depending on the setting and its resources. Expanding risk assessment in outpatient settings is necessary but will be insufficient if barriers to care access are not addressed at the health care–systems level.