Abstract
Background:
Suicide is a leading cause of death for adults aged 18-64 in the United States, and suicide risk is highest among those with previous attempts. Two evidence-based treatments for suicide prevention exist: dialectical behavioral therapy (DBT), a year-long intensive treatment, and cognitive behavioral therapy (CBT), a brief (10-12 session) treatment. To our knowledge, no direct comparison of the two treatments yet exists
Aims:
To analyze the cost-effectiveness of DBT compared to CBT, in terms of both cost and quality of life, for the prevention of future suicide attempt among previous attempters.
Methods:
A Markov Model was developed to estimate the incremental cost-effectiveness ratio (ICER) in 2022 US dollars per quality-adjusted life-year (QALY). Published literature was reviewed to identify parameter estimates. The target population was US adults aged 18-64 with a previous suicide attempt. A time horizon of one year was used, and costs were calculated from the health care system perspective.
Results:
Compared to CBT, DBT is associated with an estimated incremental cost of $26,362 per QALY gained. One-way sensitivity analysis (OWSA) revealed consistent results, with DBT being cost-effective in most cases at a maximum threshold of $50,000 per QALY.
Discussion:
Results suggest that, compared to CBT, DBT is associated with comparable costs and mortality but higher quality of life. Due to the limited evidence base, caution is recommended when interpreting and generalizing results.
Implications for Health Care Provision and Use:
Given comparable cost and efficacy, patient preference for CBT or DBT must be incorporated in treatment selection. Ensuring patients play an active role in treatment selection has the potential to lead to improved clinical and health system outcomes.
Implications for Health Policies:
As manualized treatments, both DBT and CBT are appropriate for trained master’s level clinicians to deliver. Allowing master’s level clinicians to provide evidence-based care for suicide prevention may be a cost-effective strategy for quality service provision.
Implications for Future Research:
Future research grounded in conceptual theories of suicide that distinguish suicide risk from more general psychiatric risks are needed.
Introduction
The prevention of suicide is a salient and pressing issue[1]. Death by suicide is a growing problem in the United States. From 2000 through 2018, the age-adjusted suicide rate increased 35.2% [2]. Since 2008, suicide has ranked as the 10th leading cause of death for all ages in the United States. In 2016, suicide became the second leading cause of death for ages 10–34 and the fourth leading cause for ages 35–54 [2]. Recent trends illustrate that life expectancy in the United States is declining, unlike all other wealthy countries. Rising suicide rates among young adults is one of the primary contributors [3].
While sociodemographic correlates, such as gender, age, and sexual orientation are associated with higher risk of suicide [4], the greatest predictor of future suicide attempt is a previous attempt. 25.4% of suicide attempters re-attempt within the year [5], making their relative risk of suicide attempt 15-40 times higher than that of the general population. Suicide prevention programs specifically designed for this vulnerable population [6] have the potential for high public health impact.
Two evidence-based psychotherapeutic treatments exist for suicide prevention[7-12]. Dialectical behavioral therapy (DBT) is a year-long manualized psychotherapeutic treatment comprised of individual therapy and skills-development groups [9, 10]. Cognitive behavioral therapy (CBT) is a brief (10-12 session) manualized psychotherapeutic treatment [7, 13, 14]. Current recommendations link treatment selection to psychiatric diagnosis. DBT is recommended for the treatment of borderline personality disorder and substance use disorder, while CBT is recommended for the treatment of depression, anxiety, and some trauma-related disorders.
Although suicide is associated with a range of mental health diagnoses [15], these diagnoses only explain a small portion of suicide risk [4]. By contrast, individuals who attempt suicide may be better characterized by a common experience of intense pain, hopelessness, and social isolation [16]. Co-occurring substance use is also a better predictor of suicide completion than psychiatric diagnosis [4]. These clinical characteristics supersede diagnostic criteria and leave open the question of which treatment may be most effective when suicide, rather than psychiatric symptomology, is the imminent concern. Experimental evidence and economic evaluations indicate that either treatment is better than control conditions. All else equal, decision makers encourage brief treatment as the more cost-effective option[17]. However, to our knowledge, no direct comparison of the two treatments yet exists.
Treatment uptake and retention are crucial to real-world effectiveness. Up to a third of patients do not complete treatment [9]. Drop-out rates are often highest in the early months [11], when suicide risk is most acute [18]. Differences between the two treatments in duration, intensity, and format may impact uptake and retention. Yet, efficacy estimates use intent-to-treat analyses that mask these effects. A thorough comparison of the two treatments must take into account real-world conditions
To provide preliminary decision-making support to clinicians, patients, and health systems, we performed an economic evaluation drawing from published literature comparing DBT to CBT. We aimed to answer the question: what is the comparative effectiveness of these two treatments, in terms of both cost and quality of life, for the prevention of future suicide attempt among previous attempters, independent of the risk associated with substance use and taking into account pragmatic real-world conditions?
Methods: General
Target population
The base-case population was civilian, non-institutionalized adults aged 18-64 in the United States with at least one previous suicide attempt. The base-case population was chosen to reduce heterogeneity. Exclusion criteria included diagnosed co-occurring substance use disorder and cognitive disability. No subgroup analyses were performed.
Setting
The analysis was conducted with the US healthcare system in mind, where outcomes are intimately tied to treatment uptake, retention, and emergency department use. In the US, many behavioral health providers do not accept insurance [19]. Consequently, even insured patients may refuse treatment for financial reasons. In addition, approximately 19% of individuals with diagnosed mental illness are uninsured [20]. Only a quarter of individuals with any mental illness receive treatment [20]. Given these barriers, low levels of treatment initiation may be expected and are incorporated into the decision model.
In addition to financial barriers, individuals in need of treatment face access barriers [21, 22]. Patients face barriers related to transportation, childcare, and scheduling [11]. As a result, they may prefer brief treatments such as CBT. Conversely, more intensive support may lead to greater therapeutic impact and higher retention [9]. Retention rates thus vitally impact outcomes and are also incorporated into the decision model.
Given the insurance and payment landscape, patients in crisis turn to costly emergency departments. US emergency departments are legally required to provide care regardless of a patient’s ability to pay. For this reason, treatment following suicide attempt is often initiated in emergency rooms. The model used in this analysis consequently begins with the patient in the hospital.
Study perspective
The cost of implementing each intervention is derived from the US health sector perspective. This includes costs to third-party payers (private and public insurers) as well as treatment costs paid out-of-pocket by patients but does not include time or travel costs. This perspective was chosen to capture all costs directly attributed to the psychotherapeutic treatment.
Comparators
Long-term therapy: DBT
In its standard form, DBT involves weekly individual and group sessions for one year as well as regular clinical supervision for therapists. The treatment goal is to increase an individual’s capacity to tolerate emotional distress by identifying and practicing alternative coping strategies [9], which addresses the “pain” motivation for suicide [16].
Brief therapy: CBT
CBT is an umbrella term for a group of manualized treatments distinguished by their brief treatment duration (5-12 sessions). The treatment goal is to identify and reframe cognitive distortions [23], which addresses the “hopelessness” motivation for suicide [16].
Time horizon
This study analyzes costs and outcomes over a time horizon of one year from hospitalization following first suicide attempt. We used a one-year time frame for clinical and pragmatic reasons. Clinically, the risk of re-attempt is highest in the first year following previous attempt [5]. Pragmatically, suicide behaviors are highly sensitive to situational and life course factors [24]. A one-year time frame thus reduced uncertainty in the decision model.
The cycle length was chosen to be one month to remain consistent with the published literature from which parameters were drawn[11, 12].
Discount rate
Given the one-year time horizon, no discounting was applied.
Methods: Outcomes
Choice of health outcomes
The health outcomes for each intervention are evaluated in life years and quality-adjusted life years (QALY). Life years were chosen to capture the outcome of suicide death, the primary outcome of interest. QALYs were chosen as a more sensitive measure to capture the impact of suicide attempts that do not lead to death and the experience of active suicidal ideation.
Measurement of effectiveness
To identify relevant studies, we conducted two systematic searches of the literature using the citation database PubMed. In both searches, we included only studies classified as a clinical study, clinical trial, comparative study, controlled clinical trial, evaluation study, multicenter study, pragmatic clinical trial, and randomized controlled trial.
In the first, Mesh keywords “Suicide” and “Cognitive Behavioral Therapy” produced 147 results. An initial title screen resulted in 20 studies, abstract screen resulted in 3 studies, and a full review produced 1 study for inclusion. Primary reasons for exclusion were a different population (military personnel, adolescents and youth, bereaved, or institutionalized), different intervention (self-help, internet-based, or manual assisted), or lack of information on uptake and retention rates.
Because “dialectical behavioral therapy” was not a Mesh keyword, the second search used the query “dialectical behav* therapy” with the Mesh term “Suicide.” The initial search produced 80 results. An initial title screen resulted in 12 studies, abstract screen resulted in 6 studies, and a full review produced 1 study for inclusion. Similar to the first search, primary reasons for exclusion were a different population (co-occurring SUD, outside of the United States, no previous suicide attempt), different intervention (single session or shortened treatment), and lack of information on uptake and retention. A literature review and meta-analysis published in 2015 produced results similar to the current search strategy[25].
The two included studies were randomized controlled trials comparing the treatment of interest to treatment as usual. Base case transition probabilities were drawn directly from these studies. Brown et al [11] studied CBT (10-12 sessions) among adults 18-66 years old (n=120) hospitalized for suicide attempt. Patients were randomized in the hospital to CBT or treatment as usual (TAU) and followed for two years. The primary outcome captured was suicide attempt. Drop-out rates were reported throughout the trial time period.
McMain et al [12] studied DBT among adults 18-60 years old (n=180) diagnosed with borderline personality disorder with a history of at least two suicidal or nonsuicidal self-injurious episodes, with at least one episode in the three months prior to enrollment. Patients were randomized to DBT or general psychiatric management and followed for one year. The primary outcome was frequency and severity of suicidal and nonsuicidal self-harm episodes. Drop-out rates were reported for the duration of treatment.
Measurement and valuation of preference-based outcomes
Published literature was reviewed to answer three questions to inform preference-based outcomes. Utility estimates for untreated, or active, ideation were derived from estimates of suicidality, as opposed to psychiatric diagnoses such as major depression[26]. Utility estimates for treated ideation were drawn from mild to moderate depression symptom profiles [27].
Sensitivity analysis
Given the paucity of published literature specifically addressing efficacy and utility profiles among previous suicide attempters, we developed clinically-relevant hypothetical scenarios for sensitivity analyses. For transition probabilities, a high-effectiveness scenario postulated 100% uptake, retention, and prevention. In this scenario, all patients offered the treatment completed it, and no patients in treatment re-attempt suicide. A low-effectiveness scenario postulated 50% uptake and retention and a re-attempt rate of 5%.
For utility estimates, a low utility scenario postulated utility scores of 0 were used for active suicidal ideation and suicide attempt states, since the choice to end one’s life would imply that death (with utility score of 0) is preferable to life. A utility score of 0.5 was used for symptom management during treatment. High end estimates of utility were drawn from the highest reported utility scores in the literature.
Methods: Costs
Estimating resources and costs: Model-based economic evaluation
Quantity and cost estimates were obtained from published literature and routine sources. In its standard form, DBT is a yearlong treatment of weekly individual and group sessions and clinical supervision [9]. Informal contacts between sessions were not included in available data sources. McMain et al [12] reported average number of completed individual sessions, and this number was used as in the base case analysis. Since group sessions are held despite fluctuating attendance, we assumed that each patient had access to 50 groups during the year. Base case analysis posited a group size of 5 patients, with the cost of the therapist’s time divided by the group size. Sensitivity analysis posited group size of 3 patients (the minimal size for a group) and 8 patients (the maximum allowed in a DBT group). Similarly, we divided each therapist’s weekly hour of clinical supervision by their caseload. The base case analysis posited a caseload of 14 patients, with a caseload of 10 and 18 used for sensitivity analyses.
Brief CBT comprises 10-12 individual hour-long sessions over the course of 3 months. Base case analysis assumed 10 sessions were completed, while sensitivity analyses looked at 12 (at the high end) and 5 (at the low end). Number of sessions per month was calculated by dividing the number of total sessions by 3 months.
Treatment costs were estimated as hourly provider rates. Provider salaries were drawn from the US Bureau of Labor Statistics [28]. Hourly rates were calculated by dividing the annual salary by worked hours, assuming a full-time work schedule (40 hours/week for 50 weeks/year). Base case analysis used the annual salary of a clinical psychologist. Annual salaries for master’s level therapists and psychiatrists were used as low and high cases for sensitivity analysis.
Average hospital stay was obtained from the published literature looking at hospital stay due to suicidal ideation [29]. Hospital costs per day were derived from national estimates [30]. Total hospital costs were calculated by multiplying cost per day by number of days.
Currency and conversion
All costs are reported in 2022 US dollars.
Methods: Data Analytic Procedures
Choice of model
A cohort-level, deterministic Markov model assuming independence [31] was constructed to model the progression of treatment from hospitalization to death or recovery (Figure 1). Consistent with published literature on the treatments of interest, a Markov model was used to capture the cyclical nature of symptomology, where treatment dropout and recurrence are common and may occur at any point during treatment.
Figure 1:
Markov Model, DBT
A modified model was used to capture the 3-month nature of brief CBT (Figure 2).
Figure 2:
Markov Model, CBT
Assumptions
The model as constructed assumes that all individuals start in the hospital following suicide attempt. Individuals move from the hospital to treatment or no treatment. It is assumed that individuals do not spontaneously begin treatment, such that the only path to begin treatment is through hospital referral. As a high-acuity population, the model assumes that all individuals maintain some level of suicidal ideation throughout the year. No treatment is associated with more acute symptomatology and lower utility state. All individuals are at risk of suicide attempt at all times.
Results
Study parameters
The average cost associated with crisis hospitalization stay following suicide attempt is $4,984 ($890 per day for 5.6 days). A DBT therapist spends on average 3.89 hours per month per patient in treatment, reflecting an average of 32 individual group sessions over 12 months, a group size of 5 members, and a caseload of 14. CBT therapist spends on average 3.33 hours per month per patient, calculated by dividing 10 sessions over three months, and treatment is time-limited. The calculated hourly wage of a psychologist is $51, while that of a master’s level clinician is $23 and a psychiatrist $110.
Rates of declining treatment differ between the two treatments. In the trials from which estimates were drawn, approximately 19% of patients declined CBT when offered, while 4.3% declined DBT when offered. Drop-out rates are similar between the two therapies, while re-attempt rates are notably lower among those receiving DBT.
Patients are assumed to experience suicidal ideation throughout the year, given the symptom acuity of the target population. Utility weights are summarized in Table 3.
Table 3:
Utility parameters
Incremental costs and outcomes
On the basis of deterministic Markov Models, both treatments are associated with similar rates of completed suicide (0.3% of the population within the year). Universal provision of DBT is associated with higher QALYs than CBT (0.53 vs. 0.57). The difference in quality adjusted life years is 0.04, or approximately two weeks of quality adjusted life. Total costs associated with DBT are $7,002 per patient while total costs associated with cognitive therapies are $6,166 per patient, leading to a difference of $837. The incremental cost-effectiveness ratio (ICER) is $26,362 per QALY gained.
Characterizing uncertainty
Univariate sensitivity analyses are displayed in a tornado diagram of the most influential variables (Figure 3). In this diagram, each bar represents the impact of uncertainty in an individual variable on the ICER. Transition probabilities of DBT efficacy (re-attempt rate) and CBT acceptability (decline and dropout) have the highest impact on the ICER, followed by provider hourly rate.
Figure 3:
Tornado diagram of one-way sensitivity analysis
Discussion
Study findings
Our economic evaluation, based on two published trials with individuals who had previously attempted suicide, found that long-term DBT compared to brief CBT is associated with 0.04 higher QALYs, or approximately two weeks of quality adjusted life, with an ICER of $26,362 per QALY gained. Estimated differences in life-years gained, or attempted suicides prevented, were negligible. However, DBT was associated with both higher retention and longer treatment duration, explaining the estimated improvements to quality of life. The estimated cost difference between DBT and CBT was $837 per year, including costs of crisis hospitalization and provider salary. Deterministic univariate sensitivity analysis yielded consistent results, with DBT being cost-effective in most cases at a maximum threshold of $50,000 per QALY.
Limitations
Study parameters were extracted from two clinical trials. Results are thus suggestive and intended to provide preliminary decision-making support in the absence of a clinical trial comparing DBT and CBT among the target population. In addition, clinical trial participants represent a subset of the general population [32]. Caution is warranted when interpreting and generalizing results.
The choice to limit the target population and treatments resulted in the exclusion of numerous trials that investigated suicide outcomes in related populations and settings. This was a deliberate choice. The current study balanced the costs and benefits of introducing uncertainty via a smaller, but more homogenous target population as opposed to introducing bias via a larger, but more heterogeneous population.
Evidence suggests there is merit to treating suicide risk as distinct from other psychiatric risks [4]. The risk of future attempt is significantly different among the general psychiatric population compared to those who have made a previous attempt [15]. Yet, because suicide attempts remain relatively rare, alternate inclusion criteria are often used in clinical investigations [4]. Significant heterogeneity in psychiatric presentation, symptomology, and risk is present in the mental and behavioral health fields. Including trials conducted among more general psychiatric populations carries the risk of introducing bias. Specifically, such studies may overestimate treatment uptake, retention, and efficacy. Utility scores associated with suicidal ideation common to psychiatric disorders may underestimate the pain experienced by those who advance towards making an attempt. By definition, there is a lack of retrospective data among those who die by suicide[4], leading to survivor bias in utility estimates. Strict inclusion criteria were thus used to reduce bias at the cost of greater uncertainty.
Deterministic OWSA were used to estimate uncertainty. The described limitations in the published literature meant the information needed to conduct a probabilistic sensitivity analysis were lacking. The inclusion of additional assumptions carried the risk of introducing additional bias while providing a false level of confidence. Results from the OWSA are intended to provide rough upper- and lower-bounds to the point estimate and are presented with the caveat to interpret with caution.
The one-year time frame precluded estimates at the conclusion DBT, which is a yearlong treatment, and excluded long-term outcomes. The decision to use a one-year time frame reflected the desire to reduce uncertainty. While suicidal ideation may be chronic, suicide attempt is acute and sensitive to situational factors [24]. An extended time-frame would have required predicting the probability of future life shocks, which was beyond the scope of the current project.
Implications for healthcare provision and use
To the author’s knowledge, this is the first study comparing the two treatments in question and in relation to suicide risk specifically. Previous evaluations have shown both DBT and CBT to be clinically and cost-effective when compared to treatment as usual [33-35]. Such studies indicate that treatment is better than no treatment but provide little direction on the selection of treatments. The current study contributes to the literature by illustrating that, despite differences in duration, intensity, and format, DBT and CBT incur a similar magnitude of costs to the healthcare system over a one-year period.
The implication for healthcare systems is that shorter-term treatment does not necessarily mean lower costs[17]. From a cost standpoint, the health system should be indifferent to selection of DBT or CBT for suicide prevention. DBT was associated with higher quality of life, indicating the longer-term treatment may be preferable. However, OWSA identified treatment uptake and retention as high impact factors on cost-effectiveness. The key takeaway for health systems is that, given comparable cost and efficacy, patient preference must be incorporated in treatment selection. Ensuring patients play an active role in treatment selection is likely to lead to improved clinical and health system outcomes.
Implications for health policies
Mental health clinician licensing and scope of practice is an active area of policy discussion. The mental health workforce is comprised of psychiatrists (roughly 5%), psychiatric nurse practitioners, physician’s assistants, and psychologists (roughly 20%), and master’s level professionals (over 70%)[36]. In the medical field, allowing alternative health professionals such as physician assistants and nurse practitioners to practice at the top of their license has led to improvements in costs and utilization [37]. However, similar developments in mental health policy are lagging. Despite their large contribution to the workforce, master’s level clinicians face limitations in their scope of practice due to licensing[38] and reimbursement [39] policies.
As manualized treatments, both DBT and CBT are appropriate for trained master’s level clinicians to deliver. As long as the clinician adheres to treatment protocols, no difference in outcomes have been observed by licensure or level of education[40]. OWSA found that that the hourly wage of provider had significant impact on the ICER. The implication for health policy is that allowing master’s level clinicians to provide evidence-based care for suicide prevention may be a cost-effective strategy for quality service provision.
Implications for future research
The paucity of literature specifically on individuals with past suicide attempt highlights a crucial area for future research. Given the acuity and specific needs of this population, conclusions drawn from more heterogeneous samples may not apply. Yet, given the public health burden of suicide, greater understanding of this population is urgently needed. Future research with robust foundations in conceptual theories of suicide[16] that distinguish suicide risk from more general psychiatric risks would help fill this gap.
Conclusion
The prevention of death by suicide is an urgent public health need. Targeted interventions for individuals with previous suicide attempt are a promising direction. Results of the current analysis suggest that, compared to CBT, DBT is associated with comparable costs and mortality but higher quality of life. Potential implications for health systems and policy are discussed. Due to the limited evidence base, caution is recommended when generalizing results.
Table 1:
Cost and utilization parameters
Table 2:
Transition probabilities
| Base | High | Low | |
|---|---|---|---|
| No treatment → Re-attempt | 1.7% [7, 8, 11] | - | - |
| No treatment → No treatment | 98.3% | - | - |
| Attempt → Death | 6.0% [41] | - | - |
| Attempt → Hospital | 94.0% | - | - |
| Cognitive behavioral therapy [11] | |||
| Hospital → Treatment | 81.1% | 50.0% | 100.0% |
| Hospital → No treatment | 18.9% | 50.0% | 0.0% |
| Treatment → Re-Attempt | 2.3% | 5.0% | 0.0% |
| Treatment → Drop out (Month 1) | 10.0% | 50.0% | 0.0% |
| Treatment → Treatment (Month 1) | 87.7% | 45.0% | 97.7% |
| Treatment → Drop out (Month 2) | 0.0% | 50.0% | 0.0% |
| Treatment → Treatment (Month 2) | 97.7% | 45.0% | 97.7% |
| Treatment → Drop out (Month 3) | 3.0% | 50.0% | 0.0% |
| Treatment → Completion | 94.7% | 45.0% | 0.0% |
| Completion → Attempt | 1.7% | - | - |
| Completion → Completion | 98.3% | - | - |
| Dialectical behavioral therapy [12] | |||
| Hospital → Treatment | 95.7% | 50.0% | 100.0% |
| Hospital → No treatment | 4.3% | 50.0% | 0.0% |
| Treatment → Re-Attempt | 0.4% | 5.0% | 0.0% |
| Treatment → Drop out | 3.2% | 50.0% | 0.0% |
| Treatment → Treatment | 96.4% | 50.0% | 100.0% |
Source of funding:
Research reported in this publication was supported by the Agency for Healthcare Research and Quality (grant number T32HS013853, to the University of Washington). The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.
References
- 1.Rockett IR, et al. , Mortality in the United States from self-injury surpasses diabetes: a prevention imperative. Injury prevention, 2019. 25(4): p. 331–333. [DOI] [PubMed] [Google Scholar]
- 2.Garnett MF, Curtin SC, and Stone DM, Suicide Mortality in the United States, 2000-2020. 2022, Centers for Disease Control and Prevention. [PubMed] [Google Scholar]
- 3.Woolf SH and Schoomaker H, Life expectancy and mortality rates in the United States, 1959-2017. Jama, 2019. 322(20): p. 1996–2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Klonsky ED, May AM, and Saffer BY, Suicide, suicide attempts, and suicidal ideation. Annual review of clinical psychology, 2016. 12: p. 307–330. [DOI] [PubMed] [Google Scholar]
- 5.Spirito A, Boergers J, and Donaldson D, Adolescent suicide attempters: post-attempt course and implications for treatment. Clinical Psychology & Psychotherapy: An International Journal of Theory & Practice, 2000. 7(3): p. 161–173. [Google Scholar]
- 6.Keyes KM and Galea S, Population Health Science. 2016: Oxford University Press. [Google Scholar]
- 7.Rudd MD, et al. , Brief cognitive-behavioral therapy effects on post-treatment suicide attempts in a military sample: results of a randomized clinical trial with 2-year follow-up. American journal of psychiatry, 2015. 172(5): p. 441–449. [DOI] [PubMed] [Google Scholar]
- 8.Goodman M., et al. , A randomized trial of dialectical behavior therapy in high-risk suicidal veterans. The Journal of Clinical Psychiatry, 2016. 77(12): p. 4031. [DOI] [PubMed] [Google Scholar]
- 9.Linehan MM, et al. , Dialectical behavior therapy for high suicide risk in individuals with borderline personality disorder: a randomized clinical trial and component analysis. JAMA psychiatry, 2015. 72(5): p. 475–482. [DOI] [PubMed] [Google Scholar]
- 10.Linehan MM, et al. , Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Archives of general psychiatry, 2006. 63(7): p. 757–766. [DOI] [PubMed] [Google Scholar]
- 11.Brown GK, et al. , Cognitive therapy for the prevention of suicide attempts: a randomized controlled trial. Jama, 2005. 294(5): p. 563–570. [DOI] [PubMed] [Google Scholar]
- 12.McMain SF, et al. , A randomized trial of dialectical behavior therapy versus general psychiatric management for borderline personality disorder. american Journal of Psychiatry, 2009. 166(12): p. 1365–1374. [DOI] [PubMed] [Google Scholar]
- 13.Mewton L and Andrews G, Cognitive behavioral therapy for suicidal behaviors: improving patient outcomes. Psychology research and behavior management, 2016. 9: p. 21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sahranavard S and Miri MR, A comparative study of the effectiveness of group-based cognitive behavioral therapy and dialectical behavioral therapy in reducing depressive symptoms in Iranian women substance abusers. Psicologia: Reflexão e Crítica, 2018. 31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chung DT, et al. , Suicide rates after discharge from psychiatric facilities: a systematic review and meta-analysis. JAMA psychiatry, 2017. 74(7): p. 694–702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Klonsky ED and May AM, The three-step theory (3ST): A new theory of suicide rooted in the “ideation-to-action” framework. International Journal of Cognitive Therapy, 2015. 8(2): p. 114–129. [Google Scholar]
- 17.Giannetti V., Brief treatment and mental health policy, in Handbook of the brief psychotherapies. 1990, Springer. p. 79–90. [Google Scholar]
- 18.Suominen K., et al. , Completed suicide after a suicide attempt: a 37-year follow-up study. American Journal of Psychiatry, 2004. 161(3): p. 562–563. [DOI] [PubMed] [Google Scholar]
- 19.National Council for Behavioral Health, Advocacy Roadmap: A Path to a Healthier America. 2020. [Google Scholar]
- 20.Foundation, K.F., Facilitating Access to Mental Health Services: A Look at Medicaid, Private Insurance, and the Uninsured. 2017. [Google Scholar]
- 21.Andersen RM, Revisiting the behavioral model and access to medical care: does it matter? Journal of health and social behavior, 1995: p. 1–10. [PubMed] [Google Scholar]
- 22.Andersen R and Newman JF, Societal and individual determinants of medical care utilization in the United States. The Milbank Quarterly, 2005. 83(4): p. Online-only-Online-only. [PubMed] [Google Scholar]
- 23.Powers MB, de Kleine RA, and Smits JA, Core mechanisms of cognitive behavioral therapy for anxiety and depression: A review. Psychiatric Clinics of North America, 2017. [DOI] [PubMed] [Google Scholar]
- 24.Bagge CL, Glenn CR, and Lee H-J, Quantifying the impact of recent negative life events on suicide attempts. Journal of abnormal psychology, 2013. 122(2): p. 359. [DOI] [PubMed] [Google Scholar]
- 25.Inagaki M., et al. , Interventions to prevent repeat suicidal behavior in patients admitted to an emergency department for a suicide attempt: a meta-analysis. Journal of Affective Disorders, 2015. 175: p. 66–78. [DOI] [PubMed] [Google Scholar]
- 26.van Spijker BA, et al. , Disability weights for suicidal thoughts and non-fatal suicide attempts. Journal of affective disorders, 2011. 134(1-3): p. 341–347. [DOI] [PubMed] [Google Scholar]
- 27.Schaffer A., et al. , Utility scores of symptom profiles in major depression. Psychiatry research, 2002. 110(2): p. 189–197. [DOI] [PubMed] [Google Scholar]
- 28.US Bureau of Labor Statistics, National Industry-Specific Occupational Employment and Wage Estimates: NAICS 621330 Offices of Mental Health Practitioners (except Physicians). 2020. [Google Scholar]
- 29.Owens P., et al. , Emergency department visits related to suicidal ideation, 2006–2013: statistical brief# 220. 2006. [Google Scholar]
- 30.Karaca ZM, B.J., Costs of emergency department visits for mental and substance use disorders in the United States, 2017, in Statistical Brief, H.C.U. Project, Editor. 2020, Agency for Healthcare Research and Quality: Rockville, MD. [PubMed] [Google Scholar]
- 31.Brennan A, Chick SE, and Davies R, A taxonomy of model structures for economic evaluation of health technologies. Health economics, 2006. 15(12): p. 1295–1310. [DOI] [PubMed] [Google Scholar]
- 32.Kennedy-Martin T., et al. , A literature review on the representativeness of randomized controlled trial samples and implications for the external validity of trial results. Trials, 2015. 16(1): p. 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Krawitz R and Miga EM, Cost-effectiveness of dialectical behaviour therapy for borderline personality disorder, in The Oxford handbook of dialectical behaviour therapy. 2019. [Google Scholar]
- 34.Lynch FL, et al. , Cost-effectiveness of an intervention to prevent depression in at-risk teens. Archives of general psychiatry, 2005. 62(11): p. 1241–1248. [DOI] [PubMed] [Google Scholar]
- 35.Tarrier N, Taylor K, and Gooding P, Cognitive-behavioral interventions to reduce suicide behavior: a systematic review and meta-analysis. Behavior modification, 2008. 32(1): p. 77–108. [DOI] [PubMed] [Google Scholar]
- 36.Health Resources and Service Administration, National Projections of Supply and Demand for Behavioral Health Practitioners: 2013-2025. 2016, National Center for Health Workforce Analysis, Substance Abuse and Mental Health Services,: Rockville, MD. [Google Scholar]
- 37.Morgan PA, et al. , Impact of physicians, nurse practitioners, and physician assistants on utilization and costs for complex patients. Health Affairs, 2019. 38(6): p. 1028–1036. [DOI] [PubMed] [Google Scholar]
- 38.Page C., et al. , A Descriptive Analysis of State Credentials for Mental Health Counselors, Behavioral Health Workforce Research Center, Editor. 2017, Univeristy of Michigan: School of Public Health,. [Google Scholar]
- 39.Barrasso J., Mental Health Access Improvement Act of 2021. 2021: Washington, DC. [Google Scholar]
- 40.Davidson K., et al. , Therapist competence and clinical outcome in the Prevention of Parasuicide by Manual Assisted Cognitive Behaviour Therapy Trial: the POPMACT study. Psychological medicine, 2004. 34(5): p. 855–863. [DOI] [PubMed] [Google Scholar]
- 41.Langlois S and Morrison P, Suicide deaths and suicide attempts. Health reports, 2002. 13(2): p. 9–22. [PubMed] [Google Scholar]