Skip to main content
  • Study Protocol
  • Open access
  • Published:

Managed care updates of subscriber jail release to prompt community suicide prevention: clinical trial protocol



Recent jail detention is a marker for trait and state suicide risk in community-based populations. However, healthcare providers are typically unaware that their client was in jail and few post-release suicide prevention efforts exist. This protocol paper describes an effectiveness-implementation trial evaluating community suicide prevention practices triggered by advances in informatics that alert CareSource, a large managed care organization (MCO), when a subscriber is released from jail.


This randomized controlled trial investigates two evidence-based suicide prevention practices triggered by CareSource’s jail detention/release notifications, in a partial factorial design. The first phase randomizes ~ 43,000 CareSource subscribers who pass through any Ohio jail to receive Caring Contact letters sent by CareSource or to Usual Care after jail release. The second phase (running simultaneously) involves a subset of ~ 6,000 of the 43,000 subscribers passing through jail who have been seen in one of 12 contracted behavioral health agencies in the 6 months prior to incarceration in a stepped-wedge design. Agencies will receive: (a) notifications of the client’s jail detention/release, (b) instructions for re-engaging these clients, and (c) training in suicide risk assessment and the Safety Planning Intervention for use at re-engagement. We will track suicide-related and service linkage outcomes 6 months following jail release using claims data.


This design allows us to rigorously test two intervention main effects and their interaction. It also provides valuable information on the effects of system-level change and the scalability of interventions using big data from a MCO to flag jail release and suicide risk.

Trial registration

The trial is registered at (NCT05579600). Registered 27 June, 2023.

Peer Review reports

Data Category


Primary registry and trial identifying number


Date of registration in primary registry


Secondary identifying numbers

1P50MH127512-01A1 8577

Source(s) of monetary or material support

National Institute of Mental Health (NIMH)

Primary sponsor

National Institute of Mental Health (NIMH)

Secondary sponsor(s)


Contact for public queries

Sarah Arias; Department of Psychiatry and Human Behavior, Butler Hospital, 345 Blackstone Blvd., Providence, RI 02906. Phone: 401–455-6261. Fax: 401–455-6235. E-mail:

Contact for scientific queries

Sarah Arias; E-mail:

Public title

Suicide Prevention for Justice Involved Managed Care Subscribers

Scientific title

Managed Care Updates of Subscriber Jail Release to Prompt Community Suicide Prevention: Clinical Trial Protocol

Countries of recruitment

United States

Health condition(s) or problem(s) studied

Suicide risk


Intervention #1: Caring Contacts

Intervention #2: Reports, Re-engagement, and Training (RRT)

Key inclusion and exclusion criteria

Ages eligible for study: ≥ 18 years old

Sexes eligible for study: both

Accepts healthy volunteers: no

Inclusion criteria: adults (≥ 18 years), (For CC condition): CareSource subscriber, recently spent time in jail; (For RRT condition): Behavioral health (BH) providers/staff employed at a CareSource BH agency

Exclusion criteria: None

Study type


Allocation: Randomized

Intervention model: Factorial

Masking: None

Primary purpose: Health Services Research

Phase III

Date of first enrollment

May 2023

Target sample size

44,000 (Caring Contacts); 120 (RRT)

Recruitment status


Primary outcome(s)

Suicide attempts

Key secondary outcome(s)

Inpatient and emergency department mental health care visits; Outpatient behavioral health service linkage; Number of arrests

Date and Version number

10/17/23; Version 9


Passing through jail is a marker for trait and state suicide risk. Before COVID-19, there were more than 10 million jail bookings annually in the U.S [1]. Half of those booked (40–50%) report lifetime suicide ideation or risk behaviors and 13–20% have attempted suicide [2]. Individuals are most likely to be arrested when acutely ill (i.e., manic or psychotic), and disproportionately come from groups at high risk for suicide, such as males, persons with mental health and substance use disorders, socially disenfranchised, and those who have previously engaged in suicide behaviors [3]. Compared to demographically matched people, the suicide rate is 29–58 times as high in the months after jail release from incarceration [4, 5] and 3.4–18.2 times as high in the two years after release [4,5,6,7,8,9]. Given ~ 10% of all suicides with known causes in the U.S. occur in the context of a recent criminal or legal stressor (often arrest and jail detention) [10, 11], targeting suicide risk after jail detention could have a noticeable impact on national suicide rates.

High jail admissions and discharge volumes, short jail stays, and understaffing mean that many county jails do not have capacity to coordinate care between jails and outside health agencies for suicide prevention or other health needs [12,13,14,15,16,17,18,19,20,21]. Unlike prison, jail stays are usually brief (a few days for those in pretrial status [22]) and releases are often unscheduled. Outside jail, many people with criminal-legal (CL) involvement are supported by professionals within publicly funded systems, who themselves face resource restrictions, are often unaware that their client was in jail, and may discharge the client for missing appointments. With over ~ 3,100 jails and thousands of behavioral health (BH) agencies providing services, continuity of systems to reduce morbidity and mortality is a challenge [23,24,25,26,27,28].

This protocol paper describes a randomized controlled trial (RCT) developed in partnership with CareSource, a large nonprofit managed care organization (MCO) headquartered in Ohio. MCOs like CareSource, are ideal partners to assist with jail-community service linkage because their catchment areas span multiple health and jail systems. In 2018, CareSource partnered with a justice alert company (which scans publicly available booking and release data and provides alerts when individuals are booked into or released from jail) to track jail booking/release data from jails nationally for adult Ohio CareSource members. The linking system sits within CareSource (to protect confidential health information) and automatically crosswalks CareSource’s Ohio adult Medicaid subscriber list against publicly available national booking and release data weekly. The resulting Jail-Medicaid project alerts CareSource when its subscribers pass through jail so that CareSource can reach out and link them to community healthcare services. Medicaid members were identified for inclusion in this project due to the prevalence of complex behavioral health/medical conditions that can benefit from enhanced care coordination strategies. This novel data flagging approach, made possible and generalizable by advances in biomedical informatics, addresses long-standing challenges to making care connections following jail detention.

This trial employs: (1) intersection with jail as a marker of suicide risk in the general population; and (2) CareSource’s notification system to support suicide prevention and service linkage in a way that is feasible and sustainable at scale within the existing healthcare and MCO infrastructure. To our knowledge, this n ~ 43,000 trial will be the largest RCT for any condition in any CL involved population to date.


The study is part of a larger partnership with the National Center for Health and Justice Integration for Suicide Prevention (NCHATS), funded by the National Institute of Mental Health. This newly established suicide prevention center focuses on building information bridges between healthcare organizations and CL systems to identify individuals at risk for suicide and connect them to care.

Study design

The project involves testing of two evidence-based suicide prevention practices triggered by CareSource’s detention/release notifications in a partial factorial design, a rigorous and efficient study design that does not create bias [29]. The first phase targets ~ 43,000 Ohio CareSource Medicaid subscribers released from jail over 12 months, who are either randomized to receive Caring Contact letters (CC) from CareSource or Usual Care (UC) monthly for 6 months following jail release. The second phase (running simultaneously), Reports, Re-engagement, and Training (RRT), involves ~ 6,000 of the 43,000 CareSource subscribers who were seen at one of twelve contracted BH agencies in the 6 months prior to detention. During the RRT intervention, the BH agencies will receive: (a) notifications of jail release for their existing clients as a signal for potential suicide risk, (b) instructions for re-engaging these clients in services; (c) training in suicide risk screening (Columbia Suicide Severity Rating Scale [C-SSRS [30]]) and intervention practices (Safety Planning Intervention [SPI [31]]) for providers; (d) notification to use C-SSRS for all clients and SPI when a client meets additional risk criteria per the Mental Health Research Network (MHRN) risk prediction model [32] (calculated from CareSource claims data). Participants are followed for 6 months following jail release using claims data provided by CareSource. Outcomes include:

  1. 1.

    Effectiveness: (a) decrease in medically treated (MT) suicide attempts (primary); (b) decrease in all-cause injury and poisoning; (c) increased linkage to outpatient BH services; (d) increased use of the CareSource24® Nurse Advice line (CareSource24® line); (e) decreased inpatient and ED mental health care visits; and (f) decreased return to jail detention;

  2. 2.

    Mechanisms: service linkage and CareSource24® line use as mediators of the effects of the intervention/s on suicide outcomes;

  3. 3.

    Cost-effectiveness and return on investment: (a) program cost, (b) the cost of suicide-related and overall medical care; (c) cost-effectiveness; and (d) net cost to the MCO;

  4. 4.

    Implementation outcomes and processes: (a) scalability; (b) sustainability; (c) feasibility, acceptability, appropriateness to providers/systems; (d) actual and suggested implementation strategies.

Settings and data sources

CareSource is an MCO serving 1.9 million members in five states. This project focuses on approximately 43,000 of CareSource’s 800,000 Ohio Medicaid subscribers who spend at least one night in jail each year. Outcome data are extracted from the CareSource Jail-Medicaid database, which integrates continuously updated jail booking, release, demographic, and health information (e.g., care team) for CareSource’s Ohio Medicaid subscribers. Providers from 12 BH agencies in Ohio were targeted for weekly reports and training in suicide screening and prevention because they treated CareSource subscribers involved in the CL system.

All information will be treated as confidential material and will be available only to authorized research staff. All paper-based materials, if applicable, will be kept in locked files. In addition, the University research servers maintain secure environments for storing and processing clinical data that adhere to or exceed HIPAA principles. The Internal Privacy Department at CareSource will review and approve all data de-identification procedures to ensure compliance with the Safe Harbor Method as well as all internal agency protocols prior to the release of any deidentified data for research purposes. Any results reported with the project will be reported in aggregate.

Participants and randomization

All adult (age 18 + years) CareSource Medicaid subscribers in Ohio who are released from jail over the 12-month randomization period (n ~ 43,000) are eligible for study inclusion, randomized in a 1:1 ratio to the CC intervention or UC control. A subset of the 43,000 subscribers included in the CC intervention who have been seen in one of twelve contracted BH agencies in Ohio in the 6 months prior to their jail detention are included in the RRT part of the study (~ 6,000). From each of the 12 participating BH agencies, 10 providers (for a total of 120 providers overall) will be recruited and consented to participate in the RRT Intervention. These providers potentially treat individuals that are part of the ~ 6,000 of the 43,000 subscribers passing through jail who have been seen at one of these 12 BH (i.e., mental health and/or substance use) agencies in the 6 months prior to their jail detention.

The 6-month timeframe was chosen because CareSource data indicates that these subscribers are still likely to have open cases at the BH agency (~ 85%); therefore, CareSource felt comfortable asking providers to call to try to re-engage them in care. The 12 BH agencies were randomly assigned to four cohorts of 3 agencies each, to begin the RRT intervention at Months 4, 6, 8, and 10 of the CC intervention (Table 1). We chose this stepped wedge design for the RRT intervention (vs. client- or clinic-level randomization) due to ethical concerns regarding giving providers instructions to reach out to some of their clients and not others. The stepped wedge design allows all 12 agencies to receive the RRT intervention over time, while still providing statistical controls. Individuals with records in two different agencies will be analyzed with the agency that receives RRT first, with sensitivity analyses to examine any effects on results. All study procedures were performed in accordance with the Declaration of Helsinki. Policies and procedures were reviewed and approved by the Butler Hospital Institutional Review Board. Informed consent is being obtained for the focus groups and RRT intervention components. For the CC intervention, due to the retrospective nature of the data and the large population, written or verbal consent is not feasible, so a waiver of consent was granted by the Butler IRB.

Table 1 Study stepped wedge design

Suicide prevention interventions

Caring contacts

CC is an evidence-based approach that has effectively and cost-effectively reduced suicide deaths, suicide attempts, and suicidal ideation in emergency department (ED), inpatient, and veteran populations [33,34,35,36,37,38,39,40,41]. CC is a light-touch, highly scalable intervention, which fits well with the goals and available resources of an MCO. CC involves sending individuals brief, non-demanding messages of care [33, 34, 37, 38]. Messages: (1) include a simple expression of care or concern; (2) are non-demanding; and (3) include a reminder of available resources/ways to connect with care (see Table 2 for examples). Although CC is often used following a care episode [37, 42, 43], it can also be used to engage individuals in care for the first time [44, 45]. CC can help engage individuals who are difficult to engage in care [33]. This study is the first RCT to use CC in a CL involved population.

Table 2 Caring contact message content*

To tailor messages toward individuals who have spent time in jail, we collaborated with stakeholders to design the CC messages, including individuals who spent time in jail in the past year, BH providers, and those within the community who engage with individuals who have spent time in jail. Thirty-two people in three separate virtual focus groups participated in the design and editing of the CC letters. Feedback included valuable insights into how specific wordings and phrases affected individual emotional and motivational responses to the developed messages. For instance, avoiding terms like “reminder” (i.e., too much like a reprimand) and “best wishes” (i.e., most people were not in a good place and this seemed insincere) were consistent comments from participants. Letters also were strongly preferred over postcards by participants. This type of feedback was critical for developing the caring, yet non-demanding messages in our CC letters.

CareSource chose to send CC by mail (vs. email or text) because it was the most feasible within their existing infrastructure. Jail detainees spend a short period of time in jail (typically days), so CareSource is more likely to have viable physical addresses for them. Previous studies on VA patients have shown that they prefer to receive CC messages by mail (vs. text or email) [38], and find a monthly mailing schedule acceptable [33, 34, 37, 38].

CareSource mails the finalized, approved CC letters to subscribers randomized to CC (~ 21,500) monthly for 6 months after automatic notification of jail release. The letters contain similar messages, but the wording for each is slightly different. Letters do not mention anything about the individual having spent time in jail or having received any specific services (e.g., mental health, substance use). They simply say that CareSource is here and available to help if the participant decides to reach out. CC messages list the CareSource24® line phone number (which is answered by CareSource nurses, who are trained to assess and connect callers to the appropriate services). This team works with individuals to help them navigate the healthcare system to get the most appropriate type and level of care when they need it, facilitate referrals to other internal CareSource resources to address urgent healthcare needs and/or social determinants of health, and facilitate connection to crisis lines/service. Individuals who are re-arrested during the 6-month follow-up period will continue to receive letters, but are not re-randomized.

Reports, re-engagement and training

Reports and re-engagement

During the months BH agencies are assigned to the RRT intervention condition, they receive weekly reports from CareSource with notifications of jail booking/release dates of any client seen by the agency in the 6 months prior to arrest. These reports were added to an existing, automated CareSource system that sends jail booking/release data to their contracted BH agencies. Reports include instructions for how to reengage the client and a reminder that recent time in jail is a marker for suicide risk and that the C-SSRS can be used for suicide risk assessment. Agency staff are expected to reach out to clients in a non-demanding way (“How can we help?”) and check in about suicide risk. Providers are not required to use any particular screening or intervention tool; however, they are being trained in evidence-based suicide risk screening (C-SSRS [30]) and intervention (SPI [31]) practices. Subscribers meeting additional risk criteria based on the MHRN suicide risk prediction model [29] (calculated using CareSource claims data) are flagged for additional outreach and action, including use of the C-SSRS and suggesting SPI as an option for clients with very high risk.


Agencies assigned to the RRT intervention receive a one-day, virtual training in: (1) elevated suicide risk among individuals leaving jail detention; (2) how to use the notifications of jail release to reach out, check in with clients, and ask them if they would like to resume behavioral health care without pressuring them; and (3) evidence-based suicide risk assessment (C-SSRS [30]) and intervention (SPI [31]). SPI is the only suicide prevention intervention that has been tested in an RCT for individuals leaving jail [46]. Training in assessment and intervention include both instruction and live practice (role plays).

Intervention fidelity

For the CC intervention, CC letters sent, the dates they were sent, and any letters returned are tracked using CareSource’s automated mailing system. Number of calls to the CareSource24® line are tracked for both CC and UC subscribers. For the RRT intervention, we can electronically track whether reports go out and whether they are timely/accurate. We are able to assess whether subscribers re-engaged in care using claims data. In addition, although there is not a scalable way to track outreach attempts for 6,000 subscribers in the RRT intervention, we meet monthly with CareSource and intervention agencies, hear their experiences, offer feedback, and help problem-solve challenges. Our team will keep structured implementation process notes from these meetings for further analysis of outreach attempts.

Statistical power

Power analysis used a simulation study with variability in release dates, numbers eligible, and introducing a small intra-cluster correlation (ICC = 0.01). We assumed 43,000 eligible subscribers for the CC intervention with 50% randomly assigned to CC. Based on previous data, we assumed 6,000 eligible subscribers for the RRT intervention, and that 10% of the 6,000 (n ~ 600) CareSource subscribers would receive an additional risk flag based on the MHRN suicide risk algorithm. We assumed an overall six-month risk of suicide attempt of 0.05 (0.034 for most, 0.204 for those with the additional MHRN risk flag, odds ratio [OR] = 5). We ran the simulation model 2001 times. Significance was assessed using a two-tailed type-I error level of 5%. For the CC intervention, we will have 85.6% power to detect an effect equivalent to an OR of 0.89. CC has been found to have much stronger effects in previous studies [41] suggesting that we are adequately powered for CC. For the RRT intervention, if we assume an intervention effect of 0.87 OR, we will have 88.3% power to detect this effect. For subscribers receiving an additional risk flag (who will be suggested to receive SPI), we will have 96.4% power to detect an OR of 0.63 [10]. Having superadequate power for this test offsets concerns about providers potentially not providing SPI as instructed in this large, real-world study.


Twelve months of historical (i.e., prior to arrest) and six months of prospective (i.e., after release) jail and Medicaid claims data will be extracted from CareSource’s Jail-Medicaid database for each participating subscriber for all measures, including demographics, diagnoses, suicide attempts, service use, number of arrests, dates of jail bookings/releases, and days incarcerated.

Primary outcome

The primary outcome is a decrease in the total number of MT suicide attempts during the 6 months following jail release. In addition to a risk prediction model (described above), the MHRN also has a well-validated algorithm for extracting MT suicide attempt information from claims data [29, 46,47,48,49,50,51,52,53,54,55,56,57,58], which will be used to assess the primary outcome. This method does not capture all suicide attempts, but it was designed and validated to make system- and state-wide projects such as this feasible.

Secondary outcomes

Our secondary outcomes include a decrease in the number of all-cause injury and poisoning events (capturing overdoses, etc. of unclear intent) extracted from claims data using ICD-10 codes [29, 46,47,48,49,50,51,52,53,54,55,56,57,58]; an increase in the number of outpatient BH (i.e., mental health or substance use) visits extracted from claims data; fewer mental health inpatient hospitalizations and ED visits extracted from claims data using MHRN algorithms [29, 46,47,48,49,50,51,52,53,54,55,56,57, 59]; increased use of the CareSource24® line; and a lower number of arrests 6 months after the start of the study when compared to the 12 months prior, extracted from the continuously updating Jail-Medicaid database.

Additional outcomes


We will assess service linkage (y/n and number of outpatient BH visits) and reaching out for help (y/n and number of calls to the CareSource24® line) as mechanisms of effects of interventions on suicide outcomes.


Grant accounting will capture the costs of the CC mailings. Treatment received (split into suicide-related and overall behavioral health care and other medical care) as part of intervention and UC conditions (outpatient, inpatient, emergency department, BH visits) will be tracked using claims data. Costs of UC will be computed as CareSource payments plus co-pay and deductibles. We will add amortized training costs for providers but exclude other research costs that would not be incurred if RRT was standard care. The primary cost-effectiveness (CE) measure, computed from a societal perspective for each intervention arm, will be the program cost net of any change in other medical costs divided by the sum of MT suicide attempts prevented. We will also assess the net cost of the program from a MCO perspective, excluding patient payments from costs.

Implementation outcomes and processes

This project focuses on sustainability and scalability using the IHI Framework for Going to Full Scale [60, 61] as an implementation mode. In addition to assessing cost and cost-effectiveness as implementation outcomes, we are assessing, maximizing, and optimizing intervention: (1) scalability (per the Intervention Scalability Assessment Tool [62]); (2) sustainability (per the Program Sustainability Assessment Tool [63, 64]) (3) feasibility, acceptability, and appropriateness to providers and systems (per the Acceptability of Intervention Measure [65], Intervention Appropriateness Measure [65], and Feasibility of Intervention Measure [65]); and (4) actual and recommended implementation strategies (via process mapping and implementation process/case notes). These activities will be used to develop future scale-up focused implementation approaches to maximize scale-up and spread.

Data analysis

Prior to final analysis and masked to intervention condition/s, we plan to examine distributions of outcome and control variables to determine appropriate functional forms to maximize explained variation.

Missing data

The primary source of missing data is subscribers dropping off (and possibly re-enrolling) in CareSource Medicaid plans. We aim to examine patterns of missing data and attempt to characterize the probability of missingness as a function of observed variables. We use methods of multiple imputation to address missing values, a recommended approach in the context of randomized controlled trials [66], generating 20 imputations. Baseline and interim (e.g., 4 week) values of outcome variables and pre-specified covariates are used in the imputation models, following best practice recommendations [67].

Effectiveness analyses

Our descriptive analysis follows the framework that is widely accepted in the literature [9]. Overall and for each intervention condition, we present the number of people followed, the number of reincarcerations, the days of non-incarcerated exposure tracked, the number and crude rate of fatal or MT suicide acts, and a relative risk computed as that crude rate divided by the crude rate for the UC group.

The main effectiveness analysis is planned as a comprehensive model (i.e., including all 43,000 subscribers) evaluating CC and RRT intervention effects, which provides optimal power and the ability to test interactive effects of interventions. All analyses will covary baseline (i.e., past year) values of dependent variables, and will use days incarcerated in the 6 months after the index jail release as an offset. The main analysis framework will be logistic regression with the cumulative risk of claims-data-identified suicide attempts requiring medical treatment over six months post-release as the outcome. Secondary analyses will include: (1) a negative binomial model and a count outcome for the number of MT suicide attempts and (2) a survival model with time to first MT attempt. Our models will include robust standard errors due to clustering at the agency level. We will adjust for study month, and include indicator variables reflecting: (1) CC intervention condition (i.e., CC or UC); (2) inclusion in the RRT intervention; (3) RRT intervention condition (i.e., RRT or Control). Identical analyses will be conducted for the other outcomes. We will evaluate the interaction of CC and RRT interventions by including an interaction term of the two indicators.


We will test sex, race/ethnicity, past suicide attempt (y/n), past 6-month BH visit (y/n), arrest in 12 months prior to the index arrest (y/n), cumulative days of incarceration during the study period, area Deprivation Index [68, 69], Mental Health Professional Shortage Area (HPSA) score [70, 71], and per capita incarceration (by zip code) as moderators of intervention effects on suicide attempts.

Mechanism analysis

We will examine number of outpatient BH claims in the 6 months following the index jail release and number of calls to CareSource24® line as mediators of the intervention/s effects on MT suicide attempts by adding this variable to our effectiveness model. We will use bootstrap methods to estimate 95% uncertainty intervals around the mediated effect.

Cost-effectiveness & return on investment

Costs (and savings) in future years will be discounted to present value in the year of jail release [72]. We will bootstrap the 95% uncertainty interval around the CE ratios. Analyses from the MCO perspective will compute the change in claims costs by running a generalized linear model (GLM) with gamma, inverse Gaussian, or Poisson variance based on data distribution, the log link function, and robust standard errors [14]. We will use a 2-part GLM unless less than 5% of patients have zero costs. Independent variables will include treatments received, demographics, Elixhauser co-morbidities [15], and days of MCO coverage after jail discharge. We also will track the MCO separation rate post-discharge from jail by group.

Data and safety monitoring

The study investigators are responsible for implementing and maintaining quality assurance and quality control systems for this study. The protocol was reviewed and approved by the relevant IRBs prior to study start. In addition, this study utilizes a central data and safety monitoring board (DSMB), who also reviewed and approved the study protocol. Standard reporting and monitoring “adverse events” as typically defined (i.e., patient death, hospitalization, etc.) in real time is not possible because: (1) claims data lag, and (2) data are aggregated and de-identified. The DSMB will review extant safety data at 6, 9, and 12 months after the study start.


In line with the larger NCHATS goals, the current study aims to leverage data linkage for suicide prevention at points of contact with the CL system. When completed, this will be the largest RCT for any condition in any population that is CL involved of which we are aware. This study will also be the first to evaluate MCO-provided flags for BH re-engagement and suicide prevention services for recently released individuals, a large population that contributes significantly to U.S. suicide rates. Involving MCOs, who have the capacity to work on a large scale, is critical, given the sheer volume of jail detentions per year. CareSource offers the first and best case of generalizable big health-CL data linkage; few systems have this type of data linkage, and they have figured out a way to do data linkage that is generalizable to other systems and does not rely on special relationships with the jails. It will also demonstrate how to use communication to improve public health outcomes. We believe this study will provide the first assessment of the utility of that linkage and provide a replicable model for future studies. Because these algorithms and approaches are generalizable to other systems (i.e., they do not rely on platform-specific compatibilities), we can demonstrate how notification of CL contacts can be leveraged for community suicide prevention and, by extension, other health conditions, to rapidly advance the field. This study will also be the first to evaluate CC as a suicide prevention method for individuals released from jail.

Given jail detention is a marker for suicide risk, regular information to Medicaid MCOs about their subscribers’ jail bookings/releases can serve as a catalyst for identification and preventative action (i.e., suicide prevention, treatment engagement). Our approach addresses the National Association of Counties goal to better serve CL involved individuals with mental health problems in the community [73, 74] and the National Action Alliance for Suicide Prevention’s goal to reach individuals other than standard care seekers [75]. This is the first RCT in which an MCO takes a lead role in addressing and improving suicide outcomes for subscribers recently released from jail. It also illustrates the value of a jail-MCO data linkage as a means of providing interventions. Partnering with MCOs to bridge jail and community care could produce life-saving linkages. Involving end-users intimately in study and intervention design, as we have done in this proposal, helps shorten the research-to-practice pipeline [56].

Availability of data and materials

As part of the P50 NCHATS infrastructure, relevant de-identified data collected as part of this research project will be deposited into the NIMH Data Archive (NDA).



Behavioral health


Caring contacts






Columbia Suicide Severity Rating Scale


Data and safety monitoring board


Generalized linear model


Institutional review board


Managed care organization


Mental Health Research Network




National Center for Health and Justice Integration for Suicide Prevention


Randomized controlled trial


Reports, re-engagement, and training


Safety Planning Intervention


Usual care


  1. Zeng Z. Jail Inmates in 2021 – Statistical Tables, Report NCJ 304888. Washington DC: Bureau of Justice Statistics; 2022.

    Google Scholar 

  2. Office of the Surgeon General. National Action Alliance for Suicide Prevention. Publications and Reports of the Surgeon General. 2012 National Strategy for Suicide Prevention: Goals and Objectives for Action: A Report of the U.S. Surgeon General and of the National Action Alliance for Suicide Prevention. Washington (DC): US Department of Health & Human Services (US); 2012.

    Google Scholar 

  3. World Health Organization. Preventing suicide in jails and prisons. 2007. Available from

    Google Scholar 

  4. Haglund A, Tidemalm D, Jokinen J, Långström N, Liechtenstein P, Fazel S, Runeson B. Suicide after release from prison-a population-based cohort study from Sweden. J Clin Psychiatry. 2014;75(10):1047.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Cunningham R, King PT, Telfer K, Crengle S, Carr J, Stanley J, Gibb S, Robson B. Mortality after release from incarceration in New Zealand by gender: a national record linkage study. SSM Popul Health. 2022;20:101274.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Binswanger IA, Stern MF, Deyo RA, Heagerty PJ, Cheadle A, Elmore JG, Koepsell TD. Release from prison—a high risk of death for former inmates. N Engl J Med. 2007;356(2):157–65.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Pratt D, Piper M, Appleby L, Webb R, Shaw J. Suicide in recently released prisoners: a population-based cohort study. The Lancet. 2006;368(9530):119–23.

    Article  Google Scholar 

  8. Lize SE, Scheyett AM, Morgan CR, Proescholdbell SK, Norwood T, Edwards D. Violent death rates and risk for released prisoners in North Carolina. Violence Vict. 2015;30(6):1019–36.

    Article  PubMed  Google Scholar 

  9. Janca E, Keen C, Willoughby M, Borschmann R, Sutherland G, Kwon S, Kinner SA. Sex differences in suicide, suicidal ideation, and self-harm after release from incarceration: a systematic review and meta-analysis. Soc Psychiatry Psychiatr Epidemiol. 2023;58(3):355–71.

    Article  PubMed  Google Scholar 

  10. Johnson J, Jones R, Miller T, Miller I, Stanley B, Brown G, Arias SA, Cerbo L, Rexroth J, Fitting J, Russell D, Kubiak S, Stein M, Matkovic C, Yen S, Gaudiano B, Weinstock LM. Study protocol: a randomized controlled trial of suicide risk reduction in the year following jail release (the SPIRIT Trial). Contemp Clin Trials. 2020;94:106003.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Centers_for_Disease_Control_and_Prevention. National Violent Death Reporting System (NVDRS) Query page. 2019; Accessed 18 Dec 2019, 2019.

  12. Taxman FS, Belenko S. Implementing evidence-based practices in community corrections and addiction treatment. New York: Springer; 2011.

    Google Scholar 

  13. Johnson JE, Schonbrun YC, Peabody ME, Shefner RT, Fernandes KM, Rosen RK, Zlotnick C. Provider experiences with prison care and aftercare for women with co-occurring mental health and substance use disorders: Treatment, resource and systems integration challenges. J Behav Health Serv Res. 2015;42:417–36.

  14. Scheyett A, Vaughn J, Taylor MF. Screening and access to services for individuals with serious mental illness in jails. Community Ment Health J. 2009;45:439–46.

    Article  PubMed  Google Scholar 

  15. Wolff N. Community reintegration of prisoners with mental illness: a social investment perspective. Int J Law Psychiatry. 2005;28:43–58.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Human_Rights_Watch. Ill-equipped: U.S. prisons and offenders with mental illness. Washington, DC: Human Rights Watch; 2003.

    Google Scholar 

  17. VanderWaal CJ, Taxman FS, Gurka-Ndanyi MA. Reforming drug treatment services to offenders: Cross-system collaboration, integrated policies, and a seamless continuum of care model. J Soc Work Pract Addict. 2008;8(1):127–53.

    Article  Google Scholar 

  18. Daniel A. Care of the mentally ill in prisons: challenges and solutions. J Am Acad Psychiatry Law. 2007;35:406–10.

    PubMed  Google Scholar 

  19. Friedmann P, Taxman FS, Henderson CE. Evidence-based treatment practices for drug-involved adults in the criminal justice system. J Subst Abuse Treat. 2007;32:267–77.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Perdoni M, Taxman FS, Fletcher BW. Treating offenders in the community: an overlooked population and a lost public health and public safety opportunity. Perspectives. 2008;32(2):46–53.

    Google Scholar 

  21. Sayers SK, Domino ME, Cuddeback GS, Barrett NJ, Morrissey JP. Connecting mentally ill detainees in large urban jails with community care. Psychiatr Q. 2017;88:323–33.

  22. Fisher R. The Design of Experiments. London: Macmillan Publishing Co; 1935.

  23. Collins LM. Optimization of Behavioral, Biobehavioral, and Biomedical Interventions: The Multiphase Optimization Strategy (MOST). Cham: Springer International Publishing; 2018.

  24. Binswanger IA B, PJ, Mueller, SR, Stern, MF. Mortality after prison release: opioid overdose and other causes of death, risk factors, and time trends from 1999 to 2009. Ann Intern Med. 2013;159(9):592–600.

  25. Fazel S, Baillargeon J. The health of prisoners. Lancet. 2011;377(9769):956–65.

    Article  PubMed  Google Scholar 

  26. Spaulding AC, Allen SA, Stone A. Mortality after release from prison. N Engl J Med. 2007;356:1785–7.

    Article  CAS  PubMed  Google Scholar 

  27. Zlodre J, Fazel S. All-cause and external mortality in released prisoners: Systematic review and meta-analysis. Am J Public Health. 2012;102:e67–75.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Binswanger IASM, Deyo RA, Heagerty PJ, Cheadle A, Elmore JG, Koepsell TD. Release from prison- a high risk of death of former inmates. N Engl J Med. 2007;356(2):157–65.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Simon GE, Johnson E, Lawrence JM, et al. Predicting suicide attempts and suicide deaths following outpatient visits using electronic health records. Am J Psychiatry. 2018;175(10):951–60.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Posner K, Brent D, Lucas C, et al. Columbia-suicide severity rating scale (C-SSRS). New York, NY: Columbia University Medical Center; 2008. p. 2008.

    Google Scholar 

  31. Stanley B, Brown GK. Safety planning intervention: a brief intervention to mitigate suicide risk. Cogn Behav Pract. 2012;19(2):256–64.

    Article  Google Scholar 

  32. Simon GE, Shortreed SM, Johnson E, Rossom RC, Lynch FL, Ziebell R, Penfold ARB. What health records data are required for accurate prediction of suicidal behavior? J Am Med Inform Assoc. 2019;26(12):1458–65. PMID:31529095;PMCID:PMC6857508.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Carter GL, Clover K, Whyte IM, Dawson AH, D’Este C. Postcards from the EDge: 24-month outcomes of a randomised controlled trial for hospital-treated self-poisoning. Br J Psychiatry. 2007;191(6):548–53.

    Article  PubMed  Google Scholar 

  34. Carter GL, Clover K, Whyte IM, Dawson AH, D’Este C. Postcards from the EDge project: randomised controlled trial of an intervention using postcards to reduce repetition of hospital treated deliberate self poisoning. BMJ. 2005;331(7520):805–805.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Brown GK, Green KL. A review of evidence-based follow-up care for suicide prevention: where do we go from here? Am J Prev Med. 2014;47(3):S209–15.

    Article  PubMed  Google Scholar 

  36. Fleischmann A, Bertolote JM, Wasserman D, et al. Effectiveness of brief intervention and contact for suicide attempters: a randomized controlled trial in five countries. Bull World Health Organ. 2008;86(9):703–9.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Motto JA, Bostrom AG. A randomized controlled trial of postcrisis suicide prevention. Psychiatr Serv. 2001;52(6):828–33.

    Article  CAS  PubMed  Google Scholar 

  38. Motto JA. Suicide prevention for high-risk persons who refuse treatment. Suicide Life Threat Behav. 1976;6(4):223–30.

    Article  CAS  PubMed  Google Scholar 

  39. Denchev P, Pearson JL, Allen MH, et al. Modeling the cost-effectiveness of interventions to reduce suicide risk among hospital emergency department patients. Psychiatr Serv. 2018;69(1):23–31.

    Article  PubMed  Google Scholar 

  40. Luxton DD, June JD, Comtois KA. Can postdischarge follow-up contacts prevent suicide and suicidal behavior? A review of the evidence. Crisis. 2013;34(1):32–41.

    Article  PubMed  Google Scholar 

  41. Milner AJ, Carter G, Pirkis J, Robinson J, Spittal MJ. Letters, green cards, telephone calls and postcards: systematic and meta-analytic review of brief contact interventions for reducing self-harm, suicide attempts and suicide. Br J Psychiatry. 2015;206:184–90.

    Article  PubMed  Google Scholar 

  42. Landes SJ, Kirchner JE, Areno JP, et al. Adapting and implementing Caring Contacts in a Department of Veterans Affairs emergency department: a pilot study protocol. Pilot Feasibility Stud. 2019;5(1):115.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Reger MA, Luxton DD, Tucker RP, et al. Implementation methods for the caring contacts suicide prevention intervention. Prof Psychol Res Pract. 2017;48(5):369–77.

    Article  Google Scholar 

  44. FirstLink. Caring Contacts. 2020; Accessed 17 Oct 2020.

  45. Hub RHI. FirstLink Caring Contacts. 2020; Accessed 17 Oct 2020.

  46. Coleman KJ, Johnson E, Ahmedani BK, et al. Predicting suicide attempts for racial and ethnic groups of patients during routine clinical care. Suicide Life Threat Behav. 2019;49(3):724–34.

    Article  PubMed  Google Scholar 

  47. Simon GE, Yarborough BJ, Rossom RC, et al. Self-reported suicidal ideation as a predictor of suicidal behavior among outpatients with diagnoses of psychotic disorders. Psychiatr Serv. 2019;70(3):176–83.

    Article  PubMed  Google Scholar 

  48. Yarborough BJH, Stumbo SP, Ahmedani B, et al. Suicide behavior following PHQ-9 screening among individuals with substance use disorders. J Addict Med. 2021;15(1):55–60.

    Article  PubMed  Google Scholar 

  49. Simon GE, Coleman KJ, Rossom RC, et al. Risk of suicide attempt and suicide death following completion of the Patient Health Questionnaire depression module in community practice. J Clin Psychiatry. 2016;77(2):221–7.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Simon GE, Rutter CM, Peterson D, et al. Does response on the PHQ-9 Depression Questionnaire predict subsequent suicide attempt or suicide death? Psychiatr Serv. 2013;64(12):1195–202.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Mental Health Research Network (MHRN). MHRN-Central. 2020. Accessed 10–9–20, 2020.

  52. Stewart C, Crawford PM, Simon GE. Changes in coding of suicide attempts or self-harm with transition from ICD-9 to ICD-10. Psychiatr Serv. 2017;68(3):215.

  53. Ahmedani B, Stewart C, Simon GE, Lunch F, Lu CY, Waitzfelder BE, Solberg LI, Owen-Smith AA, Beck A, Copeland LA, Hunkeler EM, Rossom RC, Williams K. Racial/ethnic differences in health care visits made before suicide attempt across the United States. Med Care. 2015;53(5):430–5.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Ahmedani B, Westphal J, Autio K, Elsiss F, et al. Variation in patterns of health care before suicide: a population case-control study. Prev Med. 2019;127:105796.

    Article  PubMed  PubMed Central  Google Scholar 

  55. Yeh H, Westphal J, Hu Y, Peterson EL, et al. Diagnosed mental health conditions and risk of suicide mortality. Psychiatr Serv. 2019;70(9):750–7.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Ahmedani B, Peterson EL, Hu Y, Rossom RC, et al. Major physical health conditions and risk of suicide. Am J Prev Med. 2017;53(3):308–15.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Ahmedani B, Simon GE, Stewart C, Beck A, et al. Health care contacts in the year before suicide death. J Gen Intern Med. 2014;29(6):870–7.

    Article  PubMed  PubMed Central  Google Scholar 

  58. Tessier-Sherman B, Galusha D, Taiwo OA, et al. Further validation that claims data are a useful tool for epidemiologic research on hypertension. BMC Public Health. 2013;13(1):51.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Wible P. Love letters prevent suicides. 2018; Accessed 10–7–2020, 2020.

  60. Barker PM, Reid A, Schall MW. A framework for scaling up health interventions: lessons from large-scale improvement initiatives in Africa. Implement Sci. 2015;11:12.

    Article  Google Scholar 

  61. Parry G. Evolving the IHI Scale-up Framework. 2018; Accessed 10–2–2020, 2020.

  62. Milat A, Lee K, Conte K, Grunseit A, Wolfenden L, Nassau F, Orr N, Sreeram P, Bauman A. Intervention Scalability Assessment Tool: a decision support tool for health policy makers and implementers. Health Res Policy Syst. 2020;18:1.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Luke DA, Calhoun A, Robichaux CB, Elliott MB, Moreland-Russell S. The Program Sustainability Assessment Tool: A New Instrument for Public Health Programs. Prev Chron Dis. 2014;11:E12.

    Google Scholar 

  64. Calhoun A, Mainor A, Moreland-Russell S, Miaer RC, Brossart L, Luke DA. Using the Program Sustainability Assessment Tool to Assess and Plan for Sustainability. Prev Chronic Dis. 2014;11:E11.

    Article  PubMed  PubMed Central  Google Scholar 

  65. Weiner BJ, Lewis CC, Stanick C, Powell BJ, Dorsey CN, Clary AS, Boynton MH, Halko H. Psychometric assessment of three newly developed implementation outcome measures. Implement Sci. 2017;12:108.

    Article  PubMed  PubMed Central  Google Scholar 

  66. Ware J, Harrington D, Hunger DJ, D’Agostino RB. Missing data. N Engl J Med. 2012;367:1353–4.

    Article  CAS  Google Scholar 

  67. Enders CK. Selecting variables for imputation (section 7.6). In: Applied missing data analysis. New York: Guilford Press; 2010.

    Google Scholar 

  68. Kind A, Buckingham JR. Making neighborhood disadvantage metrics accessible - the neighborhood atlas. N Engl J Med. 2018;378(26):2456–8.

    Article  PubMed  PubMed Central  Google Scholar 

  69. University of Wisconsin Department of Medicine. Neighborhood Atlas. 2021; Accessed 28 May 2021, 2021.

  70. Health Resources and Services Administration. Shortage areas. 2021; Accessed 28 May 2021, 2021.

  71. Health Resources and Services Administration. Health Professional Shortage Areas (HPSA) - Mental Health. 2021; Accessed 28 May 2021, 2021.

  72. Gold M, Siegel JE, Russell LB, Weinstein MC, editors. Cost-Effectiveness in Health and Medicine. New York: Oxford University Press; 1996.

    Google Scholar 

  73. National Association of Counties. Data-Driven Justice: Disrupting the cycle of incarceration. 2020;,human%20services%20systems%20around%20data. Accessed 10–5–2020, 2020.

  74. Stepping Up Initiative. Stepping Up: A national initiative to reduce the number of people with mental illnesses in jails. 2019; Accessed 30 May 2019, 2019.

  75. Ahmedani BK, Vannoy S. National pathways for suicide prevention and health services research. Am J Prev Med. 2014;47(3 Suppl 2):S222–8.

    Article  PubMed  PubMed Central  Google Scholar 

Download references


Not applicable.


This work is supported by the National Institute of Mental Health (P50 MH127512, the National Center for Health and Justice Integration for Suicide Prevention [NCHATS]; Principal Investigators: Johnson, Ahmedani, and Weinstock). The content is the responsibility of the authors and does not represent the official views of the National Institute of Mental Health or the National Institutes of Health.

Author information

Authors and Affiliations



SA, KS, JJ, and SZ were involved in the conception, design, data acquisition, and manuscript preparation. RJ and TM assisted with design and description of the statistical and cost analysis framework, in addition to manuscript revisions. LW and GB contributed to the development and design of the training information and manuscript revisions. FT and BA contributed to the study design and manuscript revisions. All authors have approved the submitted version and have agreed to be personally accountable for their contributions and the accuracy and integrity of the work.

Corresponding author

Correspondence to Sarah A. Arias.

Ethics declarations

Ethics approval and consent to participate

All study procedures were performed in accordance with the Declaration of Helsinki. This study protocol has been reviewed and approved by the Butler Hospital Institutional Review Board. For the CC intervention, due to the retrospective nature of the data and the large population, written or verbal informed consent is not feasible, so a waiver of informed consent was granted by the Butler Hospital Institutional Review Board. Informed consent will be obtained for participants in the focus groups and the Reports, Re-engagement, and Training intervention.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Arias, S.A., Sperber, K., Jones, R. et al. Managed care updates of subscriber jail release to prompt community suicide prevention: clinical trial protocol. BMC Health Serv Res 23, 1265 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: