Residential substance use treatment targets the substance use and overall functioning of individuals with moderate to severe substance use disorders (SUDs) (United States. Department of Health and Human Services, 2016). These centres utilise various models such as therapeutic communites; structured drug-free settings focused on fostering client-led support and re-entry to community following exit from treatment (Goethals et al., 2011). Despite being a common model, therapeutic communities can vary in length, treatment modalities, professional supoport, and treatment goals, leading to hetereogeneity across therapeutic community programs and outcomes (Vanderplasschen et al., 2013). Individuals attending these services see improvements in higher rates of abstinence (McKetin et al., 2018), and reduced rates of overdose, substance use frequency and relapse (Andersson et al., 2019; Pasareanu et al., 2016). However, it is also important to consider healthcare service use outside of residential treatment.

Individuals with SUDs are known to utilise healthcare services at a higher rate than the general population, with higher rates of preventable hospital readmissions and recurrent hospital utilisation compared to other health disorders (Ahmedani et al., 2015; Mark et al., 2013; van Walraven et al., 2011; van Walraven et al., 2012; Vu et al., 2015; Walley et al., 2012). A systematic review and meta-analysis of health-care services, including hospitals, primary care and emergency departments (EDs), estimated that individuals who use illicit drugs utilise these services at a rate seven times higher than the general population (Lewer et al., 2020), with a more recent meta-analysis estimating the global prevalence of acute care utilisation among individuals with substance-related disorders to be 36% for ED visits and 41% for hospitalisations (Armoon et al., 2023). Additionally, those with SUDs frequently utilise mental health services (Wang et al., 2007), particularly if they have co-morbid anxiety and mood disorders, methamphetamine use or injection drug use (Duncan et al., 2022; McKetin et al., 2018). Such frequent utilisation of healthcare services is both costly and indicative of poorer outcomes (Davies et al., 2017), yet there is currently a lack of research with reliable outcome data for individuals who have attended residential treatment services for substance use.

Research investigating residential treatment is primarily self-report and largely contains no follow-up data post-discharge from treatment. Additionally, these studies often have biased treatment evaluations due to the attrition of participants with poorer in-treatment outcomes (Cutcliffe et al., 2016; de Andrade et al., 2019; Gray & Argaez, 2019; Reif et al., 2014; Vanderplasschen et al., 2013). Due to challenges encountered when following up individuals who have accessed residential treatment, one method of collecting substance use information post treatment is administrative data linkage. Using administrative data collected across these highly utilised services can provide valuable information on individuals’ substance use following separation from drug treatment.

Data linkage has emerged as a valuable alternative method of data collection that utilizes existing administrative data infrastructure to obtain longitudinal data, particularly for hard-to-reach populations (Brownell & Jutte, 2013; Willey et al., 2016). Despite its potential, few studies have employed data linkage as a statistical method to investigate outcomes of residential treatment for substance use. In a retrospective data-linkage study comparing mortality outcomes of different treatment modalities for individuals with SUDs in Australia, residential treatment was associated with the highest risk of premature death in the first-year post-discharge—possibly related to the severity of dependence (Lloyd et al., 2017). This was similarly found in a study by Tisdale et al. (2021) linking residential treatment data with the Registry of Deaths, wherein individuals attending residential treatment in Australia were almost four times as likely to experience premature death following discharge when compared to the general population.

Maughan and Becker (2019) similarly found that drug-related mortality was highest in the first 4 weeks following SUD treatment, with individuals discharging from residential treatment at substantially higher risk of mortality than other treatment modalities. Higher mortality risks following residential treatment have been linked to the severity of dependence (Lloyd et al., 2017) and observed among individuals primarily using alcohol and opioids (Maughan & Becker, 2019); however, further research investigating the factors influencing these deaths in the period following treatment is needed. Most premature deaths observed in these studies pertained to AOD-related suicide and overdose; collectively highlighting a critical period of increased risk following discharge from residential treatment.

Within other health settings, data linkage has been used to investigate individuals with SUDs to identify and quantify risk factors associated with higher rehospitalisation (Nordeck et al., 2018), SUD treatment completion (Smith, 2020) and overdose rates (Krawczyk et al., 2020). The idea of a critical period of increased risk for vulnerable populations following separation from services is seen in other high-risk populations. Among individuals who experience incarceration in Australia, Keen et al. (2020) found that non-fatal overdoses were highest 14 days following release from prison, highlighting a lack of transitional care from service to community. In a study investigating risks of mortality, the U.S. Department of Veteran Affairs health system conducted a linkage study in which inpatient mental health units were linked with the National Death Index. The study found elevated rates of non-suicidal mortality within the first 30 to 90 days from discharge, specifying patients with dementia and neurodegenerative disease as high-risk intervention targets (Katz et al., 2019). These studies demonstrate the potential of data linkage to identify critical periods of risk for vulnerable populations and develop targeted interventions to reduce the risk of negative outcomes following separation from treatment services. The use of administrative data across healthcare services can provide valuable information on individuals’ substance use following separation from drug treatment and offers intervention opportunities to improve outcomes.

Currently, there is a significant lack of research investigating the outcomes of residential treatment following discharge. Individuals with SUDs have a higher likelihood of premature death and are frequent users of health services such as hospitals, EDs and mental health services. Given the high utilisation of these services and premature death, it is possible to leverage administrative data collected during presentations to these services and following death to investigate substance use outcomes following separation from residential treatment. Using administrative data linkage from AOD treatment (ATODS), mental health facilities (CIMHA), emergency departments (ED), hospital settings (QHAPDC), as well as deaths recorded with the Registry of Deaths, we aim to examine alcohol and other drug (AOD)-related events that result in utilisation of health services or death following separation from residential substance use treatment. Specifically, we aim to (1) describe the frequencies of AOD-related events across these services and to quantify the first AOD-related service contact following discharge; (2) identify the critical period of time when individuals with SUDs are at the highest risk of first presenting to a health service for an AOD-related event following discharge from residential treatment; and (3) determine patient characteristics that predict AOD-related health service utilisation during the two-year period following discharge from to residential treatment.

Materials and Methods

Participants

Base Cohort–Residential Substance Use Treatment Services

Participants include 1056 individuals admitted to one of three residential substances use treatment centres for moderate to severe SUDs from January 1, 2014 to December 31, 2016 in Queensland, Australia. The organisation operating the residential services is a non-for-profit government funded AOD service provider in Queensland and New South Wales, Australia, with both inpatient and outpatient services. Inpatient services were government subsidised; accepting welfare-pension based stays from self-, professional- and court-mandated referrals. A therapeutic community model of care was used at these centres during the treatment period focusing on community-led roles (e.g. scheduling, cleaning, cooking) and included client- and professional-led aspects such as group therapy, workshops and education. These services required detoxification at admission and applied no smoking and/or drug policies. At the time of data collection, three residential services were operated including a site for young adults (18–25), adults (25+) and an Indigenous Australian exclusive site. Data for participants was inclusive of 2 years from final discharge at one of these residential treatment centres. Episode data was deidentified and extracted by a reporting analyst from the service provider.

Measures

Alcohol or Other Drug-Related Event

The outcome is an ‘alcohol and other drug-related event’ (AOD-related event) within 2 years of discharge from residential treatment, created by linking several datasets (QHAPDC, CIMHA, ATODS, EDIS, Registry of Births, Deaths and Marriages; see Table 1) to indicate a service presentation or death that was partially or fully due to substance use. International Classification of Disease 10th Revision (ICD-10) codes were primarily used to identify presentations attributable to substance use (wholly or partially; see Appendix 1 Table 4 and Appendix 2 Table 5 for lists).

Table 1 Description of linked databases. See Appendix 1 Table 4 for full list of AOD-related ICD-10 codes and Appendix 2 Table 5 for full list of ICD-10 codes for AOD-related deaths
Full size table

Time from Discharge

The number of days from discharge from residential treatment was used to measure the length of time until an event occurs, censored at 2 years. Individuals were censored if premature death occurred.

Predictor Variables

The residential treatment database provided data on the cohort’s demographic characteristics, treatment history, substance use, mental health and drug-abstaining self-efficacy.

Demographics

Demographic characteristics were age (under 25 years/over 25 years), sex (female/male), legal status (justice involved at admission yes/no), Indigenous Australian status (yes/no; included Aboriginal and/or Torres Strait Islander origin).

Treatment

Treatment data were the number of previous admissions (no/1/2+ previous admissions) and treatment completion (treatment completed/early discharge). A successful treatment completion was recorded by staff if a participant spent a minimum of 4 weeks within a residential substance use facility and demonstrated progress towards treatment goals with a planned exit or transition out of treatment.

Substance Use

The primary substance of concern identified by clients at admission from a list of 22 substances was recoded into four categories, including alcohol, cannabis and methamphetamine or ‘other’ drugs which included substances with low frequency (e.g. heroin, cocaine, opioids). Lifetime injecting drug use status was a binary variable (no/yes). Polysubstance use was recoded into a binary variable for individuals with three or more substances of concern (no/yes).

Mental Health

The 21-item Depression, Anxiety and Stress Scale (DASS-21) (Lovibond & Lovibond, 1995) assessed mental health in the week prior to admission. The DASS-21 measures depression, anxiety and stress through a four-point Likert scale, where scores range from 0 ‘did not apply to me at all’ to 3 ‘applied to me very much or most of the time’. We used a total score across the DASS-21 to create a composite measure of general mental health and transformed scores into z-scores to standardize. The DASS has been established in substance-using populations (Beaufort et al., 2017) and has demonstrated excellent reliability (Depression: α = 0.81, Anxiety: α = 0.89, Stress: α = 0.78), concurrent validity, convergent validity, internal validity and discriminative validity (Coker et al., 2018).

Drug Refusal Self-Efficacy

The Drug Taking Confidence Questionnaire (DTCQ-8) provides an assessment of drug refusal self-efficacy through eight items that represent high-risk scenarios for substance use (Sklar & Turner, 1999). Participants indicate their level of confidence in resisting the urge to drink excessively or use drugs on a 6-point scale, which ranges from 0% (not at all confident) to 100% (very confident). Scores less than 80% are categorized as low drug-abstaining self-efficacy, while scores of 80% or higher are categorized as high drug-abstaining self-efficacy. The reliability (α = 0.889) and validity of the DTCQ-8 has been established in substance-using populations (Vasconcelos et al., 2016).

Ethics Approval

This study was approved by the University of Queensland Human Research Ethics Committee (approval number: 2018001063) which includes written approval from the AOD-service and Queensland Health Statistical Services Branch.

Statistical Analysis

Missing data was highest for the DTCQ-8 (6.0%) and DASS-21 (4.6%) at admission. Multivariate imputation by chained equations through SPSS statistical software was used to impute missing data (Jakobsen et al., 2017), and the multiple imputed and original data were similar overall (see Appendix 3: Table 6).

A descriptive analysis of the cohort was conducted. Chi-square and t tests were used to compare the proportion with an AOD-related event by participant characteristics. Cox-regression survival analysis was used to investigate the risk of an AOD-related event and to estimate the time until an AOD-related event occurred. Analyses controlled for all covariates. Follow-up was censored at 2 years (731 days) from index discharge from residential treatment to a maximum date of 31 December 2018. We used SPSS statistical software to conduct all analyses, and RStudio for plots. Unadjusted and adjusted hazard ratios were computed with 95% confidence intervals.

Results

Participants Characteristics

Participants were primarily male (n = 695, 65.8%) with a mean age of 32 years (M = 32.06, SD = 9.55), 26.6% (n = 281) identified as Indigenous Australian. Most had alcohol (n = 403, 38.1%) or methamphetamine (n = 407, 38.5%) as their primary substance of concern (Table 2).

Table 2 Descriptive statistics for individuals who had a substance use event within 2 years after leaving residential treatment imputed over five iterations
Full size table

Presentation for an AOD-Related Event Following Residential Treatment Discharge

A total of 600 (56.8%) participants presented for an AOD-related event within 2 years of discharge from residential treatment. At 30 days post-discharge, 136 (12.9%) participants had presented, increasing to 325 (30.8%) by 6 months, 444 (42.0%) by 1 year and 600 (56.8%) by 2 years. The most hazardous period for AOD-related events was the first month, as indicated by the steepest part of the survival curve (Fig. 1). Within 2 years of discharge, 42% of individuals presented to an ED, 42% to a hospital, 32% to a drug service provider, 20% to a mental health service and 2% experienced an AOD-related death.

Fig. 1
figure 1

Survival function of alcohol or other drug-related events and time since discharge (months)

Full size image

Socio-Demographic and Individual Characteristic Correlates

Those who presented for an AOD-related event were more likely to be aged over 25 years old (n = 478, 58.9%), did not complete treatment (n = 541, 59.8%) and had low drug-abstaining self-efficacy (n = 663, 62.8%).

For those that presented to a health service in the 2 year follow-up, the mean number of days to an AOD-related event was 447.46 days (SD = 300.60, 95CI: 429.33–465.59). Time to AOD-related event is shown in Fig. 1. The cox regression model demonstrated that after adjusting for all other factors, completing treatment at a residential service (aHR = 0.49 [0.37–0.66], p < .001) and high drug-abstaining self-efficacy at admission, as measured by the DTCQ-8 (aHR = 0.60 [0.44–0.82, p = .001) were significantly associated with a reduced likelihood of an AOD-related service presentation following discharge (Table 3 and Figs. 2, 3, 4, and 5).

Table 3 Unadjusted and adjusted hazard of accessing a health service for an alcohol or other drug related event following residential treatment
Full size table
Fig. 2
figure 2

Survival function of alcohol or other drug-related events and time since discharge (months) stratified by Indigenous Australian status

Full size image
Fig. 3
figure 3

Survival function of alcohol or other drug-related events and time since discharge (months) stratified by primary substance

Full size image
Fig. 4
figure 4

Survival function of alcohol or other drug-related events and time since discharge (months) stratified by treatment completion

Full size image
Fig. 5
figure 5

Survival function of alcohol or other drug-related events and time since discharge (months) stratified by DTCQ Score

Full size image

Individuals with alcohol as their primary substance of concern (aHR = 1.58 [1.30–1.92], p < .001), received a Disability Support Pension (aHR = 1.48 [1.06–2.06], p = 0.022), and had two or more previous admissions to residential treatment (aHR = 1.31 [1.04–1.64], p = .022) were at a greater risk of presenting for AOD-related events. Individuals who identified as Indigenous Australian were at a significantly greater risk of presentation for an AOD-related event (aHR = 1.34 [1.10–1.63], p < .001). Scoring higher on the DASS-21 (aHR = 0.62 [0.45–0.84], p = 0.010) was significantly associated with a reduced likelihood of presentation.

Discussion

To understand outcomes after leaving residential treatment for SUDs, we linked data from hospitals, EDs, mental health services and AOD services to examine subsequent AOD-related events. A high proportion of individuals presented for an AOD-related event (56.8%) within the 2-year period following discharge. Residential treatment completion and high drug-abstaining self-efficacy were protective factors, while two or more previous admissions and Indigenous Australian status were risk factors for such events. Identifying the risk factors of accessing health services for AOD-related events has significant implications for the development and delivery of AOD treatment programs and policies.

Previous research has identified the first six months following discharge as a period of increased risk of death, readmission, relapse, injecting behaviour and hospitalisations (Bockmann et al., 2019; Nordeck et al., 2018; Yedlapati & Stewart, 2018). Our study suggests the first month following discharge represents an acute period of increased risk for problematic substance use as measured by greater risks of readmission, relapse, hospitalisation and premature death. This critical period of risk has been observed in other high-risk populations such as individuals separating from prisons, mental health services and hospitals (Katz et al., 2019; Keen et al., 2020; Nordeck et al., 2018). This critical period represents an opportunity to provide targeted support through high-risk transitional periods following discharge from residential treatment. Intervening during these key periods may reduce the poorer substance use outcomes that occur during this first month. Inter-service communication following the pathways of clients as they access subsequent services may support the health journey of other individuals with similar presenting problems.

Individuals that completed treatment at a residential substance use treatment service were half as likely to access a health service for an AOD-related event than individuals who discharged early from treatment. While previous research has evidenced completing residential treatment to improve substance use, treatment and health outcomes following treatment (Drake et al., 2012; McKetin et al., 2018; Pasareanu et al., 2016), these studies are often limited by poor follow-up (Cutcliffe et al., 2016; de Andrade et al., 2019; Gray & Argaez, 2019; Reif et al., 2014; Vanderplasschen et al., 2013). As individuals with unmet substance use treatment require greater hospital and ED utilisation than individuals with adequate treatment (Rockett et al., 2005), completing treatment may act as a protective factor against adverse AOD-related events. While there is support for the effectiveness of long-term (> 90 days) residential treatment in the reduction of relapse (Andersson et al., 2019), these findings contribute to the literature that residential treatment can have a significant impact on relapse with just 4 weeks of successful treatment (de Andrade et al., 2019; Mohamed et al., 2022).

We found individuals with high drug-abstaining self-efficacy were less likely to have an AOD-related event. As seen in previous research, drug-abstaining self-efficacy is a strong predictor of 1-year abstinence following residential treatment for substance use disorders (Ilgen et al., 2005), and has been found to predict abstinence and drinking frequency 5 years post treatment (Muller et al., 2019). Our findings add to the literature that drug-abstaining self-efficacy results in lower likelihood of substance use events up to 2 years following discharge and informs future research investigating the application of treatments aiming to improve drug-abstaining self-efficacy.

Indigenous Australians are impacted by AOD-related health problems at a far greater rate than non-Indigenous Australians (Australian Institute of Health and Welfare, 2018; Wynne-Jones et al., 2016). These risks extend to increased health service utilisation, experiencing greater rates of substance use treatment, hospitalisations and premature AOD-related deaths up to 5-times higher than the non-Indigenous Australian population (Al-Yaman, 2017; Nathan et al., 2020). This increased health service utilisation is due to a range of complex factors, including historical and ongoing discrimination, socioeconomic disadvantage, cultural barriers and limited access to appropriate health care services which is beyond what is captured in the current study. It is essential to develop and provide access to culturally appropriate prevention and treatment services that are designed in collaboration with Indigenous communities.

Recurrent admissions to a residential treatment service have been associated with poorer treatment and substance use outcomes following discharge from treatment (Decker et al., 2017). In the current study, individuals who admitted to two or more previous episodes were more likely to access a health service for AOD-related events. By integrating substance use intervention into healthcare episodes that indicate a return to substance use, healthcare providers can help to identify and address substance use behaviours, reducing the risk of progression and improving overall health outcomes for patients.

Attending a residential service with alcohol as the primary drug of concern was associated with a higher risk of presenting with an AOD-related event following discharge. There is a demonstrated relationship between alcohol use and increased presentations to hospitals, EDs, mental health services and death (Armoon et al., 2021; Iranpour & Nakhaee, 2019; Leung et al., 2023; Zarkin et al., 2004). Individuals who primarily use alcohol are at a greater risk of experiencing somatic health concerns, often experience greater premature mortality (Hjemsaeter et al., 2019), and are more likely to be admitted for trauma related injuries than other substance users (cocaine, heroin, cannabis) (Weintraub et al., 2001). Due to the high utilisation of hospitals and EDs attributable to alcohol use, these service contacts present missed opportunities to offer substance use support aimed at reducing the poorer health outcomes associated with alcohol use.

Individuals receiving a government DSP at admission to treatment were at a greater likelihood of AOD-related events than those not receiving a governmental pension. Receiving the DSP in Australia has been associated with poorer income-related circumstances (poverty, housing insecurity, unemployment), social factors (stigma) and health factors such as declining mental health compared to those who do not receive a pension (Kavanagh, 2020; Kavanagh et al., 2013; Kavanagh et al., 2015; Krnjacki et al., 2018; Milner et al., 2020), exacerbating the myriad risks experienced by individuals living with disabilities. These health disparities and poorer outcomes are particularly of concern when compounded with substance use, demonstrated by the high likelihood of those receiving the DSP to present to health services for AOD-related events following residential treatment. Greater continuing and aftercare is needed for those living with disabilities seeking AOD treatment to address and mitigate the heightened disadvantage experienced by this sub-population.

Previous literature has shown that individuals with co-occurring mental health concerns have worse substance use outcomes, higher rates of hospitalisations and a greater likelihood of early discharge from treatment (Gomez-Sanchez-Lafuente et al., 2022; McGovern et al., 2014; Sofer et al., 2018; Sofin et al., 2017), which we did not find. These potentially spurious findings may have occurred due to incomplete or inaccurate data, a limitation of administrative data linkage. Admission DASS-21 scores measured the cohort as having lower mental health concern than previous estimates in Australia of 47–100% (Kingston et al., 2017) with severe and extremely severe scores on depression (34.6%), anxiety (37.5%) and stress (27.5%) being below this estimate.

Improving measurement in low-resource environments such as residential treatments can be challenging due to the burden on staff. However, implementing low-burden measurement strategies into care such as routine outcome monitoring (Carlier & Van Eeden, 2017; Lambert et al., 2018; Neale et al., 2016), is important to assess and improve treatment outcomes while ensuring data is being collected accurately and consistently.

Strengths

This is one of the first studies to use data linkage to follow-up individuals accessing residential substance use treatment. Follow up of individuals following residential treatment is difficult and outcomes are often hard to track due to the hard-to-reach nature of participants; usually the result of significant social, health and economic disadvantages among this population such as unemployment, homelessness, hospitalisation and criminal justice involvement. In the current study, we were able to avoid the typical rates of high attrition by tracking individuals through administrative data of multiple services. Using multi-service administrative data to develop an objective measure of substance use, we were able to examine the outcomes of residential treatment and examine protective and risk factors associated with substance use following residential treatment. Linking multiple health services such as EDs and hospitals improved the quality of data within the current study; by reducing errors, identifying duplications, and improving accuracy between service data. These findings highlight key intervention opportunities for continued care following separation from residential treatment services, with the potential to improve substance use outcomes.

Limitations

As this was one of the first studies to utilise data linkage to investigate individuals who access residential treatment, a relatively small number of participants were included in the 2 years of inclusion. As a result, measures were recoded into categorical variables to increase power and avoid risk of reidentification, leading to a loss in specificity. All participants were included from only one alcohol and drug inpatient service provider across three locations: representing only one model of residential care and limiting the generalisability of findings. Data linkage investigates retrospective cohort data through clinical service records which by the nature of administrative systems require years to process and obtain. The current study collected residential data from ~ 7 to 9 years ago and linked administrative data from ~ 7 to 5 years. Improving current health information systems to deliver this data more rapidly to treatment services, government reporting institutes, and researchers could improve the understanding of health outcomes for vulnerable populations who frequent these services.

Despite these limitations, this is one of the first studies to demonstrate the application of administrative data linkage to this population, with findings that remain pertinent to future research, and current drug treatment and health systems. Future research should expand the inclusion period and include multiple AOD inpatient services to extend the sample size and generalisability of these findings. Capturing information regarding the time prior to residential treatment admission may provide a greater understanding of substance use patterns that lead to seeking residential treatment. Similarly, improving measurement and examination of within-treatment factors (treatment readiness and engagement, adjustment to daily routine, social relationship interruption) and continuing-care services could address how treatment transition processes impact post-treatment outcomes, especially during the first month.

We were not able to access ambulance data in the current study. Ambulatory datasets routinely capture substance use information with greater specificity than is routinely collected in other population-level datasets such as hospitals and EDs (Lubman et al., 2020). Previous research has identified that one in four ambulance service presentations are transferred and recorded in EDs in Australia (AIHW, 2018). However, when specifically estimating AOD-related ambulatory attendances in Australia, over 70% of attendances are transported to an ED (Ferris et al., 2016), indicating that the majority of AOD-related ambulatory episodes were likely to have been captured in the current study. We only quantified AOD-related events that resulted in service utilisation however many individuals may return to AOD use without the need for health service contact. Thus, our study is not a true representation of relapse following residential discharge, only a return to problematic AOD-use resulting in service utilisation.

Research implementing low burden digital data collection methods such as routine outcome monitoring (ROM) could navigate the time delays and data quality limitations observed in this study (Beck et al., 2021; Carlier & Van Eeden, 2017; Lambert et al., 2018). Future research leveraging ROM in tandem with data linkage could enable the delivery of post-treatment outcome evaluation with increased efficiency and quality while simultaneously aiming to improve these outcomes.

Conclusion

By utilising data linkage to develop a measure of substance use following residential treatment, this study provides novel insights into the period directly following treatment exit, a period typically difficult to measure. The first month after leaving residential treatment for substance use disorders is a critical period with the greatest risk of problematic substance use leading to health service utilisation and premature death. This finding has important implications for clinical practice and highlights the potential benefits of incorporating continuing care into treatment plans for individuals with substance use disorders, especially during presentations to health services. By providing ongoing support and monitoring, continuing care can help individuals maintain treatment gains and prevent relapse, thereby reducing the risk of future hospitalizations and ED visits. Future research should expand the use of data linkage in tandem with current outcome measurement methodology to examine and triangulate post-treatment outcomes. Furthermore, examination of the specific components and optimal duration of continuing care that are most effective in reducing the need for acute care among individuals with substance use disorders is needed. Overall, the findings suggest potential intervention targets and highlight continuing care as a crucial component of a comprehensive approach to treating substance use disorders and promoting long-term recovery.