Defining a recovery-oriented cascade of care for opioid use disorder: A community-driven, statewide cross-sectional assessment.

BACKGROUND: In light of the accelerating and rapidly evolving overdose crisis in the United States (US), new strategies are needed to address the epidemic and to efficiently engage and retain individuals in care for opioid use disorder (OUD). Moreover, there is an increasing need for novel approaches to using health data to identify gaps in the cascade of care for persons with OUD. METHODS AND
FINDINGS: Between June 2018 and May 2019, we engaged a diverse stakeholder group (including directors of statewide health and social service agencies) to develop a statewide, patient-centered cascade of care for OUD for Rhode Island, a small state in New England, a region highly impacted by the opioid crisis. Through an iterative process, we modified the cascade of care defined by Williams et al. for use in Rhode Island using key national survey data and statewide health claims datasets to create a cross-sectional summary of 5 stages in the cascade. Approximately 47,000 Rhode Islanders (5.2%) were estimated to be at risk for OUD (stage 0) in 2016. At the same time, 26,000 Rhode Islanders had a medical claim related to an OUD diagnosis, accounting for 55% of the population at risk (stage 1); 27% of the stage 0 population, 12,700 people, showed evidence of initiation of medication for OUD (MOUD, stage 2), and 18%, or 8,300 people, had evidence of retention on MOUD (stage 3). Imputation from a national survey estimated that 4,200 Rhode Islanders were in recovery from OUD as of 2016, representing 9% of the total population at risk. Limitations included use of self-report data to arrive at estimates of the number of individuals at risk for OUD and using a national estimate to identify the number of individuals in recovery due to a lack of available state data sources.
CONCLUSIONS: Our findings indicate that cross-sectional summaries of the cascade of care for OUD can be used as a health policy tool to identify gaps in care, inform data-driven policy decisions, set benchmarks for quality, and improve health outcomes for persons with OUD. There exists a significant opportunity to increase engagement prior to the initiation of OUD treatment (i.e., identification of OUD symptoms via routine screening or acute presentation) and improve retention and remission from OUD symptoms through improved community-supported processes of recovery. To do this more precisely, states should work to systematically collect data to populate their own cascade of care as a health policy tool to enhance system-level interventions and maximize engagement in care.

Introduction

As the drug overdose crisis in the US continues to accelerate [1], new strategies are needed to address the epidemic and to more efficiently engage and retain individuals with substance use disorder in care. Applying a cascade of care framework offers a novel approach to curb this unrelenting crisis of drug-related harms by identifying novel points for intervention at the system level [2]. Cascades of care have been used to track and improve population health outcomes for multiple complex health conditions [3-14], with the most well-known and visible of such being the continuum of care for HIV infection [4,9,10,12-14]. Applying this continuum of care framework has drastically improved health outcomes among people living with HIV infection, encouraged data-driven policy decisions, and brought about revolutionary system-level changes worldwide in how HIV infection is managed across the course of the disease [12-14].

Researchers and policymakers have called for a cascade of care framework to be applied to understand gaps in treatment engagement among those experiencing substance use disorder, and opioid use disorder (OUD) in particular [2,15,16]. Specifically, a new framework for assessing the availability and quality of care for OUD has recently been defined by Williams and colleagues using metrics defined by the National Quality Forum and the Agency for Healthcare Research and Quality [17]. Preliminary evidence suggests that improvements in health delivery processes at a system level are linked to improved health outcomes and lower mortality among patients with OUD [18]. Defining a cascade of care is a critical first step in helping local jurisdictions establish and utilize their data resources to inform policy and promote interventions that protect against the potentially deadly harms of OUD when left untreated [15,19]. Herein, we aim to adapt and apply the cascade of care defined by Williams and colleagues [20] for the use of medication for OUD (MOUD) [21], using key national and statewide datasets for Rhode Island, a small state in New England, a region of the US heavily affected by the opioid crisis [1].

Methods

Study setting

Rhode Island is an ideal setting for local adaptation of a cascade of care framework for OUD, given ongoing cross-agency, statewide policy efforts to curb the overdose crisis, and the availability of statewide health claims data to evaluate OUD care [22]. Since 2014, state agencies and community leaders have engaged in data sharing and multidimensional surveillance of the opioid crisis [23]. Beginning in 2015, the statewide Overdose Prevention and Intervention Task Force has led the creation and implementation of the Overdose Prevention and Intervention Action Plan [22], designed to monitor key health metrics and guide data-driven program delivery to help end the overdose crisis in the state [22].

Stakeholder engagement

Given that a large amount of programmatic and surveillance data were already being collected and shared on the state’s publicly accessible overdose data dashboard, PreventOverdose, RI (https://preventoverdoseri.org/), to assess progress towards the goals of the action plan [23], key state agency partners engaged with a team at Brown University beginning in June 2018 to begin developing a statewide cascade of care for people living with OUD. These evaluation activities did not meet the federal definition of research, and, as such, ethics approval was not required. The stakeholder group involved 28 members, including local experts on opioid use and its consequences, leaders from state agencies governing health and social services, directors of nongovernmental organizations providing health and social services to people living with OUD, and community advocates with lived experiences of OUD and recovery. This group met 7 times between June 2018 and May 2019 and developed the proposed cascade of care for Rhode Island through an iterative process. Prior to discussions designating the specific stages of the cascade of care, the stakeholder group agreed on a set of shared terminology (Table 1) and identified shared values to guide the framework development process. There were no prespecified plans for defining the stages of the cascade of care or the data sources used to estimate the current state of the cascade of care prior to this stakeholder engagement process. All methods discussed below have been described in line with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement (see S1 STROBE Checklist).

Table 1

Glossary of terms guiding the framework development process.

Term	Definition
Cascade of care	A cascade of care [2,14,15,24,25] is a conceptual framework to guide and track patients (people) over time through stages of medical care for a particular disease or condition, allowing for identification of key points or places to intervene and improve health outcomes [2].
Opioid use disorder (OUD)	We use OUD as the overarching medical condition to define the population of people measured in the cascade of care through an OUD diagnosis. OUD is also used to identify the healthcare systems (places) and transition points (processes) relevant for engaging people at each stage of the cascade [26].
Criteria for OUD	We use Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria for diagnosing OUD, which is defined by loss of control of opioid use, risky opioid use, impaired social functioning, tolerance, and withdrawal symptoms from opioids [27,28].
Indicators	Indicators are the “units of service” or “process measures” in the datasets, which are the details that help us define the size of each stage in the cascade of care, as compared to other stages. They also allow us to assess the quality of referrals and retention from one stage to the next [18,29–32].
Screening and assessment for OUD	Evidence-based screening instruments (secondary prevention) that can be used at point of care (such as a clinic) to help identify someone experiencing opioid misuse or OUD. Examples include the Screening, Brief Intervention, and Referral to Treatment (SBIRT) or the National Institute on Drug Abuse (NIDA)–modified Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) [33,34].

Later discussions designating the specific stages of the cascade of care were grounded in the framework proposed by Williams and colleagues in 2017, which was updated in 2019 [15,20]. This cascade includes 4 stages, beginning with stage 1, where individuals are already identified as having OUD or experiencing a nonfatal overdose [15,20]. The first stage, engagement in care, is defined as the proportion of individuals with OUD who receive specialty services in a given year [20]. The second stage, MOUD initiation, is defined as the percentage of individuals in care who receive MOUD at least once [20]. The third stage, retention, is defined as the proportion of individuals who receive MOUD who continue to do so for at least 180 days [20]. The fourth stage, remission, is defined as the proportion of retained individuals who no longer meet diagnostic criteria for OUD [20].

With this framework informing the foundation of the adapted cascade of care, stakeholders were presented with the available measures of healthcare access and quality to represent each stage [16,17] and discussed the feasibility of using these metrics to develop a “snapshot” of the current state of the cascade of care for OUD in Rhode Island in the year 2016. This process led to additional refinements to the cascade and finalization of the operational definitions of each stage while also helping to define the target population (e.g., those experiencing OUD) and the healthcare systems and data-sharing partners for the cascade [20,26]. By using national quality metrics, the cascade could be designed to enhance existing statewide policy and surveillance tools by (1) creating an annual measure of statewide engagement in care using available datasets, (2) setting targets and standards for improving linkages across stages of care, and (3) defining successful endpoints for treatment of OUD through consensus among the stakeholders [17,18,29,30].

Results

Identifying the guiding principles of a cascade of care for OUD

The stakeholder engagement process resulted in a set of guiding principles for the cascade of care. Through this community-driven process, stakeholders determined that a cascade of care for OUD should be

Measurable and achievable: The cascade of care should track statewide progress in connecting people to high-quality OUD care by identifying meaningful stages of engagement in care and defining measurable targets for engagement in care at each of its stages.

Timely and dimensional [: The cascade of care should measure statewide progress year over year using input from multiple datasets and systems of care. This includes measuring the estimated number of patients in each stage at regular intervals and assessing their progress in moving from one stage to the next.

Incremental: The cascade of care should be used to increase impact at the population level. Through this lens, interventions moving larger groups of individuals to a subsequent stage should be prioritized, rather than moving small numbers of individuals immediately to the end stages of the cascade.

Inclusive: The definition of successful progress through the cascade of care should be client-centered and inclusive of available evidence-based treatment modalities, patient requests, and provider opinions.

Voluntary: Individuals counted in the cascade should be those who voluntarily entered treatment.

Equitable: Comprehensive analyses of health delivery processes have the capacity to uncover patterns of previously unseen inequities across health systems, across demographic groups, among special populations, or across stages of care. Inequalities that appear as systemic barriers to successful progress along the cascade should be investigated and responded to as they arise.

Defining the stages of the cascade of care for OUD

In the preliminary framework proposed by Williams and colleagues [20], “remission” is defined as a finite endpoint, similar in spirit to viral load suppression in the context of the continuum of care for HIV infection [9]. However, OUD was conceptualized by our stakeholder group as a chronic brain disease [36] (rather than as a chronic infection) that has the potential for both relapse and recovery, and where persons are subject to movement in and out of systems of care. As such, our stakeholder group sought to capture more than a single clinical endpoint for the final stage, given that clinical records indicating symptom remission, or absence of evidence of recurrence (e.g., an emergency department visit for a drug overdose) did not fully capture the spectrum of engagement in care for and recovery from OUD. Further, while retention on MOUD represents one element of successful treatment, stakeholders felt that measuring engagement in recovery support services [37,38], including those offered in outpatient healthcare settings and those offered by peers in the community, was critical for measuring long-term absence of OUD symptoms.

With this perspective, the long-term goal of the cascade of care developed by the stakeholder group (hereafter referred to as the Rhode Island Cascade of Care for Opioid Use Disorder) was 2-fold. Similar to the framework proposed by Williams and colleagues [20], the organization of the cascade reflects a primary prevention goal of reducing the number of individuals at risk for OUD (stage 0), while also increasing the number who remain active in their recovery through engagement with professional and peer-based support services (stage 4). The stages of the Rhode Island Cascade of Care for Opioid Use Disorder are defined below (Fig 1).

Fig 1

Overview of the Rhode Island model for the cascade of care for opioid use disorder (OUD).

Credit: Maxwell Krieger, Brown University.

Stage 0 includes those who are considered at risk for OUD and represents the population who may benefit from targeted prevention efforts and/or early intervention services. Individuals in this stage may meet the diagnostic criteria for a clinical diagnosis of OUD at some point during their use, but many may not. Furthermore, this heterogeneous group may include treatment-naïve individuals as well as individuals who were at some point in recovery from OUD. Therefore, the long-term goal is to reduce the total population at risk for OUD over time through a combination of primary prevention efforts (e.g., improved opioid prescribing guidelines to prevent misuse of prescription opioids) as well as secondary prevention efforts (e.g., early intervention through routine screening and rapid referral) [20]. This stage was operationalized as the number of people who reported using heroin and/or misusing a prescription opioid (i.e., using a prescription opioid without a prescription or in any way other than as described by a doctor) in the past 12 months, consistent with measures in the National Survey on Drug Use and Health (NSDUH) [39].

Stage 1 represents those diagnosed with OUD. This stage is operationalized as the number of individuals with a medical claim tied to a diagnosis of OUD in a given year, based on existing national quality metrics [17].

Stage 2 represents unique individuals who have initiated MOUD. The national quality metrics suggest including only those individuals who have been on MOUD for more than 7 days at one time (representing stabilization) but have been engaged for less than 180 days [29,30]; those who have been engaged for 7 days or less would remain in stage 1. Although we acknowledge other modalities of treatment for OUD, the stakeholders chose to focus on MOUD given strong evidence supporting its effectiveness [40-42] and the available datasets on prescribing and dispensing of these medications to persons with OUD.

Stage 3 represents those who are retained on MOUD. Based on existing national quality metrics [30,43], persons who are retained have engaged with their treatment plan for 180 days or longer at one time, without a gap of more than 7 days. The goal in moving individuals to this stage is to support continuous engagement with treatment services for at least 6 months.

Stage 4 represents recovery from OUD. In the context of this cascade of care, recovery is defined as having achieved sustained remission from or resolution of symptoms of OUD using a MOUD-assisted pathway [37,38,44]. The goals for this stage are prevention of OUD symptom recurrence and support of active recovery through structural interventions (e.g., housing, employment programs) and engagement with recovery community centers and community-based peer support services. Stakeholders noted the challenges of measuring the size of this population in existing datasets, but noted that this stage may be measured by considering individuals who have engaged with MOUD for at least 180 days as synonymous with those who are in recovery in the absence of additional data, or by supporting primary data collection activities where self-reported recovery from OUD is captured in ongoing statewide health assessments.

The arrows at the bottom of Fig 2 represent possible transitions through the cascade in 1 year. For example, individuals who are diagnosed with OUD in a given year may initiate MOUD, be retained in care, and achieve recovery. Persons who meet the definition of recovery may become at risk again, thus transitioning back to stage 0 (if never formally diagnosed) or stage 1 (if ever formally diagnosed). Some people may exit the population entirely without progressing beyond stage 0 (i.e., through cessation of drug use), as shown by the leftmost arrow.

Fig 2

Results for the Rhode Island Cascade of Care.

Stages 0 and 4 represent estimates from national survey data sources. Stage 1 represents statewide claims data from the HealthFacts RI all-payer claims database (APCD). Stages 2 and 3 represent combined estimates from the Rhode Island Prescription Drug Monitoring Program (PDMP) and the Behavioral Health On-Line Database (BHOLD), which include treatment claims for methadone and records for buprenorphine prescriptions. All estimates are approximate and considered preliminary. Credit: Maxwell Krieger, Brown University.

Preliminary estimation of the Rhode Island Cascade of Care for Opioid Use Disorder

A preliminary estimation of the number of individuals included in each stage of the Rhode Island Cascade of Care for Opioid Use Disorder is displayed in Fig 2.

Identifying the number of individuals included in stage 0 leverages data collected as part of NSDUH. This survey has been conducted by the US Department of Health and Human Services since 1971. Each year, about 70,000 people aged 12 years and older are interviewed to provide self-reported information on alcohol, tobacco, and drug use; mental health; and other health-related issues [39]. The Restricted-use Data Analysis System (R-DAS) was launched by the Substance Abuse and Mental Health Services Administration (SAMHSA) as an online analytic system that allows analysts to produce cross-tabulations using restricted-use NSDUH datafiles [39]. R-DAS allows for the creation of state-level estimates of select variables using revised weights and combining multiple years of data collection [44]. Using R-DAS, we calculated an estimate for the combined total number of people who reported heroin use and/or misuse of a prescription opioid in the past 12 months in Rhode Island. This population estimate combines data from 2015 and 2016 and, as such, is meant to be representative of the average annual population across the 2 years. Based on this analysis, we estimate that there were 47,000 (95% CI 33,250–60,700) people in Rhode Island in stage 0 in 2016. This estimate serves as the cross-sectional denominator for the Rhode Island Cascade of Care for Opioid Use Disorder.

Identifying the number of individuals included in stage 1 utilized data from HealthFacts RI, the state’s APCD. The APCD stores information on enrollment, medical claims, pharmacy claims, and healthcare providers from privately insured individuals as well as Medicare and Medicaid recipients [45]. Following the Supreme Court ruling in Gobeille v. Liberty Mutual Insurance Company in 2016, the APCD does not include information on individuals with self-funded employee health plans (i.e., those who are self-insured), representing about 20% of Rhode Islanders. Individuals included in stage 1 included those with active claims in 2016 using ICD-9-CM codes 304.00–304.03 (opioid dependence), 304.70–304.73 (dependence combinations of opioid type drug with any other), and 305.50–305.53 (opioid abuse) and using ICD-10 codes F11.1x (opioid abuse), F11.2x (opioid dependence), and F11.9x (opioid use, unspecified). Based on this data source, we estimated that there were 26,000 people in Rhode Island in stage 1 in 2016, representing an estimated 55% of people at risk for OUD (stage 0).

Identifying the number of individuals in stage 2 utilized data from PDMP and BHOLD, 2 databases maintained by the Rhode Island Department of Health and the Rhode Island Department of Behavioral Healthcare, Developmental Disabilities and Hospitals, respectively [46,47]. Using these data, we calculated the number of unique individuals receiving methadone or buprenorphine for more than 7 days at one time in 2016. Based on these sources, we estimated that there were 12,700 people in Rhode Island in stage 2, representing an estimated 27% of people at risk for OUD (stage 0) and 49% of people diagnosed with OUD (stage 1).

Identifying the number of individuals in stage 3 also utilized data from PDMP and BHOLD [46,47]. Using these datasets, we identified unique individuals who were engaged with MOUD through a certified opioid treatment program or prescribed buprenorphine for at least 180 days without a gap of more than 7 days based on the date of first service. Individuals meeting the criteria for retention at any point in 2016 were included. Based on these data sources, we estimated that there were approximately 8,300 people in Rhode Island who progressed to stage 3 through the use of methadone or buprenorphine, representing an estimated 18% of people at risk for OUD (stage 0) and 65% of people who initiated MOUD (stage 2).

Identifying the number of individuals in stage 4 utilized data from the National Recovery Survey conducted in 2016 by Kelly et al. [37]. The National Recovery Survey was a national probability-based sample of adults in the US who self-identified as having resolved a significant problem with alcohol and other drugs [37]. Individuals who answered “Yes” to the question “Did you used to have a problem with drugs or alcohol, but no longer do?” were identified as being in recovery. Based on the survey response to this screening question, these individuals were estimated to represent 9.1% of the adult population of the US (95% CI 8.6%–9.6%). Among individuals who achieved recovery, 5.3% were estimated to have achieved recovery from an opioid problem (95% CI 3.8%–6.8%) using MOUD [38]. In applying these estimates to the adult population of Rhode Island, we estimated that there were 4,200 people (95% CI 2,900–5,600) in Rhode Island who achieved recovery through use of MOUD, representing 9% of people at risk for OUD and 50% of people retained on MOUD (stage 3).

Discussion

Using a combination of national and statewide databases, we were able to generate a statewide, cross-sectional cascade of care for the treatment of OUD in Rhode Island beginning with those at risk for OUD and ending with those in recovery from OUD. Initial estimates of the number of individuals in each stage indicated that Rhode Island has high initiation and retention rates for engagement with MOUD following a diagnosis with OUD. This is consistent with 2017 data from the National Survey of Substance Abuse Treatment Services, which showed that Rhode Island had a rate of 419 per 100,000 people aged 18 years and older engaged with MOUD, ranking among the top 5 states in the nation [48]. However, the results also indicate that additional efforts, such as enhanced screening, are needed to identify those at risk and engage them in care to achieve the dual purpose of mitigating potential harms of opioid use and increasing opportunities for diagnosis and treatment of OUD symptoms. In addition, promoting and supporting continuous engagement with MOUD modalities may provide additional protection against the potential harms of opioid use among those who initiate MOUD but are not retained [49,50].

In the process of developing and defining the Rhode Island Cascade of Care for Opioid Use Disorder, stakeholders expressed the importance of inclusive, community-led prevention efforts, similar to those described by Williams and colleagues [20]. Stakeholders focused on the goal of measuring those at risk for OUD (stage 0) because of the growing efforts to incorporate primary and secondary prevention interventions for this population within the treatment-oriented portions of the cascade of care. In addition, stakeholders acknowledged that movement through the cascade may not be linear, but will likely be cyclical given that both relapse and recovery occur in the context of OUD [36]. Having the ability to provide a specific target for the size of the population to be reached by system-level prevention efforts may lead to more effective means of engaging this population with both prevention and treatment services. The estimated size of this population (5.2% of people in Rhode Island) is similar to that reported in a recent study in the neighboring state of Massachusetts by Barocas and colleagues (4.6% in 2015) [51].

Initiation and continuous engagement with MOUD are critical steps in the cascade of care. Rhode Island has steadily increased capacity for MOUD enrollment as part of its Overdose Prevention and Intervention Action Plan, but full engagement with treatment programs continues to represent a statewide challenge. Prior work has identified system-level interventions to address this gap in care, including increasing low-barrier access and initiation of MOUD in settings such as the emergency department [52,53]. Other interventions focus on increasing clinical training and education on buprenorphine prescription to increase the number of settings where people can initiate MOUD [52,54]. Based on preliminary estimates, it is estimated that about one-half of individuals with a diagnosis of OUD (stage 1) initiate MOUD (stage 2). This estimate is slightly higher than that reported by Larochelle and colleagues in the neighboring state of Massachusetts, where about one-third of individuals who experienced a nonfatal overdose later initiated methadone, buprenorphine, or naltrexone use in the following 12 months [55].

Stakeholders further identified the critical step of including and therefore measuring recovery-oriented systems of care and engagement with community-based recovery supports. There is additional value added to the model put forth by Williams and colleagues [15,20] by incorporating this stage as it captures a positive endpoint of OUD for many people. However, measuring recovery remains a challenge. The estimate we included in the initial parameterization for the cascade of care comes with some uncertainty. Engagement with recovery support services could be measured more precisely through referrals to community-based recovery supports such as peer recovery coaches, recovery-friendly employment programs, or recovery housing [37,56-58]. Measuring recovery or engagement with recovery support services is important for identifying interventions and capturing the protective factors associated with achieving and maintaining such a status, including the empowerment and resource capital that is inherent in recovery [38,59-63].

Currently, there is no existing population-based data source that allows us to measure the number of people who identify as being in recovery at the state level. The Behavioral Risk Factor Surveillance System, a national survey conducted annually by the Centers for Disease Control and Prevention, could add an optional module to achieve this aim [64]. The State of Oregon is using this approach by adding a module with items from the National Recovery Survey [37,56]. This approach offers a more consistent data source collected on an annual basis using the same validated items from the National Recovery Survey. In addition, states may consider implementing assessment of recovery at the clinical level by integrating annual screening tools into existing systems of care such as the Brief Assessment of Recovery Capital (BARC-10) [63,65,66], This scale offers a snapshot of 10 recovery measures that are indicative of someone’s progression in their recovery from OUD [63,65,66]. Use of such a scale provides an opportunity to more accurately estimate the size of the population in the final stage of the cascade.

There are implications for how interventions may be applied to improve transitions between the stages, particularly if the goal is to maximize reduction in drug overdose deaths. Examples may include enhanced screening for OUD if a substantial drop between stages 0 and 1 is observed, increased initiation of MOUD for people diagnosed with OUD to address gaps between stages 1 and 2, deployment of case management interventions to improve retention in care to improve progress from stage 2 to stage 3, and providing funding for community-based recovery support services (e.g., housing, employment training programs) to move individuals from continuous use of MOUD (stage 3) to long-term recovery (stage 4) [52].

Limitations

The primary goal of the current study was to develop and operationalize a statewide cascade of care for OUD, resulting in a “breadth of scope” hybrid model of unlinked data sources [26] offering a cross-sectional snapshot of the number of individuals in each stage. This limits our ability to fully understand the trajectories of specific individuals across all stages or account for true transitions from one stage to the next, particularly from stage 0 to stage 1 and from stage 3 to stage 4. For stage 0, we also recognize that there are limitations inherent in using self-reported measures of heroin use and prescription opioid misuse, including social desirability bias, where individuals who use heroin or misuse prescription opioids may report not doing so to avoid perceived judgment. Furthermore, recovery (stage 4) is often a self-identified status that is difficult to measure through administrative claims data due to potential overlap with other stages (i.e., those with long-term engagement with MOUD in stage 3) [67]. Future analyses will focus on improved linkages across datasets that account for movement between the outlying stages (stages 0 and 4) and the system of care transition stages (stages 1, 2, and 3), and the time-varying nature of the transition of individual patients from one stage to the next [67]. Given the challenges with unrelated, cross-sectional data sources, we ask that the descriptive data and results be interpreted with caution.

Similarly, the Rhode Island Cascade of Care for Opioid Use Disorder is not representative of detailed stages of OUD treatment nor is it inclusive of all available evidence-based treatment modalities. Our focus was to first establish the stages that stakeholders agreed upon, and for which we had access to the most complete data. For example, in stage 2 we used the number of unique initiations of MOUD statewide; however, this does not include individuals who were on MOUD while incarcerated [68]. Stages 2 and 3 also excluded extended-release injectable naltrexone, for which there were fewer than 100 records in Rhode Island in 2016. In addition, there are limitations with the current definition of recovery in stage 4 (“Did you used to have a problem with opioids and no longer do?”) [38], as this may include individuals who would not meet the diagnostic criterion for OUD or may not have progressed through the cascade to reach this stage.

Conclusion

The Rhode Island Cascade of Care for Opioid Use Disorder provides statewide data-sharing partners and policymakers with a starting point for understanding and assessing engagement in care, a system-level “snapshot” for prioritizing the datasets that would be useful in measuring the annual state of the system [14,17,26,56]. The 2 processes of undertaking a community-driven stakeholder process and prioritizing available datasets to populate this new model helped to operationalize and apply the cascade for the state. This process efficiently prioritized and identified the scope of the population and defined the essential stages of engagement with medication-based treatment for OUD. We succeeded in the original goal of operationalizing a population-level cascade orientated around promoting recovery, and we can now use this snapshot as a health policy tool to drive policy decisions and improve health outcomes across the state.

(DOC)

Click here for additional data file.

25 Jul 2019

Dear Dr. Marshall,

Thank you very much for submitting your manuscript "Operationalizing a Statewide Care Continuum for Opioid Use Disorder: Identifying Gaps to Improve Care" (PMEDICINE-D-19-02173) for consideration at PLOS Medicine for our upcoming special issue on substance mis/use.

Your paper was discussed among the editorial team, evaluated by the guest editors for the special issue, and sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

[LINK]

In light of these reviews, we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to invite you to submit a revised version that fully addresses the reviewers' and editors' comments. You will appreciate that we cannot make a decision about publication until we have seen the revised manuscript and your response, and we expect to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We hope to receive your revised manuscript within two weeks. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plosmedicine/s/submission-guidelines#loc-methods.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

Please let me know if you have any questions. Otherwise, we will look forward to receiving your revised manuscript shortly.

Sincerely,

Richard Turner PhD, for Philippa Berman, MBBS

Senior Editor, PLOS Medicine

rturner@plos.org

-----------------------------------------------------------

Requests from the editors:

In your data statement, are you able to name the "third party state institution"? If so, we would suggest including a contact point or person at this institution for readers who may wish to inquire about obtaining access to the relevant data.

Please restructure your title so that the wording falling after the colon provides a study descriptor, e.g. "a cross-sectional study".

Please combine the "methods" and "findings" subsections of your abstract. The final sentence of the new combined subsection should summarize the study's main limitations.

After your abstract, we will need to ask you to add a new and accessible "author summary" section in non-identical prose. You may find it helpful to consult one or two recent research papers published in PLOS Medicine to get a sense of the preferred style.

Please separate the "Methods" and "Results" sections of your main text.

Early in your methods section, please state whether or not your study had a protocol or prespecified analysis plan, and if so attach the relevant document as a supplementary file. Please highlight all analyses that were not prespecified.

Please add a completed checklist for the most appropriate reporting guideline (which we suspect may be STROBE or RECORD) as a supplementary document, and refer to this early in your methods section. In the checklist, please refer to individual elements by section (e.g. "methods") and paragraph number rather than by line or page numbers, as the latter generally change in the event of publication.

To your methods section, please add a statement about ethics approval, if only to note that approval was judged unnecessary given the data sources.

Are you able to add a table giving summary demographic (and relevant clinical and treatment) details for participants, early in your results section? If so, please also add a sentence on said demographic details to your abstract.

Please adapt all reference call-outs in your main text to the following style: "... treatment interventions [1].". Where multiple citations are made, please ensure that the square brackets contain no spaces.

Please remove all instances of "[Internet]" from your reference list.

Comments from the reviewers:

*** Reviewer #1:

In general, I confine my remarks to statistical aspects of articles. There really weren't statistics here, but there was methodology. Unfortunately, I think it has some serious flaws.

First, it can't be both stages and a continuum. A continuum implies continuity. Stages deny that continuity.

Second, the authors did not show any evidence that people move from stage to stage, much less that they do so sequentially. Certainly people could go from stage 0 to stage 4 without the intermediate stages.

Third, stages, to me, implies exclusivity - but here, everyone is at stage 0 and some people are also at stage 1, 2, 3 or 4.

Fourth, the authors haven't really shown evidence that these stages exist - they certainly *might* exist and they make some sense, but why these particular definitions? Are these better than others? What other definitions were considered? How were these stages arrived at?

Fifth, many other people are at risk and risk isn't a yes/no variable. Different people are at different degrees of risk.

p. 7 It's not clear that anything in here is achievable. I certainly hope it is, but no evidence is presented.

p 7 You do not want people moving from stage 0 to stage 1.

p. 7 there is a typo of stage 5.

p. 9 Focusing solely on MAT isn't a good idea. Saying that you are doing so because the data are available is a bit like the famous drunk who looked for his keys under the lamppost rather than where he dropped them, because the light was better there.

p. 11 For stage 0, what about people who deny use of opioids? Maybe stage 0 should be "denial".

p. 17 It is not really acceptable that stage 2 and 3 are defined so differently.

*** Reviewer #2:

Review of PMEDICINE-D-19-02173, "Operationalizing a statewide care continuum for opioid use disorder: identifying gaps to improve care."

This manuscript modifies an existing model of a cascade of care for OUD and then populates the cascade cross-sectionally using a range of national and state-wide data sets for the state of Rhode Island. It is difficult for me to evaluate the manuscript's importance within the existing evidence on OUD risk, treatment uptake and retention, and recovery in the U.S., as I am not an expert in this area. I do, however, have substantial experience with cascades of care, for both HIV and NCDs. I will focus my comments in this area.

1. First, unlike the cascade of care of for HIV, to which the paper refers frequently, the stages defined for this cascade do not each represent a closed cohort. This is due largely to the definition of Stage 0 and how it is populated. As far as I can tell, there is no relationship between the data set used for Stage 0 and the data set used for Stage 1. There is therefore no reason that the population achieving Stage 1 (diagnosis) should overlap to any particular extent with the population defined in Stage 0 (risk). These could be entirely different people. It's likely that there is in fact a fair bit of overlap, but it's certainly possible that Stage 1 combines some people identified in Stage 0 and some newly identified at Stage 1. More important, as the manuscript states, many if not most people in Stage 0 will never be eligible for Stage 1. Dropoff between Stages 0 and 1 is thus not meaningful. That is not how a traditional cascade of care works—you normally start with a fixed denominator in the first stage and report proportions achieving each stage after that, with the goal of 100% moving from stage to stage. In this cascade, Stage 0 is problematic, and it tells us little or nothing about retention in care (since most people in it do not start care, and many not in it do start care).

While measuring risk in the population is certainly important, it does not belong in this cascade of care, given the nature of OUD and the data available. I recommend removing Stage 0 and starting with Stage 1. A separate presentation of the proportion of those at risk whom you estimate reach Stage 1 would be fine.

2. There is no explanation of what we should expect from an OUD cascade of care. What proportions of patients are expected to reach each stage, based on what we know about treatment effectiveness and patient behavior? For non-experts, this information would place the results in context and allow interpretation.

3. While the concept of measuring program effectiveness and identifying gaps using a cascade is widely accepted and can be a valuable tool, populating it with unrelated cross-sectional data sets is problematic. Some of the data sets, such as claims data for medication, are probably quite accurate. Others, such as self-reported risk factors and self-reported recovery, are probably not. Moreover, some of the data sets represent national samples and others are state level. The authors mention these limitations, but they make no effort to consider the potential effect. At the very least, sensitivity analysis to consider the impact of higher and lower rates of achievement of each stage is needed. Ideally we'll one day have a prospective cohort that can follow individuals through the cascade; for now, it should be strongly emphasized that these are descriptive data and results and should be interpreted very cautiously.

4. I congratulate the authors on reporting proportions achieving each stage using the original denominator (Stage 0, though I would use Stage 1, as noted above). This is not done in HIV very often and can result in wildly misleading interpretations (e.g., reporting 95% virally suppressed, but only of those who know their status and started and treatment and had a viral load test—which is almost a guarantee of suppression—when in fact only half the infected population knows its status). That said, I think that this manuscript would benefit from including both—the proportion based on the original denominator and the proportion of those reaching the previous stage. The utility of the latter is that it does identify where we are losing people in the cascade. Figure 2 does not emphasize that the proportion lost between Stage 1 and Stage 2 is the most important point of dropoff and should thus likely be the main focus of intervention.

5. The explanation of where in the cascade people go who start treatment but drop off before 7 days is unclear.

6. The description of the role of the community stakeholders is also a little unclear. I get that it was due to the community that a last stage of recovery was defined as it was, but there is little explanation in the methods of who participated or exactly what they contributed.

7. Minor, but still important: the writing in this manuscript is sloppy. Many sentences are redundant, and there must be a hundred excess commas. There is no clear "Results" section. A good copy editor is needed.

8. The title of the article is misleading. The focus is not on identifying gaps, but on defining the cascade and populating it.

9. To end on a high note: the last sentence in the first paragraph headed "Defining each stage of the RI OUD care continuum" on page 8, which lists 3 purposes of the continuum, is really nice. If the paper could be structured around these three goals, it would be stronger.

*** Reviewer #3:

This manuscript reflects an ambitious and important effort to quantify at the state level the engagement of people with opioid use disorder in care. The manuscript is clearly written and the methods are well described. The results have clear implications for improving the quality of services in Rhode Island and can be a model for other states to evaluate their own services for people with opioid use disorder.

It is a strength of this analysis that the authors include a Stage 4 for recovery and define recovery broadly, but there are inconsistencies in the way Stage 4 is presented in the paper that are confusing. In the Methods Section on p10, Stage 4 is defined in several ways, including both self-reported recovery (through any possible supportive pathway) and also sustained retention (i.e. >180 days) in medication treatment. This is confusing because Stage 3 is also defined by >180d in treatment without significant interruption. In the analysis, recovery is only defined by self report from a national survey. It seems that there are opportunities to use administrative data, including those with sustained uninterrupted retention in treatment, to include people on medications in Stage 4, as the authors suggest is possible in the Methods Section. Similarly, in the discussion of Stage 4 on p18, the authors omit consideration of ways to include sustained retention in treatment as a part of Stage 4.

There is no consideration of how statistical uncertainty should be incorporated into the estimates presented. While quantifying uncertainty might not be indicated for estimates from a complete sample of the population (e.g. the complete PDMP database), uncertainty estimates should be included around point estimates from probability samples (specifically, the National Survey on Drug Use and Health and the National Recovery Survey).

A major limitation of the analysis is the exclusion of people retained in buprenorphine treatment from Stage 3, though this limitation is highlighted and discussed in the manuscript. The authors discuss how gaps between prescriptions exclude many people on buprenorphine from the meeting criteria for Stage 3 and they present the mean number of days between prescriptions and average number of refills per year. It would be more instructive to also present the number of people on buprenorphine who met criteria for Stage 3 (no treatment interruptions >7 days or some equivalent formulation based on number of fills) and number that did not. The authors could then justify whether or not to include these people in the total for Stage 3.

A minor point from the discussion: When stating that RI has high rates of initiation and retention in medication treatment for OUD, it would be helpful to reference what your standard is for determining "high" (e.g. rates from other states).

-

Alexander R. Bazazi, MD PhD

***

Any attachments provided with reviews can be seen via the following link:

[LINK]

26 Aug 2019

Submitted filename: Response to Reviewer Comments (PMEDICINE-D-19-02173)-FINAL.docx

Click here for additional data file.

27 Sep 2019

Dear Dr. Marshall,

Thank you very much for re-submitting your manuscript "Defining a Recovery-Oriented Cascade of Care for Opioid Use Disorder: A Community-Driven, Statewide Cross-Sectional Assessment" (PMEDICINE-D-19-02173R1) for consideration at PLOS Medicine for our upcoming special issue on substance mis/use.

I have discussed the paper with editorial colleagues and our academic editor, and it was also seen again by three reviewers. I am pleased to tell you that, provided the remaining editorial and production issues are dealt with, we expect to be able to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

Our publications team (plosmedicine@plos.org) will be in touch shortly about the production requirements for your paper, and the link and deadline for resubmission. DO NOT RESUBMIT BEFORE YOU'VE RECEIVED THE PRODUCTION REQUIREMENTS.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We hope to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

Please let me know if you have any questions. Otherwise, we look forward to receiving your revised manuscript shortly.

Kind regards,

Richard Turner PhD, for Philippa Berman, MBBS

Senior Editor, PLOS Medicine

rturner@plos.org

------------------------------------------------------------

Requests from Editors:

Around line 41, please cite the date that the study was done, and add a few words to summarize the groups of stakeholders involved.

At line 51, please quote at least one further limitation, e.g. use of self-report in arriving at estimates.

At line 55, please add "Our findings indicate that cross-sectional summaries ...", or similar, at the start of the sentence.

Around line 160, please add a few words to state explicitly that the study did not have a prespecified plan.

Around lines 170-75, please convert "percent of" to "proportion of".

Please refer to the attached STROBE checklist in the methods section of your main text.

Please ensure that all references match journal format. In reference 50, for example, the journal name needs to be abbreviated consistently with other citations, and the volume number adding.

In reference 13, please remove the superfluous "g".

Please spell out "Nqf" in reference 43.

We suggest substituting a panel for figure 1, which should suffice to define the individual stages.

As suggested by one referee, please adapt figure 2 to add to or adapt the arrows to indicate additional possible participant pathways.

Please remove the logos from figure 2.

Please adapt your STROBE checklist so that individual items are referred to by section (e.g. "Methods") and paragraph number rather than by page or line numbers, as the latter generally change upon publication.

Comments from Reviewers:

*** Reviewer #1:

I am checking "proceed without recommendation" because I do not know what to do here.

The authors did respond to each of my comments - thank you! - but I am not at all convinced the end result does what it claims to do. However, since the reasons for my concerns are more substantive than statistical, I am going to leave it to other reviewers and the editors to decide what to do from here. I'm not expert enough to render a final judgment on this.

Peter Flom

*** Reviewer #2:

Overall this is a much stronger manuscript than the original version. The additions and revisions made by the authors do a good job of addressing most concerns, and I congratulate them for it. In particular, the description of the stakeholder engagement process is now excellent, and I am glad to see reported the incremental proportions achieving each stage (personally I'd add these to the figure as well, but reporting them in the text is sufficient).

I still have strong reservations about the inclusion of Stage 0, though the authors have greatly improved how this stage is presented. My own preference remains that it be disconnected from the stage and, if desired, described separately, so that the cascade starts at current stage 1. I will defer to the editors' preferences on this, however. If Stage 0 is retained, I'd encourage a revision of Figure 1 to show that people can drop out of stage 0 entirely, and perhaps that they can move directly from stage 1 or stage 2 to stage 4. Cascades are often illustrated with arrows to show non-linear movement or losses to follow up; in this case, such a presentation would be helpful.

Other comments:

Author summary: Statement "Using national survey estimates and statewide administrative claims databases, we found that 47,000 Rhode Islanders were at risk for OUD (Stage 0) in 2016 and about half were diagnosed (Stage 1)" is misleading, as it implies that everyone in Stage 0 should be diagnosed but only half are. Please just state how many are in Stage 1, without the implication that everyone moves from Stage 0 to Stage 1.

Author summary: I'm also uneasy about the statement, "Engagement with a diverse group of stakeholders can result in the development of a cascade of care to assess and measure the success of statewide health systems in delivering interventions that reduce the number of individuals at risk for OUD and increase the numbers of individuals with OUD who are able to achieve long-term recovery." It's a complicated sentence that could easily be understood to mean that developing a cascade of care is what reduces number at risk and increases numbers who recover. I'd rephrase it if possible.

Lines 233-236: I do not understand the statement, "The cascade includes a primary prevention goal of reducing the number of individuals at risk for OUD (Stage 0), while also increasing the number who remain active in their recovery through engagement with professional and peer-based support services (Stage 4)." The cascade is a tool for presenting information. The goals mentioned are goals of the services and service providers, not the cascade.

There are still some writing errors, e.g. in line 760 "criteria" should be criterion, and in line 766 the newly added "to" should be dropped. PLOS Med's copy editor may catch these.

*** Reviewer #3:

The authors have adequately addressed my concerns in their revisions.

***

Any attachments provided with reviews can be seen via the following link:

[LINK]

11 Oct 2019

Submitted filename: Response to Editorial Comments.docx

Click here for additional data file.

14 Oct 2019

Dear Dr. Marshall,

On behalf of my colleagues and the academic editor, Dr. Margarita Alegria, I am delighted to inform you that your manuscript entitled "Defining a Recovery-Oriented Cascade of Care for Opioid Use Disorder: A Community-Driven, Statewide Cross-Sectional Assessment" (PMEDICINE-D-19-02173R2) has been accepted for publication in PLOS Medicine.

PRODUCTION PROCESS

Before publication you will see the copyedited word document (in around 1-2 weeks from now) and a PDF galley proof shortly after that. The copyeditor will be in touch shortly before sending you the copyedited Word document. We will make some revisions at the copyediting stage to conform to our general style, and for clarification. When you receive this version you should check and revise it very carefully, including figures, tables, references, and supporting information, because corrections at the next stage (proofs) will be strictly limited to (1) errors in author names or affiliations, (2) errors of scientific fact that would cause misunderstandings to readers, and (3) printer's (introduced) errors.

If you are likely to be away when either this document or the proof is sent, please ensure we have contact information of a second person, as we will need you to respond quickly at each point.

PRESS

A selection of our articles each week are press released by the journal. You will be contacted nearer the time if we are press releasing your article in order to approve the content and check the contact information for journalists is correct. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact.

PROFILE INFORMATION

Now that your manuscript has been accepted, please log into EM and update your profile. Go to https://www.editorialmanager.com/pmedicine, log in, and click on the "Update My Information" link at the top of the page. Please update your user information to ensure an efficient production and billing process.

Thank you again for submitting the manuscript to PLOS Medicine. We look forward to publishing it.

Best wishes,

Richard Turner

Senior Editor

PLOS Medicine

plosmedicine.org