RESEARCH DOSSIER
Addiction Recovery, Remission, and Recovery Capital
Abstract. Addiction recovery is a longitudinal process involving changes in substance use disorders, health, functioning, social relationships, and opportunities for sustained participation. This fundamental critical narrative monograph examines the boundaries among abstinence, reduced harm, DSM-defined remission, lived recovery, and recovery capital. It reconstructs the history of recovery science; analyzes clinical course, coping and recurrence, personal and social resources, housing and employment, recovery identity, mutual help, continuing care, medications, survival, and quality of life; and critically evaluates cohort, survey, randomized, and psychometric evidence. Primary studies and systematic reviews establish heterogeneous pathways to problem resolution, the importance of sustained clinical treatment where indicated, and substantial variation in definitions and recovery-capital instruments. Causal attribution, measurement invariance, missingness, unequal resources, and long-term mortality remain central research challenges. Digital support and human–artificial interactions are addressed as emerging research contexts requiring independent validation. This is a critical narrative synthesis of published literature, not a registered systematic review, meta-analysis, original experiment, or individualized clinical guidance.
Keywords: addiction recovery; substance use disorder remission; recovery capital; abstinence; harm reduction; relapse and recurrence; sustained remission; longitudinal recovery; continuing care; mutual-help support; peer recovery support; recovery identity; housing stability; employment and purpose; quality of life; survival and overdose prevention; recovery capital measurement; BARC-10; recovery management checkups; health equity; Artificial Era
Recovery as a Multidimensional Research Object
Addiction recovery is the evolving process through which people reduce substance-related harm, resolve or manage disorder-related symptoms, regain practical agency, and establish conditions for sustained health and participation. It unfolds in a person-environment system comprising biology, learned behavior, relationships, material opportunity, professional care, and community institutions. A longitudinal conception requires specifying both the person's changing condition and the supports that make change possible. The research object of this monograph is the relation among diagnostic remission, meaningful recovery outcomes, and recovery capital, including periods of improvement, recurrence, and renewed engagement with care. SAMHSA (n.d.) Betty Ford Institute Consensus Panel (2007) Cloud & Granfield (2008)
Three kinds of questions organize the inquiry. Classification asks whether a previously established substance use disorder currently meets remission criteria. Outcome research asks whether substance-related problems, mortality risk, distress, functioning, and quality of life improve over particular intervals. Explanatory research asks what changes in resources, social identities, treatment access, behavioral skills, and environmental opportunities make improvements more likely to persist. These are distinguishable observable questions. A negative drug screen cannot independently establish residential security, and a subjective recovery identity cannot by itself establish clinical remission. Different indicators may be jointly valuable without being interchangeable. Hasin et al. (2013) Laudet (2007) Hennessy (2017)
The present work is a critical narrative scientific monograph based on original clinical and population studies, theoretical accounts, systematic reviews, measurement-validation studies, and official guidance. It assesses design, sample selection, operational definitions, temporal order, and clinical applicability. The selection is purposeful rather than an exhaustive, preregistered systematic search; no new experiment or quantitative meta-analysis is claimed. In this scope, alcohol and other drug recovery provide the principal evidence. Generalizing instruments or rates to gambling and other behavioral disorders requires independent validation. Philosophical questions of agency and Artificial Era assistance appear only where they generate distinct, testable human recovery questions. Bunaciu et al. (2024) Kelly et al. (2017) SAMHSA (n.d.)
Historical Formation of the Recovery Research Field
Scientific interest in natural and assisted recovery developed through social science accounts of people changing alcohol and drug use outside formal treatment, the study of mutual-help participation, and longitudinal research on the course of dependence. These traditions expanded inquiry from short treatment episodes toward the full life course. Sociological work by Robert Granfield and William Cloud introduced recovery capital to describe the personal and social resources mobilized in overcoming substance problems; their later elaboration included depleted and negative forms of capital. This formulation connected addiction outcomes to unequal access to material security and social relationships. Cloud & Granfield (2008) Hennessy (2017)
An important definitional milestone was the 2007 Betty Ford Institute Consensus Panel, which proposed recovery through sobriety, personal health, and citizenship. Its formulation established a common conversation among researchers, clinicians, and communities while explicitly remaining a consensus proposal rather than a direct empirical discovery. Alexandre Laudet's interviews with people who identified as being in recovery showed why abstinence-focused endpoints captured only part of respondents' lived experiences. In a longitudinal cohort of 289 inner-city residents with resolved crack or heroin dependence, recovery was described in terms of broader personal change and rebuilding. Betty Ford Institute Consensus Panel (2007) Laudet (2007)
Recovery-oriented services subsequently emphasized continuity after discharge, peer support, family and community resources, and health-system responsibility. SAMHSA's four dimensions—health, home, purpose, and community—illustrate how service models can translate broad recovery goals into distinct areas for evaluation. Measurement expanded from clinical symptom counts to assessments of recovery resources, including the Assessment of Recovery Capital and its shorter derivatives. Population-probability surveys moved the field beyond clinical samples by capturing people who resolved earlier problems through assisted and unassisted routes. The resulting history is one of enlarging units of observation and stronger attention to who is actually included in evidence. SAMHSA (n.d.) Groshkova et al. (2013) Kelly et al. (2017)
Conceptual Boundaries: Use, Disorder, Remission, and Recovery
Substance use denotes behavior; hazardous use indicates elevated risk; substance use disorder is a clinical syndrome defined by applicable diagnostic criteria; recovery describes sustained processes and valued outcomes following substance-related difficulties. These terms require separate operationalization. Physiological tolerance and withdrawal may occur during appropriately supervised medical treatment without establishing a substance use disorder. Addiction is often used for severe, persistent, compulsive patterns but does not replace a complete substance-specific diagnostic assessment. A history of disorder is essential to classifying its subsequent remission. Hasin et al. (2013) SAMHSA (n.d.)
Abstinence is an observable pattern of non-use over a specified interval and often a valuable personal or clinical goal. Reduced use, avoidance of overdose, reduced disease burden, and improved functioning are separate outcomes with their own metrics. A research paper reporting fewer use days has not automatically demonstrated improved mortality or social reintegration. Conversely, maintenance pharmacotherapy for opioid use disorder is an evidence-based route through which risk and functioning may improve; treatment participation does not constitute an exclusion from recovery. Researchers should report the substance, period, treatment exposure, previous diagnostic status, and definition of success. Sordo et al. (2017) CDC (2024) Betty Ford Institute Consensus Panel (2007)
The ambiguity of recovery also concerns identity. A person may state that a previous alcohol problem is resolved without adopting the identity of someone in recovery. In the 2017 National Recovery Survey, an estimated 9.1% of U.S. adults reported having resolved a former alcohol or drug problem; only 46% of that subgroup identified as in recovery. This survey measured self-reported problem resolution, not a clinically verified lifetime-remission rate. Its importance lies in demonstrating that the respondent's language and the investigator's category can select different populations. Kelly et al. (2017)
Clinical Remission and Diagnostic Course
For DSM-5 and DSM-5-TR substance use disorder course specifiers, early remission requires at least three but fewer than twelve months without meeting the disorder's criteria other than craving, following a previously established disorder. Sustained remission requires at least twelve months under the same exception. These are time-and-criteria definitions; they are not statements that all substance exposure has permanently ceased or that social functioning has fully recovered. The substance and diagnostic context matter, and the status should be established through adequate assessment rather than an isolated questionnaire result. Hasin et al. (2013) SAMHSA (2023)
Clinical course specifiers help researchers distinguish current symptomatic burden from the history of a disorder. A person meeting the remission threshold can still experience craving or continue evidence-based maintenance treatment. A person who has reduced use considerably but continues to meet certain diagnostic criteria may show meaningful harm reduction without yet meeting remission. Such categories are particularly important in clinical datasets because coded diagnosis, reported symptoms, test results, and patient-defined recovery status frequently refer to different time windows. Measurement validity depends on preserving those distinctions instead of treating a single discharge label as a complete longitudinal outcome. Hasin et al. (2013) CDC (2024)
Remission terminology from DSM literature must not be casually mapped to ICD-11 or national surveillance instruments: each framework has its own classifications and purposes. As research definitions vary, comparative reviews need to document the specific classification, operational threshold, and follow-up period. Researchers should also identify whether periods in protected settings, loss to follow-up, incarceration, or death change who remains observable. A reported increase in remission among retained participants can reflect changes in underlying health, changing risk exposure, or selective retention. These rival explanations require explicit sensitivity analyses. Bunaciu et al. (2024) Kelly et al. (2017)
Longitudinal Trajectories, Lapses, and Recurrence
Recovery trajectories are sequences rather than categorical before-and-after events. Their possible features include declining use frequency, repeated short abstinent intervals, sustained symptom remission, fluctuating psychological distress, improvements in housing, and changes in treatment engagement. A lapse is often used descriptively for a limited return to substance use; recurrence or relapse generally denotes a more consequential return to problematic use, but study-specific definitions vary. Observing a single use event cannot determine whether the diagnostic criteria of substance use disorder have recurred. The measure, observation window, consequences, and subsequent course should be reported together. Larimer et al. (1999) Hasin et al. (2013)
Marlatt's cognitive-behavioral relapse-prevention model organized risk in relation to high-risk situations, coping responses, outcome expectancies, the effects of interpreting a lapse as failure, and broader antecedents such as lifestyle imbalance. Its historical importance is conceptual and clinical: it made recurrence a subject for analyzing antecedents and modifiable processes. Empirical tests of particular interventions require separate evaluation; endorsement of a model does not establish the magnitude of benefit in every substance or population. Larimer et al. (1999)
Retrospective counts of attempts can illuminate heterogeneity but are sensitive to memory and to what a respondent considers a serious attempt. Kelly and colleagues' 2019 U.S. survey of adults who had resolved an alcohol or drug problem reported a highly skewed distribution: a median of two serious recovery attempts and a mean of 5.35. These are descriptive statistics from survivors who reported resolution, not a prescription for how many attempts a particular person will need or a prospective estimate of transition probability. The skew illustrates why medians, distributional shape, and selection criteria are indispensable to studying recovery. Kelly et al. (2019)
Continuing Care and Recovery Management
An acute treatment episode can initiate change, while many determinants of recovery remain active after discharge. Continuing-care models therefore emphasize planned follow-up, rapid response to renewed needs, and practical continuity across medical, psychological, and social services. This perspective is informed by chronic-care models but should be matched to actual heterogeneity: some people resolve problems without formal treatment, while others require repeated, prolonged, or intensive clinical support. The appropriate unit of evaluation includes pathways into care, retention, reentry when indicated, and longitudinal health outcomes. Dennis & Scott (2012) Kelly et al. (2017)
Recovery Management Checkups provide a directly tested approach to proactive engagement. Dennis and Scott's four-year randomized trial followed 446 adult substance users assessed quarterly, comparing regular checkups and linkage to care with a control condition. The intervention group returned to treatment sooner and more often when needs were identified and experienced favorable differences on measured substance-related outcomes. The intervention was tested in a defined population with chronic problems; its results cannot be automatically transported to every community, adolescent sample, or treatment setting. Dennis & Scott (2012)
Continuing care also raises implementation and equity questions. Effective follow-up requires accessible services, acceptable communication, confidentiality, and continuity when housing, insurance, migration, employment, or caregiving conditions change. Repeated contact may be useful only if it reliably enables timely services. Evaluation should separate process outcomes such as appointments and referrals from person-centered endpoints such as safety, functioning, quality of life, and retention. High attendance may reflect clinical access rather than recovery itself; low attendance may reflect structural exclusion instead of low motivation. Dennis & Scott (2012) SAMHSA (n.d.)
The Recovery Capital Construct and Its Levels
Recovery capital is the set of resources a person can access and mobilize in pursuit of sustained change from substance-related problems. It includes capabilities, relationships, safe environments, financial and practical opportunities, and institutional support. Its components operate at interacting levels: personal and health resources; social relationships and networks; community conditions and cultural meanings; and material or institutional access. The construct has explanatory value only if the resources are defined separately from outcomes. Defining capital as whatever produces success and then demonstrating that successful people have more capital would make the claim circular. Cloud & Granfield (2008) Hennessy (2017)
Resources are not always beneficial. A large network may normalize continued hazardous use; economic capacity may increase access to substances as well as to health care; housing may be stable yet unsafe. Cloud and Granfield's elaboration explicitly examined negative recovery capital and how opportunities vary by age, gender, health, mental health, and incarceration. What matters is availability, quality, social meaning, fit with personal goals, and ability to convert assets into protective action at particular times. Recovery capital is therefore better understood as a changing ecology of opportunities and constraints than a single fixed trait. Cloud & Granfield (2008)
Resource distributions connect recovery psychology to structural conditions. Two people with similar symptoms and motivation may have profoundly different opportunities for affordable treatment, transportation, healthy housing, employment, and legal security. An explanatory model that attributes these differences solely to individual perseverance will misclassify environmental disadvantage as a personal deficit. In research, capital indicators should therefore distinguish resources possessed, resources perceived, resources accessible in practice, and resources actually used. Such distinctions support testable mediation and moderation hypotheses across place, time, and population. Hennessy (2017) Bunaciu et al. (2024)
Personal, Physical, and Psychological Resources
Personal recovery resources include health literacy, future-oriented planning, coping repertoires, behavioral flexibility, self-efficacy, and knowledge of how to obtain assistance. Their effects depend on context and on changing substance-related reward and stress processes. Cognitive skills may enable avoidance of high-risk situations or management of competing goals, but the same skills can be harder to enact under sleep deprivation, untreated psychiatric distress, unstable housing, or ongoing threat. Conceptual models should state whether a proposed resource precedes improved outcomes or is itself an early consequence of them. Larimer et al. (1999) Volkow et al. (2016)
Physical health is a central part of recovery rather than an ancillary issue. Chronic conditions, infectious disease, pain, nutrition, sleep, and access to preventive services can shape both capacity and perceived quality of life. For opioid use disorder, the evidence for medications that reduce overdose and mortality risk makes treatment access itself an important resource and outcome context. Research must not interpret medication dependence during clinically indicated treatment as evidence of failed recovery. Official CDC guidance explicitly distinguishes effective medication treatment from detoxification alone, which carries increased risk of return to use and fatal overdose. CDC (2024) Sordo et al. (2017)
Psychological resources require careful measurement. Self-efficacy items may correlate with sustained change because they capture confidence derived from recent success. If measured simultaneously, that association cannot establish causal priority. Repeated within-person measurements, preplanned temporal lags, and experimentally altered access to coping or support can improve inference. A useful research design records symptom status, recent substance-use exposure, psychological functioning, and independent resource indicators before subsequent outcomes are observed, reducing the risk of measuring the same improvement several times under different labels. Groshkova et al. (2013) Bunaciu et al. (2024)
Relationships, Family Support, and Network Composition
Social recovery capital is defined by the availability of relationships that provide dependable information, practical assistance, recognition, and participation in valued roles. Network size alone is a poor substitute for these functions. A network can contribute encouragement and transport to services while simultaneously exposing its members to substances or coercion. The same relationship may have supportive and destabilizing effects across time. Studies should measure network composition, tie strength, reciprocal obligations, substance-related norms, isolation, and changes in affiliation rather than treating social support as a single universal exposure. Cloud & Granfield (2008) Hennessy (2017)
Families participate in recovery as social systems with their own histories, resource limits, and responsibilities. Support can involve reliable accommodation, care coordination, child care, and interpersonal repair; some households also involve violence, financial strain, stigma, or conflicting expectations. Research needs to distinguish the benefits of chosen supportive relationships from assumptions that family reunification is always feasible or safe. The relevant outcomes include the person's agency, physical safety, relationship quality, and the experiences of other household members. SAMHSA (n.d.) Laudet (2007)
Networks may change through participation in treatment, mutual-help groups, employment, education, or community activities. Because people choose affiliations partly according to prior motivation and severity, observational associations between a recovery-oriented network and sustained improvement are vulnerable to selection and reverse causation. Longitudinal network studies, appropriate comparison groups, and interventions that genuinely alter access can help test proposed protective mechanisms. Reports should acknowledge that network-based programs may also select people with more resources and greater capacity to attend. Kelly et al. (2020) Hennessy (2017)
Housing, Employment, Education, and Material Security
Safe housing gives recovery a spatial and temporal foundation: privacy, sleep, predictable routines, secure storage of prescribed medications, continuity of contact, and a location from which to attend services. Housing insecurity can disrupt appointments, expose people to unsafe environments, and compromise family responsibilities. Studies of recovery capital often record housing as a resource, yet a single housed-versus-unhoused variable captures neither safety nor affordability, security of tenure, autonomy, overcrowding, or exposure to violence. Measurement should distinguish these dimensions and assess whether the housing arrangement is sustainable. SAMHSA (n.d.) Bunaciu et al. (2024)
Employment and education can contribute income, structure, competence, social participation, and future opportunity. Their relationship with recovery is reciprocal. Improved health can enable work, while meaningful work can sustain routines and reduce material insecurity. Equally, precarious employment, discrimination, hazardous conditions, and inflexible scheduling may constrain treatment engagement. Longitudinal research must address bidirectional causation and avoid interpreting absence from the labor market as personal noncompliance, particularly for those with disability, caregiving obligations, or criminal-legal barriers. Hennessy (2017) SAMHSA (n.d.)
Material security is shaped by transport, identification documents, benefits, insurance, wages, debt, and access to legal services. A study that identifies low recovery capital should ask whether the measured deficits can be changed through accessible public services rather than treating capital scores as a fixed individual prognosis. A defensible intervention theory specifies which resources will be supplied, how uptake is supported, and which separately measured outcomes should follow. Resource provision is also an ethical matter: success should include whether assistance increases autonomy and reduces preventable harm. Cloud & Granfield (2008) Bunaciu et al. (2024)
Community, Culture, and Recovery Identity
Community resources include welcoming places, peer-led services, recovery organizations, local health care capacity, cultural recognition, transportation and public institutions. The term community may refer to a geographical neighborhood, an affinity network, a family of choice, or a professional and peer network extending across places. These formations offer different kinds of belonging and practical help. A recovery-community center may support relationships and referral, whereas a local medical clinic provides a distinct function. Evaluation should name the institutional service and its measurable contribution. SAMHSA (n.d.) Hennessy (2017)
Recovery identity is the interpretation through which an individual integrates previous substance-related difficulties, present commitments, and expected future roles. It can supply shared language and access to affirming groups. It can also be rejected or reshaped by people who resolve problems without identifying with a recovery movement. The National Recovery Survey illustrates this difference between reported problem resolution and adoption of a recovery label. Researchers should allow self-description to be measured in respondents' own terms and recognize variation across generations, religious or secular settings, cultures, and medical histories. Kelly et al. (2017) Laudet (2007)
Belonging may depend on experiences of stigma, cultural safety, language and accessible services. A community intervention that improves outcomes in one region may be less acceptable where membership criteria, religious assumptions, or institutional trust differ. Research should measure acceptability, representation and exclusion alongside substance outcomes. Recovery identities and community relations are best evaluated as potentially useful but heterogeneous mechanisms whose value depends on personal agency and access to meaningful alternatives. Hennessy (2017) SAMHSA (n.d.)
Psychometric Assessment of Recovery Capital
The Assessment of Recovery Capital developed by Groshkova, Best, and White represented an effort to operationalize resources in addiction recovery rather than rely on a broad conceptual label. Their 2013 psychometric evaluation reported acceptable concurrent validity and moderate one-week test-retest reliability, alongside a factor-analytic structure in their studied samples. These findings support the instrument as a potentially useful adjunct to clinical and social assessment; they do not make its scores a direct measure of future remission or a substitute for an individualized multidomain evaluation. Groshkova et al. (2013)
The Brief Assessment of Recovery Capital, BARC-10, was developed and validated by Vilsaint and colleagues in 2017 to offer a shorter research and practice instrument. Abbreviating measurement reduces response burden while also placing greater weight on which domains the retained items represent. A brief score may facilitate repeated observation, yet any comparison across populations requires investigating reliability, factor structure, cultural meaning, responsiveness, and invariance. A score that predicts one outcome in a treatment cohort need not yield the same interpretation in a correctional, adolescent, rural, or international sample. Vilsaint et al. (2017) Bunaciu et al. (2024)
Bunaciu and colleagues' 2024 systematic review identified 69 eligible studies of quantitative recovery-capital measurement published from 2016 through 2023. Ten questionnaires were identified, collectively covering 41 distinct constructs, and the authors concluded that stronger psychometric evidence across diverse populations and settings remained necessary. This variety demonstrates conceptual richness and the challenge of score comparability. Differences in reference periods, response scaling, treatment settings, and cultural conditions can create apparent between-group variation unrelated to true resource availability. Investigators should report instrument versions, missing data, observed distributions, and justification for combining domains. Bunaciu et al. (2024)
Construct Validity and the Risk of Circular Measurement
Recovery capital creates a specific measurement problem when indicators of resources overlap with the outcomes the concept is intended to explain. Suppose a questionnaire awards points for current abstinence, general well-being, and stable housing, and a study then uses the total score to predict an endpoint containing the same items. The association will partly reflect shared content rather than a demonstrated causal contribution of available resources. An adequate model separates predictors, mediators, contextual moderators, and outcomes both conceptually and temporally. Hennessy (2017) Bunaciu et al. (2024)
Social support perceived during a period of success may be reported more favorably because relationships have already improved. Similar reverse causation can affect employment, self-efficacy, and overall life satisfaction. Experimental provision of an intervention, instrument-specific longitudinal change, and validated external outcomes can help determine whether added resources precede change. Even then, an intervention may influence results through multiple mechanisms. Mediation analysis requires explicit assumptions and reliable measurement of intermediate variables, rather than labeling every post-treatment improvement a recovery-capital effect. Groshkova et al. (2013) Vilsaint et al. (2017)
Researchers should also assess who is excluded from questionnaire development. Instruments constructed among adults in treatment may inadequately capture recovery processes among adolescents, older people, people with disabilities, individuals outside formal services, or populations facing housing and legal insecurity. Translations require cultural adaptation and evidence of measurement invariance, not only a literal rendering of items. A scientific assessment program should combine resource measurement with patient-defined goals, functional indicators, diagnostic status, and evidence about institutional access. Bunaciu et al. (2024)
Population Research and Diverse Routes to Problem Resolution
The U.S. National Recovery Survey provides a crucial complement to treatment-based cohorts. Kelly and colleagues used a probability-based adult sampling strategy to examine whether respondents had formerly experienced a problem with alcohol or drugs that they considered resolved. The weighted prevalence estimate of 9.1% belongs to the study's survey population and wording; it is not a global prevalence estimate, nor a measure of all DSM-confirmed remissions. Among respondents reporting resolution, 53.9% reported having used an assisted pathway involving formal treatment or recovery supports. The existence of unassisted pathways is a central empirical finding, subject to the survey's recall and selection limitations. Kelly et al. (2017)
Assisted and unassisted groups are not exchangeable without further work. People with more severe or recurrent disorders can be more likely to use formal care, and individuals receiving assistance may therefore show more prior attempts, higher burden, or different current outcomes. A naïve comparison that treats these differences as treatment effects risks confounding by indication. Research must document diagnosis, severity, comorbidity, treatment history, and access barriers. The normative implication is a plural service system that preserves evidence-based clinical care for people who need it while making space for self-directed and community-supported change. Kelly et al. (2017) Kelly et al. (2019)
Generalizations must also account for people absent from sampling frames: institutionalized populations, people experiencing severe housing instability, those who died before follow-up, and people whose problems have not resolved. A cross-sectional study of survivors with self-reported resolution cannot directly estimate the prospective probability of recovery among all people newly diagnosed with a disorder. Administrative datasets add other forms of selection through insurance eligibility, care-seeking, and coding practices. Population recovery science should combine surveys, clinical cohorts, death registries, and community-based recruitment. Kelly et al. (2017) Sordo et al. (2017)
Mutual-Help Organizations and Peer-Based Recovery Support
Mutual-help groups and peer services can supply social integration, practical information, shared norms, and recurring opportunities for participation. A research question is whether a particular intervention or linkage approach leads to outcomes beyond those expected from self-selection into supportive networks. Mechanisms may include shifts in contact patterns, abstinence-related self-efficacy, accessible role models, and meaningful contribution to others. These explanations remain distinct from the empirically measured efficacy of a program. Hennessy (2017) Kelly et al. (2020)
The Cochrane review by Kelly, Humphreys, and Ferri (2020) examined Alcoholics Anonymous and Twelve-Step Facilitation in alcohol use disorder. The review found high-certainty evidence that manualized AA/TSF interventions improved continuous abstinence compared with certain established treatments, including cognitive behavioral approaches, while other alcohol-related outcomes were often comparable. Interpretation requires specifying the particular comparison, intervention standardization, and endpoint; it does not imply that every mutual-help format works for every disorder or that attendance is required for recovery. Kelly et al. (2020)
Peer workers and mutual-help participants occupy different roles from clinicians. Appropriate evaluation considers training, boundaries, confidentiality, referral pathways, supervision, and safety. Recovery may be supported by secular organizations, culturally specific groups, professional treatment combined with peer work, or chosen informal relationships. Claims about mutual help should respect a person's preferences and acknowledge access barriers affecting transport, caregiving, disability, gender safety, and religious congruence. An effective support system is judged by patient-important outcomes and opportunities for voluntary participation. SAMHSA (n.d.) Kelly et al. (2020)
Recovery-Oriented Clinical Care and Medication Continuity
Recovery-oriented clinical care connects validated treatment for substance use disorders to continuing support for functioning and safety. Appropriate care differs by substance and clinical presentation. Psychotherapies, medications, harm-reduction services, and coordinated treatment of other mental and physical health conditions can be combined according to evidence and individual circumstances. Recovery capital does not replace these interventions: it describes the resources needed to access them and sustain the benefits they may offer. Treatment selection and medication management require clinicians working with the particular patient. CDC (2024) SAMHSA (2023)
Opioid use disorder illustrates the practical consequence of differentiating abstinence rhetoric from meaningful clinical outcomes. CDC guidance identifies buprenorphine, methadone, and naltrexone as FDA-approved medications for OUD, with medication treatment associated with reduced overdose and overall mortality; it warns against detoxification alone without ongoing medications for OUD. The 2017 Sordo and colleagues synthesis found markedly higher all-cause mortality during periods outside opioid substitution treatment, while noting observational confounding and heterogeneity. Maintaining appropriate medication is compatible with recovery and may be essential to safety. CDC (2024) Sordo et al. (2017)
Clinical continuity can also depend on insurance, prescriber availability, pharmacy access, transportation, trust, and coordination across institutions. Research that records medication use as a deficit in recovery capital would invert its clinical meaning for many patients. Likewise, medication discontinuation must not be presumed a superior endpoint. Studies should report retention, treatment interruption, overdose, serious adverse events, physical health, and quality of life while respecting established diagnostic categories. Particular pharmacotherapy regimens and adverse-effect management are outside this dossier's remit and belong to formal clinical guidelines. CDC (2024) Sordo et al. (2017)
Survival, Overdose Prevention, and Competing Recovery Outcomes
Survival is an indispensable endpoint in addiction recovery research. A cohort reporting high abstinence or remission among respondents still alive and reachable can overlook deaths that occur during follow-up. For opioid use disorder in particular, overdose risk is affected by changes in tolerance, exposure to unpredictable drug supplies, availability of medications, and interruptions in care. A recovery research program must therefore define death and overdose as consequential outcomes and account for them in analysis rather than simply treating them as unobserved substance-use status. Sordo et al. (2017) CDC (2024)
The systematic review and meta-analysis by Sordo and colleagues combined 19 eligible cohorts comparing mortality during and after opioid substitution treatment. Among methadone-treated populations, pooled all-cause mortality rates were 11.3 per thousand person-years in treatment and 36.1 outside treatment, a crude out-to-in rate ratio of 3.20. These figures summarize heterogeneous observational cohorts; they cannot be interpreted as an individually guaranteed treatment effect. Their public-health implication is that maintaining effective treatment and managing high-risk transitions are fundamental elements of recovery-oriented care. Sordo et al. (2017)
Recovery programs should separately record nonfatal overdose, emergency care, infectious complications, functional improvement, treatment access, and patient-reported safety. A narrow focus on zero use may obscure health gains from lower-risk behavior and timely clinical engagement. Equally, reduced harm alone may not answer every patient-defined recovery goal. A transparent evaluation pre-specifies a hierarchy or multivariate set of endpoints while allowing participants' own priorities to inform interpretation. CDC (2024) SAMHSA (n.d.)
Quality of Life, Well-Being, and Functional Recovery
Quality of life captures how people evaluate their circumstances across physical health, emotional experience, social relationships, and daily functioning. It extends beyond simple symptom counts while requiring a clear instrument and reference period. An improvement in one domain may coexist with unresolved problems in another. Research should disaggregate physical function, psychiatric symptoms, meaningful activities, intimate relationships, finances, housing, and life satisfaction. Treating a total score as the sole indicator of recovery risks hiding clinically important combinations of gains and persistent deprivation. Laudet (2007) SAMHSA (n.d.)
Laudet's qualitative and longitudinal research documented that people with resolved severe drug problems frequently interpreted recovery as a transformation of self and daily life. Such testimony identifies relevant domains and language for outcome development but does not, by itself, establish efficacy of a particular intervention. A strong mixed-methods program uses interviews to determine what participants value, then validates longitudinal quantitative measures against these domains. The concepts of citizenship and purpose in major recovery frameworks are similarly meaningful when operationalized through observable participation and self-determined goals rather than externally imposed ideals. Laudet (2007) Betty Ford Institute Consensus Panel (2007)
Functional trajectories should be evaluated against baseline circumstances and available opportunities. Someone with stable finances may experience a change in self-regulation without a measurable employment transition; another may improve safety substantially despite unemployment caused by disability or discriminatory hiring. Measures should recognize caregiving, education, voluntary activities, social connection, and self-directed roles. Research can examine whether improvements in symptom remission precede changes in functioning, whether resources permit these changes, and whether functional changes in turn sustain later health. SAMHSA (n.d.) Hennessy (2017)
Psychiatric Comorbidity, Trauma, and Integrated Recovery
Substance use disorders frequently intersect with depressive symptoms, anxiety, trauma exposure, sleep disturbances, and other health problems. The presence of comorbidity complicates both access to care and interpretation of outcomes. A person may achieve remission of a substance use disorder while continuing to experience significant depression, and improvement in emotional health may affect return to hazardous use. Recovery research should assess co-occurring conditions with appropriate validated methods and specify whether analyses concern symptoms, formal diagnoses, or reported treatment histories. Volkow et al. (2016) Kelly et al. (2019)
Associations between psychiatric diagnoses and multiple recovery attempts do not demonstrate that a particular disorder directly caused each attempt or recurrence. In Kelly and colleagues' national sample, history of psychiatric comorbidity was associated with more reported attempts, yet the design was cross-sectional and based on retrospective recall. Illness severity, access to resources, exposure to trauma, and the likelihood of entering formal services can contribute to the observed pattern. Clinical conclusions therefore require prospective information and careful attention to temporal sequencing. Kelly et al. (2019)
Integrated care examines both substance-related symptoms and other conditions in a coordinated service framework. Recovery-support interventions should be evaluated for mental-health outcomes and risk of adverse effects, including whether treatment coordination improves engagement without introducing excessive burden. Trauma-informed practice recognizes the significance of interpersonal safety and choice, though particular claims about treatment effects must be supported by condition-specific trials. This dossier treats comorbidity as a contextual determinant and points detailed diagnostic and psychotherapeutic protocols to their own clinical literature. SAMHSA (n.d.) CDC (2024)
Adolescence, Aging, and Life-Course Variation
Recovery processes depend partly on developmental opportunity. Adolescents may face changing peer relationships, school environments, family dependence, and distinct capacities for confidential access to services. Adults may balance employment, parenting, and caregiving; older adults may face chronic illness, bereavement, polypharmacy, and limited mobility. A recovery-capital instrument validated in treatment-seeking adults should not be presumed to measure the same constructs in these populations. Research must address developmental stage through instrument adaptation, subgroup-specific interpretation, and ethical recruitment. Bunaciu et al. (2024) Hennessy (2017)
Life-course designs should distinguish age, birth cohort, historical period, and duration since the onset of substance-related difficulties. Age-related associations in cross-sectional studies may reflect survivorship or changes in exposure, rather than maturation alone. Longitudinal cohorts can ask how education, work transitions, family formation, disability, retirement, or changes in local services alter recovery trajectories; they should also record individuals who are lost to follow-up or die. Recovery is therefore a changing relation between person and context across time, not a fixed developmental sequence. Kelly et al. (2017) SAMHSA (n.d.)
Preventing stigma in age-related analysis requires avoiding assumptions about the capacity for change based on age or number of past attempts. A person's priorities for employment, intimate relationships, autonomy, and community participation may change markedly over the life course. Valid study outcomes should be meaningful to the age group and responsive to differential access to support. Questions of developmental recovery can be tested by comparing prospective trajectories and evaluating interventions with age-appropriate measures. Bunaciu et al. (2024) Kelly et al. (2019)
Cultural Variation, Structural Inequality, and Stigma
Recovery capital is distributed through social and political institutions as well as personal histories. Residential segregation, discriminatory employment, insurance exclusions, punitive policies, language barriers, and limited health-service coverage can shape the opportunities available for sustained improvement. A recovery-capital score therefore combines aspects of individual circumstances with the effects of institutions and community environments. Interpreting a low score as a personal weakness would conceal the structural causes of resource deprivation. Researchers should define disadvantage concretely and measure access conditions independently of stated motivation. Cloud & Granfield (2008) Hennessy (2017)
Recovery pathways also vary by cultural expectations, collective identity, relationships to professional authority, and beliefs about acceptable help. Instruments based on one country's treatment system may use concepts such as citizenship, self-reliance, family support, or community participation differently in another setting. Cultural adaptation requires discussion with the target population and quantitative tests of item functioning. A measure may be reliable within groups yet fail to permit direct comparisons of absolute scores between groups. Bunaciu et al. (2024) SAMHSA (n.d.)
Stigma may operate through employment, housing, health care, and internalized expectations. Its proposed influence on recovery requires attention to mechanisms, including service avoidance and reduced social support, but causal inference remains difficult where stigma and prior illness severity are jointly associated. Policy research can exploit changes in access rules and service conditions to test whether removing specific barriers alters health and functional outcomes. A fair evaluation reports results separately for marginalized groups and examines whether an apparent aggregate improvement masks worsening disparities. Hennessy (2017) Kelly et al. (2017)
Research Designs and the Hierarchy of Evidence
Recovery research uses cross-sectional surveys, prospective cohorts, randomized clinical trials, qualitative interviews, linked registries, observational treatment comparisons, and systematic evidence syntheses. These designs answer different questions. A representative survey estimates the distribution of self-reported experiences in its sampling frame. A randomized trial can estimate effects of its assigned intervention under its design and adherence conditions. A longitudinal observational cohort can characterize trajectories, although causal inference requires addressing time-varying confounders, selection and measurement error. A qualitative investigation elucidates meanings and mechanisms that need further testing. Kelly et al. (2017) Dennis & Scott (2012) Laudet (2007)
Studies of recovery capital often seek an explanatory link between baseline resources and later functioning. To make such a link interpretable, investigators should measure baseline substance-use severity, prior recovery status, psychiatric comorbidity, socioeconomic conditions, treatment access, and the exposure time of each resource. Clinical samples may show restricted score ranges, while populations found through online recovery forums or mutual-help organizations reflect their own forms of selection. Attrition can be informative if people who relapse, become incarcerated, lose housing, or die are harder to contact. Bunaciu et al. (2024) Hennessy (2017)
A scientifically credible synthesis distinguishes findings supported by randomization, coherent longitudinal temporal evidence, converging observational studies, and expert consensus. These sources can complement one another when their strengths and limits remain explicit. Absence of a trial does not prove that a resource is unimportant; repeated positive correlations do not prove an intervention will work. Policymakers should consider applicability, implementation burden, adverse effects, population equity, and the relevance of measured outcomes before translating association into service design. Kelly et al. (2020) Sordo et al. (2017)
Causal Inference, Mediation, and Time-Varying Confounding
Recovery capital can be both cause and consequence of improving symptoms. Better health may make work possible; employment may supply income and routine that support later health. Standard regression adjusting for employment measured after treatment may inadvertently block part of the treatment pathway or introduce collider bias. A causal diagram should explicitly distinguish preexisting resources, intervention exposure, intermediate changes, time-varying substance use, and final endpoints. Researchers should articulate an estimand: for example, the effect of providing housing assistance, rather than the association between having housing and reporting recovery. Hennessy (2017) Bunaciu et al. (2024)
Time-varying confounding is especially consequential where earlier treatment changes later access to services and that access changes subsequent treatment decisions. Marginal structural models, appropriate longitudinal g-methods, and carefully specified dynamic models can sometimes address these structures when assumptions and measurement permit. Such techniques cannot correct entirely for unobserved determinants or poor construct validity. Pre-registration of the analysis plan and sensitivity to plausible hidden bias help clarify what the observational data support. Sordo et al. (2017) Dennis & Scott (2012)
Mediation claims require independent observations of proposed mechanisms. A peer-support intervention may affect social ties, self-efficacy, use of clinical services, and hope simultaneously. Showing that a post-intervention capital score statistically explains an outcome is insufficient when score changes include outcome components. Researchers should compare alternative models, report the timing of mediator and outcome assessment, and examine potential adverse or unintended effects. This discipline is central to advancing recovery capital from a useful organizing concept toward an empirically discriminating explanatory theory. Vilsaint et al. (2017) Bunaciu et al. (2024)
Replication, Transportability, and Missing Evidence
Evidence should be evaluated for independent replication and transportability. A clinical intervention tested with adults in one metropolitan health network may not deliver comparable effects in rural environments with sparse transport or in settings where peer services are organized differently. A questionnaire validated in English may change meaning under translation or among people with different cultural experiences of recovery. Systematic measurement reviews reveal domain heterogeneity, and the very existence of multiple instruments complicates pooled estimates that assume they operationalize an identical underlying quantity. Bunaciu et al. (2024) Hennessy (2017)
Outcomes may be selectively reported. A study can emphasize abstinence while omitting quality of life or overdose, or report improved average capital while neglecting participants whose circumstances deteriorated. Publication bias may favor favorable intervention studies, while service-based recruitment misses people who never enter treatment. Missing not at random is especially plausible when incarceration, housing loss, severe recurrence, or death disrupt follow-up. Studies should disclose recruitment, attrition by outcome-related risk, pre-specified endpoints, analytic exclusions, and all relevant adverse events. Sordo et al. (2017) Kelly et al. (2017)
Transportability analysis begins with mechanisms rather than superficial similarity. If transport is a prerequisite for regular clinical contact, an intervention that improves access to rides may have different effects depending on local transit. If group participation works partly through social network composition, a mandatory group could alter social meaning and reduce acceptability. Replication should therefore assess both outcomes and mechanisms, including ecological context, intervention fidelity, participant choice, and institutional conditions. Kelly et al. (2020) SAMHSA (n.d.)
Intervention Evaluation and Long-Term Recovery Support
Recovery interventions should be evaluated as precisely specified changes to the person's opportunities, care, skills, or social environment. A housing initiative, peer-linkage intervention, medication program, and behavioral therapy have different active components and plausible pathways. Trials should identify who receives the intervention, what constitutes the comparator, how exposure is delivered, and whether outcomes are measured long enough to detect sustained benefit and harm. A multidomain recovery framework complements this design by ensuring that efficacy is assessed beyond immediate attendance or a short reduction in use. Dennis & Scott (2012) SAMHSA (n.d.)
Evidence from regular Recovery Management Checkups demonstrates that proactive detection of renewed treatment need and linkage can influence service reentry and later outcomes. In contrast, strong evidence from alcohol-use-disorder research supports manualized Twelve-Step Facilitation for improving abstinence in defined comparisons. These findings address interventions with different mechanisms and populations. They should be integrated through clinical needs and patient preferences, not averaged into an imaginary universal recovery intervention. Health systems need both accessible treatment and the social resources that make continuing care feasible. Dennis & Scott (2012) Kelly et al. (2020)
Implementation should also consider unintended consequences. Excessive surveillance, involuntary disclosure, participation requirements incompatible with employment, and punitive responses to recurrence can undermine trust and access. Study protocols should separate supportive follow-up from coercion and report acceptability, withdrawals, adverse events, and equity. Recovery-oriented care remains an empirical field in which the value of a resource is tested by its effects under identifiable conditions and by the agency of the people who use it. SAMHSA (n.d.) Hennessy (2017)
Recovery Capital and the Limits of Prediction
Interest in recovery capital has encouraged attempts to use composite scores to forecast future abstinence, remission, or retention. Prediction models can be useful if the target is specified, the available predictors are observed before the forecast interval, and performance is measured in independently sampled data. Calibration and transportability matter at least as much as discrimination. A score associated with recovery in an existing dataset may have little predictive value for a new population if measurement properties, treatment access, or baseline risks differ. Bunaciu et al. (2024) Vilsaint et al. (2017)
Composite scores create an additional risk of disguising modifiable shortages as permanent individual risk. A low score driven by unstable housing or inaccessible care is a description of constrained opportunities; it is potentially a reason to provide support. If used to exclude people from treatment, the score could reinforce the very disadvantage it measures. Transparent assessment should separate personal skills from structural conditions, avoid using recovery labels as proxies for worthiness, and include tests for differential accuracy across relevant groups. Cloud & Granfield (2008) Hennessy (2017)
A stronger research program asks whether adding a measured resource genuinely improves outcomes, for whom, and through which pathway. Predictions should be revised as circumstances change, evaluated prospectively, and interpreted alongside clinical assessment and patient goals. The same recovery-capital dataset may support descriptive resource mapping without being suitable for individual prognostic decisions. Scientific integrity requires keeping those purposes separate. Groshkova et al. (2013) Bunaciu et al. (2024)
Digital Recovery Support and Human–Artificial Relations
Digital systems can extend access to reminders, educational materials, remote appointments, peer connections, and symptom tracking. Their possible contribution to recovery capital is an empirical question about timely access, usability, confidentiality, reliability, and sustained health outcomes. A digital service may provide a useful channel while being inaccessible to people without reliable connectivity, private space, or devices. Analysis must measure who can use a system, whether use improves service continuity, and what changes occur beyond the screen. The availability of an interface is not evidence of clinical effectiveness. SAMHSA (n.d.) Bunaciu et al. (2024)
Within the Artificial Era, artificial conversational and decision-support systems may mediate information, planning, and access to human care. Their psychological relevance concerns users' expectations of responsiveness, anthropomorphic attributions, privacy decisions, and the possibility that apparent advice influences help-seeking. Claims that an artificial system acts as a recovery therapist or creates durable recovery capital require prospective evaluation, appropriate safety governance, and clearly defined human oversight. A conversational relationship should not displace clinically indicated medication, emergency services, or qualified care. CDC (2024)
Research questions include whether digital reminders improve attendance, whether personalized information increases access, whether online peer connections strengthen social resources, and whether algorithmic classification creates unequal or stigmatizing decisions. Studies must address safety events, data retention, model drift, differential access, and the potential for overreliance. Cross-system philosophical classifications associated with Artificial Sapiens are separately attributable conceptual work; the present empirical subject remains human addiction recovery in environments increasingly mediated by artificial systems. SAMHSA (n.d.) Hennessy (2017)
A Multilevel Synthesis of Recovery Mechanisms
The literature supports a multilevel interpretation of recovery as coordinated change across symptoms, learned behavior, social relationships, material conditions, and clinical access. Biological and psychological models clarify why craving, cue exposure, stress responses, and action control may continue to matter after diagnosis enters remission. Recovery-capital approaches specify resources through which new behaviors can be enacted and supported. Longitudinal studies and service trials examine how these elements unfold in actual populations. No single domain explains all pathways or captures the whole outcome. Volkow et al. (2016) Cloud & Granfield (2008) Dennis & Scott (2012)
Conceptually, one can distinguish four interacting sequences: changing disorder-related symptoms and risks; mobilizing resources that enable treatment and daily stability; expanding meaningful roles and social belonging; and adapting services to changing needs. These sequences are proposed analytical distinctions rather than a newly validated theory. Their relevance derives from the fact that symptom course, social resources, and functional outcomes are separately measurable and may exhibit reciprocal relations. The testable question is whether specific changes precede later changes in other domains after accounting for common causes. SAMHSA (n.d.) Bunaciu et al. (2024)
Scientific synthesis requires preserving disagreement. Abstinence-centered, harm-reduction, clinical-remission, and broader wellness frameworks prioritize different endpoints. Each can be useful when matched to a transparent question and the safety profile of the disorder. Recovery capital provides a language for capabilities and opportunities but has heterogeneous measurement and limited causal confirmation across settings. Strong conclusions concern the diversity of routes to resolution, the importance of reliable access to effective clinical care, and the need for multidimensional, longitudinal assessment. Kelly et al. (2017) Sordo et al. (2017) Bunaciu et al. (2024)
Critical Scientific Discussion
The first major unresolved issue is conceptual plurality. A clinician evaluating diagnostic remission, a patient evaluating quality of life, and a policymaker evaluating overdose deaths may reasonably select different indicators. These differences can support complementary descriptions if explicitly stated, but a study cannot claim comprehensive recovery from one narrow endpoint. Comparisons across studies must retain definitions, time windows, population restrictions, and the distinction between patient-reported problem resolution and clinically assessed diagnostic status. Hasin et al. (2013) Kelly et al. (2017) Laudet (2007)
A second issue concerns causal attribution. People who possess more resources often have better outcomes, but resources may be both consequences and causes of health, and unmeasured social and clinical factors influence both. Current recovery-capital instruments are varied and developed predominantly in adult settings. Their psychometric merits are uneven across cultures and population groups, and some item content risks circularity. Rigorous longitudinal models, measurement-invariance tests, randomized resource interventions where ethical, and sensitivity analyses are needed before broad causal or individual-prognostic claims. Bunaciu et al. (2024) Groshkova et al. (2013)
Third, treatment evidence and recovery social science operate on different inferential scales. Medication studies may establish important differences in mortality risks, whereas narrative interviews clarify what a meaningful life after dependence entails. Neither form of evidence eliminates the contribution of the other. Finally, structural inequality shapes both recovery opportunities and the availability of data. More inclusive research must attend to mortality, incarceration, displacement, and the people who never enter established treatment and recovery organizations. Sordo et al. (2017) Laudet (2007) Hennessy (2017)
Open Research Questions and Testable Priorities
The field needs prospective studies that distinguish recovery capital as a preexisting resource from capital generated during recovery. Such designs should repeatedly measure clinical criteria, actual substance-use exposure, material conditions, social networks, and patient-defined quality of life. They can then ask whether improvements in housing or access to care predict later reductions in harm and whether symptom improvement subsequently changes those resources. Missingness, time-varying treatment, and competing risk of death should be addressed explicitly. Bunaciu et al. (2024) Sordo et al. (2017)
Intervention trials should test modifiable components rather than a vague total capital score: transportation, service navigation, housing stability, flexible appointment access, peer linkage, and protection against stigma. Comparators should reflect real alternative services, and endpoints should include safety, sustained function, acceptability, and distributional equity. Robust subgroup analysis is essential, especially where the studied intervention may interact with geography, disability, age, and treatment accessibility. Dennis & Scott (2012) SAMHSA (n.d.)
International psychometric collaboration should establish which recovery-capital constructs are generalizable and which require locally defined indicators. Open instruments, predeclared validation plans, longitudinal responsiveness, and transparent reporting can improve synthesis. Research on artificial and digital supports should use experimentally testable mechanisms and independent safety evaluation, with people in recovery involved in choosing meaningful endpoints. The broader scientific task is to explain how improvements become durable across changing lives, not merely to rename favorable outcomes as recovery. Hennessy (2017) Bunaciu et al. (2024)
Conclusion
Addiction recovery is a longitudinal, multidimensional process whose scientific investigation requires sustained attention to disorder course, survival, functional change, and the resources through which people pursue meaningful lives. Diagnostic remission answers a specific clinical question. Recovery capital identifies personal, social, material, and institutional opportunities that may facilitate improvement. Their concepts are related through time and environment, and each must retain an independent operational definition. Hasin et al. (2013) Cloud & Granfield (2008) SAMHSA (n.d.)
Evidence establishes several major foundations. Population research documents heterogeneous assisted and unassisted routes to resolving substance-related problems. Randomized recovery-management studies demonstrate that structured continuing contact and linkage can improve selected long-term outcomes. Clinical mortality research demonstrates the importance of remaining connected to effective opioid treatment. Research-capital measurement has advanced through validated questionnaires while revealing unresolved problems of construct scope, temporal causation, population transportability, and cultural fit. Kelly et al. (2017) Dennis & Scott (2012) Sordo et al. (2017) Bunaciu et al. (2024)
The resulting research agenda treats people as participants in changing social and clinical systems. It evaluates safety alongside symptoms, opportunity alongside individual skill, and self-defined flourishing alongside standardized outcomes. This architecture supports more precise theory, better clinical and public-health evaluation, and more equitable evidence about what sustains recovery across time. Laudet (2007) Hennessy (2017)
References
- Betty Ford Institute Consensus Panel. (2007). What is recovery? A working definition from the Betty Ford Institute. Journal of Substance Abuse Treatment, 33(3), 221–228. https://doi.org/10.1016/j.jsat.2007.06.001
- Bunaciu, A., Bliuc, A.-M., Best, D., Hennessy, E. A., Belanger, M., & Benwell, C. (2024). Measuring recovery capital for people recovering from alcohol and drug addiction: A systematic review. Addiction Research & Theory, 32(3), 225–236. https://doi.org/10.1080/16066359.2023.2245323
- Centers for Disease Control and Prevention. (2024, April 9). Opioid use disorder: Treating. Overdose Prevention. https://www.cdc.gov/overdose-prevention/hcp/clinical-care/opioid-use-disorder-treating.html
- Cloud, W., & Granfield, R. (2008). Conceptualizing recovery capital: Expansion of a theoretical construct. Substance Use & Misuse, 43(12–13), 1971–1986. https://doi.org/10.1080/10826080802289762
- Dennis, M. L., & Scott, C. K. (2012). Four-year outcomes from the Early Re-Intervention (ERI) experiment using Recovery Management Checkups (RMCs). Drug and Alcohol Dependence, 121(1–2), 10–17. https://doi.org/10.1016/j.drugalcdep.2011.07.026
- Groshkova, T., Best, D., & White, W. (2013). The Assessment of Recovery Capital: Properties and psychometrics of a measure of addiction recovery strengths. Drug and Alcohol Review, 32(2), 187–194. https://doi.org/10.1111/j.1465-3362.2012.00489.x
- Hasin, D. S., O'Brien, C. P., Auriacombe, M., Borges, G., Bucholz, K., Budney, A., Compton, W. M., Crowley, T., Ling, W., Petry, N. M., Schuckit, M., & Grant, B. F. (2013). DSM-5 criteria for substance use disorders: Recommendations and rationale. American Journal of Psychiatry, 170(8), 834–851. https://doi.org/10.1176/appi.ajp.2013.12060782
- Hennessy, E. A. (2017). Recovery capital: A systematic review of the literature. Addiction Research & Theory, 25(5), 349–360. https://doi.org/10.1080/16066359.2017.1297990
- Kelly, J. F., Bergman, B., Hoeppner, B. B., Vilsaint, C., & White, W. L. (2017). Prevalence and pathways of recovery from drug and alcohol problems in the United States population: Implications for practice, research, and policy. Drug and Alcohol Dependence, 181, 162–169. https://doi.org/10.1016/j.drugalcdep.2017.09.028
- Kelly, J. F., Greene, M. C., Bergman, B. G., White, W. L., & Hoeppner, B. B. (2019). How many recovery attempts does it take to successfully resolve an alcohol or drug problem? Estimates and correlates from a national study of recovering U.S. adults. Alcoholism: Clinical and Experimental Research, 43(7), 1533–1544. https://doi.org/10.1111/acer.14067
- Kelly, J. F., Humphreys, K., & Ferri, M. (2020). Alcoholics Anonymous and other 12-step programs for alcohol use disorder. Cochrane Database of Systematic Reviews, 2020(3), CD012880. https://doi.org/10.1002/14651858.CD012880.pub2
- Larimer, M. E., Palmer, R. S., & Marlatt, G. A. (1999). Relapse prevention: An overview of Marlatt's cognitive-behavioral model. Alcohol Research & Health, 23(2), 151–160. https://pubmed.ncbi.nlm.nih.gov/10890810/
- Laudet, A. B. (2007). What does recovery mean to you? Lessons from the recovery experience for research and practice. Journal of Substance Abuse Treatment, 33(3), 243–256. https://doi.org/10.1016/j.jsat.2007.04.014
- Sordo, L., Barrio, G., Bravo, M. J., Indave, B. I., Degenhardt, L., Wiessing, L., Ferri, M., & Pastor-Barriuso, R. (2017). Mortality risk during and after opioid substitution treatment: Systematic review and meta-analysis of cohort studies. BMJ, 357, j1550. https://doi.org/10.1136/bmj.j1550
- Substance Abuse and Mental Health Services Administration. (2023). Counseling approaches to promote recovery from problematic substance use and related issues (Treatment Improvement Protocol [TIP] Series 65; Publication No. PEP23-02-01-003). https://www.ncbi.nlm.nih.gov/books/NBK601482/
- Substance Abuse and Mental Health Services Administration. (n.d.). About recovery. https://www.samhsa.gov/substance-use/recovery/about
- Vilsaint, C. L., Kelly, J. F., Bergman, B. G., Groshkova, T., Best, D., & White, W. L. (2017). Development and validation of a Brief Assessment of Recovery Capital (BARC-10) for alcohol and drug use disorder. Drug and Alcohol Dependence, 177, 71–76. https://doi.org/10.1016/j.drugalcdep.2017.03.022
- Volkow, N. D., Koob, G. F., & McLellan, A. T. (2016). Neurobiologic advances from the brain disease model of addiction. New England Journal of Medicine, 374(4), 363–371. https://doi.org/10.1056/NEJMra1511480