Information sources

The databases Scopus, Web of Science (WoS), ScienceDirect, PubMed, and PubMed Central (PMC) were used to find relevant articles on the risk of recidivism among substance-using individuals for this systematic review. The selection of these databases was based on their extensive coverage of peer-reviewed literature in several fields, ensuring a complete and interdisciplinary review process. Following the PRISMA guidelines, the search strategy was rigorously constructed to retrieve findings published between 2004 and 2024. To include the most recently published and relevant research on recidivism and substance use, a 20-year timeframe was selected that reflected current trends and theoretical developments in the area. The study attempts to give a current synthesis of the psychosocial factors of recidivism among people with a history of substance use and imprisonment by concentrating on these particular databases and periods.

Search strategy

The search method was created especially to find studies on the risk of recidivism among substance-using individuals. This strategy comprised a variety of search phrases and Boolean operators with regard to the psychological and social elements that impact recidivism. The search strategy includes broad but pertinent phrases such as ‘psychological factors,’ ‘social factors,’ and ‘recidivism risk.’ This method was used to prevent inadvertently limiting the scope and to include research that looked at various psychosocial influences, even if they employed different terminologies. Following data extraction and synthesis, specific variables, including social support, mental health, and peer influence, were discovered. Our search strategy targeted abstracts, keywords, and, where supported by the database, full text (e.g., ScienceDirect, PMC). In databases such as WoS and PubMed, searches were limited to abstracts and keywords. The following search expression was used, with adaptations to meet each database’s requirements: (“psychological factors” OR “psychosocial factors” OR “mental health” OR “psychiatric disorders” OR “psychological disorders”) AND (“social factors” OR “social environment” OR “socioeconomic status” OR “social support” OR “family support”) AND (“recidivism” OR “reoffending” OR “criminal recidivism” OR “drug offenders” OR “substance abuse”) AND (“risk factors” OR “risk”).

Eligibility and screening

The data collection approach for this systematic review was rigorously planned to ensure that significant data were extracted comprehensively and accurately from the included studies. Several steps were taken to screen the literature using Covidence software. These procedures were aided by the establishment of inclusion and exclusion criteria, which ensure that a rational selection of papers is selected for review. These criteria were adapted to ensure that the most relevant research on the psychosocial causes of recidivism among substance users was chosen. This allowed the researcher to precisely achieve the study’s objective. The final criteria, shown in Table 1, represent a thorough and methodologically sound assessment procedure.

The Covidence software program was used to gather all recognized records from the database searches. The integrated feature of Covidence was used to automatically identify and eliminate duplicates. A total of 34,766 records had been identified across several databases during the initial search on July 11, 2024. Prior to screening, 33,908 records were eliminated, of which 21,722 were deemed ineligible, 12,179 were duplicates, and 7 were eliminated for non-linguistic reasons. 858 records were evaluated in the screening phase; 570 of those were discarded because they were not full-text or did not have open access. 288 reports were then assessed for eligibility. Of these, 34 were deemed to be grey literature, 126 were found to lack pertinent data, and 98 did not explicitly address the relationship of interest. Finally, 30 papers matched the pre-established criteria and were included in the analysis. Fig 1 shows the PRISMA Flowchart that illustrates this selection procedure in detail.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA Flowchart.

https://doi.org/10.1371/journal.pone.0327810.g001

Quality assessment

To ensure the validity and reliability of the findings presented in this systematic review, we used rigorous methodologies to analyze the risk of bias in the included studies. The evaluation was carried out using Covidence. Covidence enables numerous reviewers to conduct independent assessments, offering a platform for rapid and reliable data extraction and risk of bias examination.

To ensure impartiality and reduce the possibility of reviewer bias, each study in this review was critically evaluated for bias by two reviewers working separately. Reviewers assessed the studies according to predetermined standards, paying particular attention to participant selection, research design, and outcome. If there were disagreements amongst the reviewers, they were resolved by discussion or, if required, consultation with a third reviewer.

To assess the methodological quality of the studies further, we used the Mixed Methods Appraisal tool [31] (see Supplementary Material 1). MMAT is especially used to appraise the quality of studies using diverse research approaches, such as qualitative, quantitative, and mixed methods. Two independent reviewers evaluated each study, and disagreements were resolved through discussion. The agreement degree between reviewers was assessed through Cohen’s Kappa (K). The observed value (K = 0.86) was considered almost perfect [32]. We were able to ensure that the findings were supported by solid and well-conducted research by using this technique to appraise the quality of the included papers carefully. Overall, the risk of bias was minimal in the included research since most followed rigorous methodological criteria.

Outcome measure

The key outcome of interest in this review was recidivism, which is often measured using administrative database records of rearrest, reconviction, or reincarceration. The primary objective, however, was to assess and synthesize the psychosocial elements that influence recidivism in people with a history of substance use and incarceration. A uniform effect measure was not possible due to the inconsistencies in study designs and recidivism definition. A narrative synthesis approach was used to incorporate information about psychological vulnerabilities, mental health concerns, social support, family dynamics, and structural disadvantages.

Description of studies

This evaluation looked at 30 studies on the risk of recidivism among individuals with a history of substance use and incarceration by year of publication, country of origin, nature of study, risk factor type, and length of follow-up to provide a thorough synthesis (see Table 2). The studies were published between 2004 and 2024, with peaks in 2010 (n = 4) and 2011 (n = 3). Most years provided one study apiece, demonstrating an ongoing research interest across two decades. The majority of the research was done in the United States (n = 18), with Sweden and the Netherlands (n = 3 each) and Argentina, Australia, Finland, Ghana, Malaysia, and Spain (n = 1 each) also represented.

The methodological approach was primarily quantitative (n = 27), including retrospective and longitudinal cohort studies, with only three studies using qualitative approaches such as interviews and focus groups.

The gender composition of the samples in the 30 analyzed studies showed a preponderance of male participants. Eleven studies specifically focused on male incarcerated individuals, frequently seen in adult or juvenile populations. However, there was just one study that exclusively examined the female population in prison. The majority of research (n = 18) had male and female participants, with varying sample sizes and gender distributions. These mixed-gender samples included large national datasets (i.e., 47,326 prison population, consisting of 43,840 males and 3,486 females) and small-scale studies (i.e., 30 recidivists evenly divided by gender). The populations under study were diverse and included young, incarcerated individuals, adults with mental illnesses or forensic psychiatric conditions, and homeless former incarcerated individuals.

In terms of the types of risk factors investigated (see Table 3), ten studies focused solely on psychological variables [33–42], three assessed social aspects alone [43–45], and 17 studies combined psychological and social dimensions [4,5,46–60]. The period of follow-up utilized to determine recidivism, which spanned from 6 months to 15 years, was a key differentiating feature. A substantial number of studies utilized short- to mid-term follow-ups (e.g., 3–15 months, 1–3 years), which allowed for the evaluation of immediate post-release risks. One study followed individuals for up to 15 years, whereas others used longer tracking periods of 5–10 years.

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Findings of the study.

https://doi.org/10.1371/journal.pone.0327810.t003

Psychological factors influencing recidivism risk among incarcerated individuals with a history of substance use

Psychological dysfunction was consistently linked to an increased risk of recidivism across studies, particularly when it co-occurred with substance use disorders (SUD). Seven studies found that mental illnesses, such as mood disorders, significantly contribute to reoffending, especially when untreated [33,35–39,42]. Three of those studies discovered that the impact was enhanced by co-occurring SUD, and that dual diagnoses were linked to a greater risk than either condition alone [35,36,42]. Particularly, individuals with bipolar disorder were identified as high-risk because of their impulsivity and mental instability, which were made worse by substance use [35,36].

Additionally, two studies found that recidivism is predicted by mentally ill individuals with certain psychological diagnoses, such as personality disorders and psychosis, particularly among violent or sexual offenders [37,39]. These disruptions were frequently linked to increased impulsivity, poor coping, and trouble regulating emotions. Incarcerated individuals who are mentally ill tended to exhibit different patterns of conduct, even in cases when overall reoffending rates were the same [37]. Stahler et al. [41] corroborated these results by demonstrating that substance usage by itself raised the risk of reincarceration by about 40%. In a similar vein, Ojansuu et al. [40] discovered that during a follow-up period of five to fifteen years, incarcerated individuals with SUD had a reoffending rate that was more than twice as high. Notably, over half of these recurrent misconducts were violent, where people with untreated psychological and substance-related conditions exhibit persistent behavioral instability.

This risk was further increased by unmet psychological needs during reentry. Begun et al. [33] found a significant discrepancy between those who received services to address their psychological problems (15%) and those who anticipated post-release mental health needs (56.5%). Hepburn [38] also noted psychological barriers like low motivation or denial that decreased engagement even in the presence of services.

Lastly, it was shown that early psychological stress was also an essential component. Two studies [34,35] found that trauma and abuse throughout childhood were associated with a higher likelihood of committing crimes later on, especially when drug use was present. While symptoms such as depression or suicidal thoughts did not predict reoffending on their own, Benner et al. [34] discovered that those with a history of abuse had a higher likelihood of being arrested as adults (42%), compared to youth (27%), where a delayed but long-lasting psychological impact contributes to adult reoffending.

Social factors influencing recidivism risk among incarcerated individuals with a history of substance use

Social determinants such as familial support, parental criminality, and neighborhood disadvantage were consistently associated with recidivism risk. A protective element across studies was family emotional support, particularly during the sensitive reintegration phase. Higher levels of emotional support from family members were linked to lower recidivism rates and longer time between incarcerations, according to three studies [5,43,49]. Two of them discovered that the protective influence of family relationships and support networks offered emotional stability and lessened dependence on unhealthy coping strategies [43,49]. Support alone, however, was not always sufficient, as demonstrated by the findings of Barrick et al. [43]. Even among individuals with strong emotional ties, reoffending happened when additional risk factors, including socioeconomic instability or unresolved psychological difficulties, went untreated [50].

Conversely, unfavorable social circumstances were linked to a higher probability of recidivism. According to two study [44,45], those who were exposed to parental criminality or structural poverty at the community level had a higher likelihood of committing crimes again. These environments frequently reduced possibilities for employment or community reintegration, normalized illicit behaviors, and weakened access to prosocial role models. Specifically, parental criminality and living in a socially poor neighborhood were found to be important predictors of juvenile reoffending by Grunwald et al. [44]. Similarly, Kubrin and Stewart [45] found that rearrest rates differed significantly amongst neighborhoods, with greater rates concentrated in areas with poor social cohesiveness and inadequate access to support systems.

Interplay between psychological vulnerabilities and social environment

Research repeatedly shows that psychological and social elements do not work in isolation, but rather frequently combine in intricate ways to influence incarcerated individuals’ possibility of reoffending. Across the studies reviewed, both psychological and social factors were consistently linked to recidivism risk among those with a history of substance use. Substance use has been shown to dramatically increase the risk of recidivism in four studies [54,56–58], especially when paired with exposure to unstable environments, criminal peer associations, or inadequate family support. For instance, one of these studies found that violent repeat offending was much more common among incarcerated individuals who had substance use issues and troubled social relationships [54]. Another research [58] found that high-risk people had greater rates of drug issues and were more likely to originate from single-parent families or non-traditional family structures.

Five studies [46,48,53,54,60] evident that social risks often exacerbate psychological vulnerabilities, including pro-criminal attitudes, poor self-control, mental illness, or personality pathology. For example, two studies highlighted the importance of gang membership and mental health decline as reinforcing processes, wherein societal and psychological elements contributed to persistent criminal conduct [48,60]. One study [53] discovered that incarcerated individuals with antisocial or emotionally dysregulated traits were more likely to experience repeated misconduct issues and interpersonal conflict within institutions, to which psychosocial maladjustment perpetuates both in-prison and post-release risk.

Furthermore, three studies [49,50,60] showed that untreated mental illness, specifically depression, psychiatric instability, and emotional dysregulation, was a significant predictor of recidivism when it co-occurred with criminogenic networks, family conflict, or financial stress. One of this research [50] discovered that depressive symptoms indirectly enhanced reoffending through higher social strain, whilst two others revealed that psychological changes and emotional support after release might reduce these risks [49,60].

Four research [4,55,59,60] identified social instability, including homelessness, inadequate parenting, and disturbed peer or family connections, as a risk factor. In Karlsson and Håkansson’s [4] study, homelessness, in particular, stood out as a key predictor. When coupled with mental health issues, its influence was exacerbated, further solidifying the psychosocial vulnerability factor. The remaining research showed that exposure to peer rejection, familial dysfunction, or disengagement from school frequently happened in conjunction with internal vulnerabilities, including maladaptive attitudes, poor impulse control, or mental illness, increasing the likelihood of violent reoffending.

Three studies [5,49,59] showed that recidivism rates were decreased by solid family or partner ties, systematic supervision, emotional support, and a positive sense of self. One longitudinal research [5], for instance, discovered that those who preserved strong familial relationships and supported prosocial identities had a higher chance of being conviction-free for ten years.

Additionally, the three qualitative studies [47,51,52] highlighted that substance use was not only a personal habit but rather a component of a larger psychosocial path to identity creation, emotional regulation, and survival in challenging situations. Two studies [47,51] found that emotional discomfort, portrayed as guilt, despair, and poor motivation, was frequently exacerbated by social isolation, character contamination from the cell population, and a lack of relationship support. In both situations, a lack of prosocial options and a lack of interest in recovery led to substance use as a coping strategy. One study [51] found that post-release problems such as stigma, rejection, and a lack of social scaffolding enhanced emotional distress. In Mohd Alif et al.’s [52] study, substance use was depicted as symbolically meaningful, where it provided a sense of control and belonging in the face of persistent external constraints such as poverty, peer normalization of crime, and unmet mental health needs.

Limitations and potentialities

While this review synthesizes significant psychosocial determinants of recidivism, some limitations warrant acknowledgment. One major drawback of this review is the variability of the research populations, notably in terms of gender and age. Several studies focused solely on adult male populations, while others included juvenile offenders or female participants, frequently without breaking down results by subgroup. This heterogeneity makes it challenging to draw generalizations across different demographic groups since the psychological and social variables that influence recidivism may show up differently in males and females or adolescents and adults. Furthermore, the operationalization of recidivism was inconsistent. Most research used administrative data, such as arrest records, reconviction rates, or re-incarceration, as proxies for recidivism. Although they are widely used indicators, they cannot not adequately account for the complexities of recidivism, including changes in the judicial system or unreported criminal activities. Direct comparisons are made more difficult by the variation in sample composition and outcome measurement. Moreover, the variation in recidivism categories among the included studies adds complexity to the analysis, since different crimes may include various risks. However, this is a deliberate methodological decision rather than a constraint, since it allows for a more thorough understanding of the common psychosocial pathways that contribute to recidivism. This more comprehensive approach is consistent with data indicating that criminogenic demands are developed by social and psychological processes, such as substance use patterns and psychosocial factors, rather than being intrinsic.

While some studies concentrated only on sexual recidivism, all of them included background information on substance use, which adds complexity to the analysis of recidivism and makes it challenging to compare results across different kinds of offenses. Furthermore, there were notable differences in the duration of time at risk for recidivism amongst studies, ranging from brief to extended follow-ups. Though this might potentially point to possible biases in reporting or measurement, the results frequently converged despite this heterogeneity, showing some stability in recidivism patterns. While cross-sectional data is useful for providing a quick overview, it restricts the ability to draw conclusions about the long-term interactions between social and psychological elements. Even though the review contains research from a variety of geographic and cultural contexts, this diversity both enhances our understanding of the subject and draws attention to the difficulties associated with generalizing results to other groups. As research with more noticeable findings is more likely to be published, publication bias is a factor that should be taken into account, as it might affect the conclusions that are made overall. Despite these drawbacks, the analysis provides a strong basis for future study and useful application in the area by synthesizing a wide variety of information and identifying significant factors impacting recidivism.

While this systematic review synthesizes data from research on various crimes, it is essential to note that their fundamental basis lies in shared psychosocial components. The current review emphasizes how psychosocial variables, including family support, economic stability, mental health, trauma, and drug use, consistently shape recidivism risk, even in the face of variations in legal classifications and particular criminal typologies. This convergence implies that interventions aimed at these variables may not only affect certain crime categories but also have a wide-ranging effect on recently released adults. One of this review’s main potentialities is that it supports the idea that criminogenic demands are socially acquired and reinforced throughout time rather than being innate characteristics. This review highlights how long-term social vulnerabilities, including substance use, mental health treatment inaccessibility, and marginalization, progressively translate into psychological functioning that perpetuates criminal cycles. Specifically, as a maladaptive coping strategy influenced by larger psychosocial stresses, substance use appears as both a cause and an effect of criminal activity. This analysis emphasizes the premise that successful recidivism prevention requires addressing societal factors in addition to individual rehabilitation efforts.

Implications

The findings of this analysis have important implications for both practice and policy in the criminal justice and public health sectors. The necessity for treatment options that address both the specific psychological needs and their wider social milieu is highlighted by the substantial correlation shown between psychological issues, such as maladaptive coping strategies, and social factors, such as peer influence. For example, social support interventions that focus on peer connections and community reintegration might be incorporated with cognitive behavioral therapy programs designed for people with a history of substance use. The research also points to the necessity of continuity of care for people with a history of imprisonment when they are released from prison, with a focus on ensuring that they have access to social support networks, drug treatment centers, and mental health initiatives that reduce their risk of recidivism.

Future research

Given the identified gaps in the literature, future research should focus on investigating the interplay of social and psychological components in a variety of cultural and socioeconomic circumstances. Future research ought to emphasize addressing the issues raised in this article while exploring areas that have not been investigated. Longitudinal studies, which follow incarcerated individuals over protracted periods and offer a more thorough knowledge of how psychological and social variables impact recidivism over time, are one important path. Individuals with a history of imprisonment may be hesitant to commit to long-term engagement. To overcome these constraints, future research should look into alternate methodologies, such as employing administrative longitudinal datasets or combining mixed-methods designs with periodic follow-ups that respect participant autonomy. Using peer researchers and forming community-based collaborations may help improve retention and trust in these investigations. This method may make it easier to understand the causal relationships and temporal dynamics between psychosocial factors impacting the risk of recidivism among people with a history of substance use and incarceration. Future research should also aim to unify recidivism definitions and metrics while taking into account the biases caused by different kinds of crimes and time-at-risk.

Future research should consider stratified analyses to better capture population-specific patterns and risk profiles. Future reviews might benefit from taking a more intersectional approach, looking at the ways in which psychological and social characteristics interact with variables like race, gender, and socioeconomic position to impact recidivism. Furthermore, research comparing the efficacy of various intervention techniques is needed to determine the best methods for lowering recidivism among people with a history of substance use. This is especially relevant for interventions that combine psychological treatment with social support networks. Future research can help establish more focused and efficient treatments by filling these gaps, which will eventually increase the effectiveness of rehabilitation programs and lower recidivism rates.