Background
Substance use disorders (SUD) are very common among those incarcerated by the criminal justice system: more than half of people in prison (58%) and nearly two-thirds of people in jail (63%) meet criteria for a SUD [
1]. Individuals transitioning from jail or prison back into the community face many challenges, including being required to meet probation or parole conditions and expectations, reestablishing housing and employment that were lost while incarcerated, and stigma [
2‐
5]. Those with a SUD also must contend with increased risk of mortality and returning to environments that increase the likelihood of substance use [
2,
4,
6‐
8]. Additionally, 41% of all probationers are required to attend alcohol or drug treatment [
9]. Although there is some evidence that legal pressure may improve retention in treatment, namely for those in short- or long-term residential treatment [
10], many individuals may not seek treatment at all and still will drop out of outpatient treatment prematurely, despite severe consequences. For example, in a sample of 926 individuals who opted to attend addiction treatment in lieu of jail or prison, 59% dropped out of treatment, of whom 25% were reincarcerated [
11]. Addiction care specifically geared towards the needs of justice-involved individuals may improve treatment engagement and mitigate retention challenges, reduce risk for returning to substance use, and help to prevent individuals’ continued involvement with the justice system [
2].
In the general population of U.S. residents 12 years and older, approximately 8.1% have a SUD, but only 11–14% of those with a SUD seek or get the treatment they need [
12,
13]. Multiple barriers to treatment have been identified in general samples of persons with a SUD not specific to those who are justice-involved. For example, among two large, national samples of individuals with a SUD in the United States who were not seeking treatment [
12,
14] and a sample of individuals with a SUD presenting to treatment, [
15] barriers included not being ready to stop using alcohol or drugs, lacking resources (e.g., insurance), and not knowing where to seek treatment [
12,
14,
15].
Individuals with a SUD who are justice involved face similar treatment barriers to individuals with a SUD more generally, including lack of health care and costs of treatment [
5,
16] and not seeing their substance use as a problem [
3,
5,
17]. These barriers were experienced across various groups of justice-involved individuals not in SUD treatment, including those who were incarcerated [
7] or on probation or parole [
5,
17] and those with both alcohol and other drug use disorders [
3,
5]. Despite these similarities, justice-involved individuals with a SUD appear to face additional barriers, including stigma related to having a legal history [
3,
7] or criminal justice agencies’ preference to provide “drug-free treatment” that exclude pharmacotherapies for SUD [
18].
However, there are limitations of prior studies of barriers to addiction treatment for justice-involved individuals. For example, some of these studies focused on specific subgroups, such as those with a drug use disorder [
17] or women [
7], when justice-involved individuals comprise a heterogeneous population. Other studies only examined barriers to medication-assisted treatments [
18,
19]. Further, studies often queried participants on a finite selection of barriers rather than using open-ended questions (e.g., [
5,
7]), which may not capture the full range of barriers experienced by this group. Given the low rates of treatment receipt among justice-involved individuals with a SUD and the subsequent consequences they may experience (e.g., re-arrest or reincarceration), there is a need for a more comprehensive understanding of barriers to addiction treatment among this group to inform more effective interventions that improve treatment retention and outcomes, and decrease recidivism. Thus, exploration of a broader range of barriers to treatment among justice-involved individuals with a SUD may be needed to help treatment programs and court systems reduce these barriers and/or directly address them with their clients.
To improve our understanding of barriers to addiction treatment experienced by justice-involved individuals with a SUD more generally, the purpose of the current pilot study was to describe barriers to addiction treatment reported among a small sample of recently incarcerated individuals with a SUD, including responses provided to an open-ended question of additional barriers. Given the small sample size and broad assessment of barriers to treatment that has not been done previously with justice-involved individuals with a SUD specifically, there were no a priori hypotheses and the study was exploratory.
Results
Among the 28 participants who completed at least one follow-up interview, 11 completed only the 1-month interview, 6 completed only the 3 to 6-month interview, and 11 completed both. Responses to the open-ended question of additional barriers were examined for all completed interviews (
n = 39), of which four were left blank (10.3%). Just over half of participants included in the present study were randomized to the motivational intervention condition (
n = 15, 53.6%). Sample characteristics are provided in Table
1.
The means and standard deviations, as well as the number and percentage of participants endorsing items in each of the seven BTI domains are provided in Table
2. The most commonly endorsed barriers to addiction treatment were related to the domain Absence of Problem: “I do not think I have a problem with drugs” (42.8%) and “My drug use is not causing any problems” (32.2%), and Privacy Concerns: “I do not like to talk about my personal life with other people” (35.7%) and “I do not like to talk in groups” (32.2%). No participants endorsed four of the five barriers from the Negative Social Support domain: “I will lose my friends if I go to treatment”, “People will think badly of me if I go to treatment,” “Someone in my family does not want me to go to treatment,” or “My family will be embarrassed or ashamed if I go to treatment.”
Table 2
Descriptive information from the Barriers to Treatment Inventory
Absence of Problem
| 2.6 | 1.1 | | | |
I do not think I have a problem with drugs | 3 | 1.5 | 12 (42.8) | 4 (14.3) | 12 (42.8) |
No one has told me I have a problem with drugs | 2.5 | 1.5 | 17 (60.7) | 3 (10.7) | 8 (28.6) |
My drug use is not causing any problems | 2.7 | 1.6 | 15 (53.5) | 4 (14.3) | 9 (32.2) |
I do not think treatment will make my life better | 2.3 | 1.2 | 18 (64.3) | 6 (21.4) | 4 (14.3) |
I can handle my drug use on my own | 2.9 | 1.4 | 12 (42.8) | 8 (28.6) | 8 (28.6) |
I do not think I need treatment | 2.5 | 1.3 | 17 (60.7) | 4 (14.3) | 7 (25.0) |
Negative Social Support
| 1.8 | 0.5 | | | |
I will lose my friends if I go to treatment | 1.8 | 0.7 | 23 (82.1) | 5 (17.9) | 0 |
Friends tell me not to go to treatment | 2.0 | 1.0 | 23 (82.1) | 2 (7.1) | 3 (10.7) |
People will think badly of me if I go to treatment | 1.8 | 0.6 | 25 (89.3) | 3 (10.7) | 0 |
Someone in my family does not want me to go to treatment | 1.6 | 0.6 | 26 (92.9) | 2 (7.1) | 0 |
My family will be embarrassed or ashamed if I go to treatment | 1.5 | 0.6 | 26 (92.9) | 2 (7.1) | 0 |
Fear of Treatment
| 1.9 | 0.8 | | | |
I have had a bad experience with treatment | 2.1 | 1.2 | 21 (75.0) | 2 (7.1) | 5 (17.8) |
I am afraid what might happen in treatment | 1.9 | 0.9 | 23 (82.1) | 3 (10.7) | 2 (7.1) |
I am afraid of the people I might see in treatment | 1.8 | 0.9 | 23 (82.1) | 3 (10.7) | 2 (7.1) |
I am too embarrassed or ashamed to go to treatment | 1.8 | 0.9 | 23 (82.1) | 3 (10.7) | 2 (7.1) |
Privacy Concerns
| 2.7 | 1.2 | | | |
I do not like to talk in groups | 2.6 | 1.4 | 18 (64.3) | 1 (3.6) | 9 (32.2) |
I hate being asked personal questions | 2.6 | 1.3 | 17 (60.7) | 3 (10.7) | 8 (28.6) |
I do not like to talk about my personal life with other people | 2.8 | 1.3 | 16 (57.2) | 2 (7.1) | 10 (35.7) |
Time Conflict
| 2.2 | 0.9 | | | |
I have things to do at home that make it hard for me to get to treatment | 2.1 | 1.0 | 21 (75.0) | 3 (10.7) | 4 (14.3) |
It will be hard for me to find a treatment program that fits my schedule | 2.3 | 1.0 | 17 (60.7) | 8 (28.6) | 3 (10.7) |
Poor Treatment Availability
| 2.0 | 0.9 | | | |
I am moving too far away to get treatment | 2.0 | 1.0 | 23 (82.1) | 3 (10.7) | 2 (7.1) |
I do not know where to go for treatment | 1.9 | 0.8 | 22 (78.6) | 5 (17.9) | 1 (3.6) |
I have difficulty getting to and from treatment | 2.1 | 1.2 | 20 (71.4) | 2 (7.1) | 6 (21.5) |
Admission Difficulty
| 2.6 | 1.1 | | | |
I will have to be on a waiting list for treatment | 2.7 | 1.2 | 12 (42.8) | 8 (28.6) | 8 (28.6) |
I have to go through too many steps to get into treatment | 2.5 | 1.1 | 16 (57.1) | 6 (21.4) | 6 (21.5) |
Results of the qualitative data analysis are presented in Table
3. Coding identified responses consistent with a priori domains based on Rapp et al.’s factor analysis [
15] as well as emergent domains. Within a priori domains, barriers related to Absence of Problem (“Don’t need it”), Privacy Concerns (“I’m not completely comfortable with speaking in a group setting”), and Time Conflict (“Just pending court dates”) were endorsed. Three emergent domains were identified in the data, including Ambivalence (e.g., “The fact that I enjoy the drug”), Seeking Informal Assistance (e.g., “I have yet to go to treatment cause [sic] I want to try and see if the support im [sic] getting from my family will be enough…”), and Other, which included open-ended responses that were difficult to interpret or code (e.g., “AA meetings” and “The fact that alcohol is advertised all around”).
Table 3
Examples of additional barriers identified in responses to open-ended question, organized by a priori and emergent domains
Absence of Problem | “Don’t need it” “I just don’t want to go to treatment. I’m fine on my own” |
Negative Social Support | None listed |
Fear of Treatment | “Don’t want to, it hasn’t helped in the past” “I just don’t like the sound of it” |
Privacy Concerns | “I’m not completley comfortable with speaking in a group setting” “Myself—hard to get myself open like that (hard to trust)” |
Time Conflict | “Just pending court dates” “Not wanting to take the time to go” |
Poor Treatment Availability | “Transportation and time avail. Ex: You have a 8 am drug treatment class in Bernalillo Cty-while I now live in Santa Fe city [sic]” “Transportation is an issue” |
Admission Difficulty | “Just my misdemeanor situation prevents inpatients treatments to accept me. That’s about it as far as I can see” |
Emergent domains
|
Ambivalence | “I don’t have the motivation to go. I’m lazy” “The fact that I enjoy the drug” |
Seeking Informal Assistance | “I have yet to go to treatment cause [sic] I want to try and see if the support im [sic] getting from my family will be enough” |
Other | “AA meetings” “The fact that alcohol is advertised all around” “I have heard it all” |
No barriers | “Currently attending treatment. I am ready to change my life for the better” “Currently I have no barriers” |
Post-hoc analyses
To explore potential influences of reported barriers to treatment, independent samples t tests were conducted to examine between-group differences in endorsements of BTI items across study condition (from the parent trial, motivational intervention: n = 15, educational intervention: n = 13) and between those who did (n = 10) and did not (n = 18) report engaging in any treatment seeking behaviors after release from jail. Because we considered these between-group comparisons exploratory and not hypothesis-driven, there were no a priori power analyses conducted, and results are to be used for hypothesis generation.
Exploratory comparisons across study conditions suggested that, relative to those who received the educational intervention, participants randomized to receive the motivational intervention were more likely to agree with barriers related to Absence of Problem: “I do not have a problem with drugs” (M = 3.5, SD = 1.3 vs. M = 2.4, SD = 1.5; t = 2.169, p < .05) and “I can handle my drug use on my own” (M = 3.4, SD = 1.4 vs. M = 2.2, SD = 1.1; t = 2.491, p < .05; see Appendix A). There were no other between group differences by study condition identified in exploratory analyses.
Participants who reported seeking any post-release treatment appeared less likely than those who did not seek treatment to endorse one barrier from Absence of Problem: “I do not think treatment will make my life better” (M = 1.6, SD = 0.7 vs. M = 2.7, SD = 1.2; t = 2.501, p < 0.05), and two barriers related to Privacy Concerns: “I do not like to talk in groups” (M = 1.9, SD = 0.9 vs. M = 3.0, SD = 1.5; t = 2.454, p < 0.05) and “I hate being asked personal questions” (M = 1.9, SD = 0.7 vs. M = 3.1, SD = 1.3; t = 2.993, p < 0.05; see Appendix B). No other differences in reported barriers to treatment among those who sought versus did not seek treatment after release from jail were identified in post hoc analyses.
Discussion
This exploratory study used a survey with closed- and open-ended responses to describe barriers to addiction treatment in a small sample of recently-incarcerated adults with a SUD, a population among whom addiction treatment may be of paramount importance, is underutilized, and rarely tailored to their needs. Barriers to addiction treatment were frequently endorsed, but there were no barriers universally endorsed by this sample. Findings can be considered hypothesis-generating regarding barriers to treatment that might be specific to this population. These findings may be useful in informing future research and clinical efforts for individuals with a SUD with recent or current justice involvement.
Previous studies among justice-involved populations have identified lack of healthcare and related costs, limited availability of qualified healthcare providers, long wait times, stigma, and criminal justice agencies’ procedures and preferences (e.g., exclusion of medication-assisted treatments due to preference to be “drug-free”) as barriers to treatment [
4,
5,
7,
18,
19,
29]. There were similarities in findings between studies on barriers to treatment with other justice-involved samples and the current study. Among justice-involved individuals choosing to attend treatment in lieu of incarceration, [
17] 23.9% expressed no desire for treatment, which is similar to 25.0% of the current sample reporting “I do not think I need treatment”. Rose et al. (2014) found that women incarcerated in jail also reported long wait lists as a barrier to treatment, although this was more common among this sample of women than the current sample (58.9% vs. 28.6%) [
7]. Unfortunately, it often is difficult to make comparisons across studies due to variability in assessing barriers and differing samples. The present study is small, but potentially broader in inclusion criteria than previous studies, and adds to these previous studies by expanding the assessment of barriers. Indeed, we found that barriers not previously identified may be important to consider for this population, including perceiving Absence of Problem and Privacy Concerns. Knowledge of these potential barriers in combination with those previously identified in justice-involved individuals, such as Ambivalence, may provide counselors and criminal justice staff with more nuanced information to assist their clients in overcoming obstacles to seeking or staying in treatment.
A previous study also used the BTI, but with a sample of general treatment-seeking individuals with a SUD, [
15] which presented the opportunity to compare rates of barriers to the current sample of justice-involved individuals with a SUD. Among this sample of treatment-seeking individuals, barriers to treatment included the Fear of Treatment, Time Conflict, and Poor Treatment Availability, which also were endorsed by similar proportions of the current study sample. It is noteworthy, however, that participants in the present study sample endorsed Absence of Problem at a markedly higher rate than what was reported by Rapp et al., such as, “I do not think I have a problem with drugs” (42.8% vs. 12.5%) and “My drug use is not causing any problems” (32.2% vs. 8.0%). These discrepancies may be because the sample from Rapp et al. already had initiated the process of seeking addiction treatment, which often includes an acknowledgement of engaging in problematic substance use, compared to the current sample, who were not necessarily treatment-seeking. Alternatively, the present study selected for people reporting moderate to high alcohol use who may not have had much drug involvement. Because we did not alter the original scale to include both drugs and alcohol in these items, we may have inadvertently overestimated the degree to which our participants felt they had an “Absence of Problem.” However, half of the sample met criteria for a current drug use disorder, suggesting that many individuals in this sample still were experiencing problems related to their drug use.
Though our sample was small, an interesting finding was the relatively few endorsements of barriers in the Negative Social Support domain. Specifically, participants in the present study did not endorse four of the five items, and only a small number endorsed “Friends tell me not to go to treatment” (
n = 3, 10.7%). In contrast, these barriers were more frequently endorsed by the treatment-seeking sample of individuals with a SUD reported by Rapp et al., although still in low proportions (proportions of agreement ranged from 7.4 to 16.0%) [
15]. This finding suggests the possibility that justice-involved individuals with a SUD experience barriers related to Negative Social Support less frequently. While future research is needed to further explore this finding, these results suggest a possible intervention pathway—via the involvement of friends and family to encourage treatment engagement. Additionally, these findings may indicate that resources should be diverted to addressing other barriers, where they may have more impact.
In exploratory, post hoc analyses comparing barriers between participants randomly assigned to treatment conditions, barriers generally were endorsed by similar proportions. Groups did, however, differ on two items from the Absence of Problem domain (“I do not have a problem with drugs”, “I can handle my drug use on my own”), with participants assigned to the motivational intervention reporting these barriers more frequently than those assigned to the education intervention. These findings are hypothesis-generating in suggesting that motivational interventions may not be helpful for increasing problem recognition among justice-involved individuals with a SUD, as they are designed to do, or may be unhelpful in this domain. Other studies with jail populations found null results of motivational interventions on other outcomes such as treatment engagement [
30]. Of note, our results should be interpreted with caution, given the small sample, post hoc nature of these analyses, multiple comparisons, and that the parent study did not include a no-treatment control condition.
Similarly, in post hoc analyses comparing barriers reported between groups based on post-release treatment seeking, groups largely were similar. However, those who reported seeking treatment reported less apprehension about talking in groups or being asked personal or intrusive questions as a barrier. These findings highlight privacy as a potential concern among those who did not seek treatment, which may have clinical implications. This non-treatment-seeking group may prefer individual treatment or find pharmacotherapy less invasive and more beneficial than psychotherapy. However, research is needed in larger samples of justice-involved adults with a SUD who have and have not sought treatment to further explore this issue, and examine if pharmacotherapy could be a viable option for those with Privacy Concerns.
Findings from the present study suggest the possibility that barriers beyond those commonly identified may exist in this population. Two participants specifically referenced issues related to their justice involvement in the reporting of additional barriers that were coded under a priori domains (“Just pending court dates” coded as Time Conflict; and “Just my misdemeanor situation prevents inpatients treatments to accept me” coded as Admission Difficulty). Though justice-involvement was not, on its own, coded as an emergent barrier domain, it is possible that this population experiences specific enhanced barriers to treatment that were not cued by the generally phrased open-ended question we used regarding potential additional barriers to treatment. Additionally, it would be useful to better understand potential barriers identified with more in-depth qualitative work. For example, it would be informative to know if “pending court dates” refer to multiple demands on time or waiting to start treatment to know exact sentencing requirements, as well as to know how a “misdemeanor situation” may prevent someone from entering inpatient treatment (e.g., if someone is precluded from treatment while they wait for court dates). We had limited ability to assess these issues due to both small sample size and administration of one open-ended question without the opportunity to probe. Nonetheless, our study is hypothesis-generating regarding the need for broadening measures of barriers to treatment for persons involved with the criminal justice system, which can then inform future interventions for this population.
The present study is limited in several additional ways. First, this study had a small sample and thus the results presented here should be considered descriptive and hypothesis-generating. Given the small sample, we likely were underpowered to detect between-group differences across treatment condition assignment and report of post-release treatment receipt. The multiple exploratory comparisons we conducted between groups increase the risk of making a Type II error. Moreover, findings may not be representative of justice-involved individuals with a SUD more generally (e.g., those who are arrested without ever being incarcerated) because the sample only included individuals who were sentenced, had a substance use-related charge, were being released from a large Southwest detention center, and agreed to participate in a clinical trial, and recruitment targeted those with an alcohol use disorder (e.g., some may not have identified with needing “drug treatment”); this study also excluded those in the methadone maintenance program. The study recruited for individuals with an alcohol use disorder, but many questions of the BTI use language referring to drug use specifically (e.g., “I do not think I have a problem with drugs”), which may have influenced responses to these BTI items and resulted in overestimates of these barriers. Findings also may not be generalizable to women, and there may be additional barriers experienced by women not reflected in these results (e.g., competing basic needs related to childcare, added stigma [
7]). Findings also may be biased by loss to follow-up. Though we compared participants across follow-up status and identified no differences, participants who did not complete either follow-up interview may experience more or different barriers to treatment (e.g., lacking reliable access to a phone, which may make it difficult to seek out treatment services). Further, many participants already had made changes to their alcohol and/or drug use by the time they completed the follow-up interview, which may have influenced their perceived need for treatment. Because this study was a secondary analysis of a pilot parent study that was not designed to investigate issues related to barriers to treatment, we may have missed some barriers to treatment (e.g., barriers to addiction treatment related to cost and lack of healthcare coverage did not emerge), which are commonly reported in the literature [
5,
16]. We also did not assess some key issues that may influence barriers to treatment, such as conditions of release or probation or parole requirements (e.g., having more probation appointments interfering with ability to attend treatment) or provide external incentives to engage in treatment (e.g., SUD treatment is a condition of probation and compliance is associated with remaining out of jail or prison). Finally, participants were not given prompts or further opportunity to expand on potential barriers to treatment, as the BTI is a self-report measure rather than an interview.
Authors’ contributions
MDO was the PI for the parent study and was involved in designing and conducting the pilot trial, and analyzing data for the current study and writing the manuscript. JAC assisted with qualitative analyses, interpreting results, and reviewing and writing the manuscript. TLS and CT assisted with interpreting results, and reviewing and writing the manuscript. ECW served as senior mentor for MDO and supervised qualitative analyses, as well as assisted with interpreting results, and reviewing and writing the manuscript. All authors read and approved the final manuscript.