Criminal problem-solving and civil dependency court policies regarding medications for opioid use disorder

Abstract

Background:

Criminal problem-solving courts and civil dependency courts often have participants with substance use disorder (SUD), including opioid use disorder (OUD). These courts refer participants to treatment and set treatment-related requirements for court participants to avoid incarceration or to regain custody of children. Medications for opioid use disorder (MOUD) are the most effective treatment for OUD but are underutilized by court system participants. Little is known about variation in court policies for different MOUDs. Also, more information is needed about types of policies for each MOUD, including whether participants may begin MOUD, continue previously begun MOUD, or complete the court program with MOUD.

Methods:

An online survey was distributed to criminal problem-solving and civil dependency judges in Florida in 2019 and 2020, yielding data from 58 judges (a 24% response rate). We used nonparametric statistics to test hypotheses with ordinal data. A Friedman’s test for related samples or Cochran’s Q was used to make within-group comparisons between policies and MOUDs.

Results:

We found considerable policy variation, with more permissive policies for naltrexone than buprenorphine or methadone, and more permissive policies for continuing MOUD than for initiating MOUD or completing a court program with MOUD. For each medication, less than one quarter of judges indicated their court always permits MOUD, with most indicating that MOUD is permitted sometimes or usually.

Conclusion:

Because respondents rarely chose “never” or “always” for any MOUD policy, most courts appear to be making MOUD decisions on a case-by-case basis. A clearer understanding of this decision-making process is needed. Some court participants may be required to discontinue MOUD before completing a court program, even if they were permitted to start or continue MOUD treatment. Discontinuation of MOUD without medical justification is contrary to the standard of care for individuals with OUD and increases their risk of overdose.

Introduction

People with substance use disorder (SUD), including opioid use disorder (OUD), are overrepresented in criminal problem-solving courts and civil dependency courts.^1–5 Criminal problem-solving courts, (e.g., adult drug courts) and civil dependency courts (e.g., early childhood courts), refer court participants to OUD treatment and set treatment-related requirements for participants to avoid incarceration or to regain custody of children.^1–5 Medications for opioid use disorder (MOUDs) are the gold standard treatment for OUD⁶ and have a strong evidence base for decreasing opioid misuse and overdose.^7–10 Nevertheless, fewer than 1 in 20 justice-referred individuals in specialty OUD treatment receive buprenorphine or methadone treatment,¹¹ which have the strongest evidence base among MOUDs.⁷

Several potential reasons exist for MOUD underutilization among court-involved individuals. One such reason may be lack of preexisting justice system collaborations with MOUD providers, which are critical for facilitating MOUD access in a timely and effective manner.^12,13 For example, one study found that, as compared to jails and prisons, drug courts were more likely to cite lack of qualified medical personnel as a barrier to MOUD provision.¹⁴ Perhaps for this reason, judges and other court team members in some courts have made treatment plan decisions (e.g., type of medication, duration of medication treatment) without input from individuals trained in medication prescribing.^13,15 The National Association of Drug Court Professionals (NADCP), the leading professional organization that provides training and standards for treatment courts (e.g., drug courts), recommends that courts partner with only one or two healthcare agencies;¹⁶ if these partnering healthcare agencies do not provide MOUD, then referred individuals might lack access to MOUD. Furthermore, while partnering healthcare agencies can refer court participants to outside MOUD providers, our prior work has found that many court team members feel their partnering agencies do not encourage buprenorphine or methadone treatment.¹⁷

Some court participants may lack the funds to pay for MOUD treatment,¹³ particularly if they are not eligible for Medicaid or if MOUD providers do not accept Medicaid reimbursement. One previous study found that drug courts are less likely than jails or prisons to fund MOUD.¹⁴ Additionally, court team members may harbor common misconceptions about MOUD, such as the idea that MOUD is no more effective than psychosocial interventions, or treats OUD by replacing an addiction to one type of opioid with another,^13,18,19 which could likewise deter MOUD referrals from the court system. Finally, some courts have policies explicitly unsupportive of MOUD (e.g., prohibitions on starting or continuing MOUD).^13,18,20 Such policies could limit access even if MOUD could be funded and providers were available in the court’s geographic area.

It is important to examine existing court policies related to MOUD not only because they can influence court participant MOUD access, but also because changes in court policies, if problematic, could be influenced by state and federal government actions (e.g., through prohibitions on funding to courts that disallow MOUD).^21,22 However, few studies have examined MOUD-related court policies, including across a range of different types of courts that mandate SUD treatment (e.g., adult drug courts, veterans courts, family dependency drug courts).^1–3 Furthermore, few studies have compared court policies across all three MOUDs (i.e., methadone, buprenorphine, naltrexone),²⁰ even though court team members may harbor more negative attitudes toward methadone and buprenorphine than toward naltrexone.^13,20,23,24 The largest study to date on court MOUD policies, published in 2013, found that half of adult drug courts in a national sample prohibited buprenorphine or methadone treatment but it did not examine policies for naltrexone treatment.¹⁸ Finally, limited research has compared different types of policies between courts for MOUD, including allowing participants to start MOUD, allowing participants to continue previously begun MOUD, or allowing participants to complete a court program while taking prescribed MOUD. Existing research suggests that courts permitting participants to begin MOUD do not necessarily allow completion of the court program (e.g., graduation) while utilizing MOUD,^13,20 which could potentially cause court participants to terminate MOUD earlier than is medically advisable.

As part of a larger and more comprehensive effort to examine beliefs, social norms and barriers related to MOUD treatment in the Florida court system, we used an online survey to ascertain the prevalence of a range of MOUD policies among a convenience sample of Florida judges in criminal problem-solving and civil dependency courts. We chose to focus on judges for two reasons: 1) they are likely to be aware of the MOUD policies in their courts, and 2) one court typically has only one judge. As a secondary aim, we sought to compare differences in policies by type of medication (i.e., methadone, buprenorphine and naltrexone) and to compare differences by type of MOUD policy (i.e., permitting continuation of MOUD, starting MOUD and completing a court program with MOUD). Based on prior research regarding court team members’ MOUD attitudes, we hypothesized that courts would have more stringent policies for methadone and buprenorphine than for naltrexone.^20,23,24 We also hypothesized that courts would be more likely to allow participants to continue a previously begun MOUD than to start treatment with a new MOUD, because a 2019 federal circuit case (not directly applicable to Florida) ruled that discontinuing buprenorphine in jails may violate the Americans with Disabilities Act^25,26 by depriving people of a treatment proven to be a medical necessity given their diagnosed condition.

Methods

Ethics

Institutional Review Board approval was obtained for this research from the University of Central Florida. An Explanation of Research was provided at the beginning of the online survey instrument.

Data collection

We created an online survey in Qualtrics²⁷ software asking court staff and judges about MOUD policies in the primary court in which they work. Questions examined the following three policies for each MOUD: 1) court participants may start the medication in the court program, 2) court participants may continue previously begun medication, and 3) court participants may complete the program/graduate with the medication. Each item was formatted as a four-point ordinal scale (“never” to “always”) with additional options for “I don’t know” and “I choose not to answer.” Respondents were also asked to indicate their court role (e.g., judge, case manager, etc.), court type (e.g., adult drug court, veterans court, family dependency drug court) and the urbanity/rurality of their court. Because oral naltrexone has limited efficacy for OUD due to low adherence and retention^28,29 and is, therefore, very rarely used to treat OUD, we restricted our examination of naltrexone policies to the extended release, depot injection formulation (i.e., XR-naltrexone, hereinafter “naltrexone”). Given the relative novelty and limited implementation of injectable and implantable formulations of buprenorphine, we restricted our examination of buprenorphine policies to the oral/sublingual formulations (hereinafter “buprenorphine;” See Supplementary Appendix A for relevant questions from the survey instrument).

After piloting the survey with a small sample of court staff using a cognitive interviewing process,³⁰ the survey was distributed to all criminal problem-solving and civil dependency judges and court staff in Florida twice: once in Summer 2019 and once in Summer 2020. Once an individual’s email address was entered into Qualtrics, the software sent a reminder email each week for one month.²⁷ This distribution sample included 261 judges using contact information obtained from the state agency that oversees all courts from 2019. The state agency was unable to provide an updated contact list in 2020 but sent a recruitment message on the research team’s behalf that year, stating that judges/court staff who had not received a survey link from the research team should contact the principal investigator indicating their desire to participate. A few court staff/judges who were not included in the agency’s contact information were also provided the survey link after they contacted the research team wishing to participate. For this analysis, our final sample was limited to judges, because each court typically only has one judge, judges are expected to be knowledgeable about court policies, and the survey did not provide a means to link responses from court staff working within the same courtroom as the judge. Because our recruitment method provided the means for a judge to complete the survey in both 2019 and 2020, our analysis excluded judges who took the survey in both years.

Data analysis

The descriptive statistics for the respondent sample included court type and urbanity/rurality, and the frequency of specific MOUD policies. For purposes of analysis, courts were separated into two categories: criminal problem-solving courts (i.e., adult drug court, driving under the influence courts, veterans courts, mental health courts, juvenile drug courts) and civil dependency courts (i.e., family dependency drug courts, early childhood courts, and general dependency courts). Despite different target populations, each of these court types has large percentages of participants with SUD.^1–5

Because court policies were measured on an ordinal scale, nonparametric statistics were used to test our hypotheses. A Friedman’s test for related samples or Cochran’s Q was used to make within-group comparisons between MOUD policies and medications. A Mann Whitney U was used to make between-group comparisons between types of courts, and a Spearman’s rho was used to examine the correlation between urbanicity and MOUD policies. A Cochran’s Q for related samples was run to compare to whether judges were familiar with MOUD policies based on medication and type of MOUD policy. A Mann Whitney U test was run to compare the difference between criminal problem-solving court MOUD policies and civil dependency court MOUD policies. “I don’t know” and “I choose not to answer” responses were excluded from the analyses except for when we compared the frequency of “I don’t know” responses to ordinal ratings.

Results

Participant characteristics

After limiting the sample to judges and excluding judges who finished the survey in both years, respondents consisted of 58 judges. A total of 243 judges received the survey invitations, and 58 fully completed the survey, a response rate of approximately 24%. No random checking was completed to examine the number of judges still in positions in 2020 from the 2019 contact list. Of these respondents, 48.3% (n = 28) came from civil courts, 44.8% (n = 26) from criminal courts, and 6.9% (n = 4) were in the “other” category. Respondents in the civil court category are broken down into the following subtypes: general dependency courts (39.66%), early childhood courts (1.72%), family dependency courts (3.44%), and other types of civil dependency courts (3.44%). Criminal problem-solving courts are comprised of the following subtypes: adult drug courts (24.14%), veteran’s courts (5.17%), juvenile drug courts (1.72%), DUI courts (1.72%), and other types of criminal drug courts (12.07%). More than half (57%) of respondents were female, 55.2% indicated their court was in an urban or mostly urban area, and 44.8% indicated their court was in a rural or mostly rural area.

Descriptive statistics of policies

Induction into MOUD-based treatment while in the court program

For methadone, 70% allowed court participants to begin treatment (14% always, 22% usually, 34% sometimes), 2% never allowed participants to begin treatment, 21% did not know what their policy was, and 7% chose not to answer. For buprenorphine, 74% allowed participants to begin treatment (17% always, 16% usually, 41% sometimes), 2% never allowed participants to begin treatment, 21% did not know, and 3% chose not to answer. For naltrexone, 71% allowed participants to begin treatment (17% always, 21% usually, 33% sometimes), none reported not allowing participants to begin treatment, 26% did not know, and 3% chose not to answer.

Continuation of previously begun medication

For methadone, 63% allowed continuation of previously begun treatment (16% always, 33% usually, 14% sometimes), 3% never allowed continuation, 31% did not know, and 3% chose not to answer. For buprenorphine, 72% allowed continuation (19% always, 31% usually, 22% usually), 2% never allowed continuation, 22% did not know, and 3% chose not to answer. For naltrexone, 70% allowed continuation (22% always, 29% usually, 19% sometimes), none reported never allowing continuation, 26% did not know, and 3% chose not to answer.

Completion of the court program while enrolled in MOUD-based treatment

For methadone, 55% of judges indicated that their court allowed completion (12% always, 12% usually, 31% sometimes), 7% never allowed completion, 31% did not know, and 7% chose not to answer. For buprenorphine, 61% allowed completion (9% always, 19% usually, 33% sometimes), 7% never allowed completion, 29% did not know, and 3% chose not to answer. For naltrexone, 62% allowed completion (16% always, 17% usually, 29% sometimes), 5% never allowed completion, 29% did not know, and 3% chose not to answer.

Extreme responses

We examined the frequency of judges who selected “always” for each of the three policies for each medication and the frequency of judges who selected “never” for each of the three policies for each medication. For methadone, 5.2% of judges reported “always” in response to the three policies (i.e., allowing participants to begin, continue, or complete the court program with methadone), while 1.7% responded “never” to all three policies for methadone. For buprenorphine, 6.9% of judges reported “always” allowing all three policies, and 1.7% responded “never” to all three policies for buprenorphine. For naltrexone, 6.9% of judges reported “always” allowing all three policies, and no judges responded “never” to all three policies for naltrexone.

Differences between policies by type of medication

We did not find a significant difference in the type of MOUD that courts would allow court participants to begin MOUD by medication type, χ²(2) = 4.769, p = .092, N = 34.

The likelihood that courts allowed participants to complete the court program while using a MOUD did depend on the type of medication, χ²(2) = 7.448, p = .024, N = 33. Courts were more likely to permit participants to complete the court program if they used naltrexone (M_rank = 2.18) than either methadone (M_rank = 1.91) or buprenorphine (M_rank = 1.91). There was no difference in the policy between methadone and buprenorphine.

The likelihood that courts allowed participants to continue using MOUD if they began it prior to entering the court program also depended on the type of mediation, χ²(2) = 7.600, p = .022, N = 32. Courts were more likely to allow participants to continue using naltrexone (M_rank = 2.12) than methadone (M_rank = 1.89), p = .034. There was no difference in the policy between buprenorphine (M_rank = 1.98) and either methadone or naltrexone.

Differences between types of policies for each medication

There was a significant difference between court policies with respect to methadone, χ²(2) = 6.750, p = .034, N = 28. Courts were more likely to allow participants to continue methadone if they began treatment prior to entering the court program (M_rank = 2.27) than to begin methadone in the court program (M_rank = 1.95) or complete the court program while using methadone (M_rank = 1.79). However, there was no difference in the likelihood that courts would allow participants to begin using methadone and complete the court program while using methadone.

There was a significant difference between court policies with respect to buprenorphine, χ²(2) = 10.452, p = .005, N = 36. Courts were more likely to allow participants to continue to use buprenorphine if they began it prior to entering the court program (M_rank = 2.25) than to begin using buprenorphine (M_rank = 2.00) or complete the court program while using buprenorphine (M_rank = 1.75). However, there was no difference in the likelihood that courts would allow participants to begin using buprenorphine and complete the program while using buprenorphine.

There was a significant difference between court policies with respect to naltrexone, χ²(2) = 7.444, p = .024, N = 36. Courts were more likely to allow participants to continue to use naltrexone if they began it prior to entering the court program (M_rank = 2.25) than to begin using naltrexone (M_rank = 1.94) or to complete the court program while using naltrexone (M_rank = 1.81). However, there was no difference in the likelihood that courts would allow participants to begin using naltrexone and complete the program while using naltrexone.

Knowledge about MOUD policies

Because a large minority of judges responded “I don’t know” to court MOUD policy questions, we examined differences in the likelihood of a judge selecting “I don’t know” for a specific policy by medication type and type of policy. We found no significant differences between judges’ knowledge of MOUD policies based on either medication type or policy type (Table 1). Judges who indicated that they did not know the policies for one medication expressed similar unfamiliarity with other medications. Likewise, those who indicated that they did not know one type of MOUD policy (e.g., continuing MOUD) expressed similar unfamiliarity with other policies (e.g., beginning MOUD). Table 1 includes the Cochran W (χ²) values, p values, and sample sizes for each comparison.

Table 1.

Statistics for Cochran’s Q comparing judge’s knowledge of MOUD policies.

Comparison	c2	p	n
Differences between MOUDS for beginning treatment	0.182	.913	53
Differences between MOUDS for completing the court program/graduate while utilizing treatment	1.143	.565	53
Differences between MOUDS for continuing treatment if they began it prior to entering the court program	4.545	.103	55
Differences between court policies for methadone	3.500	.174	51
Differences between court policies for buprenorphine	3.429	.180	55
Differences between court policies for naltrexone	2.333	.311	55

Relationship between urbanicity and court policies

There was no relationship between urbanicity and policy for any of the three types of MOUD policies for the three medications. Table 2 includes the Spearman correlations, p values, and sample sizes for each policy type and medication.

Table 2.

Correlations between levels of urbanicity and policies.

Policy	rho	P	n
Begin methadone treatment	.161	.320	40
Begin buprenorphine treatment	.132	.405	42
Begin naltrexone treatment	.136	.397	41
Complete the court program/graduate while utilizing methadone treatment	.117	.502	35
Complete the court program/graduate while utilizing buprenorphine treatment	.078	.641	38
Complete the court program/graduate while utilizing naltrexone treatment	.062	.711	38
Continue methadone treatment if they began it prior to entering the court program	.133	.445	35
Continue buprenorphine treatment if they began it prior to entering the court program	.055	.735	40
Continue naltrexone treatment if they began it prior to entering the court program	.036	.827	40

Differences between MOUD policies in criminal problem-solving courts and civil dependency courts

There was no relationship between court type (i.e., criminal problem-solving court or civil dependency court) and policy for any of the three types of MOUD policies for the three medications. Table 3 includes the mean ranks for each group and significance tests.

Table 3.

Differences in court policies between criminal and civil courts.

Policy	Mean rank for criminal courts (n)	Mean rank for civil courts (n)	U	p
Begin methadone treatment	17.00 (18)	20.00 (18)	135.00	.406
Begin buprenorphine treatment	20.31 (21)	19.64 (18)	195.50	.856
Begin naltrexone treatment	22.02 (24)	16.77 (15)	228.50	.163
Complete the court program/graduate while utilizing methadone treatment	17.26 (19)	16.64 (14)	138.00	.872
Complete the court program/graduate while utilizing buprenorphine treatment	18.95 (20)	17.94 (16)	169.00	.789
Complete the court program/graduate while utilizing naltrexone treatment	19.87 (23)	17.57 (14)	181.00	.546
Continue methadone treatment if they began it prior to entering the court program	15.60 (15)	18.17 (18)	114.00	.464
Continue buprenorphine treatment if they began it prior to entering the court program	17.75 (20)	21.44 (18)	145.00	.317
Continue naltrexone treatment if they began it prior to entering the court program.	20.07 (23)	18.63 (15)	185.50	.701

Discussion

Our study is the first to quantitatively compare several types of court MOUD policies (i.e., starting MOUD, continuing MOUD, completing a court program with MOUD) for three types of MOUD (i.e., methadone, buprenorphine, naltrexone) in criminal problem-solving and civil dependency courts. Even though our sample was not representative, this preliminary work can inform the development of a nationwide survey regarding MOUD policies.

Our descriptive findings showed that MOUD policies varied across courts. Importantly, it was more common for judges to select “sometimes” and “usually” to a policy than “always” or “never,” which suggests there is discretion in the application of these policies. It is surprising and concerning that, even though naltrexone has no risks of misuse or diversion, less than 1/4 of respondents selected “always” for any of the court policies related to naltrexone. This may indicate a lack of knowledge about misuse or diversion risks among judges/court staff, which could be addressed through targeted education. Alternatively, the result could indicate a philosophical preference for “abstinence only” treatment among judges and court staff wherein court participants are only considered in recovery if they do not use treatment medications of any kind, regardless of the medications’ lack of diversion or misuse potential – a finding in one previous qualitative study.²⁰

Importantly, our study found significant differences between the same policy across different types of medication. As hypothesized, we found that courts had more permissive policies for naltrexone than for buprenorphine or methadone, because several previous studies have identified more positive attitudes toward naltrexone than toward buprenorphine and methadone among court team members.^20,23,24 Also, a previous study of Florida court staff and judges found they were more likely to receive training directly from manufacturers of XR-naltrexone as compared to manufacturers of buprenorphine or methadone, potentially contributing to a more favorable perception of naltrexone that manifests in more permissive policies.³¹

For all forms of MOUD, court policies in our study appear more favorable toward continuation of previously begun MOUD than permitting court participants to begin MOUD in the court program or to complete the court program while utilizing MOUD. This result may reflect judicial knowledge of a 2019 federal case, Smith v. Aroostook.^25,26 In that case, the federal first circuit court of appeals affirmed the district court’s opinion that a jail’s prohibition of continuing previously begun buprenorphine could constitute a violation of the Americans with Disabilities Act.^25,26 Even though the case applied to a jail rather than a court setting, and a federal first circuit decision does not set precedent for Florida judges, respondents in our survey may feel the ruling’s rationale could nonetheless be applied to Florida’s justice system at some later date. Therefore, they may feel it is advisable to permit continuation of a previously begun MOUD. Further qualitative research into this rationale is necessary, however.

Because only 9–16% of respondents in our study said their court always permits court participants to complete the court program (e.g., graduate) while utilizing MOUD, it appears that some courts are requiring participants to stop MOUD as a condition of court program completion. In the criminal justice system, court program completion can result in clearing of the associated criminal charge and/or avoidance of incarceration.³ In the civil court system, court program completion is predictive of reunification between parents and children who were removed due to parental drug use.^4,32 Therefore, our study suggests that some court participants may be required to stop MOUD to avoid incarceration/criminal charges or reunite with their children, both of which are problematic from ethical and public health standpoints. Early or medically arbitrary discontinuation of MOUD-based treatment is contrary to the standard of care for individuals with OUD³³ and could result in adverse outcomes for participants, including relapse and overdose.^34–37 These results suggest that state or federal policies designed to increase MOUD utilization among court participants cannot simply ban prohibitions on starting MOUD; they must also address the relationship between participants’ MOUD utilization and court program completion.

Only 5% of respondents selected “always” for all methadone-related policies (i.e., starting treatment, continuing treatment, completing the court program with the treatment) and only 7% selected “always” for all buprenorphine or XR naltrexone-related policies. This suggests that court participants with OUD cannot always utilize MOUD per court policy. However, no judges responded “never” to all three policies for XR naltrexone, and only 1.7% responded “never” to all three policies for buprenorphine and methadone. Although these “never” answers are not necessarily indicative of a court-wide MOUD ban, the National Association of Drug Court Professionals’ (NADCP) Best Practice standards prohibit court-wide bans on MOUDs and encourage decision-making on a case-by-case basis after obtaining input from qualified healthcare practitioners.^16,38

Our results, therefore suggest many court MOUD policies surveyed in our sample were in line with these NADCP best practices – that is, if decision-making in individual cases was based on input from qualified healthcare practitioners.^16,38 However, a previous study of Florida court staff and judges found that fewer than half believed the healthcare providers with whom they work encourage treatment with buprenorphine or methadone.¹⁷ Consequently, even if Florida courts are following NADCP best practices, attitudes of healthcare providers with whom they collaborate could, in practice, reduce court participants’ access to the most effective forms of MOUD, which are medicines with agonist properties. Significantly more research, especially qualitative work, is needed regarding court staff and judges’ relationships with healthcare providers and how these relationships affect the decision-making process for MOUD policy.

There are several examples of successful collaborations between community buprenorphine/methadone providers and criminal justice agencies other than courts (e.g., law enforcement, prisons, jails, community corrections^12,39–41). Nevertheless, some studies have found hostility toward buprenorphine or methadone in jails, prisons, and community corrections programs,^14,42,43 including preferences for naltrexone,⁴⁴ which can affect MOUD referrals and access.⁴² Because strong concerns about buprenorphine and methadone diversion/misuse have been found in courts and other justice institutions,^13,14,20 it is possible that preferential naltrexone policies are due to those concerns. While we are unaware of studies examining judges’ knowledge of buprenorphine/methadone diversion and misuse risks, it is possible that these risks are overestimated by judges. MOUD biases and policies from noncourt criminal justice settings could spill over into courts, because problem-solving and dependency court teams are interdisciplinary and include law enforcement officers, corrections officials, and other criminal justice personnel. Therefore, it is also possible that positive MOUD attitudes and policies in court programs could spill over into noncourt settings, potentially leading to greater MOUD access in jails, prisons, law enforcement agencies, and corrections programs.

Interestingly, between 21–33% of judges selected “I don’t know” when asked about MOUD policies for a specific medication. Because respondents could select “I choose not to answer,” it is unclear why judges would not know the policies regarding MOUD in their own court. It is possible that the judge had not yet had a court participant with OUD who was seeking to utilize MOUD, and thus had not been faced with a policy decision of this nature. It is also possible that the judge was new to the court or that other types of SUDs (e.g., stimulant use disorder) were more common than OUD among court participants. Because we did not ask respondents to identify their specific court, in cases where a judge did not know the policies, we could not triangulate them using the responses of other court staff, which is a limitation of this study. We likewise do not know what percentage of court participants in each court have OUD – information that could help us interpret the reason for almost one-quarter to one-third of judges selecting “I don’t know” for questions. Nevertheless, we believe that the large percentage of judges who selected this response option is, in and of itself, an interesting result requiring further research.

Our study has limitations, including a low response rate that yielded a nonrepresentative sample. Because the survey was voluntary, judges who opted in may be more likely than those who did not opt in to have policies in line with NADCP best practices, indicative of a social desirability bias in electing to participate or not. Additionally, the study included judges who took the survey in one of two separate years: 2019 or 2020. Therefore, it is possible that a judge who took the survey in 2019 had different policies in their court room in 2020, which were not captured in the survey. Finally, we do not know what percentage of court cases in the judge’s court involve SUD, including OUD.

Conclusion

Alternatives to conventional criminal and civil court proceedings involving people with SUD, including OUD, are becoming more common across the nation.³ Evidence indicates not all OUD treatment is equally effective,^7,10 and yet the most effective methods can be misunderstood or stigmatized in ways that limit their implementation and use.

Future research should use a nationally representative sample to understand the distribution of MOUD policies across a wide range of courts, including how MOUD knowledge and attitudes shape them. This study found variability in court MOUD policies and results suggest that many MOUD decisions are being made by court staff/judges on a case-by-case basis. A clearer understanding of what motivates court MOUD policy adoption and factors involved in case-by-case MOUD decision-making is needed. States should prohibit blanket bans on starting MOUD in courts. States should also require courts to permit court program completion while utilizing MOUD and should prohibit policies more favorable to naltrexone than methadone or buprenorphine.

Data availability statement

Deidentified survey data, with small cell sizes redacted, are available from the corresponding author by request.