Characteristics of Substance Use Screening at Intake in a Sample of U.S. Jails

Abstract

Background. Despite high rates of substance use among justice-involved populations, the use of substance screening tools in justice settings varies. Methods. Data are from the National Jail Health Care Study, which surveyed jails across the U.S. about their health care practices (n=371). Jails were asked to voluntarily submit their medical intake forms. A content analysis of intake forms (n=63) specific to questions about substance use was completed. Results. Seventy-three percent (73%) of intake forms used non-standardized questions to assess current substance use, and 27% did not ask any questions about substance use. Alcohol use was most assessed (52%), followed by tobacco (30%), and marijuana (22%). Less than 11% of jails asked about use of opioids and 40% of forms asked about withdrawal history. Conclusions. The lack of adequate substance use screening in jails hinders identification of substance use disorders, potential for withdrawal symptoms, and appropriate connection to treatment resources.

Nearly two-thirds of individuals in jails have a substance use disorder (SUD), and 54% of sentenced people in jails reported drug use in the month prior to their incarceration, compared with 14% of the general population.¹ After release from carceral facilities, the risk of overdose and death is increased.^2–4 Due to a disproportionate burden of physical and mental health comorbidities among justice-involved populations,⁵ jails can be important public health venues for health screenings for a variety of conditions, including sexually transmitted diseases, HIV, hepatitis C virus, SUDs, and suicidality.^6–8 Mass incarceration is a social determinant health with a multitude of adverse outcomes, and jails are often the first touchpoint in the carceral system. Prisons house people convicted for stays of one year or longer, compared with jails where individuals are detained who are convicted for shorter sentences, or are simply awaiting charges, a trial date, or sentencing. This results in many more individuals having contact with jails than prisons. Prisons tend to have more robust substance use screening processes,⁹ while a failure to implement adequate processes in jails affects millions of medically underserved populations who come into contact with jail systems.^5,9

Substance use screening upon entry to jails is important for several reasons including (1) assessment for recent use of substances and potential for withdrawal syndromes, and (2) identification of people who could benefit from connection to SUD treatment services in jail and following release. In 2019, 1,200 people died in a jail—and the majority had not been convicted but rather were awaiting trail (pre-trial detention).¹⁰ Data are not provided on number of deaths or complications arising from complicated withdrawal from alcohol or drugs, but an archival analysis of data from 2000–2013 found 103 fatalities related to withdrawal.¹¹ Estimates suggest that in any given year approximately one million people in jails, approximately 9% of the yearly U.S. jail population, are at risk for untreated alcohol or opioid withdrawal.¹² High-risk withdrawal is a concern among people who use alcohol, sedatives, and opioids as well as among people with co-morbid health conditions^12,13 such as heart-related problems and psychiatric comorbidities that are disproportionately found among justice-involved populations.^14,15 Screening is the first step to assessment, improving connections to appropriate services during incarceration, and pre-release services aimed at reducing relapse risk and post-release mortality.^16–18 The risk of overdose post-release is acute,^2,19 but individuals who are entered onto and subsequently retained on medications for opioid use disorder, for example, are at lower risk of post-release mortality than those who are not.²⁰ Jails represent a special population, as many individuals are held for short periods of time. Retention on community medications such as methadone and buprenorphine for substance use disorders, as opposed to forced detox, can increase the likelihood of reengagement with treatment post-release,²¹ and reduce overdose risk.

Screening for substance use upon intake to jails and prison is recommended by the Substance Abuse and Mental Health Services Administration and the National Commission on Correctional Health Care.^8,22 There is a lack of data describing substance use screening practices in jails, but data suggests screening may be inadequate at identifying substance use needs; the most recent estimates indicate only 19% of individuals in jail with a substance use disorder participated in treatment during their jail incarceration¹ and only 28% of jails provided medically supervised detoxification.¹² The purpose of this study was to understand the prevalence and characteristics of substance use screening in a sample of U.S. jails.

Methods

Sample

Data from the National Survey of Health Care in U.S. Jails²³ were analyzed. The study survey closely models the Bureau of Justice Statistics (BJS) National Survey of Prison Health Care.²⁴ An expert advisory panel of members from the American Jail Association (AJA), jail administrators, and jail physician assisted in the development and adoption of the survey. A cross-sectional quantitative survey asked jails about their provision of a range of services, including medical intake, non-urgent care, medical services and staffing, and substance abuse treatment. Jails were told, "Medical intakes are the screening questions and activities conducted by health or custody staff after admission, before assigning someone to a housing unit. Medical intake does NOT include: screening you do at the front door, while the arresting officer is still present, to decide whether you will accept or divert the arrestee; a more in depth evaluation that is done by a nurse or practitioner, typically called the 14-day health assessment." Jails were then asked to voluntarily submit copies of their medical intake forms.

Jails were sampled using AJA membership data. The inclusion criteria for the study were (1) member of the AJA; (2) valid contact information; (3) included in the 2013 BJS Census of Jails;²⁴ and (4) were not located in states with combined jail/prison systems (Alaska, Connecticut, Delaware, Hawaii, Rhode Island, and Vermont). Jails operated by the federal government or Bureau of Indian Affairs were excluded so that only county and regionally operated jails were included. This sampling frame represents 45.7% of all county jails included in the BJS Census of Jails. A total of 1,126 jails met inclusion criteria and were sent (via email or postal mail) the survey in June 2018, a second wave of surveys was sent to non-responders in January 2019. A total of 376 jails completed the survey; 81 jails declined to participate, and 674 did not respond/had an unknown outcome. Five surveys were discarded due to errors for a total sample of 371. Seventeen percent (n=64) of responders also submitted copies of their medical intake forms. One medical intake form was not included in the analysis as it contained no codable information and appeared to be more of a manual/instruction guide. The current study analyzed the medical intake forms to understand the prevalence and characteristics of substance use screening questions.

Using American Association for Public Opinion Research criteria²⁵ to calculate the minimum response rate, the overall rate was 32.8% and the adjusted response rate (accounting for cases where absolutely no information ever reaches the researcher about the outcome of the contact) was 40.7%. The minimum response rate ranges from 28.8% among jails with fewer than 50 inmates to 45.9% among jails with 1,000 or more inmates. The adjusted response rate ranges from 34.4% to 51.7%, respectively. The AJA reported a previous response rate of 10% for emailed surveys to its membership (personal communication).

Analysis

The 63 medical intake forms were coded by the first author using NVivo software.²⁶The content analysis was specific to the questions about substance use asked directly to individuals incarcerated at the jail (i.e., inmates). We operationalized substance use questions as questions asked to a respondent about their drug and alcohol use. We descriptively coded the specific components of the substance use questions by collecting the specific substances screened and the type of question (e.g., dichotomous, open-ended). We also coded questions related to substance use, such as questions about route of administration, substance use history (e.g., withdrawal history, treatment), and medications for substance use disorder treatment. After the intake forms were coded, we examined the questions to determine if any items were from standardized screening instruments. We did so relying on our knowledge as substance use researchers, and any plausible questions were reviewed by the fourth author who is an expert in the field of drug-use screening. We only coded respondent-facing questions. That is, there were some questions where the person completing the form was to use their judgement to answer questions about substance use (e.g., are there signs of alcohol/drug withdrawal), however we did not code these questions as the recommended approach to screening for unhealthy substance use is through participant-reported questionnaires.²⁷

Results

The 63 medical intake forms are in the following 29 states: Arizona, California, Colorado, Florida, Georgia, Indiana, Iowa, Kansas, Kentucky, Louisiana, Michigan, Minnesota, Mississippi, Montana, Nevada, New Hampshire, New York, North Carolina, North Dakota, Ohio, Oklahoma, Oregon, South Dakota, Tennessee, Texas, Utah, Virginia, Washington, and Wisconsin. The jails differed in size: 14 jails (22%) had 500 or more inmates (large), 24 jails (38%) had 100–499 inmates (mid), and 25 (40%) jails had 50–99 inmates (small). In comparison, for the total sample of 371, 26% of those surveyed had 500 or more inmates (large), 37% had 100–499 inmates (mid), and 37% had fewer than 99 inmates (small). Overall, the subsample of jails sending medical intake forms was comparable to the larger survey population. The jails reported that the screening process was completed by a correctional officer (46%), a registered nurse (26%), a license practical nurse (21%) or other personnel (6%).

Table 1 contains descriptive results of the medical intake forms. Of the 63 analyzed forms, 73% (n=46) had questions regarding substance use. Almost 21% of forms asked

Table 1.

SUBSTANCE USE CHARACTERISTICS OF JAIL INTAKE FORMS (N=63)

about specific substances used, while 27% included dichotomous questions such as, Do you use drugs? or included drug addiction or drug dependency in a list of medical conditions individuals could endorse. One-quarter of the forms had open-ended questions about substance use. Often these open-ended questions began with, Do you use drugs? but followed-up with What kind? or What type? The remaining 27% (n=17) did not contain any respondent-facing questions about substance use behaviors. None of the questions about drug use were from standardized screeners or assessment tools.

Twenty-four percent of forms (n=15) asked about route of administration broadly or by substance, with 11% asking open-ended (e.g., What method?) or providing multiple consumption methods for selection (i.e., smoke, snort, ingest, inject), and 13% asked about injection drug use only. Three of the eight forms that asked about injection drug use also asked if the individual had ever shared syringes, (on forms "needles"). Fourteen percent of forms requested information on the individual's history of substance use treatment, ranging from simply asking if the individual had ever/recently received treatment to some including more specifics such as when or where. Only one form asked about overdose history.

The most prevalent question, on nearly 40% of forms, asked about withdrawal history. Forms asked if individuals had a history of withdrawal symptoms upon stopping alcohol and/or drugs, as well as specifics into the type of withdrawal symptoms or about a history of delirium tremens. While several forms asked about current medications an individual was taking in an open-ended manner, only two specifically asked about medications for opioid use disorder; both asked about including methadone and one Suboxone (specifically, by name). Both questions were in a section marked applicable for pregnant women only.

Table 2 contains an in-depth examination of the substances examined among the 46 forms that contained substance use questions. The most common substance screened was alcohol use, with 52% of forms that examined substance use including a question about alcohol. Tobacco was asked about in 30% of the forms, and marijuana in nearly 22%. The screening of other substances was more sporadic. In general, less than 11% of forms screened for opioid use including heroin and prescription opiates.

Discussion

The current research sought to understand the prevalence and characteristics of substance use screening among a national sample of U.S. jails. Overall, we found that the use of standardized substance use screening instruments was nonexistent in the medical intake forms obtained for this study, and there was little consistency across screening forms that were used by jails. Moreover, almost one-third of jails did not include any screening questions about substance use. This estimate is similar to a 2007 finding from Taxman et al;⁹ they found that 34.1% of jails did not screen for substance use. There are several implications to consider.

The fact that slightly more than one in four jails did not have any substance use screening questions is alarming. Justice-involved populations have a high burden of unhealthy substance use, with 50% of sentenced persons in jail reporting substance use in the 30 days prior to their incarceration.^1,27 At a minimum, screening for substance

Table 2.

SPECIFIC SUBSTANCES SCREENED IN THE 46 INTAKE FORMS THAT INCLUDED SUBSTANCE USE QUESTIONS^a

use at entry to jail allows for appropriate intervention for possible high-risk withdrawal. While individuals at jails who are not screened may still receive medical attention should withdrawal occur, screening upon entry provides an opportunity to assess and provide treatment to prevent withdrawal before it becomes a medical emergency. Screening should occur for high-risk withdrawal from alcohol, benzodiazepine, opioids, substance use during pregnancy, and substance use among individuals with underlying health conditions. For these populations, failure to provide medically supervised detoxification creates a heightened risk of death.^13,28

Moreover, identifying those at highest risk of overdose upon release is critical for making treatment referrals and targeting distribution of harm-reduction supplies (e.g., naloxone and fentanyl test strips) upon release to the community. Neglecting to capture prior overdose events (which was the case in all but one of the responding jails) at intake is a costly omission, given that the number of prior nonfatal overdose events has been shown to be the strongest predictor of a subsequent fatal overdose.²⁹

Among jails in this study that did include screening in their intake forms, there were no standardized instruments in use and the type of substance use information captured varied substantially. For example, at the time these data were collected, the U.S. was in the throes of an opioid overdose crisis, however only 11% of jails screened explicitly for opioid use. Adequate screening allows for further assessment, connections to medications for opioid use disorder either through jail or community programs, and improved reentry processes including connection to treatment, provision of pre-release naloxone, and other attempts to reduce post-release overdose mortality. However, it is important to note that many prisons and jails do not have services to which they can refer individuals; many jails do not have medications for opioid use disorder available and lack referral processes either due to organizational barriers or a lack of community treatment options.^30–33 It may be that jails exclude questions from screening as they feel they have no services to offer individuals or are reticent to identify the extent of unmet treatment needs in their populations.

Some jails included several questions that the study team judged as inadequate; the use of brief standardized screeners would take the same amount of time and effort but provide more detailed/improved information. In contrast with the forms examined in this study, standardized screeners include assessments of frequency with pre-determined timeframes and answer categories, include a consistent and broad range of substances, and assess substance-related problems. Research has suggested that screening need not be burdensome or lengthy to provide details that are helpful for assessment and planning purposes.³⁴ Examples of brief screening instruments that are recommended for screening for substance use in justice settings by the Substance Abuse and Mental Health Services Administration include the Alcohol, Smoking, Substance Involved Screening Test (ASSIST³⁵); the Texas Christian University Drug Screen 5 (TCUDS-V³⁶); and the Simple Screening Instrument (SSI³⁷). Screening via self-report is preferred over toxicology as the latter is invasive, has a restricted detection timeframe depending on the method, and may undermine assessment of needs most acutely related to high-risk withdrawal or connection to treatment services. Moreover, in the carceral setting attention must be given not only to the type of screeners used, but to who is administering the screening.³⁸ In this study, the majority of screeners were completed by correctional officers, a screening process that has been found to result in underreporting of substance use in jails.³⁸ As the majority of individuals enter jails without a conviction, individuals may be hesitant to disclose illegal drug use to an officer. Clinic staff or staff viewed as impartial should ideally complete intake, but this ideal may be particularly difficult to implement in smaller jails with limited staff, and more guidance is needed on the promise of using computer self-administered screeners that report aggregate risk scores or results to jail staff.³⁹ Notably, guidance about who should screen individuals, the strengths of computer self-administered screening, and the potential for bias are absent from the best practices recommended by the Substance Abuse and Mental Health Services Administration.

Note that the Taxman et al.⁹ study found only 7% of jails self-reported that they did not use any standardized screening or assessment measures. The discrepancy between that study and the current one could be due to several factors. The prior study relied on jail self-report using a survey, and it may be that jails were prone to social desirability bias in their reported use of the screeners. It is also possible that jails use standardized screening and assessment tools at other touchpoints in the facility. Most jails in that study (59%) reported "some" use of a standardized substance instrument and only 20% reported that at least 75% of their jail population receive a standardized tool, so the substance use tools may be administered to select groups (such as individuals who enter certain treatment programs) as opposed to the screening tools used during medical intake. While jails may have a high burden of individuals with short stays, they should not neglect this important opportunity to address health disparities.

The current research provided an understanding of the characteristics of substance use screening in a national sample of U.S. jails. The rigorous methodological approach to gather the sample is a strength of this study, however, there are limitations to consider. The response rate for the current study was low but comparable to other published rates for voluntary surveys administered by mail, email, and or phone. The study requested copies of medical intake forms, so jails with computerized programs that may be more complex may not have shared their forms. It is also possible that jails did not perceive substance use screening as part of the request for medical intake forms. While we asked for medical intake forms to capture the initial screening process at entry to jail, ultimately the interpretation of the request was up to the jail administrator completing the survey. Submitting medical intake forms was a voluntary component of the survey, and only a small percentage of jails chose to do so. Some jails reported that they were not permitted to share their forms, but it was unknown why this was so. Therefore, results should be considered within these limitations and indicate that the overall prevalence and quality of substance use screening in jails remains largely unknown.

Conclusion

The current research examined the substance use screener portions of medical intake forms of a national sample of U.S. jails. Findings from the current study indicate that there is substantial room for improvement in the collection of these data upon intake to jail. Specifically, jails should screen at minimum for recent substance use that is known to lead to high-risk withdrawal syndromes, using standardized screening questionnaires at intake. A lack of substance use screening can also hinder appropriate referral processes for SUD treatments in jails and in the community and have implications for post-release overdose risk. Given their impact on public health, jails may merit support to implement improved screening and intake processes.

Funding

This research was supported NIDA K01DA053435, R25DA037190, P30AI073961, P30AI42853, and P20GM125507. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Conflict of Interest/Disclosures

AB—None to disclose

KN—None to disclose.

DF—Received study medications from Alkermes and Indivior.

JM—None to disclose

CB—None to disclose