A peer-support, group intervention to reduce substance use and criminality among persons with severe mental illness

This study compared the effectiveness of two interventions in reducing alcohol use, drug use, and criminal justice charges for persons with severe mental illnesses: first, a community-oriented group intervention with citizenship training and peer
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  PSYCHIATRIC SERVICES  ♦  ♦  July 2007 Vol.58 No.7 955   E stimates of co-occurring alco-hol and drug use disordersamong people with mental ill-ness range from 41% to 70% (1,2).Risk factors for this population in-clude poverty, homelessness, incar-ceration, and unemployment (3). Ap-proximately 800,000 persons withmental illnesses enter jails and pris-ons each year (4), and 70% have a co-occurring alcohol or substance usedisorder (5).Nationally, efforts are under way toimprove the lives and conditions of persons with co-occurring disordersand a criminal justice history (1,2,4).Programs developed to address theneeds of individuals with co-occurringdisorders include integrated treat-ment, dual recovery therapy, cogni-tive-behavioral therapy, case manage-ment, Alcoholics Anonymous (AA) 12-step integration, motivational inter- viewing, outreach and engagement,assertive community treatment,“staged” or “readiness” approaches,and community reinforcement (6,7).Most of these programs use a combi-nation of approaches from the alcoholand drug abuse treatment and mentalhealth fields. In Drake and colleagues’(6) review of 26 dual-disorder pro-grams, the outcomes for recovery, re-duced alcohol use, and hospitalizationsimproved as the degree of program in-tegration increased.  A Peer-Support, Group Intervention to Reduce Substance Use and Criminality  Among Persons With Severe Mental Illness Michae   l Rowe, Ph.D.Chyrel   l Bel   lamy, Ph.D., M.S.W.Made   lon Ba   r    a   noski, Ph.D.Melissa Wiela   nd, Ph.D., M.P.H.Maria J. O’Connell, Ph.D.Patricia Benedict, B.A.Larry Davidson, Ph.D. Josephi   ne Bucha   na   n, B.A. Dave Sel   ls, Ph.D.   With the exception of Ms. Benedict, the authors are affiliated with the Department of Psy-chiatry, Yale School of Medicine. Ms. Benedict is with Columbus House, Inc., NewHaven, Connecticut. Send correspondence to Dr. Rowe at the Department of Psychiatry,Yale School of Medicine, 319 Peck St., Building 6W, Suite 1C, New Haven, CT 06513 (e- mail: Objective: This study compared the effectiveness of two interventions inreducing alcohol use, drug use, and criminal justice charges for persons with severe mental illnesses: first, a community-oriented group inter-vention with citizenship training and peer support that was combined with standard clinical treatment, including jail diversion services, andsecond, standard clinical treatment with jail diversion services alone.  Methods:  A total of 114 adults with serious mental illness participated ina 2×3 prospective longitudinal, randomized clinical trial with two levelsof intervention (group and peer support for the experimental conditionand standard services for the control) and three interviews (baseline, sixmonths, and 12 months). Self-report questionnaires assessed alcoholand drug use, and program databases assessed criminal justice contacts.The authors used a mixed-models analysis to assess alcohol and druguse, repeated-measures analysis of covariance to assess criminal justicecharges, and correlational analyses to assess the relation between inter-vention participation and outcome variables.  Results: The experimentalgroup showed significantly reduced alcohol use in comparison with thecontrol group. Further, results showed a significant group-by-time in-teraction, where alcohol use decreased over time in the experimentalgroup and increased in the control group. Drug use and criminal justicecharges decreased significantly across assessment periods in bothgroups. Conclusions: Of the outcomes, only decreased alcohol use wasattributable to the experimental intervention. Although this may be achance finding, peer- and community-oriented group support andlearning may facilitate decreased alcohol use over time. (  PsychiatricServices 58:955–961, 2007)  Less studied is the rate of criminalbehavior as an outcome of programparticipation. Essock and colleagues(7), comparing assertive community treatment and standard case manage-ment for individuals with co-occur-ring disorders, found similar reduc-tions in substance use for bothgroups but also found that partici-pants in both groups spent aboutthree times as many days incarcerat-ed as in the hospital. Calsyn and col-leagues (8) found similarly poor re-sults regarding criminality for home-less persons with co-occurring disor-ders who received assertive commu-nity treatment, integrated treatment,or standard treatment. In addition,participants with more serious sub-stance abuse (nonprescribed drugsand alcohol) problems reported moresubstance use offenses and weremore likely to be arrested.Jail diversion programs were devel-oped to reduce incarceration rates forpersons with psychiatric disorders who commit low-level offenses by di- verting them from the criminal jus-tice to the mental health system (9).Diversion is accomplished throughclinician negotiation (with clientagreement) with court personnel tohave charges dropped or stayed in ex-change for the client’s agreement topursue mental health treatment andthrough related approaches (10,11).Outcome studies show mixed results,some indicating a decrease in sub-stance use (12) and arrests (13,14),others showing similar results or nodifferences between those who wereand were not diverted (10), and oth-ers showing continued high rates of criminal recidivism after initially suc-cessful diversion (15).Our own approach to addressingthe substance use and criminality problems of persons with seriousmental illness is built on a theoreti-cal framework of “citizenship” de-rived from our research on outreachto persons with mental illness andhomelessness (16–18). We definecitizenship as a measure of thestrength of people’s connections tothe rights, responsibilities, roles,and resources available to peoplethrough public and social institu-tions and through the informal, “as-sociational” life of neighborhoodsand local communities (16,18). Thisframework draws on social sciencetheories of citizenship that empha-size civic participation as a measureof one’s involvement in society (19)and the need to create participationopportunities for members of mar-ginalized groups (20). It also drawson social capital theory, which em-phasizes the importance of socialnetworks in enhancing people’s par-ticipation in society (21,22). In addi-tion, the citizenship frameworkshares common ground with com-munity mental health approaches,such as self-efficacy for persons withmental illness (23,24), the strengths-based emphasis of psychiatric reha-bilitation (25,26), and social integra-tion, with emphasis on supportingclients’ access to housing, work,friends, and public and social activi-ties (27). Finally, the concept of cit-izenship as full participation in soci-ety has been applied recently in re-search regarding civic reintegrationof criminal offenders (28).Our approach also draws on theory and recent research suggesting thatpeer staff, with their personal knowl-edge of coping with psychiatric dis-abilities and attendant stigma in soci-ety, have a special ability to engageclients and support them in their ownrecovery (29–31). We used the citizenship framework with integrated peer support to devel-op an intervention for persons whosemental illnesses, alcohol and druguse, and criminality posed seriousbarriers for their community stability and participation. We hypothesizedthat those receiving the citizenshipintervention along with standard serv-ices, which included jail diversion, would show lower levels of alcoholuse, drug use, and criminal chargesover time in comparison with thosereceiving standard services with jaildiversion only. Participants in bothconditions received treatment fromthe same pool of clinicians. To evalu-ate these hypotheses we conductedan investigation from June 2002 toNovember 2003, which used a 2×3prospective longitudinal, randomizedclinical trial design with two levels of intervention, comprising the citizen-ship intervention plus standard serv-ices (experimental) and standard ser- vices only (control) across three eval-uation periods (baseline, six months,and 12 months). Assessment proce-dures included interviewing partici-pants concerning alcohol and druguse and reviewing public databaseson criminal charges. Methods  Participants and procedure  We received approval for this study from the institutional review boardof our university. Participant selec-tion criteria included adults with se- vere mental illness who had criminalcharges within the two years beforeenrollment in the study. We provid-ed information on the study throughpostings at a local social rehabilita-tion center and emergency shelter,in social service agency newsletters,and at an information table at a localmental health center. We enrolled114 participants, with an average ageof 39.8±8.8. Seventy-eight (68%) were men. Participants’ racial andethnic background included 66African Americans (58%), 35 Cau-casians (31%), three Native Ameri-cans (3%), nine who endorsed an“other” category (8%), and one per-son who elected not to identify anancestry (1%). Seventeen partici-pants endorsed Hispanic ethnicity (15%). All were receiving outpatienttreatment. After giving informedconsent, participants in both condi-tions completed interviews at base-line, six months, and 12 months andresponded to questionnaires that in-cluded considerations of alcohol anddrug use. Drug testing was not con-ducted as part of the study.Forty-one participants (36%) wererandomly assigned to the control, or“standard services” condition, whichincluded individual and group treat-ment with medication management,case management, and jail diversionservices in which clinicians assignedto the local criminal court worked with defendants, public prosecutors,and judges to divert defendants withmental illness to mental health treat-ment (10). Seventy-three participants(64%) were diverted to the experi-mental condition, which consisted of the standard services described aboveand the citizenship intervention (dis-cussed below). Randomization re- PSYCHIATRIC SERVICES  ♦  ♦  July 2007 Vol.58 No.7 956  flected a 2:3 control-intervention pro-cedure designed to maintain suffi-cient numbers in the group compo-nents of the intervention.A total of 111 participants (97%)had either a primary or secondary di-agnosis of psychiatric illness, 48(42%) had a primary or secondary drug use diagnosis, 35 (31%) had analcohol use diagnosis, and 80 partici-pants (70%) had co-occurring psychi-atric andsubstance or alcohol use di-agnoses. As detailed in Table 1, pro-portional diagnostic distributions wereequivalent across the two groups. APearson chi square test yielded no sys-tematic relationship between condi-tion and diagnosis. All participantshad a recent criminal history, withcharges ranging from violations tofelonies.  Program context   We conducted our research at an ur-ban public mental health center inConnecticut. The experimental inter- vention involved a group componentconsisting of classes with topics relat-ed to social participation and commu-nity integration (citizenship classes)followed by projects designed to fos-ter participants’ acquisition of valuedsocial roles (valued-roles projects), with wraparound peer mentor sup-port to participants. Both citizenshipclasses and valued-roles projects wereongoing components of this program.Participants in this study who hadprevious or ongoing involvement withthe program were invited to attend(repeat) classes as well, with their pri-or attendance added to the classcount under the current study. Partic-ipants received $10 stipends for at-tendance at each class and valued-roles project meeting.Intervention participants wereeach assigned a peer mentor. Of thesix peer mentors who worked with theproject during the study period, all were diagnosed as having a seriousmental illness and were in treatment.The three who worked most inten-sively with participants had co-occur-ring drug or alcohol use disorders,and two had a criminal justice history.Peer mentors with co-occurring dis-orders were required to have at leastone year of sobriety. All peer mentorscompleted an intensive training pro-gram concerning confidentiality, theclient engagement process, culturalcompetence, and the distinctive rolesof criminal justice and mental healthtreatment system personnel. Workingpart-time and meeting with clients anaverage of once weekly over the in-tervention period of four months, with less frequent and formal follow-up contact in some cases after the in-tervention, peer mentors supportedparticipants by helping them to iden-tify goals and set priorities for achiev-ing them, sharing their own perspec-tives and coping strategies as people who have “been there,” and advocat-ing for participants’ access to socialservices, employment, education, andhousing.Peer mentors encouraged partici-pants to maintain their sobriety by of-fering examples of their own strugglesand recovery work and providing so-cial support and friendship to them.Peer mentors appeared to combinethe functions of case manager withconsumer experience, role model,and “paid friend” in a distinctive way that facilitated relationships less for-mal than in case management butmore formal than in friendship. Withparticipants’ permission, mentors oc-casionally contacted clinicians to dis-cuss plans regarding participants’housing, criminal justice obligations,and other issues or in cases of psychi-atric crisis. The peer mentors, howev-er, did not take on monitoring func-tions regarding participants’ criminal justice obligations, and the interven-tion was not used as a stipulation formeeting those obligations.The primary objectives of the citi-zenship classes component of the in-tervention were to enhance partici-pants’ knowledge of available com-munity resources, their problem-solv-ing and other life skills for daily living,and their ability to establish socialnetworks based on mutual trust andshared interests. A project directorfacilitated twice-weekly two-hourclasses of six to ten participants overan eight-week period. The classes, which were held at a local church thathouses a community soup kitchen fre-quented by many of the participants,shared similarities with social rehabil-itation and social skills programs yetemphasized both group support andcommunity contacts via class presen-tations by community members.Class topics included negotiatingthe criminal justice system, vocation-al and educational programs, localhousing options, problem solving,public speaking, relationship build-ing, social integration, and advocacy and self-help groups, including AAand Narcotics Anonymous (NA).Participants developed classroomrules and norms and helped to shapethe content of the classes through re-quests for outside speakers or addi- PSYCHIATRIC SERVICES  ♦  ♦  July 2007 Vol.58 No.7 957   Table 1 Diagnoses of 114 participants with mental illness who received standard jail diversion services or standard services plus citizenship training and peer support Diversion and citizenship trainingDiversion onlyOverallDiagnosis a N%N%N%Primary Psychotic disorder263617424338Major mood disorder294017424640Alcohol use disorder461254Drug use disorder573787Other disorder912371210Secondary Psychotic disorder231233Major mood disorder682587Alcohol use disorder172313323026Drug use disorder273713324035Other disorder12165121714 a Ninety-eight participants (86%) had a secondary diagnosis.  tional topics. The citizenship classesfocused on full and valued participa-tion in community life and encour-aged participants’ interaction withoutside speakers. Of the 73 partici-pants, 33 attended 16 or more class-es, 21 attended one to 15 classes, and19 did not attend any classes. Over-all, the mean participant attendanceat citizenship classes—includingthose whose repeat participation ex-ceeded the 16 “standard” class count—was 10.6±8.5.A second eight-week valued-rolescomponent encouraged participantsto contribute to their communities by drawing on their life experiences andskills gained through the classes andcontacts with community presenters.Participants designed and participat-ed in education-focused projects,such as teaching police cadets aboutconsumers’ encounters with policeofficers and conducting a fundraiserfor a local nonprofit organization. Valued-roles projects gave partici-pants an opportunity to challengetheir underestimations of their capa-bilities and explore their interests in asupportive environment. Community members learned that participantscan fulfill valued roles in society. Al-though most valued-roles projects were completed successfully, partici-pants sometimes felt their communi-ty audience did not respond as favor-ably or as quickly as they hadhoped—not donating as generously to a food drive project as initially an-ticipated, for example. However, suchexperiences also became a source of group support and learning about thefrustrations of translating personalgrowth into positive action and ac-ceptance in the social world. Of the73 participants, 15 attended 16 ormore valued-roles meetings, 20 at-tended one to 15 meetings, and 38did not attend any meetings. Overall,the mean participant attendance at valued-roles meetings was 6.7±8.4.  Measures  Addiction Severity Index (ASI). TheASI is a structured interview forgauging the degree of potential treat-ment barriers across domains typical-ly affected by alcohol and drug usedisorders, including psychiatric andsocial considerations (32). For this in- vestigation, we used only the alcoholand drug use ASI subscales. The alco-hol use subscale concerns the fre-quency and severity of use in the past30 days, whereas the drug use sub-scale concerns the frequency andseverity of use of various substances(such as cocaine and heroin) duringthe same period. The ASI has beenrigorously assessed within similarclient populations and shown to beboth a reliable and valid way to assessalcohol use and drug use and theirconsequences (33). Criminal justice data.  We obtainedcriminal justice data from a nonconfi-dential state court docket manage-ment system containing informationon criminal charges, arraignment anddisposition dates, and dispositiontypes for all persons within the Con-necticut criminal justice system. Wecategorized criminal charges in termsof severity: felony, infraction, misde-meanor, or violation. For the purpos-es of this study we focused on crimi-nal charges, where each was weightedequally and counted only once re-gardless of charge type. Statistical analyses To assess our hypotheses concerningalcohol and drug use, we used gener-al linear mixed-models procedures, which were particularly appropriategiven the multiple missing data points within these data sets (34), a problemcommon to longitudinal research in- volving persons with co-occurring dis-orders and criminal justice back-grounds (35,36). Given completedata, we used a repeated-measuresanalysis of covariance (ANCOVA)toassess our criminal justice chargeshypothesis. We also used correlation-al analyses to assess the relation of in-tervention class attendance to out-come. For the alcohol and drug usedata, gathered via participant inter- view as noted above, the overall sam-ple showed 23% attrition from time1, with 20 participants missing thetime 2 (six-month) interview but re-turning for the time 3 (12-month) in-terview and 19 participants missingthe time 3 interview. Chi squareanalyses confirmed that attrition andmissing data were not systematic tocondition. Analyses controlled forbaseline variable levels, and withinmixed-models procedures concern-ing alcohol and drug use, each analy-sis controlled for both baseline alco-hol and drug use, given their docu-mented association (37). Results Controlling for baseline levels of alco-hol and drug use, mixed-modelsanalysis showed that those who re-ceived the citizenship interventionhad overall significantly lower levelsof alcohol use across six- and 12-month follow-up periods than thosein the control group (F=12.12, df=1and 227, p<.005,  η 2 =.05) (Table 2).Moreover, the analysis also yielded asignificant interaction, where inter- PSYCHIATRIC SERVICES  ♦  ♦  July 2007 Vol.58 No.7 958   Tabl    e 2 Results of Addiction Severity Index alcohol use subscale among intervention andcontrol groups by evaluation period a Alcohol useN% b MSDTime 1 (baseline)Intervention7364.14.18Control4136.09.13Total114100.12.17Time 2 (6 months)Intervention4160.10.18Control2740.10.13Total68100.10.16Time 3 (12 months)Intervention4058.07.13Control2942.11.16Total69100.09.14 a Possible subscale scores range from 0 to 1, with higher scores indicating greater problem severity. b Percentages are calculated relative to the respective time period totals.   vention participants showed decreas-ing levels of alcohol use across follow-up periods and control group partici-pants showed increasing levels of al-cohol use across the same follow-upperiods (F=3.90, df=2 and 227, p<.05,  η 2 =.03). Baseline ASI alcohol usecomposite scores were noted to be just slightly lower than that of norma-tive data concerning persons withboth criminal justice backgroundsand substance use disorders (38).Again controlling for baseline levelsof drug and alcohol use, mixed-mod-els analysis concerning drug use re- vealed a significant main effect fortime (F=4.17, df=2 and 227, p<.05,  η 2 =.04), where both groups showeddecreases in their nonalcohol druguse across assessment periods (Table3). Baseline ASI drug use compositescores were noted to approximatethose of persons with substance usedisorders receiving community-basedoutpatient services (38).Controlling for baseline levels of criminal justice charges, repeated-measures ANCOVA yielded a signifi-cant main effect for time (F=4.30,df=1 and 111, p<.05,  η 2 =.04), withboth groups decreasing in the num-ber of new criminal charges from zeroto six months and from six to 12months (Table 4).The intervention showed no maineffect for drug use or criminal justiceinvolvement.Partial correlational analyses of outcome variables at times 2 and 3 with attendance at intervention class-es controlling for baseline variablelevels within the experimental grouplargely showed no significant relation-ship between outcome and class at-tendance level, with one exception:criminal charges at time 2 were nega-tively correlated with attendance at valued-roles classes (Table 5). Discussion Our initial research suggests that thecitizenship intervention may be effec-tive in reducing alcohol use amongpersons with severe mental illnessand a criminal history. Findings didnot support our hypothesis that theintervention group would have signif-icantly less nonalcohol drug use andfewer criminal justice charges thanthe control group, where both groupsposted decreases over time. Findingssuggest the possibility that the inter- vention, combined with mental healthtreatment and jail diversion services,can contribute to supporting morestable community tenure for the tar-get population.The findings noted above couldcarry a particular significance for per-sons with co-occurring severe psychi- PSYCHIATRIC SERVICES  ♦  ♦  July 2007 Vol.58 No.7 959   Table 3 Results of Addiction Severity Index drug use among intervention and controlgroups by evaluation period a Drug useN% b MSDTime 1 (baseline)Intervention7364. 09. 09Control4136. 05. 06Total114100. 07. 08Time 2 (6 months)Intervention4160. 04. 06Control2740. 07. 09Total68100. 05. 07Time 3 (12 months)Intervention4058. 04. 05Control2942. 04. 07Total69100. 04. 06 a Possible subscale scores range from 0 to 1, with higher scores indicating greater problem severity. b Percentages are calculated relative to the respective time period totals.   Table 4  Means and standard deviations for criminal justice charges among interventionand control groups by evaluation period a Control (N=41)Intervention (N=73)Total (N=114)ChargeMSDMSDMSDMisdemeanor b 6 months before time 1.781.33.931.69.881.57Time 2.461.03.891.50.741.36Time b 6 months before time  b 6 months before time Violation b 6 months before time charges b 6 months before time 1 c 1.001.531.402.381.252.12Time 2.761.501.181.871.031.75Time a Possible scores reflect counts of criminal justice charges, with higher scores indicating more crim-inal charges. b Time 1, baseline; time 2, baseline to 6 months; time 3, six to 12 months c Represents baseline values
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks