Treatment Compliance in Patients With Co-Occurring Mental Illness and Substance Abuse

January 1, 2006
Ivn D. Montoya, MD, MPH
Volume 25, Issue 1

Treatment compliance is a crucial determinant of the outcome of any disease. Poor treatment compliance can worsen the prognosis and significantly increase health care costs. Effective methods to improve treatment compliance for individuals with comorbid mental illness and SUDs will translate in better outcome for the patients and significant health care cost savings.

Compliance is a crucial determinant of the treatment outcome of any medicalcondition. Poor treatment compliance may affect the therapeutic alliance;create skepticism in both therapist and patient; create resistance; worsen thedisease or the prognosis; and increase health care costs (Osterbergand Blaschke, 2005). Unfortunately, poor treatmentcompliance is often associated with blame, and noncompliant patients aresometimes punished with involuntary administrative discharge from treatment.

There are multiple indicators of treatment compliance that can be measuredusing direct or indirect methods. Among the direct methods, investigators andclinicians have used actual attendance to therapy sessions, direct observationor video-recording of sessions, measurement of medication blood levels,surrogate markers of medication, or expected changes in laboratory values. Theindirect methods include self-reported compliance, pill counts, evidence orabsence of expected side effects, and electronic monitoring devices.Unfortunately, the direct methods are expensive, and the indirect ones can besubject to biases.

In psychiatry, treatment compliance may be affected by factors associatedwith the therapist's characteristics, the service, the nature of the treatmentand the patient's idiosyncrasies. The therapist may not adhere to therecommended treatment guidelines or the therapy manual. The services may affectcompliance if they are hard to access or have long wait times, long lapsesbetween appointments or complex administrative procedures. Treatments thatinvolve complex procedures, are hard to follow, have unpleasant side effects,take a while to produce the desired effect, and are either unavailable ordifficult to access may increase the chances of poor compliance. Thecharacteristics of the patient, such as the presence of comorbidmental illness and substance use disorders (SUDs),can greatly affect treatment compliance.


The interest in psychiatric comorbidity increasedwith the publication of results from the Epidemiologic CatchmentArea Study (Regier et al., 1990) and the National Comorbidity Study (Kessler et al., 1994). More recently,Kessler et al. (2005) showed that the relative magnitude of associations ofhaving at least one substance use disorder in the past 12 months wassignificant for all but two mental conditions. The prevalence of mental healthservice use in the past year was only 41.1% and 38.1% for individuals who hadany mental disorder or any substance use disorder in that time period,respectively (Wang et al., 2005).

Unfortunately, little is known about the proportion of individuals in thegeneral population who use mental health services and actually adhere to theirtreatment plan. A survey of psychiatrists showed that 40% of their patientswith SUDs had treatment compliance problems (Herbeck et al., 2005). Both clinical and nonclinical factors appeared to be associated withtreatment compliance problems. Among the clinical factors, patients with lowtreatment compliance were more likely to have personality disorders, lowerglobal assessment of functioning scores and medication side effects than thosewithout treatment compliance problems (Herbeck etal., 2005).

It has been reported that the rates of completion of clinical trials forchronic medical conditions are only between 43% and 78% (Osterbergand Blaschke, 2005). It is likely that for those withpsychiatric disorders, particularly with comorbid SUDs, this percentage may be even lower. According to ameta-analysis of medication compliance, the mean compliance rate for patientswith physical disorders was 76%, whereas the ratio for patients takingantidepressants was 65% and 58% for antipsychotics(Cramer and Rosenheck, 1998). It has been estimatedthat medication noncompliance accounts for about 40% of re-hospitalizations ofpatients with schizophrenia (Weiden and Olfson, 1995).

The concurrence of mental illness and SUD seems to have a negativesynergistic effect. It has been suggested that comorbidpsychiatric disorders can further increase the risk of relapse and can haveimportant implications for predicting treatment outcomes (Compton et al.,2003). The rates of treatment compliance among patients with SUD vary greatly,depending on the type of SUD and treatment, severity of the disorder, degree ofpsychosocial support, and the presence and severity of psychiatric comorbidity. Patients with only alcohol use disorders havesignificantly higher treatment retention rates (42%) than those with drug usedisorders (20%) or combined alcohol and drug use disorders (26%) (McCaul et al., 2001).Furthermore, clinical trials of treatments for cocaine and other stimulant usedisorders have difficulty retaining participants. In contrast, clinical trialsof opioid agonist medications have better compliancerates (De Castro and Sabate, 2003).

Some psychiatric comorbidity factors that canaffect SUD treatment include the type of psychiatric disorder, severity of the comorbid psychiatric condition, early onset of illness andlevel of cognitive impairment (Broome et al., 1999; Rowe et al., 2004). Theclinical features of some types of psychiatric disorders may become riskfactors for poor treatment compliance. For example, among patients withschizophrenia, the lack of awareness of the illness, paranoid ideation,persecutory delusions and lack of initiative can worsen treatment compliance(Owen et al., 1996).

One of the psychiatric aspects that can greatly affect SUD treatmentcompliance is the patient's level of neurocognitivefunctioning (Aharonovich et al., 2003; Fals-Stewart and Schafer, 1992a). Attention, mentalreasoning and spatial processing are some of the cognitive domains that aresignificantly affected among noncompliant patients (Aharonovichet al., 2003).

More severe comorbidity has been associated withpoorer therapeutic alliance, and treatment alliance can predict patientretention (Barber et al., 2000). Patients with moderate and severe psychiatricproblems are not likely to remain in treatment unless they develop a strongtherapeutic relationship with their therapist (Petryet al., 2001).

Psychiatric Disorders

Studies have documented that treatment compliance can be affected by theconcurrent presence of substance abuse (Keck et al., 1997; Lambert et al.,2005; Olfson et al., 2000). In patients withschizophrenia, medication noncompliance has been significantly associated withsubstance abuse, this in turn with a greater symptom severity than other groups(Olfson et al., 2000; Owen et al., 1996).

For patients with bipolar disorder (BD), noncompliance was significantlyassociated with the presence of a comorbid SUD, andthe most common reported reason was denial of need for treatment. In a study byWeiss et al. (1998), compliance was significantly associated with treatmentwith divalproex (Depakote),compared to treatment with combined lithium (Eskalith,Lithobid). The study also showed that patients withBD and SUD who were prescribed benzodiazepines, neurolepticsand tricyclic antidepressants tended to take moremedication than prescribed.

With regard to comorbid personality disorders and SUDs, results are not conclusive. However, borderline andantisocial personality disorders predict lower treatment retention rates(Marlowe et al., 1997). In addition, the concurrent presence of Axis IIdisorders on top of the Axis I disorders and SUD seems to worsen treatmentcompliance even more. A study showed that inpatients with triple comorbidity (Axis I and Axis II disorders plus SUD) wereless likely to be compliant with the treatment plan than those without triple comorbidity (Ross et al., 2003).

Adolescent Compliance

Adolescents with comorbid mental illness andsubstance abuse are a unique clinical population because of the high risk thatthe problems will continue or worsen during adulthood. A retrospective recordreview of one year of admissions to a residential adolescent substance abusetreatment program showed that patients with attention-deficit/hyperactivitydisorder and those with conduct disorder had the lowest treatment compliance(Wise et al., 2001).

In another study, investigators showed that adolescents with SUD who have comorbid affective and adjustment disorders have bettercompliance, whereas patients with conduct disorders have poorer compliance.Furthermore, patients who received psychotropic medications have bettertreatment compliance (Kaminer et al., 1992).


Successful interventions to improve treatment compliance can belabor-intensive but ultimately cost-effective (Haynes et al., 2002). Theyshould address factors such as the characteristics of the therapist, theservice, the nature of the treatment and the patient.

Interventions to enhance treatment compliance have been categorized into theaffective, behavioral and cognitive (ABC) domains (Schaffer and Yoon, 2001).They require an understanding of the therapeutic regimen; counseling about theimportance of adherence; organizing medication-taking; rewarding andrecognizing the patient's efforts to follow the regimen; and enlisting socialsupport from family and friends (Haynes et al., 2002)

Some of the most promising strategies for improving treatment compliance arebased on cognitive/motivational interviewing and behavioral techniques such asreinforcement (Carroll et al., 2005; Zygmunt et al.,2002). A review of the literature of measures to enhance treatment adherenceamong patients with BD showed that cognitive-behavioral therapy, interpersonalgroup therapy, group sessions for partners and education about the illness areeffective in improving treatment adherence (Owen et al., 1996; Sajatovic et al., 2004).

Contingency management interventions that reinforce treatment compliancehave been shown to improve SUD treatment outcome. These interventions canreinforce one or multiple aspects of the treatment plan such as therapeuticactivities, counseling, attendance at AlcoholicAnonymous meetings or simply compliance with prescribed medications (Petry et al., 2001). Good attendance to individualstandardized interpersonal cognitive psychotherapy has been positivelycorrelated with objective measures of treatment outcome (e.g., urine drugtesting) in patients with comorbid cocaine and heroindependence (Montoya et al., 2005). Given that poor compliance is likely torecur, booster interventions are needed to reinforce and consolidate gains (Zygmunt et al., 2002).

Patients with comorbid mental illness and SUD canbenefit from the simultaneous treatment of both disorders. A study of substanceabusers dually diagnosed with obsessive-compulsive disorder showed that thosewho received a combined intervention that addressed their obsessive-compulsivesymptoms and substance abuse stayed longer in treatment than a second groupthat received only substance abuse treatment (Fals-Stewartand Schafer, 1992b).

Clinicians, services, patients and treatment regimens should work in tandemto meet the particular clinical needs of patients with comorbiddisorders. Patients need to be supported and not blamed or punished withadministrative discharges for poor treatment compliance. Innovative approachesto improve treatment compliance for patients with comorbidmental illness and SUD can be cost-effective and make a significant publichealth contribution.





Aharonovich E, Nunes E, Hasin D (2003), Cognitive impairment, retention andabstinence among cocaine abusers in cognitive-behavioral treatment. DrugAlcohol Depend 71(2):207-211.


Barber JP, Connolly MB,Crits-Christoph P et al. (2000), Alliance predicts patients' outcome beyondin-treatment change in symptoms. J Consult Clin Psychol 68(6):1027-1032.


Broome KM, Flynn PM, Simpson DD (1999), Psychiatric comorbiditymeasures as predictors of retention in drug abuse treatment programs. Health Serv Res 34(3):791-806.


Carroll KM, Ball SA, Nich C et al. (2005),Motivational interviewing to improve treatment engagement and outcome inindividuals seeking treatment for substance abuse: a multi-site effectivenessstudy. Drug Alcohol Depend. Available online Sept. 28.


Compton WM 3rd, Cottler LB, Jacobs JL et al.(2003), The role of psychiatric disorders inpredicting drug dependence treatment outcomes. Am JPsychiatry 160(5):890-895.


Cramer JA, Rosenheck R (1998),Compliance with medication regimens for mental and physical disorders. Psychiatr Serv 49(2):196-201 [seecomments].


De Castro S, Sabate E (2003), Adherence to heroindependence therapies and human immunodeficiency virus/acquired immunodeficiencysyndrome infection rates among drug abusers. ClinInfect Dis 37(suppl5):S464-S467.


Fals-Stewart W, Schafer J (1992a), The relationship between length of stay in drug-free therapeuticcommunities and neurocognitive functioning. J Clin Psychol 48(4):539-543.


Fals-Stewart W, Schafer J (1992b), The treatment of substance abusers diagnosed withobsessive-compulsive disorder: an outcome study. J SubstAbuse Treat 9(4):365-370.


Haynes RB, McDonald H, Garg AX, Montague P (2002),Interventions for helping patients to follow prescriptions for medications.Cochrane Database Syst Rev 2002(2):CD000011 [update].


Herbeck DM, Fitek DJ, Svikis DS et al. (2005), Treatment compliance in patientswith comorbid psychiatric and substance usedisorders. Am J Addict 14(5):195-207.


Kaminer Y, Tarter RE, BuksteinOG, Kabene M (1992), Comparison between treatmentcompleters and noncompleters among dually diagnosedsubstance-abusing adolescents. J Am Acad Child Adolesc Psychiatry 31(6):1046-1049.


Keck PE Jr, McElroy SL, StrakowskiSM et al. (1997), Compliance with maintenance treatment in bipolar disorder. Psychopharmacol Bull 33(1):87-91.


Kessler RC, Chiu WT, Demler O et al. (2005),Prevalence, severity, and comorbidity of 12-monthDSM-IV disorders in the National Comorbidity SurveyReplication. [Published erratum in Arch Gen Psychiatry62(7):709.] Arch Gen Psychiatry 62(6):617-627 [see comment].


Kessler RC, McGonagle KA, Zhao S et al. (1994),Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in theUnited States. Results from the National ComorbiditySurvey. Arch Gen Psychiatry 51(1):8-19.


Lambert M, Conus P, LubmanD et al. (2005), The impact of substance use disorderson clinical outcome in 643 patients with first-episode psychosis. Acta Psychiatr Scand112(2):141-148.


Marlowe DB, Kirby KC, Festinger DS (1997), Impactof comorbid personality disorders and personalitydisorder symptoms on outcomes of behavioral treatment for cocaine dependence. JNerv Ment Dis 185(6):483-490.


McCaul ME, Svikis DS, Moore RD (2001),Predictors of outpatient treatment retention: patient versus substance usecharacteristics. Drug Alcohol Depend 62(1):9-17.


Montoya ID, Schroeder JR, Preston KL et al. (2005), Influence ofpsychotherapy attendance on buprenorphine treatmentoutcome. J Subst Abuse Treat 28(3):247-254.


Olfson M, Mechanic D, HansellS (2000), Predicting medication noncompliance after hospital discharge amongpatients with schizophrenia. Psychiatr Serv 51(2):216-222.


Osterberg L, Blaschke T(2005), Adherence to medication. N EnglJ Med 353(5):487-497 [see comments].


Owen RR, Fischer EP, Booth BM, Cuffel BJ (1996),Medication noncompliance and substance abuse among patients with schizophrenia.Psychiatr Serv47(8):853-858.


Petry NM, Tedford J,Martin B (2001), Reinforcing compliance with non-drug-related activities. J Subst Abuse Treat 20(1):33-44.


Regier DA, Farmer ME, Rae DS et al. (1990), Comorbidity of mental disorders with alcohol and other drugabuse. Results from the Epidemiologic CatchmentArea (ECA) Study. JAMA 264(19):2511-2518 [see comments].


Ross S, Dermatis H, LevounisP, Galanter M (2003), A comparison between duallydiagnosed inpatients with and without Axis II comorbidityand the relationship to treatment outcome. Am J DrugAlcohol Abuse 29(2):263-279.


Rowe CL, Liddle HA, Greenbaum PE,Henderson CE (2004), Impact of psychiatric comorbidityon treatment of adolescent drug abusers. J SubstAbuse Treat 26(2):129-140.


Sajatovic M, Davies M, HroudaDR (2004), Enhancement of treatment adherence among patients with bipolardisorder. Psychiatr Serv55(3):264-269.


Schaffer SD, Yoon SJ (2001), Evidence-based methods to enhance medicationadherence. Nurse Pract 26(12):44, 50, 52, 54.


Wang PS, Lane M, Olfson M et al. (2005),Twelve-month use of mental health services in the United States: results from theNational Comorbidity Survey Replication. Arch GenPsychiatry 62(6):629-640 [see comments].


Weiden PJ, Olfson M(1995), Cost of relapse in schizophrenia. SchizophrBull 21(3):419-429.


Weiss RD, Greenfield SF, NajavitsLM et al. (1998), Medication compliance among patients with bipolar disorderand substance use disorder. J Clin Psychiatry59(4):172-174.


Wise BK, Cuffe SP, Fischer T (2001), Dual diagnosisand successful participation of adolescents in substance abuse treatment. J Subst Abuse Treat 21(3):161-165.


Zygmunt A, Olfson M,Boyer CA, Mechanic D (2002), Interventions to improve medication adherence inschizophrenia. Am J Psychiatry 159(10):1653-1664 [seecomments].