Is the collaborative chronic care model effective for patients with bipolar disorder and co-occurring conditions?

BACKGROUND: The effectiveness of bipolar collaborative chronic care models (B-CCMs) among those with co-occurring substance use, psychiatric, and/or medical conditions has not specifically been assessed. We assessed whether B-CCM effects are equivalent comparing those with and without co-occurring conditions. METHODS: We reanalyzed data from the VA Cooperative Study #430 (n=290), an 11-site randomized controlled trial of the B-CCM compared to usual care. Moderators included common co-occurring conditions observed in patients with bipolar disorder, including substance use disorders (SUD), anxiety, psychosis; medical comorbidities (total number), and cardiovascular disease-related conditions (CVD). Mixed-effects regression models were used to determine interactive effects between moderators and 3-year primary outcomes. RESULTS: Treatment effects were comparable for those with and without co-occurring substance use and psychiatric conditions, although possibly less effective in improving physical quality of life in those with CVD-related conditions (Beta=-6.11;p=0.04). LIMITATIONS: Limitations included multiple comparisons and underpowered analyses of moderator effects. CONCLUSIONS: B-CCM effects were comparable in patients with co-occurring conditions, indicating that the intervention may be generally applied. Specific attention to physical quality of life in those with CVD maybe warranted.

Factors associated with prospective long-term treatment adherence among individuals with bipolar disorder.

OBJECTIVE: Clinical characteristics, adverse effects of medication, and treatment attitudes have been associated with adherence in bipolar populations in cross-sectional studies. The aim of this secondary analysis from a larger study was to identify the association between baseline variables and average treatment adherence over a subsequent three-year period. METHODS: Veterans with bipolar disorder were evaluated on self-reported adherence status at baseline and every six months over a three-year period. The sample was dichotomized into two clinically relevant categories: those who were primarily adherent and those who were primarily nonadherent. Demographic and clinical variables were examined for the two groups of patients in relation to their average adherence over the three-year period. RESULTS: The study recruited a sample of 306 persons with severe bipolar disorder. The sample was predominantly male (278 men, or 91%), with a mean+/-SD age of 46.6+/-10.1 years. A total of 240 individuals (78%) were largely adherent to treatment, and 37 individuals (12%) were largely nonadherent to treatment. Nonadherent individuals were less likely to be on intensive somatotherapy regimens (p=.001); experienced more barriers to care, including lack of telephone access (p<.05) and life obligations and commitments (p<.05); and had more prior suicide attempts (p=.003). CONCLUSIONS: Nonadherent individuals with bipolar disorder received less intensive pharmacologic treatments, had more suicide attempts, and experienced more barriers to care than adherent individuals. Nonadherence may have system as well as patient components. Consideration of nonadherence as a function of both patient factors and system factors will enhance our ability to understand nonadherence and intervene more effectively.

Collaborative care for bipolar disorder: part I. Intervention and implementation in a randomized effectiveness trial.

Outcome for bipolar disorder remains suboptimal despite the availability of efficacious treatments. To improve treatment effectiveness in clinical practice, a Veterans Affairs study team created a care model conceptually similar to the lithium clinics of the 1970s but augmented by principles of more recent collaborative care models for chronic medical illnesses. This intervention consists of improving patients' self-management skills through psychoeducation; supporting providers' decision making through simplified practice guidelines; and enhancing access to care, continuity of care, and information flow through the use of a nurse care coordinator. In this article, which is part I of a two-part report, the authors summarize the conceptual background and development of the intervention, describe the design of a three-year, 11-site randomized effectiveness trial, and report data describing its successful implementation. Trial design emphasized aspects of effectiveness to support generalizability of the findings and eventual dissemination of the intervention. Part II (see companion article, this issue) reports clinical, functional, and overall cost outcomes of the trial.

Collaborative care for bipolar disorder: Part II. Impact on clinical outcome, function, and costs.

OBJECTIVE: The study addressed whether a collaborative model for chronic care, described in part I (this issue), improves outcome for bipolar disorder. METHODS: The intervention was designed to improve outcome by enhancing patient self-management skills with group psychoeducation; providing clinician decision support with simplified practice guidelines; and improving access to care, continuity of care, and information flow via nurse care coordinators. In an effectiveness design veterans with bipolar disorder at 11 Veterans Affairs hospitals were randomly assigned to three years of care in the intervention or continued usual care. Blinded clinical and functional measures were obtained every eight weeks. Intention-to-treat analysis (N=306) with mixed-effects models addressed the hypothesis that improvements would accrue over three years, consistent with social learning theory. RESULTS: The intervention significantly reduced weeks in affective episode, primarily mania. Broad-based improvements were demonstrated in social role function, mental quality of life, and treatment satisfaction. Reductions in mean manic and depressive symptoms were not significant. The intervention was cost-neutral while achieving a net reduction of 6.2 weeks in affective episode. CONCLUSIONS: Collaborative chronic care models can improve some long-term clinical outcomes for bipolar disorder. Functional and quality-of-life benefits also were demonstrated, with most benefits accruing in years 2 and 3.

Self-reported medication treatment adherence among veterans with bipolar disorder.

OBJECTIVE: Treatment nonadherence has been identified as an understudied, yet frequent, cause of relapse among patients with bipolar disorder. This study evaluated factors related to treatment adherence. METHODS: This study was a cross-sectional analysis of patient characteristics, features of the patient-provider relationship, and barriers to care as they related to self-reported treatment adherence among veterans with bipolar disorder. Data were collected before patients were randomly assigned to treatment in a multicenter trial. Participants were enrolled in the study from 1997 to 2000. RESULTS: This analysis found that individuals with bipolar disorder who were adherent to medication (N=113) differed from those who were not (N=71) on patient characteristics and variables related to the patient-provider relationship. Individuals with bipolar disorder who were not adherent to medication were more likely than those who were adherent to have a current substance use disorder. A past substance use disorder was not associated with treatment nonadherence. Adherence to medication was not predicted by symptoms, overall health status, functional level, or gender. Individuals who were adherent to medication took a greater number of different medications than those who were nonadherent. CONCLUSIONS: Treatment adherence is a multidimensional entity and includes not only patient characteristics but also factors related to patient-provider interactions. Therapeutic approaches that address or incorporate an individual's comorbid conditions and are able to optimize intensity of treatment may have the greatest effect on treatment adherence attitudes and behaviors.

Racial differences in outcome and treatment effect in congestive heart failure.

BACKGROUND: In congestive heart failure (CHF), it is unknown whether race affects mortality and whether the effect of treatments differs by race. METHODS: This study was a post hoc analysis of data from the DIG study that evaluated the effect of digoxin on morbidity and mortality in CHF. RESULTS: Investigators followed 897 black and 6660 white participants for a mean of 37 months. Compared with whites, blacks were younger (60 +/- 13 vs 65 +/- 11 years). Total mortality was 34.2% in blacks and 33.6% in whites; hospitalization for worsening CHF occurred in 39% of blacks and 28% of whites. Cox regressions with race as the only covariate showed no effect of race on risk for death (relative risk = 1.04, 95% CI 0.93-1.18, P = .49) but an increase in CHF hospitalization in blacks (relative risk = 1.52, 95% CI 1.35-1.70, P = .0001). Multivariate Cox regression showed no difference by race in risk for death or death/hospitalization for CHF and no difference in the effect of digoxin on either end point. CONCLUSION: Race is not an independent predictor of mortality in CHF. The effect of digoxin on morbidity and mortality in CHF does not differ in blacks and whites.

Impact of childhood abuse on the course of bipolar disorder: a replication study in U.S. veterans.

BACKGROUND: The association between early childhood abuse and the course of illness, including psychiatric comorbidities, in adults with bipolar disorder has not been examined in a predominantly male or veteran population. METHODS: As part of the VA Cooperative Study 430, "Reducing the Efficacy-Effectiveness Gap in Bipolar Disorder," 330 veterans (91% male) with bipolar I or II disorder who were enrolled in a 3-year prospective study were examined for baseline data obtained at study entry. Diagnoses were determined by the use of the SCID. A semistructured interview designed to elicit data about exposure to childhood physical, sexual, or combined abuse was conducted as part of baseline demographic and clinical information. Other reports from this data set have not addressed the issues of childhood adversity. RESULTS: Childhood abuse was reported by 48.3% of the subjects (47.3% of men). Any abuse (AA) was reported by 48.3%; sexual abuse without physical abuse (SA) was reported by 8%, physical abuse without sexual abuse (PA) by 20.7%, and both types of abuse (combined abuse, CA) by 18.7% of the male subjects. Female veterans reported more SA (27%) and less PA (6.7%). AA subjects were more likely to have current PTSD and lifetime diagnoses of panic disorder and alcohol use disorders. CA was associated with lower SF-36 Mental scores, higher likelihood of current PTSD and lifetime diagnoses of alcohol use disorders, as well as more lifetime episodes of major depression and higher likelihood of at least one suicide attempt. Younger age at study entry was associated with AA and PA. LIMITATIONS: Potential limitations include generalizability beyond the male, veteran population of patients with bipolar disorder and the methodology used to elicit abuse histories. CONCLUSIONS: Similar to studies of predominantly female nonveteran samples, this study extends the finding that a history of childhood abuse acts as a disease course modifier in male veterans with bipolar disorder. Clinicians should routinely seek information regarding abuse and be aware that these patients may be more difficult to treat than bipolar patients who have no abuse histories.

Perceived barriers to health care access in a treated population.

OBJECTIVE: Health care access may be a significant contributor to health outcome. However, few data exist on perception of barriers by patients in treatment, and attending a clinic visit does not mean that no barriers exist. Understanding barriers for treated populations is particularly important in optimizing care for high vulnerability populations, such as those with mental illness and the elderly. METHOD: A structured interview, demographic questionnaire, and SF-12 were administered to 324 veterans presenting for primary care or mental health appointments at a Veterans Affairs medical center. Principle components analysis was performed and relationships to vulnerability characteristics were identified. RESULTS: Most interview items showed modest mean levels but high variance. Responses were stable over three to six weeks. As hypothesized, perceived total barriers were greater in participants from several vulnerable populations: those receiving treatment for mental health problems, those with disabilities, and those with worse physical and mental function. Minority participants did not perceive greater barriers. An "inverted-U" relationship with age was found. Principal components analysis assigned 18 items across six clinically meaningful subscales. Participants with mental health treatment perceived greater barriers in three subscales including provider communication. Curvilinear relationships were again seen between subscales and age. CONCLUSIONS: Even individuals "in care" perceive barriers. Members of vulnerable populations, particularly those receiving mental health treatment, perceive greater barriers. Data support a multi-dimensional conceptualization of perceived barriers, and different subgroups experience different patterns of barriers.

Medical comorbidity and health-related quality of life in bipolar disorder across the adult age span.

BACKGROUND: Little is known about medical comorbidity or health-related quality of life (HRQOL) in bipolar disorder across the adult age span, especially in public sector patients. METHODS: We obtained cross-sectional demographic, clinical, and functional ratings for 330 veterans hospitalized for bipolar disorder with Mini-Mental State score > or = 27 and without active alcohol/substance intoxication or withdrawal, who had had at least 2 prior psychiatric admissions in the last 5 years. Structured medical record review identified current/lifetime comorbid medical conditions. SF-36 Physical (PCS) and Mental (MCS) Component Scores, measured physical and mental HRQOL. Univariate and multivariate analyses addressed main hypotheses that physical and mental function decrease with age with decrements due to increasing medical comorbidity. RESULTS: PCS decreased (worsened) with age; number of current comorbid medical diagnoses, but not age, explained the decline. Older individuals had higher (better) MCS, even without controlling for medical comorbidity. Multivariate analysis indicated association of MCS with age, current depressed/mixed episode, number of past-year depressive episodes, and current anxiety disorder, but not with medical comorbidity, number of past-year manic episodes, current substance disorder or lifetime comorbidities. LIMITATIONS: This cross-sectional design studied a predominantly male hospitalized sample who qualified for and consented to subsequent randomized treatment. CONCLUSIONS: Medical comorbidity is associated with lower (worse) physical HRQOL, independent of age. Surprisingly, younger rather than older subjects reported lower mental HRQOL. This appears due in part to more complex psychiatric presentations, and several mechanisms are discussed. Both results suggest that age-specific assessment and treatment may enhance HRQOL outcome.

Prevalence and distinct correlates of anxiety, substance, and combined comorbidity in a multi-site public sector sample with bipolar disorder.

BACKGROUND: Recent data indicate high prevalence of both anxiety and substance comorbidity in bipolar disorder. However, few studies have utilized public sector samples, and only one has attempted to separate contributions of each type of comorbidity. METHODS: 328 inpatient veterans with bipolar disorder across 11 sites were assessed using selected Structured Clinical Interview for DSM-IV modules and self-reports. RESULTS: Comorbidity was common (current: 57.3%; lifetime: 78.4%), with multiple current comorbidities in 29.8%. Substance comorbidity rate was comparable to rates typically reported in non-veteran inpatient samples (33.8% current, 72.3% lifetime). Selected anxiety comorbidity rates exceeded those in other inpatient samples and appeared more chronic than episodic/recurrent (38.3% current, 43.3% lifetime). 49% of PTSD was due to non-combat stressors. Major correlates of current substance comorbidity alone were younger age, worse marital status, and higher current employability. Correlates of current anxiety comorbidity alone were early age of onset, greater number of prior-year depressive episodes, higher rates of disability pension receipt, and lower self-reported mental and physical function. Combined comorbidity resembled anxiety comorbidity. LIMITATIONS: This is a cross-sectional analysis of acutely hospitalized veterans. CONCLUSIONS: Distinct patterns of substance and anxiety comorbidity are striking, and may be subserved by distinct neurobiologic mechanisms. The prevalence, chronicity and functional impact of anxiety disorders indicate the need for improved recognition and treatment of this other dual diagnosis group is warranted. Clinical and research interventions should recognize these divergent comorbidity patterns and provide individualized treatment built "from the patient out."

Pharmacists' and technicians' perceptions and attitudes toward dispensing buprenorphine/ naloxone to patients with opioid dependence.

OBJECTIVE: To assess the perceptions and attitudes of pharmacists and pharmacy technicians involved in an office-based opioid dependence treatment program using buprenorphine/naloxone. DESIGN: Cross-sectional attitudinal assessment. SETTING: Community, outpatient hospital, and clinic pharmacies. PARTICIPANTS: Pharmacists and technicians participating in a clinical trial of opioid dependence treatment using buprenorphine/naloxone. INTERVENTION: Written and telephone surveys followed by interviews with open-ended items. MAIN OUTCOME MEASURES: Attitudes and perceptions regarding opioid-dependent patients and use of buprenorphine/naloxone for treatment of opioid dependence. RESULTS: Pharmacies in seven states (New York, Virginia, Illinois, Florida, Texas, California, and Washington) participated in the clinical trial. A total of 40 pharmacists and pharmacy technicians responded to the initial written survey, representing 27 of the 32 pharmacies (84%). Follow-up interviews were obtained from one individual at 30 of those pharmacies (93.8%). Most pharmacy personnel (77.5%) involved with this study were not more concerned about theft or break-ins and would be willing to participate in opioid dependence treatment as the medication became available commercially (70%). The majority of respondents (85%) indicated that patients did not cause problems at their pharmacies. Compared with their experiences in administering other narcotic medications, most respondents did not express increased concern regarding prescription forgery (75%) or diversion (80%) of buprenorphine/naloxone. CONCLUSION: The majority of respondents expressed positive attitudes and perceptions regarding patients treated for opioid dependence with buprenorphine/naloxone.

Clinical, psychosocial, and treatment differences in minority patients with bipolar disorder.

OBJECTIVES: The clinical profile of minorities with bipolar disorder has been largely unexplored. We compared the clinical (e.g. psychiatric and substance use comorbidity), psychosocial, and treatment characteristics between white and minority patients with bipolar disorder (minorities were defined as black or other minority, which included Hispanic, Asian-American, or Native-Americans). METHODS: We collected demographic, diagnosis, and treatment information using the Structured Clinical Interview for DSM-IV (SCID) from 330 inpatients with a current major affective episode across 11 Veterans Affairs (VA) Medical Centers enrolled in the VA Cooperative Study (Reducing the Efficacy-Effectiveness Gap in Bipolar Disorder). RESULTS: Twenty-four percent (n=80) were minority; 9% (n=30) were women, 4% (n=20) were >or=65 years old; and the majority (87%, n=286) had bipolar type I. Minorities compared with whites were no more likely to have a current episode of psychosis (30% versus 37%, respectively; p=0.28). However, minorities were more likely than whites to have a cocaine use disorder (adjusted odd's ratio, OR=2.2; 95% CI: 1.4-3.5; p<0.01) or current alcohol abuse disorder (adjusted OR=1.8; 95% CI: 1.1-3.9;p<0.05). Further breakdown by race/ethnicity revealed that cocaine use disorder was most prevalent among blacks (n=14, 29%), compared with all other minorities (n=2, 6%) or whites (n=10, 4%; p<0.001). Other minorities compared with blacks or whites were more likely involuntarily committed during some part of their index hospitalization (adjusted OR=2.47; 95% CI: 1.1-5.7; p=0.04). CONCLUSIONS: Minorities with bipolar disorder may be a more vulnerable population because of higher rates of substance use disorder and higher rates of involuntary psychiatric commitment. Moreover, the specific profile of vulnerability may differ across minority groups.

Clinical nurse specialist care managers' time commitments in a disease-management program for bipolar disorder.

OBJECTIVES: As part of a cost-effectiveness analysis for Department of Veterans Affairs Cooperative Studies Program #430, 'Reducing the Efficacy-Effectiveness Gap in Bipolar Disorder,' we conducted a time and motion study to quantify the time psychiatric clinical nurse specialist (CNS) care managers spent providing care for patients. METHODS: Clinical nurse specialist care managers completed activity logs in which they recorded time spent implementing the Bipolar Disorders Program (BDP) during a 1-week period in spring, summer, fall and winter over a 1-year period when caseloads were at steady state. Mean service time was estimated by use of univariate analysis of means and by multivariable regression analysis. RESULTS: On average CNS care managers spent 40% of their clinical time in activities that typically are reimbursed (e.g. clinic visits) and spent the remaining 60% of their time in activities that are typically unreimbursed. Total clinic time increased as the number of visits per day increased; however, this increase got smaller with each additional visit per day. CONCLUSIONS: As with other chronic illness management programs, CNS care managers expend a substantial portion of their clinical effort for the BDP in activities that are typically unreimbursed. Their activities have a fixed component per day as well as a component that systematically varies with the number of visits per day. These findings should be considered when costing out and disseminating psychiatric and other medical chronic illness management programs.

A multivariate model for predicting mortality in patients with heart failure and systolic dysfunction.

BACKGROUND: Heart failure is a leading cause of morbidity and mortality, but there are no reliable models based on readily available clinical variables to predict outcomes in patients taking angiotensin-converting enzyme (ACE) inhibitors. METHODS: A multivariate statistical model to predict mortality was developed in a random sample (n = 4277 patients [67%]) of the 6422 patients enrolled in the Digitalis Investigation Group trial who had a depressed ejection fraction (

Hemorrhagic complications during long-term postoperative warfarin administration in patients undergoing lower extremity arterial bypass surgery.

Lower extremity bypass procedures restore function and prevent amputation in many patients with severe peripheral arterial occlusive disease. The regular postoperative use of aspirin offers the dual benefit of extending bypass patency and patient survival. Previous trials of adjunctive oral anticoagulant therapy with warfarin have infrequently combined warfarin with aspirin. We hypothesized that the addition of oral anticoagulant therapy would further enhance the benefits of aspirin but may increase the risk of clinically important bleeding. Eligible patients (N = 831) scheduled for elective lower extremity arterial bypass surgery were randomized to receive either warfarin plus aspirin (WA) (n = 418) or aspirin alone (n = 413). At monthly intervals, the warfarin dose was adjusted to a target international normalized ratio (INR) of 1.4 to 2.8; both groups received aspirin (325 mg/d). The end point of major hemorrhagic events, defined as intracranial hemorrhage or bleeding that required intervention, is reported, and INR values and compliance with warfarin therapy are presented. Major hemorrhagic events occurred more frequently in the WA group (35 in the WA group vs 15 in the aspirin group; p = .02) during a mean follow-up of 38 months. In the WA group, an intracranial hemorrhage occurred in six patients (two had an INR > 3.0), of whom four died; one subdural hemorrhage occurred in the aspirin group. Transfusions and interventions for bleeding were more frequent in the WA group, as were minor bleeding events. Of the 8,946 INR determinations, 58% were in the target range, whereas a higher value occurred in 10% and a lower value in 32%. Compliance with warfarin was maintained in 65% of the patients after the first year of observation. In patients with elective lower extremity bypass procedures, the postoperative adjunctive use of warfarin with aspirin increased the risk of major hemorrhagic events. Most of these events occurred when the INR was in the target range.

The role of the data coordinating center in the DIG trial.

The Digitalis Investigation Group (DIG) trial was a large simple trial (LST) begun in 1990 as a collaboration between the National Heart, Lung, and Blood Institute and the Department of Veterans Affairs (VA) Cooperative Studies Program (CSP). Its primary objective was to determine whether digitalis had beneficial, harmful, or no effect on total mortality in patients with congestive heart failure and an ejection fraction < or =0.45. The Perry Point VA CSP Coordinating Center served as the trial's data coordinating center (DCC). The DCC was involved in all phases of the study from planning and design, organization and start-up, and patient recruitment and follow-up through closeout, final analyses, and manuscript preparation. While DCC responsibilities for an LST are basically the same as for other multicenter randomized clinical trials, their size and the inclusion of many inexperienced research sites can add a complexity that the DCC must be prepared to handle from the beginning. This paper describes the role of the DCC in the DIG trial.

Overview of the DIG trial.

Congestive heart failure is a major public health problem in the United States, Canada, and other Western countries. The Digitalis Investigation Group (DIG) trial was a randomized, double-blind placebo-controlled trial that evaluated the effects of digoxin on all-cause mortality and on hospitalization for heart failure in patients with heart failure and left ventricular ejection fraction < or =0.45 with normal sinus rhythm. It was designed as a large simple trial. There were 6800 patients entered into the main study over a 31.5-month recruitment period at 302 participating centers in the United States and Canada. All patients were followed for a minimum of 28 months. In order for this study to succeed, many groups had to work together successfully. In this supplement, we present practical aspects of organizing and conducting a large simple trial such as DIG.

The role of the pharmacy coordinating center in the DIG trial.

Large simple trials (LSTs) emerged in response to the need for large sample sizes to answer important clinical questions in which treatments have a moderate effect on clinical endpoints. Between 1991 and 1996 the National Heart, Lung, and Blood Institute and the Department of Veterans Affairs (VA) Cooperative Studies Program conducted an LST entitled "Digitalis Investigation Group (DIG): Trial to Evaluate the Effect of Digitalis on Mortality in Heart Failure." The VA Cooperative Studies Program Clinical Research Pharmacy Coordinating Center served as the DIG pharmacy coordinating center (PCC). As a direct result of involvement in the DIG trial, the PCC identified the need for an increased emphasis on computerization and automated support of clinical trials, especially LSTs.

The role of the data coordinating center in the IRB review and approval process: the DIG trial experience.

Before any clinical trial can begin to recruit patients, participating clinical centers must obtain approval from their institutional review board (IRB). When studies are federally funded, such as by the U.S. Department of Health and Human Services (DHHS), centers must also have or obtain a federal compliance agreement from the Office of Human Research Protections (formerly the Office for Protection from Research Risks [OPRR]). The Digitalis Investigation Group trial was a large, international, double-blind, DHHS-funded randomized trial on the effect of digoxin on mortality in heart failure. Due to the anticipated number of centers (>200), the study's data coordinating center (DCC) was requested to assume additional responsibilities that included: (1) acting as a liaison between the OPRR and all study centers; (2) reviewing and correcting all assurance statements before submission to the OPRR; (3) reviewing and approving all centers' informed consent forms; and (4) helping the many research-inexperienced centers to establish IRBs or to locate an IRB in their region that would accept IRB responsibility for them. Although a heavy burden was placed on the DCC, the IRB and OPRR approval process was probably shortened by many weeks at those centers not already possessing a federal compliance agreement. This enabled the study to be completed on schedule and within budget.

Office-based treatment of opiate addiction with a sublingual-tablet formulation of buprenorphine and naloxone.

BACKGROUND: Office-based treatment of opiate addiction with a sublingual-tablet formulation of buprenorphine and naloxone has been proposed, but its efficacy and safety have not been well studied. METHODS: We conducted a multicenter, randomized, placebo-controlled trial involving 326 opiate-addicted persons who were assigned to office-based treatment with sublingual tablets consisting of buprenorphine (16 mg) in combination with naloxone (4 mg), buprenorphine alone (16 mg), or placebo given daily for four weeks. The primary outcome measures were the percentage of urine samples negative for opiates and the subjects' self-reported craving for opiates. Safety data were obtained on 461 opiate-addicted persons who participated in an open-label study of buprenorphine and naloxone (at daily doses of up to 24 mg and 6 mg, respectively) and another 11 persons who received this combination only during the trial. RESULTS: The double-blind trial was terminated early because buprenorphine and naloxone in combination and buprenorphine alone were found to have greater efficacy than placebo. The proportion of urine samples that were negative for opiates was greater in the combined-treatment and buprenorphine groups (17.8 percent and 20.7 percent, respectively) than in the placebo group (5.8 percent, P<0.001 for both comparisons); the active-treatment groups also reported less opiate craving (P<0.001 for both comparisons with placebo). Rates of adverse events were similar in the active-treatment and placebo groups. During the open-label phase, the percentage of urine samples negative for opiates ranged from 35.2 percent to 67.4 percent. Results from the open-label follow-up study indicated that the combined treatment was safe and well tolerated. CONCLUSIONS: Buprenorphine and naloxone in combination and buprenorphine alone are safe and reduce the use of opiates and the craving for opiates among opiate-addicted persons who receive these medications in an office-based setting.