Methods
The study was registered at ClinicalTrials.gov (Identifier: NCT00057135), approved by Human Subjects Committees at each of the 4 participating study sites, and involved human participants between November 18, 2002, and September 30, 2005. Study data and findings are reported in accordance with the Consolidated Standards of Reporting Trials.
Figure 1 outlines the recruitment of study participants. Using VA administrative data, we identified 4394 patients prescribed antipsychotic medications with ≥2 outpatient mental health visits in the prior 12 months. Research staff reviewed patients' chart notes, pharmacy data, and consulted with their psychiatrists to confirm that patients met other eligibility criteria, including having clinical diagnoses of schizophrenia, schizoaffective, or bipolar disorder; a treatment plan that included long-term antipsychotic treatment, antipsychotic medication possession ratios (MPRs) of <0.8 in the prior 12 months; and no clinical contraindications to study participation. Clinical rather than semi-structured research diagnoses were used as entry criteria, as clinical diagnoses are more likely to be used in any clinical implementation. For patients on more than one antipsychotic, the weighted average antipsychotic MPR during the prior 12 months had to be <0.8.
(Enlarge Image)
Figure 1.
Patient Enrollment.
Patients were excluded if they: no longer appeared to be in active VA care (eg, no visits in the prior 6 months and no visits scheduled or chart notes indicating a transfer of care), were receiving antipsychotic medications outside the VA system, or had a life expectancy of less than a year. Patients were also excluded if they were receiving "stronger interventions" to promote adherence (eg, the VA adaptation of Assertive Community Treatment, staff administration of medications, depot antipsychotics, or clozapine with its attendant close supervision). Living situation was not a specific inclusion/exclusion criterion, as long as paid staff did not assist patients in managing their medications.
Because of Institution Review Board requirements, study staff stopped reviewing charts as soon as any exclusion criterion was noted; 837 patients were identified as potentially eligible for participation. These patients were sent letters informing them about the study and were subsequently contacted by telephone. Patients who were interested in participation after a scripted telephone discussion were invited to an in-person meeting to complete an informed consent process; 154 patients attended this in-person meeting.
During the informed consent process, visual aids were used to explain the key elements of the study, and standardized queries were made to ensure that patients understood these elements. If patients had guardians, their guardians were informed about the study and consent forms mailed. Four patients were excluded at this time because they failed to understand key study elements or did not have a mailing address or telephone access. An additional 32 patients were excluded because their most recent MPR exceeded 0.8.
In all, 118 patients were randomized to UC alone or UC plus Meds-Help. Randomization was completed centrally at the coordinating site, using a blocked randomization scheme by site based on the patient's level of adherence in the prior 12 months (MPR of <0.4, 0.4–0.59, or 0.6–0.79) and the presence of concurrent substance use. Substance use was considered "present" if any patient chart note in the prior 12 months indicated difficulties with drugs or alcohol or if the patient reported that in the prior 6 months he/she had used any illicit drugs, had an episode of binge drinking, or consumed ≥7 drinks per week (for women) or ≥14 drinks per week (for men). Site research associates (RAs) called into the central research coordinator to receive newly enrolling patients' treatment assignment.
UC consisted primarily of treatment in VA outpatient mental health clinics and included psychiatrist visits, non-MD mental health visits, and group visits. During the study period, patients completed an average of 8 visits with psychiatrists; 49% had visits with non-MD mental health clinicians and 23% had group visits.
The Meds-Help intervention consisted of (1) unit-of-use packaging that included all patients' medications for psychiatric and general medical conditions (figure 2), (2) a medication and packaging education session, (3) refill reminders mailed 2 weeks before scheduled refill dates, and (4) notification of clinicians when patients failed to fill antipsychotic prescriptions within 7–10 days of a fill date. Meds-Help staff served as contacts for patient questions regarding pharmacy services or doctors' prescriptions. Pharmacy technicians, with oversight by pharmacists, completed many of these intervention components.
(Enlarge Image)
Figure 2.
Unit-of-Use Packaging.
The medication education session was conducted by a pharmacist, usually in-person, but occasionally by telephone. During this session, the pharmacist reviewed patients' prescribed medications, including treatment indications. The pharmacist also explained unit-of-use medication packaging (eg, package usage and warning labels, lack of child proofing) and plans for interim use of pill boxes when medication changes were made by clinicians before the next shipment of medication packages.
Measures
The primary outcome measure was medication adherence as measured by the MPR. A more stringent "composite adherence measure" (CAM) was also assessed.
The MPR is the ratio of the number of outpatient days' supply of medication that a patient has received during the designated time period divided by the number of days' supply they needed to receive to take their prescribed dose of antipsychotic continuously during noninstutionalized days. The MPR was the primary outcome measure because it is based on data (pharmacy fills) that can be collected unobtrusively, making it less likely to affect adherence behaviors; it can be calculated for patients even if they stop actively participating in the study; and it is associated with important outcomes in observational studies.
MPRs were calculated for 3 periods: the 12 months prior to enrollment (baseline), 0–6 months following enrollment, and 6–12 months following enrollment, unless participants had fewer than 90 noninstitutionalized days in a specified time period because of death, transfer of care outside the VA, or long-term inpatient stays. If depot antipsychotic medications were instituted after enrollment, we did not calculate MPRs after the switch.
Because patients may fill prescriptions but not ingest their antipsychotic medications, the more stringent CAM was based on multiple data sources. Patients were considered adherent on the CAM only if: (1) their MPR during the study time periods was ≥0.8, (2) they reported they "always" took their antipsychotics or only missed antipsychotics "a couple of times" in response to questions from Schizophrenia Outcomes Module, and (3) their blood test indicated the presence of some antipsychotic medication. The rationale for requiring patients to meet all 3 criteria is that an indication of poor adherence on any of these measures is likely associated with less-than-optimal adherence. Our a priori research plan was to consider blood levels to be negative when patients refused blood draws or did not show for follow-up assessments. If a physician discontinued antipsychotic medication because a patient refused it, blood levels were not drawn and assumed to be negative; if a physician discontinued antipsychotics because the medications were no longer considered necessary, the CAM was assessed only to the point of discontinuation.
Secondary outcomes were psychiatric symptoms as measured by the Positive and Negative Symptom Scale (PANSS), quality of life as measured by the Quality of Well-Being Scale (QWB), and patient satisfaction as measured by Client Satisfaction Questionnaire (CSQ-8) scale. The PANSS and QWB were completed by the research associates at each of the study sites. All these scales are widely used and have demonstrated reliability and validity. Exploratory outcomes included VA psychiatric hospitalizations, ascertained through chart reviews.
As with many health services interventions, patients could not be blinded to study assignment. Research associates were also not blinded due to the costs and logistics of hiring blinded assessors for each site and the likelihood that assessors would be "unblinded" by patient comments. However, the primary outcomes of adherence were based on longer term pharmacy fill patterns and less likely to be affected by subtle biases on the part of patients or interviewers.
Prior to enrolling patients, RAs received training in PANSS administration and scoring using standardized videotapes. RAs also received periodic training during the study, using tapes of study patients. For the total PANSS score, the inter-rater reliability, as assessed by an intraclass correlation coefficient (ICC), was 0.94 for the first 2 RAs enrolling patients; at a second time point, the ICC was 0.77 for 4 study RAs, and additional training was provided. At a third time point, the ICC for the 4 RAs was 0.93. In-person follow-up assessments were conducted in a private room at the patient's local VA facility.
Data Analyses
Analyses were conducted on an intent-to-treat basis. We describe means (±SD) for the continuous MPR measure and frequencies for the CAM categories (fully adherent/poorly adherent) at baseline, 6, and 12 months. The primary outcome of MPR was examined at 6 and 12 months using multiple linear regression; the CAM was evaluated as these time points using logistic regression analyses. Multivariable analyses included the baseline MPR as a covariate to adjust for levels of adherence prior to enrollment. We also adjusted for potential confounding variables of race, age, psychiatric diagnosis, substance use, and study site. Analyses for secondary outcomes of psychiatric symptoms, quality of well-being, and satisfaction were conducted similarly, with multiple linear regression analyses that adjusted for race, age, psychiatric diagnosis, and study site. The study was powered to detect a medium-sized effect in the primary outcome of MPR.