Comparison of pharmacist-led outpatient depression management to current prescriber-led depression management
Abstract
Introduction
Considering the expected decrease in psychiatrists due to retirements and fewer individuals entering the field, it is important to examine the pharmacist’s role in improving mental health outcomes. This study evaluated pharmacist treatment of depression in a behavioral health-integrated virtual (BHIV) clinic where patients were generally followed twice monthly compared with the standard of care, twice-yearly primary care management.
Methods
A descriptive report from January 1, 2020, to May 31, 2023, identified patients diagnosed with depression managed in the pharmacist-run BHIV clinic and the internal medicine (IM) clinics. Patients were excluded if they were established with a psychiatric specialist, diagnosed with a severe psychiatric disorder or postpartum depression, pregnant, or only prescribed trazodone at doses less than 100 mg. One hundred forty-two patients were included, with 71 in each group. The primary outcome was the final Patient Health Questionnaire-9 (PHQ-9) score.
Results
The average initial PHQ-9 score was equal in both groups (13 ± 4.5; 13 ± 5.5; P = 0.8). The average final PHQ-9 score was similar between groups (7 ± 5.6; 6 ± 6.6; P = 0.7). The average time to remission and response in weeks was shorter in the BHIV group (9 ± 6.4; 27 ± 20.5; P ≤ 0.001; 7 ± 5.9; 29 ± 26.0; P ≤ 0.001).
Discussion
Remission and response were identified more rapidly in the BHIV group than in the IM group. Clinical outcomes were comparable between cohorts at the final assessment. Overall, this shows that pharmacists can provide close follow-up and improve patients’ quality of life by providing rapid symptom resolution.
Introduction
MDD is one of the most prevalent mental health diagnoses worldwide. Over the past year, 10% of the United States population was diagnosed with MDD.1 It is thought that 39% of adults with MDD do not receive treatment.2 The availability of psychiatrists continues to decrease, and a 32% shortfall is expected by 2030.3-5 As a result, the National Council for Behavioral Health Medical Director Institute recommends that the number of advanced practice providers, including psychiatric pharmacists, be increased.3,4
Psychiatric pharmacists have the experience and knowledge to work under collaborative practice agreements (CPAs) which establish partnerships between pharmacists and other healthcare providers, enabling pharmacists to perform patient care tasks that extend beyond their usual dispensing role.6 Studies indicate that pharmacists providing medication management can improve adherence, increase patients’ knowledge about their prescribed therapy, and increase health-related quality of life.7-9 Pharmacist interventions for the management of depression in these randomized controlled trials have not demonstrated improvement in clinical outcomes in patients compared with the standard of care.7-9 No studies to date have compared the outcomes of pharmacist medication intervention in depression with provider-led medication intervention.
At Saint Francis Health System (SFHS), the Behavioral Health Integration Virtual (BHIV) Clinic assists providers in managing mental health medications by providing pharmacist assessments and treatment within a collaborative practice agreement. A grant funds the clinic and is free of cost to patients. Pharmacists working in the clinic are either in their second year of postgraduate psychiatric pharmacy residency or have completed their residency. The BHIV clinic receives referrals from internal medicine (IM) providers in SFHS’s Network. Patients eligible include those 18 years of age and older who can independently participate in their own care. One of the major advantages of the BHIV clinic is the short time from referral to first appointment and the ability to schedule frequent follow-up visits with patients. The scope of practice of the pharmacists in the clinic includes the management of depression, anxiety, neurocognitive disorders, and substance use disorders. The number of BHIV referrals tripled from 2022 to 2023. The growth was primarily driven by strong marketing efforts from clinic pharmacists to internal medicine providers within the health system, as well as increased provider familiarity with the clinic over time. The goal of this study was to assess the difference in initial and final PHQ-9 scores, time to PHQ-9 remission and response, and prescribing patterns for BHIV patients compared with IM patients.
Methods
Study Design and Patient Population
This study was conducted at Laureate Psychiatric Clinic and Hospital and the Warren Clinic at SFHS, a nonprofit organization. A retrospective chart review was performed of outpatients receiving treatment with antidepressants between January 1, 2020, and May 31, 2023. Patients were included if they received antidepressant treatment during the study period, were 18 years of age or older, diagnosed with depression at or before the initial visit, and had a documented PHQ-9 score within 4 weeks of psychiatric medication initiation or adjustment in the internal medicine group or within 4 weeks of the first visit for the BHIV cohort. Patients were excluded if they were established with a psychiatric specialist, diagnosed with a severe psychiatric disorder, pregnant, diagnosed with postpartum depression, or prescribed only trazodone at a dose of less than 100 mg. Diagnoses of severe mental health disorders that were excluded consisted of schizophrenia, other psychotic disorders, and bipolar disorder.
Clinical Data Collection
Clinical data were collected via manual review of patient charts in the electronic health record. Data retrieved included patient demographics, psychiatric comorbidities, prior and current psychiatric medications prescribed, benzodiazepine use at initiation, initial and final PHQ-9 score, time to response and remission in weeks, participation in and referrals to therapy, and referrals to higher level of care. Literature suggests using benzodiazepines alongside antidepressants may be linked to worse treatment outcomes in depression.10
Variable Definitions
The primary outcome was the difference in the initial and final PHQ-9 score. The initial score was defined as the most recent score recorded within 4 weeks after antidepressant initiation or adjustment in the IM group and the most recent score recorded within 4 weeks of the first appointment in the study period for BHIV patients. The final score was defined as the most recent score recorded 6 months to 1 year after the initial psychiatric medication initiation or adjustment in the IM group and the score recorded 6 months after the first BHIV visit or the score at graduation from the clinic in the BHIV group (patients were not followed after graduation). Graduation was based on symptom improvement and patient presentation in a clinical pharmacist interview.
Secondary outcomes included time to response and remission in weeks. Time to response was defined as the time to achieve 50% or more reduction in symptoms based on the PHQ-9 score. Time to remission was defined as the time to achieve a PHQ-9 score of 5 or less. Concomitant psychotherapy was categorized as those who had participated in at least 1 psychotherapy visit after medication initiation or adjustment in the IM group or after clinic enrollment in BHIV. A psychiatric drug change for depression management was identified as follows: addition, change, or discontinuation of medications for depression management, or a change in the time of drug administration. Antidepressant at the initial visit included the medication prescribed directly before the first BHIV visit, or the medication prescribed before psychiatric medication initiation or adjustment in the IM group. Antidepressants at final visit included the medication prescribed at graduation or 6 months after the first BHIV visit in the BHIV group or 6 months after the initial psychiatric medication initiation or adjustment in the IM group.
Statistical Analysis
Descriptive statistics were used for demographic and clinical presentation data. All analyses for categorical data were performed using Fisher’s exact test; continuous variables were compared using a 2-sided Student’s t-test or Mann-Whitney U test. Statistical significance was set at P < 0.05.
Patient Consent Statement
The SFHS institutional review board reviewed and found the study exempt from qualifying as human subject research.
Results
All 173 of the BHIV patients during the 2.5-year study period were reviewed; of these, 102 were excluded, and 71 were included in the final data analysis. The most common reason patients from this group were excluded was due to not having a qualifying PHQ-9 value. IM patients prescribed an antidepressant and diagnosed with depression during the period, totaling 22 114, and were chronologically screened. A total of 202 nonrandomized patients were screened for inclusion, yielding 71 patients in each group. The most common reason patients from this group were excluded was because they received a psychiatry referral. Aside from a greater number of benzodiazepine users at the initial visit (20 vs 5; P = 0.002) and a greater number of prior psychiatric medication trials (4 ± 1.8 vs 2 ± 1.8; P < 0.001) in the BHIV group, the groups were similar at baseline (Table 1). BHIV patients were most commonly prescribed bupropion (19/81, 23%) or duloxetine (17/81, 21%), and IM patients sertraline (15/55, 27%) when presenting for the initial visit. Escitalopram (15/73, 21%) and bupropion (15/73, 21%) were the most prescribed medications at the final visit in the IM group, and bupropion (25/92, 27%) and sertraline (18/92, 20%) in the BHIV group. The initial PHQ-9 scores were similar between the BHIV and IM groups, respectively (13 ± 4.5 vs 13 ± 5.5; P = 0.8) (Table 2). The primary outcome of the difference in final PHQ-9 score between the groups was not statistically significant (7 ± 5.6 vs 6 ± 6.6; P = 0.7) (Table 2). The percentage of patients achieving response (41/71, 58% vs 35/71, 49%) and remission (36/71, 51% vs 32/71, 45%) was greater in the BHIV group. Time in weeks to identify response (7 ± 5.9 vs 29 ± 26; P < 0.001) and remission (9 ± 6.4 vs 27 ± 20.5; P < 0.001) differed significantly (Table 2). Though not statistically significant, a greater number of patients were referred to therapy in the BHIV group (16/71, 23%) despite a significant majority of patients already having a prior therapy referral compared with the IM group (23/71, 32% vs 9/71, 13%; P = 0.008) (Table 2). On average, BHIV patients had twice as many visits as the IM group (6 ± 3.9 vs 3.0 ± 1.8; P < 0.001) and a greater total number of medications at graduation (2 ± 0.9 vs 1 ± 0.8; P = 0.025) (Table 2). The most common reason for drug change for both groups was ineffectiveness, with a significantly greater number of IM patients changing for ineffectiveness than BHIV patients (55/71, 77% vs 69/71, 97%; P < 0.001) (Table 2). Time from referral to initial BHIV visit was 22 days (Table 3).
Discussion
Our findings suggest that the BHIV clinic can provide management of depression symptoms despite potentially managing complicated patients who have trialed more medications for depression. Our results are similar to those of other studies that found pharmacists improved health-related quality of life but did not statistically improve clinical outcomes.7-9
The definition of treatment-resistant depression is often referred to as a major depressive episode that does not respond satisfactorily to at least 2 trials of antidepressant monotherapy.11 Although there were no significant differences in the final PHQ-9 score identified between the 2 groups, the BHIV group may have included more individuals with treatment resistance. We believe this difference between the groups is reflective of IM providers referring patients with depression who were more difficult to manage. Higher benzodiazepine use at the initial visit in the BHIV group compared with the initial visit in the IM group may have indicated that patients in the BHIV group had other psychiatric comorbidities, including generalized anxiety disorder. Regardless of pharmacists’ primary recognition as medication experts, a greater number of individuals in the pharmacist-managed group received comprehensive depression management by placement of a psychotherapy referral. Increased access to care because of mental health medication management by a pharmacist demonstrated rapid time to response, remission, and completion of more patient visits.
Limitations of this study include the retrospective, single-center design and the small sample size. Most of the population assessed were middle-aged White females, which limits the generalizability of the results, as the response to treatment can vary between populations. The number of PHQ-9 scores obtained was affected by the unavailability of documentation, which is likely to be a source of bias toward providers who consistently document PHQ-9 scores. Capturing response and remission results from PHQ-9 scores would also be biased toward providers who have more frequent patient visits.
Regardless of its limitations, this is the only study to date that has compared medication therapy management for depression by a pharmacist with IM management. The average time in weeks to observe response and remission was 7 and 9 weeks in the BHIV group and 29 and 27 weeks in the IM group. In some offices, time to see a psychiatrist may take up to 90 days, whereas in the BHIV group, pharmacists were able to see patients within 22 days.12 On average, the time to remission in the pharmacist group was 9 weeks, demonstrating that pharmacists can have patients in remission before a psychiatrist sees them.
Conclusion
This retrospective review of patients in the BHIV and IM clinic demonstrated no statistical difference in the final PHQ-9 score. However, it resulted in documentation of measured achievement of response and remission more rapidly in the pharmacist group. On average, patients in the BHIV group had more prior antidepressant medication trials, likely reflecting a more treatment-resistant patient population. Pharmacists can maintain close follow-up and improve patients’ quality of life by providing earlier symptom resolution and comparable symptom control to that of an IM physician.
Contributor Notes
Disclosures: The authors have no conflicts of interest to disclose.
Financial support. The authors did not receive any funding to conduct this study.