Changes in PHQ-9 score after clinical pharmacy practitioner intervention in primary care mental health integration clinic

Abstract

Introduction

The Lt. Col. Luke Weathers, Jr. Veterans Affairs Medical Center (Memphis VAMC) expanded clinical pharmacy mental health services to Primary Care Mental Health Integration (PCMHI) in July 2023. This expansion provides an opportunity for further evaluation of the specialized Clinical Pharmacy Practitioners’ (CPP) impact on providing patient-centered healthcare and improving mental health outcomes. This study aimed to assess the impact of adding a CPP in PCMHI.

Methods

A retrospective chart review of computerized medical records from the Memphis VAMC was conducted from August 1, 2023, to August 1, 2024, in patients seen by the PCMHI CPP who had pre- and post-intervention PHQ-9 scores. The primary objective of this study was to assess changes in PHQ-9 scores. Secondary objectives were to analyze pharmacotherapy agents used, concomitant psychotherapy usage, changes in other measurement-based care (Generalized Anxiety Disorder-7 item scale [GAD-7] and Posttraumatic Stress Disorder Checklist for the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition [PCL-5]), reported side effects and adherence, and disposition after treatment. Descriptive statistics were used to analyze demographic data and outcomes.

Results

A total of 376 patients were screened, and 199 patients enrolled. PHQ-9 scores improved in 78.4% of patients, with a median score improvement of 5 points (interquartile range [IQR] 1-10). In patients with documented pre- and post-CPP intervention GAD-7 (n = 173) and/or PCL-5 (n = 8) scores, the median score improvement was 5 points (IQR 1-10) and 19 points (IQR 8.75-28.5), respectively.

Discussion

PHQ-9, GAD-7, and PCL-5 scores improved after incorporating a CPP in the PCMHI clinic. These findings suggest patients benefit from CPP expertise in assessing the efficacy of mental health pharmacotherapy and improving mental health symptomology.

Introduction

One of the many challenges veterans face is accessing mental health care resources, and less than 50% of veterans in need receive any mental health treatment.¹ The 2024 National Veteran Suicide Prevention Annual Report stated that the suicide rate for veterans was more than double the rate for non-veteran US adults (34.7 vs 17.1 per 100 000).²

To improve veterans’ access to mental healthcare, Veterans Affairs (VA) established the Primary Care Mental Health Integration (PCMHI) model in 2007, and this service is now required across all patient-aligned care teams at VA medical centers, including community-based outreach clinics. The PCMHI model provides a streamlined approach to treating common mental health conditions, including, but not limited to, depression, anxiety, posttraumatic stress disorder (PTSD), and alcohol use disorder. PCMHI consists of a multidisciplinary team of mental health experts, including nurse care managers, psychologists, social workers, licensed mental health counselors, and pharmacists, embedded within the patient’s primary care team.³

Before the incorporation of PCMHI, primary care providers would often refer patients to specialty mental health clinics due to time constraints hindering adequate mental health assessments, in addition to addressing multiple chronic disease states. As a result of increased referrals to specialty mental health clinics, patient load and wait times increased significantly. PCMHI addresses these barriers by offering same-day appointments, reducing specialty clinic referrals and wait times. PCMHI providers manage patients with mild to moderate complexity mental health conditions until they are stable, then discharged back to their primary care provider.^3,4

Clinical Pharmacist Practitioners (CPPs) working within their scope of practice in PCMHI clinics may prescribe evidence-based pharmacotherapy, enhance medication adherence, and provide in-depth patient education. Currently, few retrospective studies report improved patient outcomes when integrating CPPs into PCMHI.^5-7 Most of these studies reported remission based on improved rating scores after employing assessment tools such as the PHQ-9 for depression, the Generalized Anxiety Disorder Questionnaire (GAD-7), and the PTSD Checklist for the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition, or the DSM-5 (PCL-5).^6,7 Mental health clinicians widely use these screening tools to diagnose and measure the severity of depression, anxiety, and PTSD. Pharmacists can specifically use these tools to assess patient response and determine the efficacy of psychotropic medication regimens. The PHQ-9 consists of 9 questions that correspond to each DSM-5 criterion for MDD and is scored on a scale of how frequently symptoms occur, ranging from 0 (not at all) to 3 (nearly every day). This results in a total score ranging from 0 to 27. A score of 0 to 4 is considered minimal or no depression, 5 to 9 is mild, 10 to 14 is moderate, 15 to 19 is moderately severe, and 20 to 27 is severe depression. The National Committee for Quality Assurance (NCQA) defines response as a 50% or greater reduction in PHQ-9 score and remission as a score of less than 5, documented at 4 to 8 months after the initial elevated PHQ-9 score. As these are validated tools recommended in current literature, their use is necessary in future research as a standardized approach to evaluating remission and the impact CPPs have on treating patients with mental health disorders.^8,9

The Lt. Col. Luke Weathers, Jr. Memphis VAMC expanded clinical pharmacy mental health services to PCMHI in July 2023. This expansion provides an opportunity to further evaluate the specialized impact of CPPs on providing patient-centered healthcare and improving mental health outcomes through timely access to psychotropic medication management.

Methods

A retrospective chart review of electronic medical records of veterans seen by PCMHI CPP between August 1, 2023, and August 1, 2024, at the Lt. Col. Luke Weathers, Jr. Memphis Veterans Affairs Medical Center (VAMC) was conducted. This study was approved and conducted in compliance with the Lt. Col. Luke Weathers, Jr. VAMC institutional review board.

Patients were initially identified by performing a structured query language of patients prescribed psychiatric medications by a PCMHI CPP. Each patient’s medical record was screened to ensure that the study’s inclusion and exclusion criteria were met. Patients aged 18 years or older were included if they were seen by the PCMHI CPP and had pre- and post-intervention PHQ-9 scores documented. Patients were excluded if they were receiving hospice care, pregnant/lactating, or if they were seen exclusively for substance use disorder or insomnia treatment.

After determining the study population, each patient record was assessed, and baseline data, including age, sex, race, medical comorbidities, and psychiatric diagnosis, were collected. Baseline and any follow-up PHQ-9 scores were recorded to assess changes from the initial CPP intervention to the time of disposition or the most recent follow-up, if the patient was still receiving care. The date of the initial CPP visit, the pharmacotherapy agents used, and whether concomitant psychotherapy was provided were recorded. Confirmation of medication tolerability and efficacy was primarily completed via chart review of clinician notes. Patient-reported adherence and refill history were documented by the CPP at each follow-up visit using a standardized note template. A patient was considered adherent if they consistently reported taking the medication as prescribed and had no discrepancies in the fill history, as confirmed by refill dates in their medical record. The disposition of each patient was documented to identify those still receiving care versus those discharged to either a primary care provider (due to treatment success), referred to a specialized mental health clinic, or lost to follow-up.

Descriptive statistics, including means, medians, and percentages, were used to analyze demographic data. Parametric continuous variables were described as a mean ± standard deviation (SD). Nonparametric continuous variables were described as medians with interquartile ranges (IQRs).

Results

A total of 375 patients were seen by the PCMHI CPP in the defined timeframe and screened for enrollment. Of patients, 176 were excluded, mainly because a PHQ-9 was not documented before CPP intervention or because follow-up scores were not available after treatment was initiated, leaving a total of 199 patients enrolled in the study (Figure 1). The mean age ± SD of the patients enrolled was 53.4 ± 14.8; 167 patients were male, and 135 were African American. Of patients, 71.9% had multiple psychiatric diagnoses, but of the patients who did not have multiple psychiatric diagnoses, depression was the most common diagnosis. The median baseline PHQ-9, GAD-7, and PCL-5 scores were 15 (IQR 10-18), 13 (IQR 8-18), and 55 (IQR 34-62), respectively (Table). The median final PHQ-9, GAD-7, and PCL-5 scores were 8 (IQR 4-11), 6.5 (IQR 4-11), and 45 (IQR 26.75-49.5), respectively. For the primary outcome, improvement in PHQ-9 score was observed in 78% of the 199 enrolled patients. Of patients, 18% had worsened scores, and 4% had no change in scores after CPP intervention. Overall, including patients with worsened scores, there was a median score improvement of 5 points (IQR 1-10) on the PHQ-9 scale. Of enrolled patients, 173 also had pre- and post-CPP intervention GAD-7 scores documented. A score improvement was seen in 78% of patients; 14% had worsening scores, and 8% had no change in scores after CPP intervention, resulting in an overall median score improvement of 5 points (IQR 1-10) on the GAD-7 scale. Only 8 patients had pre- and post-CPP intervention PCL-5 scores documented. In this group, 87.5% of patients showed score improvement, and 12.5% had documented worsening scores, resulting in an overall median improvement of 19 points (IQR 8.75-28.5).

FIGURE 1.

Patients screened for enrollment.

TABLE.

Patient characteristics

Baseline Characteristics (n = 199)
Age, yr, mean±SD	53.4±14.8
Sex, n (%)  Male  Female	167 (83.9) 32 (16.1)
Race, n (%)  Black or African American  White  Unknown  Asian	135 (67.8) 49 (24.6) 14 (7.1) 1 (0.5)
Mental Health Diagnosis, n (%)  Multiple diagnoses  Depression  Other  Anxiety  PTSD	143 (71.9) 29 (14.6) 15 (7.5) 8 (4.0) 4 (2.0)
Baseline Score, Median (IQR)  PHQ-9 (n = 199)  GAD-7 (n = 173)  PCL-5 (n = 8)	15 (10-18) 13 (8-18) 55 (34-62)

PTSD = Posttraumatic Stress Disorder; GAD-7 = Generalized Anxiety Disorder-7 item scale; PCL-5 = PTSD Checklist for the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition.

Of the 199 patients included, 222 total antidepressant prescriptions were either continued, optimized, or initiated by the CPP, indicating that some patients were receiving multiple antidepressants. The most common antidepressant used was sertraline, with 128 prescriptions. Duloxetine and fluoxetine were the second-most-common agents used (Figure 2). Of patients, 68.3% used concomitant psychotherapy in addition to pharmacotherapy, and the remaining 31.7% were treated solely with pharmacotherapy. Patients were considered to have participated in concomitant psychotherapy if they had undergone at least 1 session of PCMHI brief treatment or Behavioral Health Interdisciplinary Program psychotherapy at any point during their medication management with the PCMHI CPP. The median improvement in PHQ-9 score was 5 points in both groups (IQR 1-10 in the psychotherapy group vs IQR 1-10.5 in the non-psychotherapy group). Adherence was measured by a patient’s verbal report directly to the PCMHI CPP and documented in the chart at each follow-up visit. Refill history discrepancies were also noted. Of patients, 88.4% were adherent to their pharmacotherapy regimen prescribed by the CPP, and 13.1% of patients experienced an adverse drug reaction, resulting in discontinuation of treatment. After treatment by the CPP, the most common disposition was discharge back to the patient’s primary care provider, reflecting treatment success in 48.2% of patients. There were 34.2% of patients who were still receiving care from the PCMHI CPP at the time of data collection. Lastly, 9.5% were referred to the Behavioral Health Interdisciplinary Program, which is a service for patients requiring a higher level of specialty mental healthcare, and 8.1% of patients were lost to follow up, meaning they did not show up for an appointment with the PCMHI CPP at some point following medication initiation and did not respond to subsequent attempts (2 phone calls and 1 letter) to reschedule (Figure 3).

FIGURE 2.

Antidepressant medications used.

FIGURE 3.

Disposition after treatment.

Discussion

As previously mentioned, our results showed a median improvement in PHQ-9 score of 5 points (IQR 1-10) after patients received treatment prescribed by the CPP. Based on the NCQA definition, this score improvement does not meet the definition of a response because the initial median score was 15 (IQR 10-18) and the final score was 8 (IQR 4-11), and therefore does not constitute a 50% reduction. However, the initial median score of 15 was in the moderately severe range, and the final median score of 8 is considered in the mild range on the PHQ-9 scale, indicating a clinically significant improvement in depression severity after CPP intervention.⁸ Although response and remission thresholds have not been validated for the GAD-7, the initial median score was considered in the moderate range at 13 (IQR 8-18), and the median final score improved to 6.5 (IQR 4-11), which is considered in the mild range on the GAD-7 scale.¹⁰ Considering scores decreased by 50% from baseline and extrapolating from the response definition for PHQ-9, this indicates a clinically significant change. A study by Marx and colleagues¹¹ indicates a “reliable change” on the PCL-5 scale as a score reduction of 15 to 18 points, and a score of 28 or less indicates a clinically significant change. As stated above, we observed a median score improvement of 19 points on the PCL-5 scale among the 8 patients, exceeding the definition of a reliable change in score. Overall, these results further enhance and build on preexisting smaller-scale VA studies, whose findings also support the incorporation of clinical pharmacists into PCMHI teams to improve patient access and mental health outcomes.^5-7 A similar study by Herbert et al⁷ defined depression response as a 50% reduction in PHQ-9 score from baseline and remission as a PHQ-9 score of less than 5. Of the 136 patients evaluated, 46% (n = 62) achieved response, and 31% (n = 42) achieved remission. Using this definition of response and remission, our study found the following similar rates: 45% (n = 89) achieved response, and 28% (n = 55) achieved remission. Additionally, the low rate of specialty mental healthcare referrals observed in our study (9.5%) aligns with the referral rate reported by Herbert et al (9.9%),⁷ further strengthening the evidence that CPPs play an integral role in improving access to mental healthcare.

Improvements discussed above were achieved through the use of antidepressant treatment, most frequently sertraline. Other antidepressants were often used following a failed trial or intolerance to sertraline. The frequency of medication use is shown in Figure 2. The most common adverse drug reaction that resulted in discontinuation was sexual dysfunction, which was most often seen with sertraline, which may be, in part, due to higher rates of prescribing.

Of note, the addition of psychotherapy did not result in a significant PHQ-9 score improvement compared with the group of patients who did not receive concomitant psychotherapy. Although this was an unexpected finding, it indicates that medication management was effective regardless of psychotherapy use.

This project has several strengths and limitations affecting the interpretation and generalizability of the results. To our knowledge, this study enrolled the largest sample in current publicized research evaluating the impact of CPPs on patient outcomes within PCMHI clinics. It also looked at multiple secondary outcomes. Data were collected from a standardized note template that involved comprehensive documentation (ie, adherence and refill history documented, adverse effects, etc.), which limits potential documentation error and undetected bias.

Patients were enrolled in this study as long as they had a PHQ-9 score documented before starting treatment and at least 1 after, regardless of how long they had been receiving a given treatment. For example, a patient who only recently referred to the PCMHI CPP may have just had 1 follow-up PHQ-9 score after initiating a new psychotropic medication, and it is possible that the full effect of that medication was not yet seen at the time of data collection. Referring back to the NCQA definition for response and remission, it is recommended to obtain a repeat PHQ-9 score 4 to 8 months after the initial elevated score.⁹ Additionally, results from the Sequenced Treatment Alternatives to Relieve Depression landmark trial indicated that medication trials of at least 8 weeks with moderately aggressive dosing are often necessary to achieve remission.¹² Therefore, the lack of standardized PHQ-9 score collection intervals could have impacted results by underestimating the degree of score improvement, especially if patients’ treatment was not yet optimized at the time of data collection. Similarly, it is likely that some patients in the psychotherapy group had not yet experienced the full benefits of therapy, especially because patients were included even if they had participated in only 1 session. The fact that the data were based on subjective patient reports of symptoms is another potential limitation of this study. Although these scales are validated in current literature, the use of subjective scoring tools may contribute to patient variability in reported symptoms. With no comparator group, it is difficult to establish the true impact of the CPP compared with clinicians treating mental health conditions outside of PCMHI. Future studies can address this limitation by assuming that clinicians use the same standardized measurement-based care tools, which are currently less used/documented in settings outside PCMHI. Additionally, only 8 patients had pre- and post-intervention PCL-5 scores documented, thus making it difficult to generalize these results to patients with predominant PTSD symptoms. Future studies are needed to confirm the positive results observed in this group.

Conclusion

These results showed improvements in PHQ-9, GAD-7, and PCL-5 scores after PCMHI CPP intervention. This indicates there is benefit from integrating mental health CPP expertise in primary care to manage mental health pharmacotherapy, encourage adherence, and improve symptoms of mental health conditions.