Impact of pharmacist‐led chronic disease management in a Federally Qualified Health Center

There is growing interest in pharmacist‐led chronic disease management (CDM) for underserved populations such as those treated within Federally Qualified Health Centers (FQHCs).


| INTRODUCTION
Chronic disease poses a significant healthcare burden in the United States, affecting six in ten adults and accounting for 90% of the nation's annual healthcare expenditures. 1 Addressing chronic illness is a major challenge in the United States (US); however, with appropriate management, morbidity and mortality can be mitigated.
A common component of chronic disease management (CDM) is through long-term pharmacotherapy. In 2018, nearly 70% of outpatient physician visits in the United States involved drug therapy, and over 860 million drugs were provided or prescribed in this setting. 2 Similarly, almost half of the US population reported using at least one prescription drug in the last 30 days, 3 and retail outlet sales of prescription drugs totaled 456.3 billion US dollars. 4 Those living with chronic illness utilize more prescription drugs than those without chronic illness, and utilization and cost of prescriptions increase with the number of concurrent chronic illnesses that an individual has. 5 Despite their frequent use, prescription medications can be complicated, both from the perspective of the prescriber as well as the patient, and they are not without the ability to cause harm. According to 2013-2014 data, an estimated four emergency department visits for adverse drug events occurred per 1000 individuals annually, 6 and in 2018, there were 861 000 emergency department visits in the United States associated with a primary diagnosis related to adverse drug events, drug poisoning, or underdosing of drugs. 7 Some key challenges encountered when using prescription medications include ensuring they are prescribed for an appropriate indication, utilizing the safest and most effective option, and promoting medication adherence. Due to these challenges, pharmacists are often incorporated into chronic disease models of care and have served as the lead in these initiatives. 8 As medication experts, pharmacists are well suited to provide medication management services to enhance CDM by promoting safe and effective medication use in collaboration with other healthcare professionals to achieve optimal patient outcomes.
When delivered in outpatient settings, medication management by pharmacists can improve medication appropriateness and adherence and reduce medication dosing, hospitalizations, and healthcare costs. 9 There is growing interest in incorporating pharmacist services in Community Health Centers like Federally Qualified Health Centers (FQHCs) for CDM. FQHCs are community-based healthcare providers that receive funds from the Health Resources and Services Administration Health Center Program to provide primary care services in underserved areas. 10 Much of the available literature describing pharmacist services within FQHCs focuses on the management of diabetes mellitus and associated cardiovascular disease and results indicate that when pharmacists are added to the healthcare team, clinical outcome measures such as glycated hemoglobin (A1C), blood pressure, and cholesterol levels improve. [11][12][13][14][15][16][17] Such studies are of variable quality and include limitations such as being retrospective, lacking comparison groups, being descriptive in nature without statistical analyses, focusing on only one disease state, and/or including a small number of patients. [11][12][13][14][15][16][17] As such, there is an ongoing need to provide additional evidence which describes the provision of CDM services by pharmacists to further define the ways in which they can impact care in the FQHC setting.
Henry J. Austin Health Center (HJAHC), an FQHC in Trenton, NJ, is a system of 8 sites and one mobile health unit that provides care to over 18 000 patients throughout Mercer County. In 2020, 91% of patients were at or below 200% of the federal poverty level, 31% were uninsured, most patients with insurance had governmentfunded insurance, and 93% of patients identified as racial and/or ethnic monitories, with 51% identifying as Black/African American and 48% identifying as Hispanic/Latino. Being a patient-centered medical home, HJAHC follows a team-based healthcare delivery model to provide high-quality comprehensive and accessible medical care. Each medical team is comprised of a unit receptionist, medical assistant, nurse, and primary care provider (PCP). HJAHC also follows an integrated behavioral healthcare model whereby behavioral health services are embedded within the medical teams. At the time of the study, the behavioral health team included behavioral health consultants and a behavioral therapist. In addition to its core primary care services, HJAHC offers dental, podiatric, nutrition, social work, and pharmacy services. Pharmacy services were first established at HJAHC in August 2014 with the support of grant funding to assess the impact of adding a pharmacist to the healthcare team. Via this grant, the pharmacists were responsible for conducting a pilot study in which Medicaid-eligible patients with uncontrolled type 2 diabetes mellitus (T2DM) were randomized to receive standard of care (control arm) or standard of care plus the care of the pharmacist (intervention arm) for 12 months, the results of which have been published elsewhere. 18 As pharmacy services were first introduced via a specific protocol in which pharmacists proactively identified eligible patients and services were restricted based on insurance status and disease state, it is unclear for which disease states the clinical staff of HJAHC would seek assistance from pharmacists and if pharmacists could help to improve clinical outcomes for HJAHC patients across a variety of disease states. To further define the role of the pharmacists based on the needs of the health center and to reach a larger population of patients, the pharmacy team developed a new protocol for CDM by which any member of the healthcare team could refer a patient to receive CDM from a pharmacist, the details of which are described in this paper.

| METHODS
The main purpose of this study is to assess the impact of adding a pharmacist to the healthcare team within an FQHC for CDM. This non-randomized, 6-month prospective pilot study was approved by the Rutgers University and HJAHC institutional review boards. The primary outcome of this study is the change in disease control at the end of 6 months of pharmacist management via CDM. The research team determined a consistent clinical endpoint for each referred disease state which served as the marker for change in disease control.
As this was an exploratory assessment and referrals were not limited to specific disease states, the primary clinical endpoint could not be predetermined for all referred disease states before initiating the study. However, the research team conducted an internal review at HJAHC and hypothesized which disease states were likely to be the reason for referral to anticipate the primary clinical endpoint for each. If a patient was referred for a condition that was not listed among the predetermined disease states, the research team would determine the most appropriate primary clinical endpoint to use.
When subjective surveys of disease control were used, such as for anxiety and depression, validated survey instruments already used at HJAHC as part of standard practice were selected as the primary clinical endpoint. Table 1  Patients were referred to receive CDM services from a pharmacist via a referral order placed within the electronic health record (EHR). A patient navigator reached out to each referred patient and scheduled the initial evaluation. The pharmacist determined if the patient met criteria for enrollment into the study by performing chart review of scheduled patients prior to their appointment.
Patients were eligible if they were 18 years or older, English-speaking, and referred for the management of at least one newly diagnosed or pre-existing disease state that is a chronic medical condition in which pharmacotherapy is indicated and expected to continue for a minimum of 6 months after the referral. Patients of all genders, races, ethnicities, socioeconomic classes, and insurance status were included.
Exclusion criteria included pediatric patients, pregnant patients, and those seen by an outside provider for management of the referred disease state. Any patient who failed to adhere to the study protocol by missing appointments, failing to reschedule missed visits, or switching to an outside PCP was administratively withdrawn, with further details provided in Appendix B. Finally, patients who were eligible for, or currently enrolled in, the T2DM randomized controlled trial were excluded from participation in this study. 18 If eligible, study participants were invited to join the research study. Either the pharmacist or patient navigator reviewed the written informed consent with the patient. If they accepted, the participant signed the informed consent, was provided a copy, and was enrolled at that time.
Upon enrollment, the patient completed a demographic form. The patient completed a Self-Rated Health Form, Healthcare Utilization Form, and Provider Satisfaction Survey at baseline and 6 months. The Self-Rated Health Form and Healthcare Utilization Form were originally developed by the Stanford University self-management resource center and were chosen as these were the same forms used by the researchers for the T2DM randomized controlled trial. 19 The baseline Provider Satisfaction Survey was regarding the care received from the PCP and the 6-month form was regarding care from the pharmacist. This is not a validated survey, and it was adapted from that already in  Table 2.
Of enrolled patients, most were referred by their PCP, with 195 (65%) referred by a nurse practitioner (NP) and 80 (26.67%) by a physician. Fourteen were referred by a behavioral health consultant, five by a registered dietician, four by a behavioral therapist, and two by a nurse.

| Primary outcome
The results of the primary outcome for the most commonly referred disease states (at least 20 participants) are summarized in Table 3.   (Table 4).

| Secondary outcomes
Patient satisfaction with care is summarized in Figure 2. Patients consistently reported higher satisfaction scores for care received from the pharmacist as compared to the PCP. When considering each question individually, pharmacists received an ideal score from a higher percentage of respondents than the PCPs, which was statistically significant for all questions but one.

| DISCUSSION
Patients were most likely to be referred to the pharmacist by their PCP as compared to other team members. This was expected as it is were referred by a nurse practitioner, which is consistent with the staffing at HJAHC, as the full-time equivalent ratio of physicians to nurse practitioners is approximately 1 to 3.
The most common disease states prompting referral include T2DM, hypertension, anxiety, and depression, which may be due to a few key reasons. Firstly, pharmacy services were originally established at HJAHC for management of T2DM more than 2 years before the start of this study, and enrollment into the T2DM randomized controlled trial continued concurrently during the first 7 months of this study; therefore, the medical teams were encouraged to make refer- When considering hypertension, the mean SBP at baseline was consistent with Grade 2 hypertension, and at the end of the study, the mean SBP was reduced to Grade 1 hypertension and met the essential hypertension goal as defined by the 2020 International Society of Hypertension Global Hypertension Practice Guidelines. 24 Like T2DM, the mean SBP at the end of the study no longer met UDS criteria for uncontrolled hypertension (SBP ≥140 mmHg). 23 In the anxiety population, the mean GAD-7 score was consistent with severe anxiety at baseline, and at the end of study, the mean score was consistent with mild anxiety. 25 Similarly, in the depression population, the mean baseline PHQ-9 score was consistent with moderately severe depression and the final mean score with mild depression. 26 In a comparable practice setting in which a pharmacist provided depression management within an FQHC, mean baseline PHQ-9 scores were like those observed in this population, and the change in PHQ-9 score was also statistically significant. 27 There are several limitations to note regarding this study and the interpretation of its results. As this was an exploratory pilot study, no power calculation or sample size was determined prior to initiating the study. There was no limit to the types of chronic disease states which could have been referred for CDM; therefore, there was variability in the primary endpoint to measure disease control. This led to many statistical analyses without the use of a correction factor which may have increased risk of alpha error. This study only occurred at one FQHC and generalizability of the results is limited. The short duration of the intervention may have led to an inability to observe the impact on disease control for some of the referred disease states as well as some of the secondary outcome measures, such as selfreported health and healthcare utilization. As median healthcare utilization was low at baseline, it is unlikely a change would be seen for Abbreviations: A1C, glycated hemoglobin; GAD-7, Generalized Anxiety Disorder-7; PHQ-9, Patient Health Questionnaire-9; SBP, systolic blood pressure; T2DM, Type 2 diabetes mellitus.

T A B L E 4 Healthcare utilization
Healthcare service Baseline End

FUNDING INFORMATION
There was no external funding for this research. more than one disease state for 63 (62.4%) patients. The average number of disease states for which MTPs were identified was 2 (SD = 1.12).
A summary of the MTPs is listed in Table A1, and the types of interventions implemented are listed in Table A2. Interventions listed in Table A2 did not require primary care provider approval prior to implementation. Only those interventions that were accepted by the patient are included. MTPs and associated interventions may be counted more than once in some cases, such as continued issues with adherence despite interventions to resolve adherence problems.

Reasons for administrative withdrawal
Patients were administratively withdrawn from the study under the following circumstances: • The patient missed three appointments during the study period; • The pharmacist was unable to schedule an appointment within 30 days of a missed appointment; • The patient did not complete a minimum of three visits within the 6-month study period or the third appointment did not occur at least 30 days prior to the ideal study close out date; • The patient was no longer being managed by a Henry J. Austin Health Center (HJAHC) primary care provider (PCP) for the referred medical condition; or • The patient became pregnant at any point during the study.
The total full-time equivalent of clinical pharmacists at HJAHC was 1.0 during this study. The clinical pharmacists had other responsibilities to HJAHC in addition to providing chronic disease management (CDM) as part of this study, and the entire 1.0 full-time equivalent was not dedicated to this study. Therefore, to allow the pharmacists to spend more time on delivering CDM and limit the time spent by the pharmacists in following up with patients who were not engaged in the study (eg, those who missed multiple appointments, did not re-schedule missed appointments, and did not complete a minimum of three appointments), the pharmacists were permitted to administratively withdrawal patients under several circumstances. To ensure the pharmacists were able to proactively make prescriptive changes to those medications for the referred diseases, patients who were no longer managed by a HJAHC PCP for the condition were administratively withdrawal. Finally, due to the variable effect of pregnancy on disease control, the researchers allowed the pharmacists to remove any patient who became pregnant during the study. However, no participant was withdrawal from the study for this reason. Decrease frequency of administration 2 0.3 Note: All interventions listed did not require primary care provider consent as pharmacist interventions for medication adjustments were pre-approved at the time of referral. All interventions listed were accepted by the patient prior to implementation.

Self-Rated Health
In general, would you say your health is…(circle one number)

Patient Experience with Medical Provider Survey
Please choose the medical profession that best describes your current primary care provider. If you do not know, ask the pharmacist for help.
Medical Provider: • Physician 1. In the past 6 months, how many times did you visit a physician? Do not include visits while in the hospital or to a hospital emergency room. Fill in with "0" or another number.
____ times 2. In the past 6 months, how many time did you go to a hospital emergency room? Fill in with "0" or another number. ____ times 3. How many different times did you stay in a hospital overnight or longer in the past 6 months? Fill in with "0" or another number. ____ times 4. How may total nights did you spend in the hospital in the past 6 months? Fill in with "0" or another number. ____ nights

• Physician Assistant
• Nurse Practitioner • Advanced Practice Nurse • Pharmacist

Patient Experience with Medical Provider Survey
Please answer each of the following questions. Think of your medical provider when you answer the questions. This information will remain confidential.
1. How often does the medical provider explain things in a way that is easy to understand? Discipline of referring team member: Primary clinical outcome measure identified as: Result of primary clinical outcome measure: What is the patient's weight?

____________ (kg)
What is the patient's height?

____________ (inches)
What is the patient's BMI?

____________ (kg/m 2 )
What is the patient's systolic blood pressure?

____________ (mm Hg)
What is the patient's diastolic blood pressure?

____________ (mm Hg)
Is the patient currently taking one or more drugs for the management of primary referred disease state? (choose one; if "Yes," then choose a corresponding year when the patient first started to use any agent for the management of hypertension, if known)