Impact of a Pharmacist-Managed Diabetes Clinic on Quality Measures
Nadia J. Aneese, PharmD; Alexandra Halalau, MD; Sarah Muench, PharmD; Daniel Shelden, DO; Janna Fett, PharmD; Colleen Lauster, PharmD
A pharmacist-managed diabetes clinic (PMDC) was created to assist with improvement of diabetes quality measures. The objective was to evaluate the PMDC impact on quality measures.
According to the CDC 2017 National Diabetes Statistics Report, diabetes was the seventh leading cause of death in 2015.1 Additionally, diabetes is associated with significant morbidity leading to a high burden on healthcare costs. Tools are used to evaluate provider performance on this costly disease. The Healthcare Effectiveness Data and Information Set (HEDIS) developed by the National Committee for Quality Assurance (NCQA) is 1 tool with defined diabetes care criteria.2 These criteria include current glycated hemoglobin testing (A1C), A1C control (<8%), current retinal or dilated eye exam, blood pressure control (<140/90 mmHg), and nephropathy monitoring.
The NCQA maintains Recognition Programs such as the Patient-Centered Medical Home designation for clinicians who adhere to medical evidence proven to provide high-quality care. Achieving high HEDIS scores is imperative for these NCQA Recognition Programs and has been associated with cost-effective practices and better health outcomes.2,3
A pharmacist-managed diabetes clinic (PMDC) was created to assist with the improvement of diabetes quality measures. The criteria developed by HEDIS were used to evaluate care in our Outpatient Clinic (OPC). Studies have shown the benefits of PMDCs across different clinical settings, including attaining significant improvement of diabetes related parameters (A1C, blood pressure, lipids, etc) and reaching A1C goals more frequently.4-9 The objectives of this study were to evaluate the impact of a PMDC on HEDIS measures and to assess adherence to other diabetes standards-of-care recommendations.
Methods Study Design
This study was approved by the Beaumont Health Institutional Review Board as a retrospective cohort analysis of patients with type 1 diabetes (T1D) or type 2 diabetes (T2D). Patients followed in the PMDC were compared to those receiving standard clinic care (Figure 1).
The outpatient clinic (OPC) is a training site for 60 internal medicine (IM) and 15 combined medicine and pediatric (Med-Peds) residents. These residents serve as primary care providers (PCPs) for patients. The OPC manages more than 800 patients with diabetes.
Prior to January 2015, 2 standard care options were available for diabetes management 1) PCPs managed their own patients with diabetes 2) referred patients to the multidisciplinary diabetes clinic (MDC). The MDC is comprised of an endocrinologist, medical residents, dieticians, and pharmacists. The focus is general diabetes care including physical exam, medication management, hypertension and lipid management, nephropathy screening, diabetic foot exams, and immunizations. Despite these services, the OPC had not been designated as a patient-centered medical home, thus identifying the need for improvement in ndiabetes-related HEDIS measures.
The PMDC was created in January 2015 with the goal of assisting the OPC in attaining improved HEDIS measures in high-risk patients (those with an A1C ≥9%). This created a third option for diabetes management. The PMDC works in tandem with the standard care that patients receive in the OPC. Pharmacists in the PMDC work under a collaborative practice agreement to conduct visits with patients. During appointments, patients are asked to describe gaps in knowledge and identify their own self-management goals (Figure 2). These goals serve as the foundation for follow up. Patient education is reviewed at appointments (Table 1). Review of topics typically occurs over multiple appointments, depending on the
Eligible patients had an A1C ≥9% between January 1, 2015, and September 30, 2015, and had to be ≥18 years old. Patients in the PMDC were identified through electronic health record (EHR) schedules. Standard clinic care patients were identified through a report that identified patients with diabetes assigned to a clinic PCP. Exclusion criteria for both groups were those that had not been seen by an OPC physician in 2015.
The following variables were collected: patient demographics, blood pressure (BP), and the following laboratory values: A1C, lipid panel, and microalbumin/creatinine ratio. Additionally, information regarding the patients’ medication profile was collected such as antidiabetic, anti-hypertensive, and lipid-lowering regimens. Finally, factors related to medical follow up, such as number of office visits and hospitalizations, were reviewed. All laboratory parameters were evaluated between 3 to 6 months and 6 to 9 months after the baseline A1C was measured in 2015.
The primary endpoints were the absolute change in A1C from baseline to 3 and 6 months and percent of patients who reached an A1C goal of less than 8% at either 3 or 6 months. If patients had readings at both 3 and 6 months, the most recent A1C was used. If patients only had readings at 3 months then that was included. Secondary endpoints included BP <140/90 mmHg measured at 3 or 6 months, current retinal or dilated eye exam, current nephropathy screening, and appropriateness of medication use with angiotensin-
converting-enzyme inhibitors (ACE-inhibitors) or angiotensin receptor blockers (ARBs) and statins. Patients were deemed to have appropriate use of medications if agents prescribed were indicated based upon current diabetes recommendations and agents not prescribed were contraindicated due to allergies or adverse effects.
Study sample size
All diabetes patients that were enrolled in PMDC between January 1, 2015, and September 30, 2015, were included. They were compared with diabetes patients that were receiving regular care, at a ratio of approximately 1:2.
Descriptive statistics were used for data collected and examined separately for both groups. Categorical variables were reported as counts and frequencies, and Pearson’s Chi-square tests or Fisher’s Exact tests were used. Continuous variables were examined for normality and non-parametric data. Wilcoxon tests were used for all continuous variables as they were not normally distributed.
A total of 110 patients were included (Figure 1); 36 patients were managed by the PMDC and 74 by standard clinic care. Patients in both groups were well matched in terms of baseline characteristics, with the exception of baseline A1C and insulin use (Table 2). PMDC patients had a higher median baseline A1C of 11.5% versus 10.6% (P = .033). Insulin use was 94.4% in the PMDC group versus 75.7% in the standard care group (P = .017).
Primary endpoints Figure 3 depicts the change in A1C for both groups. At 3 months, the A1C improved by 2.2 points in the PMDC group compared to 0.9 points in the standard group (P = .006). At 6 months, the A1C improved with 3.2 points in the PMDC group compared to 1.2 points in the standard group (P = .044). The percentage of patients with a follow-up A1C at 3 months was 67% in the PMDC group compared to 50% in the standard group (Table 3). At 6 months, 53% in the PMDC and 55% in the standard group had a follow-up A1C. Of the patients
with follow-up data, 50% in the PMDC and 30% in the standard group reached an A1C goal of <8% by 3 or 6 months. The difference between groups was not statistically significant (OR 2.31 [95% CI, .90, 5.95]; P = .08).
The difference between groups for the percent of patients reaching blood pressure goals or receiving eye exams was not significant (Table 4). A higher percentage of patients in the PMDC group had nephropathy screening (97.2%) compared to the standard group (79.7%, P = .015). Regarding medication appropriateness, 97.2% of patients in the PMDC group
and 100% of patients in the standard group were appropriately prescribed ACE-inhibitors or ARBs. Appropriate use of statins was seen in 94.4% in the PMDC group vs 91.9% in the standard group (P = 1.00) (Table 4).
PMDCs have proven to be advantageous in diabetes care.4-9 Similarly, our PMDC contributed to positive outcomes related to diabetes quality measures. Overall, patients in the PMDC had a 3.2 point decrease in A1C at 6 months. Although a significant change, the study did not show a statistically significant difference in the percent of patients reaching a A1C of <8%. The number of patients with available A1C data at 3 and 6 months (Table 3) was lower than desired in both groups. This may be related to the percentage of missed appointments (Table 5.) However, the missed appointment rate seen in the OPC was consistent with reported rates at other residency teaching clinics.10 The lack of data and the missed appointment rate may have impacted the observed difference between groups.
Pharmacists have been valuable members of the MDC for years. They provide recommendations for medication titration and patient education alongside the multidisciplinary group. The role of the pharmacist in the PMDC and MDC differ, which may explain the benefit seen with the PMDC. One difference is that the PMDC pharmacist conducts the full 30-60–minute appointment, whereas in the MDC the pharmacist spends approximately 15 minutes with the patient. This PMDC appointment may have allowed for more detailed
patient-centered diabetes education. In the PMDC, pharmacists encouraged patients to identify self-management goals and allowed patients to set the foundation for follow-up visits. The PMDC was comparable with the pharmacist-managed clinic reported by Kelly and Rodgers,6 who also incorporated a management plan with patient- and pharmacist-identified management goals.
While they were unable to show statistical significance in their A1C reduction compared to control, they did show a positive trend. As the state of the current healthcare system evolves, patient-centered care is crucial for improving chronic disease management. The percentage of patients with current retinopathy screening was lower than desired in both treatment groups. Although patients are frequently referred for diabetic retinopathy screening by their physicians or pharmacists, follow up was often low. Nephropathy screening was improved in the PMDC group compared to the standard group. This finding has also been supported in the literature.9
Pharmacists in the PMDC follow a specific checklist at the initial appointment to keep track of diabetes standards, such as nephropathy screening, which may have explained the difference between the groups.There were limitations to this study. At the evaluation time points of 3 and 6 months, a smaller amount of follow-up data was available. This may be related to the rate of missed appointments in general, but it also affects the percent of patients reaching the goal. If authors assumed that any patient without data did not reach the goal, the overall number of patients reaching the target A1C would be less. Additionally, as this study was a chart review, the results are dependent upon documentation in the institution’s charting system. Lastly, a snapshot method was used to determine if patients had reached their target BP—a practice that payers use when evaluating HEDIS measures. However, patients may have been at their target immediately before or after the snapshot value, and this may not accurately reflect overall BP control.
Based on this study, areas for process improvement were identified. To improve rates of retinopathy screening, clinic pharmacists are now working with the institution’s eye clinic to better streamline an appointment process for patients with diabetes. Efforts to schedule these annual retinopathy screenings immediately after a clinic visit were initiated. The eye clinic being located within the same building improved coordination of appointments, which may alleviate concerns of transportation, work-related time off, or other matters.
It was noted that many patients in the standard group were referred to the PMDC, however appointments to the PMDC were not made. Improving the referral process will also require coordination with clinic registration to ensure referrals made for the PMDC are carried out. Additional education to the medical residents on the PMDC referral process may improve this.
The addition of a PMDC had a positive effect on the change in A1C of patients with diabetes. A higher percentage of patients in the PMDC were able to reach a target A1C of <8%. Rates of nephropathy screening were also improved with the PMDC. Patients in both groups were ordered appropriate medications in regard to other diabetes standards of care. Overall, retinopathy screening in this patient population is open to improvements.