Abstract
Background: Literature has shown the positive impact pharmacists have on diabetic outcome measures through collaborative drug therapy management (CDTM). There is minimal literature evaluating characteristics and clinical factors of patients who benefit from CDTM diabetes clinics by pharmacists. Objective: Identify patient characteristics and clinical factors that may be associated with patients who reach goal hemoglobin A1c (HbA1c) of <7% at discharge by pharmacists practicing under CDTM agreements. Methods: This retrospective chart review included patients referred to pharmacist CDTM clinics for type 2 diabetes with an HbA1c goal of <7%. Data were extracted from the electronic medical record at enrollment and discharge. Results: Of the 228 patients included, 84 achieved a goal HbA1c of <7%. Factors predictive of patient success were Asian ethnicity (odds ratio [OR] = 19.32), baseline HbA1c of 7% to 7.9% (OR = 2.34), enrolled in clinic for 5 to 6 months (OR = 2.06), in-person visit every 4 to <8 weeks (OR = 3.06), not on insulin initially or at discharge (OR = 1.79, OR = 2.02), or discharged on a glucagon-like peptide-1 receptor agonist (OR = 1.83). Factors predictive of not reaching goal were Black or African American ethnicity (OR = 0.42), <5 encounters of any type (OR = 0.44), an encounter of any type every 8 weeks or more (OR = 0.08), or discharged on a sodium-glucose cotransporter-2 inhibitor (OR = 0.27). Conclusion: Several clinical and demographic factors were identified that influenced a patient’s ability to reach a goal HbA1c of <7%. The results of this study may be applied to further advance pharmacist-run clinics in optimizing diabetes care for patients.
Keywords: clinical pharmacy, ambulatory care, type 2 diabetes, clinical practice, disease management, medication therapy management
Introduction
The Centers for Disease Control and Prevention estimated in 2018 that 34.2 million people, or 10.5% of the US population, had diabetes; that number is expected to climb to more than 60 million by 2060.1,2 Diabetes may lead to various comorbidities such as cardiovascular disease, chronic kidney disease, lower extremity amputations, and diabetic retinopathy.3,4 It is estimated that half of people with diabetes have a hemoglobin A1c (HbA1c) value of 7% or higher, and 27.7% have a value of 8% or higher, increasing their risk of development and progression of microvascular complications.1
Given the challenges with managing patients with diabetes, interprofessional teams, including pharmacists, often provide care for these patients.5-17 Pharmacist may utilize collaborative drug therapy management (CDTM) agreements, or similar practices, to improve patient health outcomes and reduce costs for health care systems.5-17 CDTM agreements are often created between a health care provider (typically a physician) and pharmacist to apply evidence-based medicine guidelines for pharmacists to care for patients. CDTM agreements may also provide guidance for pharmacists to initiate, adjust, and/or monitor drug regimens for numerous chronic diseases (eg, diabetes mellitus, hyperlipidemia, hypertension, and depression).18-20 As pharmacist-run diabetes clinics become more prevalent, it is important for clinics to optimize the care for their patients by understanding patient characteristics and clinical factors that are associated with reaching HbA1c goals.18-20 There remains a lack of literature evaluating patient characteristics and clinical factors that may be associated with one’s ability to reach a goal HbA1c from CDTM diabetes clinics by pharmacists.
At the time of publication there are currently 2 available studies that evaluate specific characteristics in achieving HbA1c goals with pharmacist-managed diabetes clinics.15,16 However, both studies lacked CDTM agreements and were limited in the demographic information collected.15,16
Methods
This study was conducted at a multicenter, not-for-profit, independent, physician-led health care organization, Atrius Health, located within eastern Massachusetts. This health care system provides care for more than 745 000 adults who are seen by 715 primary care providers (PCPs) with an additional 425 specialized clinics. Among the specialized clinics, pharmacists provided CDTM services at 15 different sites. Each pharmacist had a Doctor of Pharmacy degree and was credentialed to provide CDTM by the Clinical Pharmacy Department. A CDTM agreement between a supervising physician and the pharmacist at the individual practice sites allowed patients referred to the pharmacist for diabetes, hypertension, and/or hyperlipidemia, to have their therapies initiated, modified, or discontinued. At the discretion of the referred pharmacist, patients were discharged from the CDTM clinic if 1 of 4 criteria were met: (1) patient reached goal, (2) patient requested discharge, (3) a plan was in place and the patient was able to receive disease management through PCP or specialty, or (4) patient refused to adhere and/or fails to show to more than 2 consecutive office visits.
This retrospective chart review was completed using the electronic medical record (EMR) for patients enrolled and discharged by a pharmacist with a CDTM agreement for type 2 diabetes mellitus (DM2). Patients included were at least 18 years old, enrolled and discharged between January 1, 2017, and January 1, 2019, diagnosed with DM2, had a documented HbA1c goal of <7% at the enrollment, and had an HbA1c of >7% within 6 months of enrollment. Patients were excluded if they did not have DM2, had an HbA1c of ≤7% within 6 months of the enrollment, had a documented HbA1c goal of >7% at enrollment, had not completed an HbA1c within 6 months of enrollment and discharge, or were being managed by an endocrinologist within the organization for their diabetes care at any point ≤1 year prior to enrollment.
For all patients, data were extracted from the EMR with automated or manual reports based on the patient’s date of enrollment and discharge. Each patient’s diabetes care and goal HbA1c was set at the discretion of the referred to pharmacist in accordance with present-time diabetes clinical guidelines set by the American Diabetes Association.21 Demographics collected included gender, age, ethnicity, marital status, primary language, medical insurance, and preferred pharmacy. Clinical and clinic-related data collected included HbA1c (within 6 months of enrollment and discharge; if multiple values were available, the closest to the date of enrollment or discharge was used), duration of diabetes, baseline body mass index, comorbidities, duration of enrollment, number of CDTM face-to-face appointments with the pharmacist (defined as scheduled, in-person visits with the pharmacist), and number of follow-ups via telephone and/or electronic messaging system with the pharmacist (extracted based on encounter type providing outreach between patient and pharmacist without appointments), and diabetes medications (medications used prior to enrollment, active at enrollment, and active on discharge). Considering patients had different lengths of enrollment in clinic, the total number of pharmacist-patient encounters was divided by the duration of enrollment in weeks to allow for a standard of comparison (encounter ratio). The primary objective of this study was to identify patient characteristics and clinical factors that may be associated with patients who were able to reach a goal HbA1c of <7% at discharge by pharmacists practicing under CDTM agreements. The recommended goal of <7% is in accordance with the American Diabetes Association guidelines for most patients.21 This study was approved by the MCPHS University and Atrius Health Institutional Review Boards.
Statistics
Patient-specific clinical and demographic factors assessed were summarized using descriptive statistics. Logistic regression analysis with odds ratios (ORs) was used to determine which of these clinical and demographic factors significantly correlate to the goal of an HbA1c of <7% at discharge. Results were considered statistically significant if the observed level of significance is P < .05. All analyses were performed using NCSS (Number Cruncher Statistical System) 2007.22
Results
A total of 419 patients were enrolled, and subsequently discharged by the clinical pharmacists between January 1, 2017, and January 1, 2019; 228 of them met the inclusion criteria. Of those excluded, more than half (58.6%) of patients were excluded because their goal HbA1c was >7% (n = 112) at enrollment. Of those included, 84 of the 228 patients (36.9%) achieved their goal HbA1c of <7% within 6 months before or after discharge, and 144 of the 228 patients (63.1%) did not reach a goal HbA1c of <7% (Figure 1).
Figure 1.
Study flowchart.
There was no statistical difference found between the groups with regard to gender, age, marital status, preferred primary language, use of other specialty providers, body mass index at time of enrollment, duration of diabetes, comorbidities at enrollment, type of medical insurance, and preferred pharmacy (Tables 1-3).
Table 1.
Demographics.
| Final HbA1c <7% | P | OR | 95% CI | ||
|---|---|---|---|---|---|
| Yes, N = 84, n (%) | No, N = 144, n (%) | ||||
| Gender | |||||
| Male | 45 (53.6) | 95 (66.0) | .068 | 0.60 | 0.34-1.03 |
| Female | 39 (46.4) | 49 (34.3) | .068 | 1.00 | Reference |
| Age (years) classification I | |||||
| <65 | 25 (29.9) | 43 (29.9) | .999 | 1.00 | Reference |
| ≥65 | 59 (70.2) | 101 (70.1) | .999 | 1.00 | 0.55-1.79 |
| Age (years) classification II | |||||
| 30-39 | 3 (3.6) | 5 (3.5) | .999 | 1.03 | 0.24-4.42 |
| 40-49 | 7 (8.3) | 16 (11.1) | .650 | 0.73 | 0.29-1.85 |
| 50-59 | 31 (36.9) | 55 (38.2) | .888 | 0.95 | 0.54-1.65 |
| 60-69 | 33 (39.3) | 50 (34.7) | .568 | 1.22 | 0.70-2.12 |
| 70-79 | 10 (11.9) | 18 (12.5) | .999 | 0.95 | 0.41-2.16 |
| Ethnicity | |||||
| White | 52 (61.9) | 78 (54.2) | .270 | 1.38 | 0.79-2.38 |
| Black/African American | 13 (15.5) | 44 (30.6) | .011a | 0.42 | 0.21-0.83 |
| Hispanic/Latino | 6 (7.1) | 10 (6.9) | .999 | 1.03 | 0.36-2.95 |
| Asian | 10 (11.9) | 1 (0.7) | .001a | 19.32 | 2.43-153.86 |
| Other | 2 (2.4) | 6 (4.2) | .713 | 0.56 | 0.11-2.84 |
| Unknown | 1 (1.2) | 5 (3.5) | .418 | 0.33 | 0.04-2.92 |
| Marital status | |||||
| Single | 18 (21.4) | 24 (16.7) | .381 | 1.36 | 0.69-2.69 |
| Married | 49 (58.3) | 76 (52.8) | .491 | 1.25 | 0.73-2.16 |
| Divorced/separated | 3 (3.6) | 12 (8.3) | .267 | 0.41 | 0.11-1.49 |
| Other | 2 (2.4) | 11 (7.6) | .140 | 0.29 | 0.06-1.36 |
| Unknown | 12 (14.3) | 21 (14.6) | .999 | 0.98 | 0.45-2.10 |
| Primary language | |||||
| English | 79 (94.0) | 136 (94.4) | .999 | 0.93 | 0.29-2.94 |
| Other | 5 (6.0) | 8 (5.6) | .999 | 1.00 | Reference |
| Medical insurance | |||||
| Medicaid | 1 (1.2) | 2 (1.4) | .999 | 0.86 | 0.08-9.58 |
| Medicare | 23 (27.4) | 33 (22.9) | .524 | 1.27 | 0.68-2.35 |
| Private | 58 (69.0) | 102 (70.8) | .881 | 0.92 | 0.51-1.65 |
| Self-pay | 1 (1.2) | 3 (2.1) | .999 | 0.57 | 0.06-5.53 |
| Unknown | 1 (1.2) | 4 (2.8) | .999 | 0.42 | 0.05-3.84 |
| Preferred pharmacy | |||||
| Chain | 53 (63.1) | 75 (52.1) | .128 | 1.57 | 0.91-2.73 |
| Independent | 0 (0.0) | 1 (0.7) | .999 | NA | NA |
| HVMA | 25 (29.8) | 47 (32.6) | .768 | 0.87 | 0.49-1.57 |
| Mail order | 4 (4.8) | 16 (11.1) | .145 | 0.40 | 0.13-1.24 |
| Other | 2 (2.4) | 5 (3.5) | .999 | 0.68 | 0.13-3.57 |
Abbreviations: CI, confidence interval; HbA1c, hemoglobin A1c; HVMA, Harvard Vanguard Medical Associates; NA, not available; OR, odds ratio.
Statistically significant, P < .05.
Table 2.
Clinical and Clinic-Related Factors.
| Final HbA1c < 7% | P | OR | 95% CI | ||
|---|---|---|---|---|---|
| Yes, N = 84, n (%) | No, N = 144, n (%) | ||||
| Baseline HbA1c (%) | |||||
| 7.0-7.9 | 19 (22.6) | 16 (11.1) | .023a | 2.34 | 1.13-4.85 |
| 8.0-8.9 | 29 (34.5) | 53 (36.8) | .776 | 0.91 | 0.52-1.59 |
| 9.0-9.9 | 14 (16.7) | 37 (25.7) | .139 | 0.58 | 0.29-1.15 |
| 10.0-10.9 | 14 (16.7) | 18 (12.5) | .431 | 1.40 | 0.66-2.98 |
| 11.0-11.9 | 5 (6.0) | 11 (7.6) | .790 | 0.77 | 0.26- 2.28 |
| 12.0-12.9 | 2 (2.4) | 3 (2.1) | .999 | 0.28 | 0.03-2.34 |
| ≥13.0 | 1 (1.2) | 6 (4.2) | .265 | 0.28 | 0.03-2.34 |
| Duration of diabetes (years) | |||||
| <1 | 8 (9.5) | 6 (4.2) | .104 | 2.42 | 0.81-7.24 |
| 2-4 | 13 (15.5) | 23 (16.0) | .921 | 0.96 | 0.46-2.02 |
| 5-9 | 27 (32.1) | 40 (27.8) | .485 | 1.23 | 0.69-2.21 |
| 10-14 | 15 (17.9) | 32 (22.2) | .432 | 0.76 | 0.38-1.51 |
| 15-19 | 13 (15.5) | 22 (15.3) | .968 | 1.02 | 0.48-2.14 |
| ≥20 | 4 (4.8) | 15 (10.4) | .136 | 0.43 | 0.14-1.34 |
| Unknown | 4 (4.8) | 6 (4.2) | .832 | 1.15 | 0.32-4.20 |
| Baseline BMI (kg/m2) | |||||
| Normal (18.50-24.99) | 6 (7.1) | 10 (6.9) | .999 | 1.03 | 0.36-2.95 |
| Overweight (25.00-29.99) | 21 (25.0) | 38 (26.4) | .876 | 0.93 | 0.50-1.72 |
| Obese (30.00-34.99) | 26 (31.0) | 43 (29.9) | .882 | 1.05 | 0.59-1.89 |
| Class II obese (35.00-35.99) | 18 (21.4) | 23 (16.0) | .372 | 1.43 | 0.72-2.85 |
| Class III obese (≥40.00) | 13 (15.5) | 30 (20.8) | .382 | 0.70 | 0.34-1.42 |
| Comorbidities at enrollment | |||||
| Dyslipidemia | 62 (73.8) | 117 (81.3) | .242 | 0.65 | 0.34-1.24 |
| Heart failure | 3 (3.6) | 1 (0.7) | .143 | 5.30 | 0.54-51.76 |
| Hypertension | 54 (64.3) | 109 (75.7) | .070 | 0.58 | 0.32-1.04 |
| Asthma | 16 (19.0) | 18 (12.5) | .184 | 1.65 | 0.79-3.44 |
| COPD | 2 (2.4) | 4 (2.8) | .999 | 0.85 | 0.15-4.76 |
| Clinical ASCVD | 4 (4.8) | 18 (12.5) | .065 | 0.35 | 0.11-1.07 |
| CKD | 8 (9.5) | 19 (13.2) | .525 | 0.69 | 0.29-2.58 |
| OSA | 13 (15.5) | 19 (13.2) | .694 | 1.20 | 0.56-2.58 |
| Insomnia | 2 (2.4) | 4 (2.8) | .999 | 0.85 | 0.15-4.76 |
| Duration of enrollment (months) | |||||
| <1 | 0 (0.0) | 3 (2.1) | .299 | NA | NA |
| 1-2 | 9 (10.7) | 10 (6.9) | .320 | 1.61 | 0.63-4.13 |
| 3-4 | 16 (19.0) | 26 (18.1) | .852 | 1.07 | 0.54-2.13 |
| 5-6 | 20 (23.8) | 19 (13.2) | .040a | 2.06 | 1.02-4.13 |
| 7-8 | 12 (14.3) | 20 (13.9) | .934 | 1.03 | 0.48-2.24 |
| 9-10 | 9 (10.7) | 14 (9.7) | .810 | 1.11 | 0.46-2.70 |
| 11-12 | 5 (6.0) | 15 (10.4) | .250 | 0.54 | 0.19-1.56 |
| >12 | 13 (15.5) | 27 (25.7) | .072 | 0.53 | 0.26-1.07 |
| Total number of encountersb | |||||
| <5 | 11 (13.1) | 37 (25.7) | .028a | 0.44 | 0.21-0.91 |
| 5-9 | 40 (47.6) | 67 (46.5) | .891 | 1.04 | 0.61-1.79 |
| 10-14 | 24 (28.6) | 27 (18.8) | .100 | 1.73 | 0.92-3.26 |
| 15-19 | 0 (0.0) | 0 (0.0) | .999 | NA | NA |
| 20-24 | 7 (8.3) | 9 (6.3) | .596 | 1.36 | 0.49-3.81 |
| ≥25 | 2 (2.4) | 4 (2.8) | .999 | 0.85 | 0.15-4.76 |
| Weeks per 1 encounterb | |||||
| ≥8 | 1 (1.2) | 18 (12.5) | .002a | 0.08 | 0.01-0.64 |
| 4 to <8 | 29 (34.5) | 57 (39.6) | .481 | 0.80 | 0.46-1.41 |
| 3 to <4 | 22 (26.2) | 33 (22.9) | .631 | 1.19 | 0.64-2.22 |
| 2 to <3 | 13 (15.5) | 13 (9.0) | .194 | 1.85 | 0.81-4.19 |
| 1.5 to <2 | 6 (7.1) | 9 (6.3) | .788 | 1.15 | 0.32-4.20 |
| Weeks per 1 office visit | |||||
| ≥8 | 36 (42.9) | 93 (64.6) | .002a | 0.41 | 0.24-0.71 |
| 4 to <8 | 40 (47.6) | 33 (22.9) | .001a | 3.06 | 1.72-5.45 |
| 3 to <4 | 4 (4.8) | 11 (7.6) | .581 | 0.60 | 0.19-1.96 |
| 2 to <3 | 1 (1.2) | 5 (3.5) | .418 | 0.33 | 0.04-2.92 |
| 1.5 to <2 | 1 (1.2) | 2 (1.4) | .999 | 0.86 | 0.08-9.58 |
Abbreviations: ASCVD, atherosclerotic cardiovascular disease; BMI, body mass index; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CI, confidence interval; HbA1c, hemoglobin A1c; NA, not available; OR, odds ratio; OSA, obstructive sleep apnea.
Statistically significant, P < .05.
Encounter included the following: in-person office visits, telephonic and electronic messaging interactions.
Table 3.
Diabetes Medicationsa.
| Final HbA1c < 7% | P | OR | 95% CI | ||
|---|---|---|---|---|---|
| Yes, N = 84, n (%) | No, N = 144, n (%) | ||||
| Biguanide | |||||
| Previous trial | 10 (11.9) | 23 (16.0) | .400 | 0.71 | 0.32-1.58 |
| Enrollment | 69 (82.1) | 111 (77.1) | .366 | 1.37 | 0.69-2.70 |
| Discharge | 74 (88.1) | 119 (82.6) | .270 | 1.55 | 0.71-3.42 |
| Sulfonylurea | |||||
| Previous trial | 14 (16.7) | 29 (20.1) | .518 | 0.79 | 0.39-1.60 |
| Enrollment | 43 (51.2) | 67 (46.5) | .497 | 1.21 | 0.70-2.07 |
| Discharge | 24 (28.6) | 46 (31.9) | .594 | 0.85 | 0.47-1.54 |
| TZD | |||||
| Previous trial | 6 (7.1) | 12 (8.3) | .748 | 0.85 | 0.31-2.34 |
| Enrollment | 2 (2.4) | 1 (0.7) | .281 | 3.49 | 0.31-39.06 |
| Discharge | 4 (4.8) | 1 (0.7) | .063 | 7.15 | 0.79-65.07 |
| DPP-IV | |||||
| Previous trial | 2 (2.4) | 6 (4.2) | .480 | 0.56 | 0.11-2.84 |
| Enrollment | 8 (9.5) | 12 (8.3) | .759 | 1.16 | 0.45-2.96 |
| Discharge | 3 (3.6) | 8 (5.6) | .500 | 0.63 | 0.16-2.44 |
| GLP-1 RA | |||||
| Previous trial | 4 (4.8) | 6 (4.2) | .832 | 1.15 | 0.32-4.20 |
| Enrollment | 6 (7.1) | 15 (10.4) | .410 | 0.66 | 0.25-1.78 |
| Discharge | 47 (56.0) | 59 (41.0) | .029 | 1.83 | 1.06-3.15 |
| SGLT2 inhibitors | |||||
| Previous trial | 2 (2.4) | 2 (1.4) | .582 | 1.73 | 0.24-12.53 |
| Enrollment | 1 (1.2) | 4 (2.8) | .430 | 0.42 | 0.05-3.84 |
| Discharge | 5 (6.0) | 27 (18.8) | .007b | 0.27 | 0.10-0.74 |
| Insulin regimen at enrollment | |||||
| Not on insulin | 60 (71.4) | 84 (58.3) | .048b | 1.79 | 1.01-3.18 |
| Basal only | 20 (23.8) | 41 (28.5) | .535 | 0.79 | 0.42-1.46 |
| Bolus only | 0 (0.0) | 0 (0.0) | .999 | NA | NA |
| Basal + bolus | 3 (3.6) | 11 (7.6) | .264 | 0.45 | 0.12-1.65 |
| Premix | 0 (0.0) | 7 (4.9) | .049b | NA | NA |
| Insulin regimen at discharge | |||||
| Not on insulin | 63 (75.0) | 86 (59.7) | .021b | 2.02 | 1.12-3.67 |
| Basal only | 20 (23.8) | 45 (31.3) | .287 | 0.69 | 0.37-1.27 |
| Bolus only | 0 (0.0) | 0 (0.0) | .999 | NA | NA |
| Basal + bolus | 0 (0.0) | 9 (6.3) | .028b | NA | NA |
| Premix | 0 (0.0) | 6 (4.2) | .088 | NA | NA |
Abbreviations: CI, confidence interval; DPP-IV, dipeptidyl peptidase 4 inhibitor; GLP1-RA, glucagon-like peptide-1 receptor agonist; HbA1c, hemoglobin A1c; NA, not available; OR, odds ratio; SGLT2, sodium-glucose cotransporter-2; TZD, thiazolidinedione.
Data presented as n (%).
Statistically significant, P < .05.
Several clinical and demographic factors were identified that influenced a patient’s ability to reach a goal HbA1c of <7% at discharge. Patients who were identified as Asian (n = 11) were more likely to achieve HbA1c goal (11.9% vs 0.7%; OR = 19.32; 95% confidence interval [CI] = 2.43-153.86); however, those who were identified as Black or African American (n = 57) were less likely to achieve HbA1c goal (15.5% vs 30.6%; OR = 0.42; 95% CI = 0.21-0.83). Patients with a baseline HbA1c between 7% and 7.9% (n = 35) were more likely to achieve HbA1c goal (22.6% vs 11.1%; OR = 2.34; 95% CI = 1.13-4.85). It also appeared that patients who were enrolled in the CDTM clinics for 5 to 6 months (n = 39) were more likely to achieve HbA1c goal (23.8% vs 13.2%; OR = 2.06; 95% CI = 1.02-4.13). When evaluating the number of in-person encounters alone, patients seen in office every 4 to <8 weeks (n = 73) were more likely to achieve HbA1c goal (47.6% vs 22.9%; OR = 3.06; 95% CI = 1.72-5.45), but those with any type of encounter (in-person or via telephone/electronic messaging) every 8 weeks or more (n = 19) or in-person only (n = 129) were less likely to achieve HbA1c goal (1.2% vs 12.5%; OR = 0.08; 95% CI = 0.01-0.64 and 42.9% vs 64.6%; OR = 0.41; 95% CI = 0.24-0.71) and those with <5 encounters of any type (n = 48) were less likely to achieve HbA1c goal (13.1% vs 25.7%; OR = 0.44; 95% CI = 0.21-0.91). There were also differences seen when assessing diabetes-specific medications. Patients who were not on insulin at enrollment (n = 144) and those not on insulin at discharge (n = 149) were more likely to achieve HbA1c goal (71.4% vs 58.3%; OR = 1.79; 95% CI = 1.01-3.18 and 75.0% vs 59.7%; OR = 2.02; 95% CI = 1.12-3.67). Patients discharged on a glucagon-like peptide-1 receptor agonist (n = 106) were also more likely to achieve HbA1c goal (56.0% vs 41.0%; OR = 1.83; 95% CI = 1.06-3.15). Patients discharged on a sodium-glucose cotransporter-2 (SGLT2) inhibitor (n = 32) were less likely to achieve HbA1c goal (6% vs 18.8%; OR = 0.27; 95% CI = 0.10-0.74).
Discussion
In this retrospective study of patients seen by a pharmacist for DM2 under CDTM services, several clinical and demographic factors were identified that influence a patient’s ability to reach a goal HbA1c of <7% at discharge.
Ethnicity appeared to be a major driving factor that indicated the rate of achievement of HbA1c goal. According to the Centers for Disease Control and Prevention, Asians are the second largest group with diabetes; the results of this study help support the importance of pharmacist care for this rising population.1 Black or African American patients represent the race most likely to have diabetes; however, they were less likely to achieve their goal.1 Similar results were seen with Price-Haywood et al11 showing less of a decline of HbA1c in Black patients compared with White.11 These findings may suggest that, in addition to seeing a pharmacist, some patients may need different approaches to diabetes care due to differences in patient-identified ethnic groups, cultural differences, economic status, education level, and/or self-efficacy and management.23 While these barriers are not unique to patients pharmacists care for, pharmacists and other health care professionals may begin to increase clinical and behavioral education and training to overcome these barriers.23,24
An interesting finding in this study showed a positive correlation in patients whose baseline HbA1c was 7% to 7.9%. This is particularly helpful as this cohort of patients represents the largest among those with uncontrolled diabetes.1 By comparison, Lam et al16 found that patients with a baseline HbA1c >9% were more likely to achieve HbA1c goal. The difference in this outcome may be explained by the variation in how each study defined achievement of HbA1c goal. Lam et al16 used a decrease in HbA1c of 2% or more as one of the methods for clinical goal in addition to reaching <8% HbA1c. When this method is applied to this study, similar results are noted: patients with a baseline HbA1c of 10% to 10.9% (n = 32) were more likely to have a decrease of 2% or more. In our study, similar to Kostoff et al,15 higher baseline HbA1c did not correlate to achieving goal HbA1c. Of note, the authors would like to address the difference in primary outcome of this study is based on achieving clinical guideline–guided HbA1c goal of <7%. This target of stricter glycemic control to an HbA1c of 7% has shown in studies to reduce microvascular complications such as retinopathy, neuropathy, and diabetic kidney disease.3,4 Therefore, the goal of <7% used in this study may more accurately associate to reduction in microvascular diabetes complications seen in clinical trials.3,4
A particularly interesting finding in this study was patients who were discharged from CDTM services at 5 to 6 months were more likely to achieve HbA1c goal. The results of this finding may be extrapolated to predict time to reach goal HbA1c. This may assist clinics in managing patient population size and pharmacist scheduling. This finding may be corroborated by the results seen by Anaya et al7 that identified the majority of HbA1c reduction in patients seeing the pharmacist occurred within 3 to 6 months of enrollment, compared with PCP alone.7 Importantly, the reasons for patient discharge may have affected these results outside of achieving HbA1c goal. Specifically, many patients were discharged prior to reaching goal HbA1c at the discretion of the individual pharmacist if a plan was in place and the patient was able to continue disease management through PCP or specialist. It was anticipated that these patients would reach HbA1c goal, but after discharge from pharmacist follow-up. And last, this result may be affected by those patients who may have self-discharged from service before reaching goal. Patients who had <5 encounters of any type with the clinical pharmacist, the patient was less likely to reach HbA1c goal. These patients may have been those who refused to adhere to recommendations, failed to show to more than 2 consecutive office visits, or requested discharge early. Interestingly, for patients who had >5 visits, all were equally as likely to achieve goal.
Reviewing the calculated encounter ratios, patients who had 1 in-person visit every 4 to <8 weeks were more likely to achieve HbA1c goal. Oddly, this trend was not replicated in patients who had any encounter once every 4 to <8 weeks. This may imply that patients do better when seen face-to-face for pharmacist visits in diabetes management, or the need for advancements in electronic communication.25,26 These findings may also help predict the ideal amount of time a pharmacist should schedule follow-up with patients. Furthermore, if the number of weeks per in-person visits decreased no significance was found. This result may be caused by patients needing more frequent follow up for issues with medication costs, less motivation, and/or poor adherence to recommendations.
Findings indicated specific diabetes medication regimens were correlated with achievement of HbA1c goal. Similar to what was found in the studies by Kostoff et al15 and Lam et al16, patients who were not on any insulin at the enrollment or at discharge were more likely to achieve HbA1c goal. This may indicate patients already on insulin have an increased difficulty reaching goal HbA1c, despite insulin being an effective glycemic-lowering agent. Additionally, when patients were discharged on GLP-1 RA, they were more likely to achieve HbA1c goal. GLP-1 RAs, when compared with other oral DM2 medications such as dipeptidyl peptidase 4 inhibitors, sulfonylureas, SGLT2 inhibitors, and thiazolidinediones are at least as effective in terms of glycemic lowering, and similar (though may not be as good) when compared with insulin. Additionally, benefits include cardiovascular risk reduction, weight loss, and options with once weekly dosing.27-29 These advantages may relate to improved medication adherence and potential reasons why patients were more likely to achieve goal HbA1c. Conversely, patients discharged on SGLT2 inhibitors were less likely to achieve HbA1c goal. Since their glycemic benefit is not as large, this may explain why these patients were less likely to achieve HbA1c goal. However, despite the smaller glycemic-lowering potential, these agents are frequently used for their oral administration and other non-glycemic benefits.29
The lack of significance among comorbidities may be due to the sizeable cohort excluded whose HbA1c goal was >7%, as patients may have higher HbA1c goals due to multimorbidity, possibly inhibiting identifying clinical significance.21 This limitation may explain the low volume of patients with atherosclerosis cardiovascular disease and chronic kidney disease, which are more common combination comorbidities among DM2 patients in the United States.30 The authors recognize several other limitations regarding the design of this study. Most notably, this study was performed retrospectively and was limited to collecting clinical and demographic factors from the EMR without assessment of medication adherence, adverse drug reactions, and/or self-motivation. This also restricted access to external data from outside clinical practices where patients may have been receiving care. Additionally, each clinical and demographic factor was evaluated in a univariate analysis, a multivariate regression was not conducted. This inhibited the statistical analysis from identifying correlations within specific statistically significant findings; therefore, the authors believe future studies are needed to consider such evaluations. Of the clinical and demographic factors collected, frequency of patient follow-up with other members of their PCP team was not collected and it is unclear how this may have affected the primary outcome. Another limitation of this study was the design of the inclusion and exclusion criteria. Although all patients started with an HbA1c goal of <7%, some patients’ goals may have changed during enrollment, if this was deemed appropriate by the pharmacist. Furthermore, this study did not include a comparator group with usual care, which may have improved the study design, validity, and assist in understanding why more patients did not reach HbA1c goal. And last, this study was performed solely within an integrated health system located within eastern Massachusetts. This may limit the applicability of applying these results to other CDTM clinics. The authors hope these limitations will help structure future study designs.
The authors would like to address the depicted low rate of patients reaching HbA1c goal of <7% since this study was not designed to identify if pharmacist can assist patients in reaching HbA1c goals. Notably, the average HbA1c reduction of all patients in this study was 1.6%. After removing patients discharged for refusing to adhere and/or failed to show to more than 2 consecutive office visits, the average HbA1c reduction was 1.9%. Additionally, 62 of 144 patients who did not reach an HbA1c of <7% were discharged from the practice with a plan in place and able to receive disease management through PCP or specialty.
Conclusion
This retrospective chart review suggests there are various patient-specific clinical and demographic factors with positive and negative associations to reaching goal HbA1c of <7% in CDTM clinics. Despite several limitations affecting the external and internal validity of this study, the results will intend to direct future prospective studies in further identifying factors associated with achievement of HbA1c goal in CDTM clinics. Until then, the results of this study may be extrapolated to further assist CDTM clinics to further improve care for all patients seen in CDTM clinics.
Acknowledgments
To Seth Robin, MBA, Director of Pharmacy Finance and Business Strategy, Luis Carpio, MBA, Data Analyst III at Atrius Health for retrieving data, and Gary Tataronis, MS, Associate Professor of Statistics and Mathematics at MCPHS University for employing statistical methods.
Footnotes
Declaration of Conflicting Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iD: Stefanos Torkos 
https://orcid.org/0000-0002-5011-9072
References
- 1. Centers for Disease Control and Prevention. National Diabetes Statistics Report, 2020. Estimates of Diabetes and Its Burden in the United States. Centers for Disease Control and Prevention; 2020. [Google Scholar]
- 2. Lin J, Thompson TJ, Cheng YJ, et al. Projection of the future diabetes burden in the United States through 2060. Popul Health Metr. 2018;16:9. doi: 10.1186/s12963-018-0166-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group [published correction appears in Lancet 1999;354(9178):602]. Lancet. 1998;352:837-853. [PubMed] [Google Scholar]
- 4. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). UK Prospective Diabetes Study (UKPDS) Group. Lancet. 1998;352:854-865. [PubMed] [Google Scholar]
- 5. Aneese NJ, Halalau A, Muench S, Shelden D, Fett J, Lauster C. Impact of a pharmacist-managed diabetes clinic on quality measures. Am J Manag Care. 2018;24(4 spec no.)SP116-SP119. [PubMed] [Google Scholar]
- 6. Benedict AW, Spence MM, Sie JL, et al. Evaluation of a pharmacist-managed diabetes program in a primary care setting within an integrated health care system. J Manag Care Spec Pharm. 2018;24:114-122. doi: 10.18553/jmcp.2018.24.2.114 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Anaya JP, Rivera JO, Lawson K, Garcia J, Luna J, Jr, Ortiz M. Evaluation of pharmacist-managed diabetes mellitus under a collaborative drug therapy agreement. Am J Health Syst Pharm. 2008;65:1841-1845. doi: 10.2146/ajhp070568 [DOI] [PubMed] [Google Scholar]
- 8. Collier IA, Baker DM. Implementation of a pharmacist-supervised outpatient diabetes treatment clinic. Am J Health Syst Pharm. 2014;71:27-36. doi: 10.2146/ajhp130200 [DOI] [PubMed] [Google Scholar]
- 9. McAdam-Marx C, Dahal A, Jennings B, Singhal M, Gunning K. The effect of a diabetes collaborative care management program on clinical and economic outcomes in patients with type 2 diabetes. J Manag Care Spec Pharm. 2015;21:452-468. doi: 10.18553/jmcp.2015.21.6.452 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kislan MM, Bernstein AT, Fearrington LR, Ives TJ. Advanced practice pharmacists: a retrospective evaluation of the efficacy and cost of Clinical Pharmacist Practitioners managing ambulatory Medicare patients in North Carolina (APPLE-NC). BMC Health Serv Res. 2016;16:673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Price-Haywood EG, Amering S, Luo Q, Lefante JJ. Clinical pharmacist team-based care in a safety net medical home: facilitators and barriers to chronic care management. Popul Health Manag. 2017;20:123-131. doi: 10.1089/pop.2015.0177 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ko JJ, Lu K, Rascati K, et al. Analysis of glycemic control of a pharmacist-led medication management program in patients with type 2 diabetes. J Manag Care Spec Pharm. 2016;22:32-37. doi: 10.18553/jmcp.2016.22.1.32 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Wang Y, Yeo QQ, Ko Y. Economic evaluations of pharmacist-managed services in people with diabetes mellitus: a systematic review. Diabet Med. 2016;33:421-427. doi: 10.1111/dme.12976 [DOI] [PubMed] [Google Scholar]
- 14. Fazel MT, Bagalagel A, Lee JK, Martin JR, Slack MK. Impact of diabetes care by pharmacists as part of health care team in ambulatory settings: asystematic review and meta-analysis. Ann Pharmacother. 2017;51:890-907. doi: 10.1177/1060028017711454 [DOI] [PubMed] [Google Scholar]
- 15. Kostoff MD, Boros ML, Moorman JM, Frazee LA. Evaluation of factors associated with achieving glycemic control in a pharmacist-managed diabetes clinic. Am J Ther. 2014;21:234-239. doi: 10.1097/MJT.0b013e31826fc47e [DOI] [PubMed] [Google Scholar]
- 16. Lam SW, Russo-Alvarez G, Cristiani C, Rothberg MB. Development of an ambulatory clinical pharmacy prioritization prediction model for patients with diabetes. Ann Pharmacother. 2017;51:33-38. doi: 10.1177/1060028016665122 [DOI] [PubMed] [Google Scholar]
- 17. Wu WC, Taveira TH, Jeffery S, et al. Costs and effectiveness of pharmacist-led group medical visits for type-2 diabetes: a multi-center randomized controlled trial. PLoS One. 2018;13:e0195898. doi: 10.1371/journal.pone.0195898 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Ourth H, Groppi J, Morreale AP, Quicci-Roberts K. Clinical pharmacist prescribing activities in the Veterans Health Administration. Am J Health Syst Pharm. 2016;73:1406-1415. doi: 10.2146/ajhp150778 [DOI] [PubMed] [Google Scholar]
- 19. Mishra P, Thomas J., 3rd Survey of collaborative drug therapy management in US hospitals. Am J Health Syst Pharm. 2017;74:1791-1905. doi: 10.2146/ajhp151058 [DOI] [PubMed] [Google Scholar]
- 20. American College of Clinical Pharmacy; McBane SE, Dopp AL, et al. Collaborative drug therapy management and comprehensive medication management—2015. Pharmacotherapy. 2015;35:e39-e50. doi: 10.1002/phar.1563 [DOI] [PubMed] [Google Scholar]
- 21. American Diabetes Association. 1. Improving care and promoting health in populations: Standards of Medical Care in Diabetes—2020. Diabetes Care. 2020;43(suppl 1):S7-S13. doi: 10.2337/dc20-S001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Hintze JL. NCSS (Number Cruncher Statistical System) software 2007. NCSS; 2007. [Google Scholar]
- 23. Walker RJ, Williams JS, Egede LE. Influence of race, ethnicity and social determinants of health on diabetes outcomes. Am J Med Sci. 2016;351:366-373. doi: 10.1016/j.amjms.2016.01.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Manolakis PG, Skelton JB. Pharmacists’ contributions to primary care in the United States collaborating to address unmet patient care needs: the emerging role for pharmacists to address the shortage of primary care providers. Am J Pharm Educ. 2010;74:S7. doi: 10.5688/aj7410S7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Xu T, Pujara S, Sutton S, Rhee M. Telemedicine in the management of type 1 diabetes. Prev Chronic Dis. 2018;15:E13. doi: 10.5888/pcd15.170168 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Tchero H, Kangambega P, Briatte C, Brunet-Houdard S, Retali GR, Rusch E. Clinical effectiveness of telemedicine in diabetes mellitus: a meta-analysis of 42 randomized controlled trials. Telemed J E Health. 2019;25:569-583. doi: 10.1089/tmj.2018.0128 [DOI] [PubMed] [Google Scholar]
- 27. Levin PA, Nguyen H, Wittbrodt ET, Kim SC. Glucagon-like peptide-1 receptor agonists: a systematic review of comparative effectiveness research. Diabetes Metab Syndr Obes. 2017;10:123-139. doi: 10.2147/DMSO.S130834 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Sheahan KH, Wahlberg EA, Gilbert MP. An overview of GLP-1 agonists and recent cardiovascular outcomes trials. Postgrad Med J. 2020;96:156-161. doi: 10.1136/postgradmedj-2019-137186 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Gurgle HE, White K, McAdam-Marx C. SGLT2 inhibitors or GLP-1 receptor agonists as second-line therapy in type 2 diabetes: patient selection and perspectives. Vasc Health Risk Manag. 2016;12:239-249. doi: 10.2147/VHRM.S83088 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Arnold SV, Kosiborod M, Wang J, Fenici P, Gannedahl G, LoCasale RJ. Burden of cardio-renal-metabolic conditions in adults with type 2 diabetes within the Diabetes Collaborative Registry. Diabetes Obes Metab. 2018;20:2000-2003. doi: 10.1111/dom.13303 [DOI] [PubMed] [Google Scholar]