Influence of Pharmacist Intervention on Re-Elevation of Glycated Hemoglobin for Diabetic Outpatients

Abstract

Purpose: The effect of pharmacist intervention on blood sugar control in diabetic outpatients in a pharmacist-managed clinic was studied by focusing on the re-elevation of the glycated hemoglobin (A1c) level defined as a continuous variable. Methods: A retrospective chart review was performed at the Mizushima Kyodo Hospital from April 2014 to March 2016. Of the 221 diabetic outpatients who were provided guidance by nurses and nutritional managers, 62 further consulted the pharmacist-managed clinic. The remaining 159 patients were enrolled in a nonintervention group. Finally, the data of 115 patients with A1c level of ≥7.5% and A1c re-elevation were extracted. Intergroup comparison was performed between the pharmacist intervention (n = 26) and nonintervention (n = 89) groups. In both the groups, the starting point (baseline) was the time when the A1c level of ≥7.5% was observed. Subsequent monitoring was performed once in every 3 months. The average cumulative level of A1c re-elevation (CARE) was compared between groups. Patients with A1c level of ≥8.0% and A1c level between 7.5% and 8.0%, and male and female patients were also compared. Furthermore, the number of days until the re-elevation of the A1c level from the baseline was also compared. Results: The CARE values were 0.89 ± 0.86% and 1.51 ± 1.25% in the pharmacist intervention and nonintervention groups, respectively, showing a significant difference (P = .0195). There were no significant differences between patients with A1c level of ≥8.0% and A1c level between 7.5% and 8.0%, or between males and females. The number of days until the re-elevation of A1c level from the baseline also showed no significant difference. Conclusion: Pharmacist intervention for diabetic outpatients in pharmacist-managed clinics significantly suppressed CARE when compared with effects of no intervention, and this could be useful for preventing the exacerbation of diabetes.

Introduction

The American Society of Health-System Pharmacists (ASHP) guidelines suggest that pharmacists can contribute to positive outcomes by educating and counseling patients to prepare and motivate them to follow pharmacotherapeutic regimens and monitoring plans. Besides, education and counseling are most effective when undertaken in a room or space that ensures privacy and opportunity to ensure confidential communication.¹ Therefore, pharmacist-managed clinics were introduced in the Japanese clinical practice.²

In Japan, pharmacist-managed clinics have already been established for anticoagulation therapy, asthma, chronic obstructive pulmonary disease,³ cancer chemotherapy,⁴ hypercholesterolemia, and pain control.⁵ The Japanese Health, Labor and Welfare Ministry estimates that there are approximately 20 million Japanese people with strongly suspected diabetes or in whom the possibility of diabetes cannot be ruled out. Medical treatment of diabetes is an important and urgent issue in the Japanese healthcare system.

The effects of pharmacist intervention through pharmacist-managed clinics in the diabetic outpatients have already been reported in other countries.^6,7 However, the relationship between pharmacist intervention in diabetic medical treatment and stabilization of blood sugar level has not been sufficiently addressed. Therefore, this study focused on the evaluation of the difference in the re-elevation of glycated hemoglobin (A1c) levels between pharmacist intervention and nonintervention groups keeping in mind that re-elevation of A1c level has been associated with disease worsening and poor glycemic control.

Although a previous study has reported changes in A1c level from the baseline,⁸ no study has defined re-elevation of A1c as a continuous variable and analyzed it epidemiologically. To clarify the relationship between pharmacist intervention in a pharmacist-managed clinic in diabetic outpatients and prevention of re-elevation of A1c, “cumulative level of A1c re-elevation” (CARE) and “elapsed days until A1c re-elevation from the baseline” (EDRB) were defined as continuous variables for re-elevation, and were compared between pharmacist intervention and nonintervention groups.

Methods

Research Objectives

A retrospective chart review was performed with the patients’ clinical characteristics, age, sex, A1c level, and type of diabetes at Mizushima Kyodo Hospital in Japan from April 2014 to March 2016. Of 221 diabetic outpatients who were provided guidance by nurses and nutritional managers, 62 further consulted the pharmacist-managed clinic. The remaining 159 patients were enrolled in the nonintervention group. In routine medical care, it is difficult to increase the sample size of the pharmacist intervention group, and hence, the sample size of the control group was fixed high to increase the power of the study. Patients with A1c level of <7.5% and patients undergoing dialysis were excluded. Patients with A1c level of <7.0% were targeted for glycemic control.⁹ On the contrary, it is said that when the A1c level exceeds 8.0%, there is an increased risk of retinopathy.¹⁰ Therefore, in this study, the A1c level of 7.5% was used as the cutoff value. In addition, patients without any re-elevation of A1c were excluded. Patients without any re-elevation had the following conditions: (1) continuous increase in the A1c level, (2) continuous reduction in the A1c level, (3) steady A1c with no change, (4) no collection of blood at the first point of re-elevation, and (5) no change in A1c from the baseline or continuous reduction after transient increase in A1c. Finally, 115 patients with A1c of >7.5% and A1c re-elevation were included. Intergroup comparison was performed between pharmacist intervention (n = 26) and nonintervention (n = 89) groups (Figure 1).

Figure 1.

Research objectives.

Note. A1c = glycated hemoglobin.

Pharmacist-Managed Clinic

Before clinical examination by the physician, medical interview and nutritional guidance were provided by a nurse and a dietician. Furthermore, the outpatients who first visited the hospital and had poor glycemic control consulted the pharmacist-managed clinic. In the pharmacist-managed clinic, pharmacists not only confirmed compliance, daily habits, quality of life, manipulation of self-injection, and efficacy and safety of medication, but also instructed on the dosage and the method of drug administration to improve the outcome. Specific interviews and guidance included names of drugs used for diabetes treatment, medication instructions for efficacy and adverse effects of drugs, explanations for hypoglycemia and sick-days, instructions on injection procedures for insulin and glucagon-like peptide-1 receptor agonists, guidance on self-measurement of blood sugar level, explanation of clinical laboratory values, medical guidance on diet and exercise, and responses to questions of patients. The pharmacists provided the physicians with feedback on patient information obtained in the clinic. The outpatients consulting the pharmacist-managed clinic were included in the pharmacist intervention group. The other outpatients were included in the nonintervention group.

Study Design

The effect of pharmacist intervention in the pharmacist-managed clinic on the blood sugar control of diabetic outpatients was studied by focusing on the re-elevation of the A1c level. In both pharmacist intervention and the nonintervention groups, the starting point (baseline) was the time when the A1c level was ≥7.5%. Subsequent monitoring was performed once in every 3 months.

The average of CARE was set as the primary endpoint. The definition of the re-elevation point, CARE, and EDRB are shown in Figure 2. The A1c levels increased after a decline from the starting point of observation. The point showing a greater increase in the A1c level than that from the last point was defined as a re-elevation point. In cases of multiple re-elevations, the sum of the difference between the A1c level at a specific point in the past and the value at the re-elevation point was defined as CARE. The number of days from the baseline to the first point of re-elevation was defined as EDRB.

Figure 2.

Definition of re-elevation.

Note. A1c = glycated hemoglobin. (1), (2): Re-elevation points; (3), (4): level of A1c re-elevation; (3)+(4): cumulative level of A1c re-elevation; (5): elapsed days until A1c re-elevation from the baseline.

The point of time when the A1c level of >7.5% was first observed was considered as the baseline. The patients were followed up for 1 year 9 months after recording the baseline value. Patients with A1c level of ≥8.0% and A1c level between 7.5% and 8.0%, and male and female patients were also compared between two groups. Furthermore, EDRB were also compared between the groups.

Ethical Considerations

The ethics committees of both Mizushima Kyodo Hospital (approval no. 20140703-1) and Shujitsu University (approval no. 103) approved the study. Information on all patients was extracted from the Mizushima Kyodo Hospital database. All data were kept anonymous to avoid identification of the patients from the database.

Statistical Analysis

The background factors and A1c levels (%) were compared between the pharmacist intervention and nonintervention groups. The continuous variables were compared by Student t test, and the categorical variables by the chi-squared test. All P values were calculated by a 2-sided test. The data were analyzed using JMP 11 for Windows (version 11.2, 2014, SAS Institute Inc, Cary, North Carolina).

Results

Background Characteristics of Patients

The demographic characteristics of the final cohort of 26 and 89 patients in the pharmacist intervention and nonintervention groups, respectively, are presented in Table 1. The baseline characteristics of two groups were not significantly different.

Table 1.

Patient Background.

	Pharmacist intervention group	Nonintervention group	P value
n	26	89
Age, mean ± SD, y	63.9 ± 12.5	66.2 ± 11.7	.3917
Sex (male/female), n	12/14	39/50	.8331
A1c, mean ± SD, %	8.6 ± 1.5	8.6 ± 1.7	.9909
Type of diabetes (type 1/2), n	2/24	1/88	.0645

Note. A1c = glycated hemoglobin.

Comparison of CARE

The CARE values were compared between the pharmacist intervention and nonintervention groups for 1 year 9 months from the baseline (Figure 3). CARE values were 0.89 ± 0.86% and 1.51 ± 1.25% in the pharmacist intervention and nonintervention groups, respectively, and the CARE value in the former group was significantly lower than that in the latter (P = .0195).

Figure 3.

Comparison of cumulative level of A1c re-elevation between pharmacist intervention and nonintervention groups.

Note. A1c = glycated hemoglobin.

For further assessments, patients with A1c level of ≥8.0% and A1c level between 7.5% and 8.0%, and male and female patients were also compared between two groups. In patients with A1c level of ≥8.0% A1c along with severe diabetes the CARE values were 1.07 ± 0.40% and 1.88 ± 0.19% in the pharmacist intervention and nonintervention groups, respectively. Although there were no significant differences between the groups, the CARE value in the pharmacist intervention group showed a lower tendency than that in the nonintervention group (P = .0697) (Figure 4). In patients with an A1c value of 7.5% to 8.0%, the CARE values were 0.74 ± 0.67% and 1.04 ± 0.75% in the pharmacist intervention and nonintervention groups, respectively. There were no significant differences between the groups (P = .1954).

Figure 4.

Comparison of cumulative level of A1c re-elevation between pharmacist intervention and nonintervention groups with A1c ≥8.0% and 7.5-8.0%.

Note. A1c = glycated hemoglobin.

The CARE values by gender were compared between two groups. The CARE values in the male patients were 1.04 ± 1.14% and 1.53 ± 1.28% in the pharmacist intervention and nonintervention groups, respectively, without any significant difference (P = .247). On the contrary, the CARE values in female patients were 0.76 ± 0.52% and 1.50 ± 1.24% for in the pharmacist intervention and nonintervention groups, respectively, with a significant difference between the groups (P = .0338) (Figure 5).

Figure 5.

Comparison of cumulative level of A1c re-elevation between pharmacist intervention and nonintervention groups in males and females.

Note. A1c = glycated hemoglobin.

Comparison of EDRB

EDRB was also compared between two groups. The EDRB values were 312 ± 130 and 291 ± 104 days in the pharmacist intervention and nonintervention groups, respectively, without any significant difference (P = .412) (Figure 6)

Figure 6.

Comparison of elapsed days until A1c re-elevation from the baseline between pharmacist intervention and nonintervention groups.

Note. A1c = glycated hemoglobin.

Discussion

CARE in the pharmacist intervention group was significant lower than nonintervention group. Although patients in the nonintervention group were provided advice by nurses and dieticians, they were never provided pharmaceutical care by pharmacists. On the contrary, patients in the pharmacist intervention group were provided pharmaceutical care by pharmacists along with advice by nurses and dieticians. The adherence rate to oral medication improved considerably, reflecting more than just temporary improvement in patients’ mind-set and blood sugar control. This resulted in the stabilization of A1c level. It was reported that patients in the pharmacist intervention group showed greater improvement in their ability of self-management and knowledge of diabetes after 6 months of patient education when compared with those in the control group.¹¹ According to a meta-analysis including 43 randomized controlled trials conducted in foreign countries, including Asian countries and the United States of America, intervention by pharmacists who provided pharmaceutical care in addition to diabetes education helped in significantly reducing the A1c level when compared with that aided by usual medical instruction given by the medical team to patients with type 2 diabetes.¹² It was also shown that pharmacist-led interventions in the community have contributed to improved adherence and better disease control.¹³ These reports support the stabilization of A1c levels by pharmacist intervention as found in this study.

Because in more severe diabetic patients having A1c level of >8.0%, the CARE was lower in the pharmacist intervention group, it was suggested that intervention by pharmacists would be effective to prevent the exacerbation of diabetes in these patients. The impact of a novel pharmacist telephone-based patient-centered intervention in patients with poorly controlled diabetes was reported.¹⁴ The patients had a baseline A1c level of 9.4%. Although there was no significant difference in adherence, pharmacist intervention significantly improved diabetes control as revealed by the as-treated analyses. Therefore, stabilization of A1c in patients with poor glycemic control appears to require pharmacist intervention, in particular.

Pharmacist’s intervention in females was more effective than that in males. Iqbal et al reported that age, gender, education, diabetes-related knowledge, and treatment satisfaction were significantly correlated with medication adherence. Especially, they reported that older males with only primary education and poor diabetes-related knowledge had the lowest adherence.¹⁵ Hence, our result might be justified.

In this study, the EDRB values in the pharmacist intervention and nonintervention groups were not significantly different, and both were about 10 months on an average (291-312 days). It is generally said that the control of blood sugar will worsen after 6 months of diabetic education.¹⁶ The extension of EDRB in the pharmacist intervention group is reasonable. However, it is necessary to further examine the reason why EDRB in the nonintervention group was extended. Furthermore, new therapeutic drugs were released successively, and the way of intake and use of these drugs got complicated. Ayele et al reported that the high complexity of the medication regimens in diabetes was associated with poor glycemic control in the adjusted analyses.¹⁷

Furthermore, aging diversifies the problem of medication management. The relationship between hypoglycemia and polypharmacy strongly urges to limit polymedication as much as possible, especially in elderly patients.¹⁸ This result encourages the strong involvement of clinical pharmacists with an aim to reduce the risk of hypoglycemia during hospitalization. Therefore, the role of a pharmacist-managed clinic in diabetes is becoming increasingly important. This study reconfirmed the usefulness of pharmacist’s involvement in the treatment of diabetic outpatients.

However, as a limitation of this study, the target patients had poor glycemic control and the A1c level was >7.5%. Therefore, because the effects of pharmacist intervention shown in this study are limited by results of poor glycemic control, the effects of intervention in patients with a relatively good glycemic control need further consideration.

Conclusions

Pharmacist intervention in diabetic outpatients in pharmacist-managed clinics could significantly suppress CARE when compared with effects of no intervention, and it would be useful for preventing the exacerbation of diabetes.