Effect of Pharmacist-Led Intervention on Progression of Diabetic Complications at Two Tertiary Care Hospitals of Malaysia.

BACKGROUND AND
OBJECTIVE: Diabetes mellitus is a chronic disease which can lead to different complications in patients if not treated properly. An appropriate intervention from health care providers is needed to stop and decrease the progression of diabetic complication in diabetic patients. This study aimed to measure the effect of pharmacist intervention on improvement in sign and symptoms and progression of diabetic complications in diabetic patients.
METHODOLOGY: Diabetic patients were randomly selected and divided into two groups from two tertiary care hospitals. Control group contained those 200 patients who were receiving usual treatment from hospitals. In contrast, the intervention group included those 200 patients who were receiving conventional treatment from hospitals together with separate counselling sessions with pharmacists from Diabetes Medication Therapy Adherence Clinic departments. The study continued for 1 year and two follow-up visits for both groups. A prevalidated data collection form was used to measure the improvement in sign and symptoms and progression of diabetic complication in diabetic patients. Statistical Package for the Social Sciences (SPSS) version 24 was used to analyze the data.
RESULTS: The average hemoglobin A1c (HbA1c) values decreased up to 1.43% in the control group and 2.82% in the intervention group. The intervention group showed significant improvement in HbA1c between groups (P < 0.05). The results of univariate and multivariate regression analysis showed that a statistically significant (P < 0.001) improvement was observed in all the predictors of diabetic complications among in the intervention group when compared with the control group.
CONCLUSION: Statistically significant reduction in the sign and symptoms of diabetic complications was observed in the intervention group at the end of 1 year. The progression of diabetic neuropathy was significantly reduced in the pharmacist intervention group. Copyright:

INTRODUCTION

Diabetes mellitus is a progressive disease which is linked with many microvascular and macrovascular complications. These complications can be diabetic neuropathy, nephropathy, retinopathy and vasculopathy.[1] The occurrence and progression of these complications are directly associated with the control of glycemic levels in patients.[2] The prevalence of diabetes mellitus is continuously increasing in all over the globe, especially in middle and low-income countries.[3] Because in these countries, it remains underdiagnosed in patients for many years as the general population is not aware of the disease.[3] Thus, at the time of diagnosis itself, the patients need treatment of diabetes along with diabetic complications.[4]

If diabetes mellitus remains uncontrolled inside the body of the patients for a few years, it can affect the eyes, resulting in diabetic retinopathy, it can be kidneys, resulting in diabetic nephropathy, and nerves resulting in diabetic neuropathy.[56] It has been proven that diabetic complications are strongly associated with the presence of chronic hyperglycemia in patients.[7] The results of many randomized trials have proven the fact that early and tight glycemic control can reduce the onset and the progression of these diabetic complications.[89] This strict glucose control cannot be achieved without patient compliance with the given treatment as reported by many trials.[9] Patient compliance can only be increased if the patients are having proper knowledge about the disease and its complication later on.[10]

Many studies throughout the globe have confirmed that patient counselling and patient education is having a positive impact on the control of diabetes mellitus in diabetic patients. The involvement of pharmacist is directly associated with an increase in patient education about disease and ultimately will result in better outcomes of diabetes mellitus.[9]

The Malaysian health-care system is one of the best health-care systems in the world;[11] therefore, in Malaysia the pharmacist directly collaborates with physicians in tertiary hospitals by the help of Diabetes Medication Therapy Adherence Clinic (DMTAC). In this system, physicians prescribe the best medication regimens and the pharmacists educate the patient about diabetes mellitus, its complications, its medication, dosage regimens, and proper use as well as storage of insulin devices.[12] Once proper treatment given to patients, the outcomes of that given therapy can be measure by various methods.[13] The current study was conducted to evaluate the impact of pharmacist educational intervention on the sign and symptoms, along with the effect of these interventions on the progression of various diabetic complications. The hemoglobin A1c (HbA1c) was used as the clinical outcome measure of diabetes mellitus as HbA1c is a recommended predictor for diabetic complication as per the Malaysian guideline for diabetes mellitus.[14]

METHODOLOGY

Study population and approvals

The current study was conducted at outpatient clinics of two tertiary hospitals in Kedah Malaysia. The present study was prospective nonclinical, multicenter randomized control study. The current study was registered with the National Medical Research Register (NMRR) with registration number NMRR-17-2381-38042. It was approved by both hospital authorities as well as the Medical Research and Ethics Committee (MREC), Ministry of Health Malaysia approval reference (KKM/NIHSEC/P18-1307 [13]). The sample size was calculated using the previous data from Butt et al.[15] to compare the mean of HbA1c between the intervention and control group. A total of 65 patients in each study group were needed to detect the difference of 0.79% (8.47% vs. 9.26% HBA1c) with 80% certainty (power) and using an alpha level of 0.05 and standard deviation (SD) is σ = 1.61. The type I error probability associated with the test of this null hypothesis is 0.05. With an additional 20% dropout rate, the sample size was about 80 samples per group. The inclusion criterion of study was diabetes mellitus for minimum of 5 years, HbA1c more than 8.0%, and receiving medicine from any selected hospital of the study.

Procedure and randomization

Initially, a total of 150 patients from each hospital signed the informed consent form. Then the patient's hospital registration number from each selected hospital was listed as a control group and similarly an interventional group. Upon completion of 150 registration numbers of patients in control and intervention group this list was entered into Microsoft Excel. The randomization was carried out by Microsoft Excel to select randomly 100 patients from 150 hospital registration numbers as the final participants of the study from each group.

The control group includes adult outpatients of diabetes mellitus who were receiving treatment from diabetic clinics in selected hospitals and the intervention group includes adult patients of diabetes mellitus who were receiving treatment from diabetic clinics with the intervention of pharmacist from selected hospitals.

Baseline data were taken for the control group and intervention group from each selected hospital. Then follow-ups were made for these patients after every 6 months in both the control group and intervention group. After baseline, a total of two follow-ups were made for both control groups and intervention groups in each hospital. At every follow-up, the laboratory parameters and the sign and symptoms of various diabetic complications were documented on data collection forms. The total duration of the study was 1 year for every patient after recruitment.

Measures

For collecting the patients’ information, a prevalidated data collection tool was used. The contents and face validation of the data collection form was done by experts consisting of members of diabetic clinic specialists and academic experts in pharmaceutical sciences faculties of different universities. They were requested to check the issues such as readability, clarity, grammatical, and typo mistakes. The final data collection form was finalized after a series of discussions with experts.

Statistical analysis

The data analysis was done using Statistical Package for the Social Sciences (SPSS) IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp. Descriptive data were expressed as mean ± SD. The normality of the data was checked by SPSS using skewness and kurtosis testing. The univariate and multivariate regression analysis was used to evaluate the association between independent variables and pharmacist intervention in the current study. A value of P < 0.05 was considered statistically significant.

RESULTS

Overall, 26.5% (intervention arm: 29.5% vs. control arm: 23.5%) patients dropped out of the study due to various known and unknown reasons.

The sociodemographic characteristics of finally included 299 patients are presented in Table 1.

Table 1

Demographics and clinical characteristics of patients at baseline

Variables	Frequency
Gender
Male	144 (48.2)
Female	155 (51.8)
Ethnicity
Malay	231 (77.3)
Chinese	48 (16.1)
Indian	20 (6.7)
Age (mean±SD)	60.25±6.31
Duration of diabetes (years)±SD	9.96±2.87
Residence status
Urban	145 (48.5)
Rural	154 (51.5)
Employment status
Unemployed	145 (48.5)
Employed	154 (51.5)
Educational status
No education	96 (32.1)
Primary	97 (32.4)
Secondary	78 (26.1)
College/University	28 (9.4)
Type of daily diet
Vegetarian	159 (52.2)
Nonvegetarian	140 (46.8)
Exercise status
Yes	78 (26.1)
No	221 (73.9)

SD: Standard deviation

The baseline characteristics of the intervention and control group can be seen in Table 2.

Table 2

Diabetic complications determinants at baseline (n=294)

Variables	Study groups		Univariate analysis		Multivariate analysis
	Control	Intervention	OR (95% CI)	P	OR (95% CI)	P
Diabetic neuropathy
No symptom	52 (17.3)	54 (18.1)	Referent		Referent
Numbness	32 (10.6)	29 (9.6)	1.261 (0.39–4.57)	0.092	-	-
Tingling	21 (7.1)	19 (6.4)	0.341 (0.32–3.62)	0.069	0.253 (0.21–2.13)	0.213
Ankle reflex	18 (6.1)	22 (7.4)	0.419 (0.51–2.15)	0.329	-	-
Dry skin	19 (6.4)	18 (6.1)	0.262 (0.43–3.12)	0.213	-	-
Loss of balance/coordination	8 (2.6)	7 (2.3)	0.146 (0.31–2.18)	0.653	-	-
Diabetic retinopathy
No symptom	83 (27.6)	85 (28.3)	Referent
Dry/irritated eyes	12 (4.1)	13 (4.3)	5.365 (3.22–6.84)	0.263	-	-
Blurred vision	27 (9.1)	28 (9.4)	5.543 (3.54–6.97)	0.621	-
Trouble in night vision	16 (5.3)	18 (6.1)	4.432 (3.41–6.17)	0.581	-	-
Maculopathy	8 (2.7)	9 (3.1)	3.543 (2.23–4.71)	0.354	-	-
Diabetic nephropathy
No symptoms	48 (16.2)	52 (17.5)	Referent
Urine dipstick for protein	11 (3.4)	13 (4.4)	1.331 (0.57–2.51)
Microalbuminuria	12 (4.1)	14 (4.7)	3.453 (2.25–4.14)	0.241	-	-
Polyuria	56 (18.7)	58 (19.4)	0.328 (0.43–3.51)	0.066	0.923 (0.73–2.42)	0.231
Swelling of feet	16 (5.3)	19 (6.3)	3.453 (2.25–5.14)	0.348	-	-
Diabetic foot
No symptoms	129 (43.1)	132 (44.1)	Referent
Ulceration	6 (2.1)	9 (3.1)	2.543 (1.92–3.31)	0.213	-	-
Callus	5 (1.7)	8 (2.7)	2.231 (1.45–2.93)	0.431	-	-
Infection	3 (0.9)	7 (2.3)	2.673 (1.96–4.29)	0.075	1.257 (1.78–2.64)	0.251

aThe assumptions of simple logistics regression were not met, there must be at least two cases for each category of the dependent. OR: Odds ratio, 95% CI: 95% confidence interval. Inclusion criteria in adjusted OR was >0.09

The inclusion criteria used in logistic regression to enter the data into multivariate regression from the univartiate regression was the adjusted odds ratio < 0.09.

Table 2 shows at the baseline, in demographic variables of the control and intervention group, there was no statistical difference (P < 0.05) observed in any predictor of the included diabetic complications in patients.

Table 3 shows at the follow-up 1, in demographic variables of the intervention group, there was some statistical difference (P < 0.05) observed in few predictors of the included diabetic complications in patients.

Table 3

Diabetic complications determinants after 6 months of baseline (n=294)

Variables	Study groups		Univariate analysis		Multivariate analysis
	Control	Intervention	OR (95% CI)	P	OR (95% CI)	P
Diabetic neuropathy
No symptom	58 (19.4)	85 (28.4)	Referent		Referent
Numbness	28 (9.3)	14 (4.7)	2.263 (1.35–6.56)	0.062	1.543 (0.56–5.45)	0.192
Tingling	20 (6.7)	17 (5.7)	0.238 (0.72–3.72)	0.051	0.361 (0.41–2.18)	0.213
Ankle reflex	17 (5.7)	15 (5.0)	0.347 (0.63–2.45)	0.619	-	-
Dry skin	18 (6.2)	16 (5.4)	0.262 (0.51–3.52)	0.252	-	-
Loss of balance/coordination	7 (2.3)	4 (1.2)	0.137 (0.58–2.71)	0.568	-	-
Diabetic retinopathy
No symptom	95 (31.8)	118 (39.6)	Referent
Dry/irritated eyes	10 (3.3)	7 (2.3)	4.313 (2.19–5.82)	0.693	-	-
Blurred vision	24 (8.0)	16 (5.3)	2.463 (1.54–6.63)	0.081	1.843 (1.02–3.42)	0.367
Trouble in night vision	12 (4.1)	8 (2.6)	3.622 (1.47–6.19)	0.079	2.412 (1.13–4.32)	0.319
Maculopathy	5 (1.7)	4 (1.3)	2.325 (1.43–3.53)	0.494	-	-
Diabetic nephropathy
No symptoms	66 (22.2)	97 (32.4)	Referent
Urine dipstick for protein	10 (3.4)	9 (3.0)	1.213 (0.87–2.74)	0.247
Microalbuminuria	11 (3.6)	12 (4.0)	2.252 (1.35–2.12)	0.276	-	-
Polyuria	44 (14.8)	31 (10.3)	1.243 (0.63–6.76)	0.023	0.923 (0.31–4.65)	0.034
Swelling of feet	12 (4.0)	7 (2.3)	2.513 (1.23–6.19)	0.098	1.973 (0.97–4.45)	0.327
Diabetic foot
No symptoms	133 (44.5)	149 (49.9)	Referent
Ulceration	5 (1.6)	3 (1.0)	2.591 (1.41–4.56)	0.053	1.435 (0.46–2.73)	0.561
Callus	4 (1.3)	2 (0.7)	2.601 (1.39–5.82)	0.061	1.781 (1.81–3.49)	0.219
Infection	1 (0.3)	2 (0.7)		NAa

aThe assumptions of simple logistics regression were not met, there must be at least two cases for each category of the dependent. OR: Odds ratio, 95% CI: 95% confidence interval, NA: Not available

Table 4 shows at the follow-up 2, in demographic variables of the intervention group, there was a statistical difference (P < 0.05) observed in all predictors of the included diabetic complications in patients, which shows the effect of pharmacist intervention on diabetic complications at the end of the study.

Table 4

Diabetic complications determinants after 1 year of baseline (n=294)

Variables	Study groups		Univariate analysis		Multivariate analysis
	Control	Intervention	OR (95% CI)	P	OR (95% CI)	P
Diabetic neuropathy
No symptom	73 (24.3)	121 (40.4)	Referent		Referent
Numbness	23 (7.7)	9 (3.1)	2.326 (0.38–8.48)	<0.001	1.543 (0.56–5.45)	0.002
Tingling	16 (5.3)	5 (1.7)	3.549 (0.93–7.81)	<0.001	1.246 (0.86–5.38)	0.003
Ankle reflex	17 (5.7)	9 (3.1)	3.482 (1.73–8.92)	0.009	2.529 (1.98–5.43)	0.051
Dry skin	14 (4.7)	6 (2.0)	2.212 (1.71–9.39)	<0.001	1.174 (0.34–6.81)	0.042
Loss of balance/coordination	5 (1.7)	1 (0.3)		NAa
Diabetic retinopathy
No symptom	103 (34.4)	140 (46.8)	Referent		Referent
Dry/irritated eyes	8 (2.7)	3 (1.0)	3.652 (2.26–9.27)	<0.001	2.341 (1.61–6.11)	0.032
Blurred vision	21 (7.1)	7 (2.3)	2.457 (1.91–8.71)	0.002	1.263 (1.01–5.61)	0.047
Trouble in night vision	9 (3.0)	3 (1.0)	3.821 (1.25–9.31)	<0.001	2.211 (1.03–6.12)	0.019
Maculopathy	3 (1.0)	2 (0.7)	1.121 (0.22–5.58)	0.032	1.012 (0.12–3.51)	0.329
Diabetic nephropathy
No symptoms	80 (26.7)	136 (45.5)	Referent		Referent
Urine dipstick for protein	8 (2.7)	3 (1.0)	3.413 (1.32–9.64)	0.041	2.549 (0.21–3.31)	0.412
Microalbuminuria	10 (3.3)	5 (1.7)	2.652 (1.21–4.42)	0.061	1.322 (0.91–3.46)	0.481
Polyuria	35 (11.7)	10 (3.3)	1.271 (0.93–9.42)	<0.001	0.733 (0.21–3.15)	0.004
Swelling of feet	10 (3.3)	2 (0.8)	2.273 (1.36–8.37)	<0.001	1.148 (0.18–4.46)	0.039
Diabetic foot
No symptoms	134 (44.9)	152 (50.9)	Referent		Referent
Ulceration	4 (1.3)	2 (0.7)	2.438 (1.29–8.41)	<0.001	1.285 (0.72–5.93)	0.061
Callus	4 (1.3)	1 (0.3)		NAa
Infection	1 (0.3)	1 (0.3)		NAa

aThe assumptions of simple logistics regression were not met, there must be at least two cases for each category of the dependent. OR: Odds ratio, 95% CI: 95% confidence interval, NA: Not available

Table 5 shows that the HbA1c decreased continuously in every follow-up visit in both of the study groups.

Table 5

Differences in outcomes in the study groups

Outcome as HbA1c	Baseline	Follow-up 1 after 6 months of baseline			Follow-up 2 after 1 year of baseline
		Mean±SD	Mean difference	Pα	Mean±SD	Mean difference	Pβ
Control group	11.15±1.32	10.10±1.04	−1.05	<0.001	9.72±1.02	−0.38	0.003
Intervention group	11.69±1.50	9.94±0.92	−1.75	<0.001	8.87±0.79	−1.07	<0.001

Paired t-test was used to find the P-values. Pα: P-value between baseline × follow-up 1, Pβ: P-value between follow-up 1 × follow-up 2. SD: Standard deviation, HbA1c: Hemoglobin A1c

DISCUSSION

The current study intended to measure the effect of pharmacist intervention on the progression of diabetic complications from the improvement in signs and symptoms of these complications in type 2 diabetic patients. This study had proven that if the patient education about the disease is increased by pharmacist intervention, then ultimately, the outcomes of the disease will be better.

In the current study, the mean HbA1c lowered from baseline to the end of the study in the control group was 0.38% only. In contrast, the mean decrease in HbA1c in the intervention group with the involvement of pharmacist was 1.07% which is statistically significant from the control group. The result of our study was supported by a randomized prospective study by Lim et al., in Malaysia, where the reduction of mean HbA1c in the intervention group was 0.90%, and the control group 0.08% only and was significant (P = 0.011) in both study groups.[12]

In diabetic neuropathy. At baseline, all the predictors of sign and symptoms of diabetic neuropathy were statistically insignificant between study groups. At 6-month follow-up, symptoms of numbness and tingling were improved. At 1-year follow-up, all the predictors of sign and symptoms of diabetic neuropathy were statistically improved in the intervention group as compared with the control group. These results were comparable to the findings of a study conducted by Keban et al., in India, according to which the patients’ knowledge about diabetic neuropathy was improved by pharmacist intervention resulted in the improvement of diabetic neuropathy symptoms in patients.[16] Similar results were also reported by a pilot study conducted in India.[17]

In the progression of diabetic retinopathy. At baseline, statistically, the nonsignificant difference was observed between both of the study groups, but at 6-month and 1-year follow-up, the improvement in the intervention group was started producing significant effects in the intervention group. The current study findings have supported a study conducted in the United States by Weber et al., according to which pharmacists-managed care has a potential role in protecting the patients from diabetic retinopathy and reduce its progression.[18] The current study shows the improvement in retinopathy signs and symptoms was started from “blurred vision” and “trouble in night vision” predictor with P = 0.021 and 0.039, respectively. Then at the 1-year follow-up, the results of univariate and multivariate regression analysis showed significant association in all signs and symptoms predictors of diabetic retinopathy. The reason behind is the education provided by the pharmacist to the patients; once the patients knew the complications of the disease, they became more conscious about the daily activities and diet modification.[9]

The current study showed the improvement in nephropathy signs and symptoms of “polyuria” and “swelling of feet” predictor with P = 0.023 and 0.098 at 6-month follow-up. In the 1-year follow-up, the results of univariate and multivariate showed that it became statistically significant in all nephropathy signs and symptoms in the pharmacist-led intervention group as compared with the control group. These findings are strongly supported by a study conducted in France in which community pharmacist-provided the educational interventions to renal impairment patients, and improvement in sign and symptoms was observed at the end of the study.[19]

The results of the present study showed that the pharmacist intervention resulted in improvement in the progression of the diabetic foot also. The total number of patients with “callus” and “infection” was only one with each symptom in the intervention group that proved the pharmacist educational intervention resulted in an improved outcome of diabetic foot in all the included patients with a diabetic foot. These findings of the current study are supported by a retrospective study by Stading et al., which proves the impact of the pharmacist on diabetic foot complication.[20]

CONCLUSION

Overall, pharmacist-led educational intervention has a positive impact on the progression of diabetic complications. This study findings gave evidence for the effectiveness of DMTAC program in Malaysia and would like to suggest that pharmacist services should enhance especially to patients with uncontrolled diabetes to avoid further diabetic complications.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.