Outcomes of Pharmacist-led Diabetes Care Interventions in Lao PDR

Abstract

Background: Diabetes is a major health problem worldwide. Despite the availability of evidence-based guidance to deliver effective diabetes care, many patients do not achieve treatment goals as per recommendations, including in Laos. Pharmaceutical care provided to diabetic patients by pharmacists together with other healthcare providers has shown to be effective in many countries. Diabetes care in Laos has been provided by doctors, nurses, and nutritionists, without the involvement of pharmacists. This study aimed to develop and validate standard tools of patient satisfaction and quality of life in Lao language and to evaluate the outcomes of pharmacists' interventions in diabetes care in Lao PDR.

Methods: Phase 1: Development and validation tools. Two questionnaires in Lao language were developed and validated in this study. They were related to patient satisfaction to diabetes management and quality of life for diabetes (D-39). Patient satisfaction questionnaire (PSQ) was developed by using systematic review of qualitative studies on patients' and healthcare providers' perspectives on diabetes management. Databases were searched including CINAHL, PubMed, Science Direct, and Web of Science from January 2001 to September 2017. Chronic care model framework was used to analyze the main themes and sub themes. The first PSQ was in the Thai language. The Thai version of Quality of life for diabetes (D-39) was used in this study. Both PSQ and D-39 were back-to-back-translated from the Thai version to Lao version. The properties of both tools were tested in 150 Type 2 diabetic patients in a university hospital in Thailand in November 2018, and in 150 Type 2 diabetic patients in a center hospital in Lao PDR in January 2019. Construct validity of both versions were tested by using factor analysis, Pearson correlation analysis, and reliability test.

Phase 2: Randomized controlled trial (RCT). The RCT with a control group was conducted to evaluate the outcomes of diabetes care led by a pharmacist in Lao PDR from June 2019 to July 2020. The mutual understanding protocol of this study was set by performing a focus group with healthcare providers in a hospital. The intervention group received pharmaceutical care from a pharmacist. The control group received standard care from the hospital clinic. Patients were randomized by permuted block. Primary outcomes measurements were HbA1c and FPG. Secondary outcomes measurements were blood pressure, lipid profiles, GFR, ASCVD 10 years' risk score, scores of PSQ and D-39. Primary and secondary outcomes of patients of both groups were measured at month 0 (pre-test), month 3 (post-test 1) and 6 (post-test 2) except for HbA1c which was measured only at month 0 and 6. Independent t-test and Mann-Whitney U test were used to test the differences between two groups. Pair t-test and Wilcoxon Signed Rank tests were used to test the differences within the group. Subgroup analysis was performed in diabetes with hypertensive patients.

Result: Phase 1. Nine main themes from the systematic review of qualitative studies were used to develop the PSQ in this study. The PSQ of 45 items was constructed by three major-dimensions (attitudes to health, satisfaction to diabetes services, and attitudes to services achievement) and 11 sub-dimensions (attitude to knowledge on self-management, attitude to family, attitude to the community, satisfaction of the standard of services, satisfaction of the type of services, satisfaction of the competency of providers, satisfaction of the competency of pharmacists, satisfaction of the communication with providers, attitude to the accessibility of service, attitude to the health service system, attitude to goal setting). Kaiser-Meyer-Olkin (KMO) measure for sampling adequacy for both versions were 0.753 and 0.850, p-value < 0.001 for the Lao and Thai versions respectively. The Cronbach’s alpha scores were 0.945 and 0.948 for the Lao and Thai versions respectively. Convergent and discriminant validity values of both versions were > 70%. Factor loadings for both versions were > 0.4 in all 45 items. Property test of Diabetes 39 (D-39) questionnaire showed KMO measure of 0.917, p-value < 0.001. The Cronbach's alpha score was 0.966. Convergent and discriminant validity values were >70%. Factor loadings were > 0.5 in all items.

Phase 2. Seventy-three diabetes patients were randomly allocated to the intervention group. Seventy-one patients were allocated to the control group. Thirteen patients of the intervention group (17.8%) and 20 patients of the control group (28.2%) were lost during follow-up. Sixty patients of the intervention group and 51 patients of the control group were analyzed. There was no significant difference between the groups for HbA1c, FPG, lipid profiles. There was a significant difference in systolic blood pressure between the groups for Month 6 (post-test). Both groups had well-controlled of HbA1c when comparing the mean of month 0 (pre-test) and month 6 (post-test), p-value < 0.001. The intervention group had well-controlled total cholesterol and the LDL-cholesterol when comparing the mean of month 0 (pre-test) and month 6 (post-test), p-value <0.001, and 0.001 respectively.

Conclusion: The PSQ and D-39 questionnaires are valid, reliable, and acceptable to use in further research in order to measure diabetic patients’ satisfaction with diabetes management. Patients who received diabetes care interventions led by a pharmacist tend to have better control of HbA1c and LDL-cholesterol. However, patients in the intervention group had poor blood pressure control.

-