INTRODUCTION

Diabetes Mellitus (DM) is a chronic and progressive disease with a global prevalence of 10.5% (536.60 million people) in 2021¹. Over the past few decades, the number of diabetic patients has risen faster in low and middle-income countries. Bhutan reported an incidence of 75-82 persons per 10,000 people in 2020².

Non adherence to medication is a major concern for all chronic diseases requiring life-long treatment³. Medication adherence is the extent to which a patient gets treatment in accordance with the prescribed interval and dose⁴. Though it is a global concern, medication adherence is generally considered poor worldwide, with minimal importance accorded to it even during hospital consultations^5,6.

Although Bhutan shares concerns on the low level of medication adherence to chronic diseases and its related consequences, there are no studies examining the level of medication adherence. With this lack of data, health consequences due to nonadherence to prescribed treatment could be significant. In order to address this knowledge gap, this study intended to study the level of adherence and risk factors affecting antidiabetic treatment in eastern region of Bhutan.

METHODS

The study was a multicenter cross-sectional observational study conducted in six district hospitals in eastern Bhutan. It was conducted from June 2019 to February 2020.

Study population and sampling technique

Diabetic patients (both DM type 1 and 2) above the age of 18 years visiting the diabetic clinics of the 6 eastern hospitals were eligible participants. Those with gestational DM and mentally challenged patients without a capacity to consent were excluded from the study. A simple randomization technique was used for recruiting participants, wherein every third or fifth diabetic patient visiting the diabetic clinic who fulfilled the inclusion criteria were approached for inclusion into the study. They were enrolled after receiving informed written consent.

The sample size was calculated based on an estimated total number of patients with DM in Bhutan of 15,000 in the year 2019 and a non-adherence rate of 50% to the treatment. A minimum sample size of 375 was calculated, considering a 95% confidence interval with a margin of error of 5%. To offset loss of subjects through refusal to consent or incomplete data, we took a sample size of 400.

Data collection

Data was collected by focal persons of DM clinic working in the six identified hospitals. Some of these hospitals operate diabetic clinics daily whereas others operate two to three days a week. Each recruited participant was provided with information about the research and consent for participation was obtained. The diabetic clinic’s focal person collected relevant data by reading out questions from a structured questionnaire.

The questionnaire consisted of socio demographic details, details about the type of DM, type of treatment regimen (mono or polytherapy), route of administration of medications, history of any adverse drug reactions, presence of any other comorbidity, and the time taken to reach the nearest health facility. The Morisky Medication Adherence Scale 8 Items (MMAS-8) was also a part of the questionnaire, to determine the level of adherence. The MMAS-8 is a tool which has been tested and validated in many languages for chronic diseases requiring long term treatment^7-9. The owner of the tool granted us the permission to use the tool via certificate Number: 1207-2004-207-31418477.

MMAS-8 consists of 8 closed-ended questions. Except for item number 8, all other questions are responded with either Yes or No. Each ‘Yes’ response is accorded a score of ‘0’ while each ‘No’ response is accorded a score of 1, except for item 5, where ‘Yes’ response is scored 1 and ‘No’ response is scored 0. Item number 8 includes a 5-point Likert response (0-4), corresponding to scores of 0, 0.25, 0.50, 0.75, and 1, respectively. As per the MMAS-8 score, the level of medication adherence can be high (=8), intermediate (6 & 7) or low (< 6)⁷.

Statistical analysis

For descriptive analysis of continuous variables, mean and SD were employed and for categorical variables, percentages were used. For inferential analysis, Chi-square test was used. Statistical analysis was conducted using EpiData Software Version 3.1. The primary outcome of this study was to find the level of adherence to anti-diabetic medications, and were reported as being low, intermediate or high.

Ethical clearance

For this study, ethical approval was accorded by the Research Ethics Board for Health, Ministry of Health, Bhutan (Ref. No. REBH/Approval/2019/022 23/06/ 2019 dated 23/06/ 2019).

RESULTS

Of the 400 participants recruited, 390 were included in the analysis. To analyse the primary outcome of level of medication adherence, only 383 were included since 7 had missing data. Figure 1 depicts the process of patient recruitment and analysis.

Figure 1. Flow diagram depicting the process of recruitment of participants in the six eastern district hospitals of Bhutan, June 2019-February 2020

The majority of the participants in this study (49.20%) were availing services from the Eastern Regional Referral Hospital (ERRH). Females outnumbered males (54.4% vs. 45.6%) and participants in the age group of 51-65 years constituted the majority (161, 41.3%). As depicted in Table 1, patients with Type 2 DM (378, 97.4%) outnumbered those with Type 1 DM and a majority of them (257, 65.90%) had an additional co-morbidity. Most of the participants (361, 94%) were on oral hypoglycemic agents and most did not report experiencing an adverse effect to treatment (314, 80.70%).

The overall mean MMAS-8 score was 6.06 (95% CI 5.91-6.22). Amongst the 383 participants included in the analysis of the primary outcome, 16.40% (n=63) reported high level of adherence, 41.30% (n=158) reported intermediate level of adherence and 42.30% (n=162) reported a low level of adherence.

Table 1. Baseline characteristics of participants in the six eastern hospitals of Bhutan, June 2019-February 2020, n=390

Variables

N (%)

Sex

Male

Female

178 (45.6)

212 (54.4)

Age group

<35 years old

35-50 years old

51-65 years old

>65 years old

13 (3.30)

122 (31.30)

161 (41.30)

94 (24.10)

Education level

Illiterate

Literate

287 (73.60)

103 (26.40)

Occupation

Employed

Retired

Farmer

Housewife

86 (22.10)

29 (7.40)

177 (45.40)

98 (25.10)

Type of DM

DM Type 1

DM Type 2

Missing data

10 (2.60)

378 (96.90)

2 (0.50)

Co-morbidity

Yes

257 (65.90)

133 (34.10)

Treatment Regimen

Monotherapy

Polytherapy

Missing data

185 (47.40)

189 (48.50)

16 (4.10)

Route of administration

Oral

Injection

Both oral and injection

Missing data

361 (92.60)

7 (1.80)

16 (4.10)

6 (1.50)

Adverse drug reaction

Yes

Missing data

75 (19.20)

314 (80.50)

1 (0.30)

Hospitals

Eastern Regional Referral Hospital

Trashigang District Hospital

Pemagatshel District Hospital

Samdrup Jongkhar District Hospital

Lhuentse District Hospital

Trashi Yangtse District Hospital

192 (49.20)

62 (15.90)

48 (12.30)

50 (12.80)

16 (4.10)

22 (5.60)

Average time taken to reach health facility

<1 hour

1-3 hours

>3 hours

191 (49.00)

179 (45.90)

20 (5.10)

Patients availing DM treatment at Samdrup Jongkhar district hospital reported the lowest mean MMAS-8 score of 4.46 while patients availing DM treatment at Trashi Yangtse district hospital reported the highest mean MMAS-8 scores of 6.73. The mean scores of patients from other hospitals ranged from 5.73 to 6.63 as reflected in Figure 2.

Figure 2. Bar graph depicting mean MMAS-8 score in 6 hospitals of Bhutan. [The numbers at the end of bar represent means scores with 95% confidence intervals within brackets.]

Adverse reactions to anti-diabetic drugs, types of treatment regimen (polypharmacy or monotherapy) and the type of DM demonstrated a statistically significant (p<0.05) association with mean MMAS-8 score, as shown in Table 2.

Table 2. Risk factors affecting the medication adherence amongst diabetic patients availing services at six eastern hospitals of Bhutan, June 2019-February 2020, n=390

Risk factors

Adherence scores

p-value

Mean score

95% CI

Age group

<30 years

35 – 50 years

51 – 65 years

>65 years

6.25

5.93

6.04

6.26

5.36 – 7.14

5.66 – 6.19

5.78 – 6.30

5.96 – 6.55

0.46

Sex

Male

Female

6.01

6.10

5.78 – 6.24

5.88 – 6.31

0.58

Marital status

Married

Unmarried

6.05

6.21

5.89 – 6.21

5.61 – 6.82

0.59

Level of education

Literate

Illiterate

6.08

6.04

5.90 – 6.26

5.72 – 6.36

0.82

Occupation

Employed

Retired

Farmer

Housewife

5.97

6.22

5.92

6.22

5.63 – 6.31

5.52 – 6.93

5.70 – 6.14

6.00 – 6.63

0.21

Treatment regimen

Monotherapy

Polytherapy

6.28

5.84

6.06 – 6.49

5.62 – 6.06

0.00

Types of diabetes mellitus

Type 1

Type 2

4.95

6.09

3.84 – 6.06

5.94 – 6.25

0.02

Route of drug administration

Oral

Injection

Both

6.08

4.93

6.02

5.92 – 6.24

2.83 – 7.03

5.28 – 6/75

0.14

Presence of co-morbidities

Yes

6.17

5.86

5.98 – 6.35

5.58 – 6.14

0.06

Average travel time to nearest health facility

<1 hour

1 – 3 hours

>3 hours

6.13

6.00

6.05

5.90 – 6.36

5.78 – 6.22

5.27 – 6.83

0.74

Are you informed on the need to take medicines / injection if you don’t feel any symptoms?

Yes

6.02

6.35

5.86 – 6.18

5.81 – 6.89

0.05 (One-way ANNOVA)

Do you know why you are taking medications?

Yes

6.10

5.64

5.95 – 6.26

5.04 – 6.24

0.10

DISCUSSION

With an overall mean MMAS-8 score of 6.06, this study reports an intermediate level of adherence to oral anti-diabetic drugs amongst diabetic patients availing services from six hospitals in the eastern part of Bhutan. However, 42% of participants reported a low level of medication adherence, which is concerning. This rate is similar to the global adherence rate of below 50%^6,10,11. Even developed countries like the USA have reported that 40% of patients with Type 2 DM are nonadherent, the United Kingdom reports that 39.9% were nonadherent, and China reports a non-adherence rate of 54.6% amongst diabetic patients^7,12,13. However, our neighboring country India has reported a lower non-adherence rate of 25.5% amongst diabetic patients availing services from 3 government hospitals in Delhi⁸.The studies in the United Kingdom and China only assessed adherence to oral diabetic agents and excluded those using insulin.

Low level of medication adherence amongst diabetic patients is associated with poor glycemic control, diabetic micro and macro vascular complications, and eventual increased morbidity and mortality¹². In the long run, it not only affects physical health of the patients but also becomes a financial burden to a country^9,13.

While comparing mean MMAS-8 scores among diabetic patients availing services from the 6 eastern hospitals, the current study noted that participants from Samdrup Jongkhar district reported the lowest level of adherence whereas participants from Trashi Yangtse district revealed the highest level of adherence. Samdrup Jongkhar has only 7 primary health care centers, 3 subpost facilities and 1 district hospital to cater to its population of 35,079. Except for one additional sub-post, the number of healthcare facilities have remained the same since 2018^14,15. This could be a potential reason why they had the lowest adherence. However, disproportionate sampling from the various hospitals could have impacted the mean adherence rates. A study in rural Kerala in India reported poor adherence in 74% of diabetic patients using the MMAS-8 tool¹⁶. Rural areas usually have poor socio-economic conditions which may influence the adherence level.

Higher levels of education have been reported to bring about a positive influence on medication adherence^10,12. However, findings of our study did not reveal a significant association of level of education with medication adherence levels. This may be because health advocacy programs are regularly conducted in local languages by health assistants in the primary health care centers in Bhutan. Thus, the level of formal schooling may have limited influence on the understanding of being adherent to medications.

This study, however, demonstrated a significant association between mean MMAS-8 score and polypharmacy. A similar finding was also revealed in other studies where multiple drug regimens were associated with reduced levels of medication adherence^9,10. The need to take multiple drugs make the treatment complex with a potential for forgetting a few doses^16,17. An adverse event with the medication has been reported to adversely affect medication adherence¹⁸. This study noted a statistically significant association between a past episode of adverse drug reaction and the level of medication adherence.

The strength of this study is that it was conducted in a free healthcare setting involving 6 district hospitals of eastern Bhutan, which represents 24% of Bhutanese population¹⁹. However, the authors also admit to have few limitations. The study experienced a few dropouts due to incomplete data and the sample size was not proportionate among the districts.

CONCLUSION

An intermediate level of adherence to oral and injectable antidiabetic medicine is noted amongst diabetic patients in the eastern region of Bhutan. While it is premature to draw any conclusion with just one study, this sets the stage to conduct other studies to address factors resulting in nonadherence to medications.

ACKNOWLEDGEMENTS

We express our gratitude to Mr. Thinley Wangdi who provided constant assistance in developing this manuscript. We would also like to thank all our patients for consenting to participate in the study.

REFERENCES

1. Sun H, Saeedi P, Karuranga S, Pinkepank M, Ogurtsova K, Duncan BB, et al. IDF Diabetes Atlas: Global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. 2022; 183:109119. [PubMed ] [Full Text] [DOI ]

2. Ministry of Health: Annual Health Bulletin 2020. Thimphu: Health Management and Information System, Policy & Planning Division, Ministry of Health, Royal Government of Bhutan; 2020; 107.p [Full Text ]

3. Rombopoulos G, Hatzikou M, Athanasiadis A, Elisaf M. Treatment compliance with fixed-dose combination of vildagliptin/metformin in patients with type 2 diabetes mellitus inadequately controlled with metformin monotherapy: a 24-week observational study. Int J Endocrinol. 2015; 2015:251485 [PubMed] [Full Text] [DOI ]

4. Raum E, Krämer HU, Rüter G, Rothenbacher D, Rujosemann T, Szecsenyi J, et al. Medication non-adherence and poor glycaemic control in patients with type 2 diabetes mellitus. Diabetes Res Clin Prac. 2012; 97(3):377-84. [PubMed] [Full Text ] [DOI]

5. Barnett NL. Medication adherence: where are we now? A UK perspective. Eur J Hosp Pharm. 2014;21:181-4. [Full Text ] [DOI]

6. Iglay K, Cartier SE, Rosen VM, Zarotsky V, Rajpathak SN, Radican L, et al. Meta-analysis of studies examining medication adherence, persistence, and discontinuation of oral antihyperglycemic agents in type 2 diabetes. Curr Med Res Opin. 2015;31(7):1283-96. [Full Text] [DOI ]

7. Wu P, Liu N. Association between patients’ beliefs and oral antidiabetic medication adherence in a Chinese type 2 diabetic population. Patient Prefer Adherence. 2016; 10:1161-7. [PubMed ] [Full Text] [DOI ]

8. Basu S, Khobragade M, Kumar A, Raut DK. Medical adherence and its predictors in diabetes mellitus patients attending government hospitals in the Indian capital, Delhi, 2013: A cross-sectional study. Int J Diabetes Dev Ctries. 2015;35(2):95-101 [Full Text] [DOI ]

9. Odegard PS, Gray SL. Barriers to medication adherence in poorly controlled diabetes mellitus. Diabetes Educ. 2008;34(4):692-7. [Full Text] [DOI ]

10. Cramer JA. A Systematic Review of Adherence With. Diabetes Care. 2004; 27(5):1218–24. [Full Text] [DOI ]

11. World Health Organization. Adherence to long-term therapies: evidence for action.Geneva, World Health Organization. 2003. 196 p. [Full Text ]

12. Samu AM, Amirthalingam PS, Mohammed OS. Assessment of patient medication adherence among the type 2 diabetes mellitus population with peripheral diabetic neuropathy in South India. J Taibah Univ Med Sci. 2017;12(2):164–8. [PubMed] [Full Text ] [DOI ]

13. Tunceli K, Iglay K, Zhao C, Brodovicz KG, Radican L. Factors associated with adherence to oral antihyperglycemic monotherapy in patients with type 2 diabetes mellitus in the United Kingdom. Diabetes Res Clin Pract. 2015;109(3):e27–31.[Full Text ] [DOI]

Factors affecting the level of medication adherence to anti-diabetic treatment: A multicenter cross-sectional study in Eastern Bhutan

INTRODUCTION

METHODS

Study population and sampling technique

Data collection

Statistical analysis

Ethical clearance

RESULTS

Figure 1. Flow diagram depicting the process of recruitment of participants in the six eastern district hospitals of Bhutan, June 2019-February 2020

Table 2. Risk factors affecting the medication adherence amongst diabetic patients availing services at six eastern hospitals of Bhutan, June 2019-February 2020, n=390

DISCUSSION

CONCLUSION

ACKNOWLEDGEMENTS

REFERENCES