Adherence to treatment and its predictive factors among adults with type 2 diabetes in northern Iran

Abstract

BACKGROUND:

Type 2 diabetes is a common chronic disease in developing countries. It has widespread effects on many aspects of the lives of patients and their families, so health care systems should prioritize the management of diabetes. Blood glucose control and prevention of complications are the primary goals of diabetes care. Achievement is possible when the patients adhere to the treatment regimen, although several factors are known to affect the adherence.

OBJECTIVE:

The aim of this study was to evaluate the adherence to treatment and identify a number of its predictive factors in patients with diabetes in northern Iran.

METHODS:

In this cross-sectional analytical study, 266 patients with type 2 diabetes were selected through convenience sampling. The data-gathering tool was a self-administered questionnaire containing questions about personal and social characteristics of patients, as well as a 40-item questionnaire to measure patients’ adherence to treatment. The questionnaire consisted of seven domains: (i) making the effort for treatment, (ii) intention to take the treatment, (iii) adaptability, (iv) integrating illness into life, (v) sticking to the treatment, (vi) commitment to treatment, and (vii) indecisiveness about applying treatment. Data collection lasted from first week of November 2017 to the first week of January 2018.

RESULTS:

Most subjects were adherent to treatment. The highest adherence was in “commitment to treatment,” subdomain and the lowest adherence was in “adaptability.” In univariate analysis age, educational achievement, occupation, Body Mass Index (BMI), duration of diabetes, searching information about diabetes, and comorbidities were significantly related to adherence to treatment. The results of multivariate analysis also showed that patients who had normal BMI, academic education, searching behaviour for information, longer history of diabetes, and more comorbidities were more likely to be adherent.

CONCLUSION:

This study revealed a number of predictive factors of adherence to treatment in patients with diabetes. This may inform the health policy makers to take appropriate actions for promoting treatment adherence among patients.

Keywords

Type 2 diabetes adherence to treatment predictive factors

1. Introduction

Type 2 diabetes is a prevalent health problem across the world. The first World Health Organization (WHO) global report on diabetes shows that the number of adults living with diabetes has almost quadrupled since 1980 to 422 million. This dramatic rise is due to the increase in type 2 diabetes and factors provoking it, such as overweight and obesity. In 2012, diabetes caused 1.5 million deaths. Its complications can lead to heart attack, stroke, blindness, kidney failure, and lower-limb amputation [1, 2].

About three-quarters of patients with diabetes are living in low- and middle-income countries, which are facing limited financial resources. This imposes an extra pressure on their health care system’s economy. The prospect is not encouraging, and the International Diabetes Federation estimates that the number of people with diabetes worldwide will increase to 693 million by 2045 [3].

In Iran, a country in Western Asia and a member of the Middle East and North Africa (MENA) region, type 2 diabetes is a prevalent and costly disease. In terms of the number of people with diabetes, Iran has about 5 million diagnosed cases (3.9 million to 6.6 million), ranking third in the MENA countries after Egypt and Pakistan. Recent studies show that a majority of Iranian people have one or more known risk factors of diabetes. According to a recent meta-analysis, the prevalence of obesity in Iran is about 21.7% (CI: 18.5% –25%) [4]. Similarly, the physical inactivity rate among Iranian people has been reported as high as 70%. The results of a study on 23,183 Iranian students, aged 6 to 18 years, showed that 23.48% of these people were physically inactive and that obesity was a significant associated factor. In a longitudinal cohort study of 9,765 Iranian adults aged 35 to 65 years, the prevalence of central obesity was as high as 46.7% [5]. The economic burden of diabetes in Iran was US$3.8 billion in 2009 and is estimated to reach US$9 billion by 2030 [6]. Despite all these issues, the substantial concern of diabetes laies in its complications [7].

Diabetes could lead to several devastating physical, psychological, and social consequences that all deteriorate patients’ quality of life [8]. In addition to the acute complications of diabetes, such as severe blood glucose fluctuations, vascular changes from diabetes lead to major illnesses, including cardiovascular disease, cerebrovascular disease, nephropathy, and retinopathy [9 –11]. The risk of death in people with diabetes is twice as high as in people without the disease [12]. The main goal of diabetes management is to prevent its complications by controlling blood glucose [13, 14]. According to the Iranian Diabetes Leadership Forum report, for each percentage of decrease in HbA1c, the risks of developing diabetes-induced microvascular complications (such as diabetic nephropathy, neuropathy, and retinopathy), diabetes-related deaths, and myocardial infarction are reduced by 37%, 21%, and 14%, respectively [15]. This shows the importance of prescribing appropriate treatment and treatment adherence for patients with diabetes. However, different studies have reported poor blood glucose control status among the patients, mainly due to patients’ nonadherence to such prescriptions [16].

The WHO defines adherence as “the extent to which a person’s behaviour—taking medication, following a diet, and/or executing lifestyle changes—corresponds with agreed recommendations from a health care provider” [8]. Adherence to treatment is linked to several disease outcomes in patients with diabetes and reduces the complications and problems associated with the disease. It is associated with lower HbA1c levels [17] and, a reduction in hospitalisation and healthcare costs [18]. Khunti et al. [19] reported that adherence might reduce the risk of mortality and hospitalization up to 30% and 10%, respectively.

Because of the nature of diabetes, most patients often have three or more other comorbid chronic conditions. This makes patients less likely to adhere to treatment, despite its benefits. The results of noncompliance with treatment may include numerous problems such as long-term complications of illness, increased health care costs, reduced quality of life, and death. According to the results of a number of studies [20 –22], the factors that may affect adherence to treatment among patients with diabetes are either internal (e.g., patients’ characteristics) or external (e.g., characteristics and other aspects of the health care system and staff, as well as economic factors).

To improve planning for diabetes management, health authorities should be informed about the treatment status of patients with type 2 diabetes. Since limited studies have been conducted on adherence to treatment in patients with type 2 diabetes in Guilan province, Iran, we decided to determine adherence to treatment and its predictive factors among these patients.

2. Methods

This is a cross-sectional analytical study. The population included all patients with type 2 diabetes referring to two clinics affiliated with university hospitals.

2.1. Sampling

To calculate the sample size for a study with 95% level of confidence (z), reasonable precision (d = 2.5 score) and a standard deviation (σ = 20.83) of similar study [23], the below formula was used: $n = \frac{z_{1 - α}^{2} σ^{2}}{d^{2}} = 266$ (1)

This sample was obtained after inviting 298 patients to take part in the study. The sampling and data gathering procedures lasted from first week of November 2017 to the first week of January 2018. All patients were eligible to be included provided that they had these inclusion criteria: (i) a history of type 2 diabetes for at least six months (ii) the ability of responding to the questionnaire (iii) giving informed consent.

2.2. Settings

The specialised clinic and endocrine wards of Razi hospital in Rasht, Iran, as well as the specialised clinic of Beesat were used as the research settings. Razi hospital serve endocrine patients with 30 active beds. The specialised clinics are also active every day except for holidays in both morning and evening shifts. More than 80 patients with diabetes from all over the province come to these clinics to receive specialised services. The rationale behind selecting these settings was to facilitate the access of the patients with diabetes.

2.3. Measurements

The data-gathering tool was a two-part questionnaire. The first part of the questionnaire included questions about socio-demographic characteristics of subjects; i.e., age, gender, height, weight, marital status, educational achievement, employment status, supplementary insurance, area of residency, household income status, living condition, and questions about the patient’s disease. In the second part, 40 items pertaining to different aspects of adherence were incorporated. The psychometric properties of this tool, which is in Persian language, were examined by Modanloo [24] and has been used a number of studies [23 , 25–27]. Modanloo Adherence to Treatment Questionnaire (MATQ) addressed seven adherence subdomains including (i) making the effort for treatment (9 items scored 0–45), (ii) intention to take the treatment (7 items scored 0–35), (iii) adaptability (7 items scored 0–35), (iv) integrating illness into life (5 items scored 0–25), (v) sticking to the treatment (4 items scored 0–20), (vi) commitment to treatment (5 items scored 0–25), and (vii) indecisiveness for applying treatment (3 items scored 0–15). The scale of measurement in this questionnaire was Likert ranging from ‘completely agree’ with a score of 5 to ‘not agree at all’ with a score of 0). The total range of score for each patient was 0 to 200. In order to make the comparison of subdomains possible, and given that the minimum and maximum scores for each subdomain were different, the odds for each subdomain were converted to percentage points (adjusted score).

The researcher completed surveys, based on the patients’ answers to the questions. The inclusion criteria were having a history of at least six months of known diabetes, ability to answer questions, and consent to take part in the study.

2.4. Data gathering process

After receiving the necessary permissions from relevant authorities, the researcher went to the research settings and began collecting data. The research data were collected during two months from first week of November 2017 to the first week of January 2018. The researcher collected the required data from eligible patients from Saturday to Wednesday in the specialised clinics and on Thursday and Friday in Razi hospital. They were given the necessary information about research objectives, the process of involving in the research project, and their rights and expectations. These patients were included in the study if they signed the informed consent. Sampling continued until sample size was complete. During the sampling period, 298 eligible patients were enrolled in the study, 32 of whom were reluctant to continue their collaboration. Of these, 5 were men and 27 were women. Finally, 266 questionnaires were completed by the researcher. In order to complete the questionnaires, items related to the patient’s personal profile and items related to adherence to the treatment were collected through interviews and information about their illnesses was extracted from the clinics or hospital medical records. In all cases, the questionnaires were read by the researcher for all patients and marked in the relevant column.

2.5. Statistical analysis

The IBM SPSS Statistics 19.0 (Chicago, Illinois) was used to analyse the data. Descriptive statistics including mean, standard deviation, and frequency distribution were used to describe the study variables. The relationship between adherence to treatment and a series of individual variables and related variables were studied using analytical statistics. According to the Kolmogorov-Smirnov test, the distribution of adherence to treatment did not follow normal distribution, so non-parametric Mann-Whitney and Kruskal-Wallis tests were used to examine the relationship between variables. Friedman test was used to analysis the adherence to treatment in subdomains. The significance level of the tests was considered with (P < 0.05). Following univariate analysis, variables with p < 0.05 were included in multivariable logistic regression analysis. Odds ratios were also calculated.

2.6. Ethical considerations.

All phases of the investigation were conducted according to ethical standards of the Helsinki Statement of 1964 and subsequent developments and individuals were included in the study after written consent was signed. This research project was reviewed by the Ethics Committee of the Vice-Chancellor for Research of Guilan University of Medical Sciences and was approved with the registration code IR.GUMS.REC.1396.315.

3. Results

The mean age of the patients was 64.05±15.45 years. Most subjects were women (69.2%), with primary level education (54.1%), married (88%) and homemaker (55.3%). Also, regarding the status of the BMI, more than 80% of these patients were overweight or obese. Most subjects had fewer than five years history of diabetes (37.6%) and most patients (80.1%) reported the history of diabetes in their first degree family members (Table 1).

Table 1
Characteristics of patients with diabetes (n = 266)

Parameters n (%)

Gender(male/female)

Male 82(30.8%)

Female 184(69.2%)

Age groups

Less than 50 98(36.8%)

50–59 122(45.9%)

60 and more 46(17.3%)

Area of residency

Urban 215(80.8%)

Rural 43(16.2%)

Semi urban 8(3.0%)

Educational achievement

Under diploma 144(54.1%)

Diploma 85(32%)

Academic education 37(13.9%)

Marital status

Single 3(1.1%)

Married 234(88.0%)

Divorced 8(3.0%)

Widowed 21(7.9%)

Profession

Homemaker 147(55/3%)

Employed 21(7.9%)

Retired 54(20.3%)

Self employed 44(16.5%)

Supplementary insurance

Yes 144(54.1%)

No 122 (45.9%)

Living condition

Single 14(5.3%)

With wife 50(18.8%)

Couple and father 183(68.8%)

With child/children 19(7.1%)

Household income status

Low 22(8.3%)

Moderate 208(78.2%)

High 36(13.5%)

Parameters	n (%)
Gender(male/female)
Male	82(30.8%)
Female	184(69.2%)
Age groups
Less than 50	98(36.8%)
50–59	122(45.9%)
60 and more	46(17.3%)
Area of residency
Urban	215(80.8%)
Rural	43(16.2%)
Semi urban	8(3.0%)
Educational achievement
Under diploma	144(54.1%)
Diploma	85(32%)
Academic education	37(13.9%)
Marital status
Single	3(1.1%)
Married	234(88.0%)
Divorced	8(3.0%)
Widowed	21(7.9%)
Profession
Homemaker	147(55/3%)
Employed	21(7.9%)
Retired	54(20.3%)
Self employed	44(16.5%)
Supplementary insurance
Yes	144(54.1%)
No	122 (45.9%)
Living condition
Single	14(5.3%)
With wife	50(18.8%)
Couple and father	183(68.8%)
With child/children	19(7.1%)
Household income status
Low	22(8.3%)
Moderate	208(78.2%)
High	36(13.5%)

Data showed 24.8% of subjects had a “very good” status in terms of adherence to treatment and 54.9% had a “good” status. The rate of “moderate” and “poor” adherence were 19.9% and 0.4%, respectively (Table 2). Comparing the adherence subdomains showed that the highest average score was related to the commitment to treatment (75.11±17.12) while the lowest one was in adaptability (60.43±19.55). This difference was statistically significant (p < 00001) (Table 3).

Table 2

Level of adherence to treatment in patients with diabetes

Level (adherence score percentage)	n (%)
High (75–100%)	66 (24/8%)
Good (50–74%)	146 (54/9%)
Moderate (26–49%)	53 (19/9%)
Poor (0–25%)	1 (0.4%)

Table 3

The score of adherence to treatment in patients with type 2 diabetes in questionnaire

Subdomain	Mean	SD	Median	Min-max acquired score	P value
Making the effort for treatment	60.55	18.50	62.22	15.56–100	0.0001
Intention to take the treatment	63.89	19.17	65.71	5.71–100
Adaptability	60.43	19.55	62.86	2.86–100
Integrating illness into life	65.04	17.23	68	8–100
Sticking to the treatment	61.24	23.07	65	5–100
Commitment to treatment	75.11	17.12	76	8–100
Indecisiveness for applying treatment	67.07	22.53	66.67	10–100

According to Table 4, the age of subjects had a positive significant (p < 0.0001) relationship with their score on adherence with treatment. Higher educational achievement is significantly associated with higher scores in adherence to treatment (p < 0.001). Furthermore, the occupation of subjects had a significant impact on their scores of adherence to treatment as retired patients had the highest scores compared to others (p < 0.007).

Table 4

Individual and social characteristics and their association with level of adherence to treatment in patients with type 2 diabetes

	Adherence score
	N	Mean	SD	P-value
Gender
Male	82	64.27	15.27	0.863
Female	184	63.95	15.57
Age groups
Less than 50	98	58.62	16.75	0.0001
50–60	122	65.99	13.67
60 and more	46	70.49	13.48
Area of residency
Urban	215	64.42	15.36	0.712
Rural	43	62.20	15.20
Semi urban	8	64.06	20.41
Educational achievement
Under diploma	144	60.93	15.67	0.001
Diploma	85	65.07	14.50
Academic education	37	73.85	12.29
Marital status
Single	3	66	14.47	0.912
Married	234	63.74	15.77
Divorced	8	66.50	14.90
Widowed	21	66.29	12.56
Occupation
Homemaker	147	63.36	15.66	0.007
Employed	21	67.17	13.74
Retired	54	69.40	14.03
Self employed	44	58.33	15.26
Supplementary insurance
Yes	144	65.77	15.36	0.051
No	122	62.03	15.38
Living condition
Single	14	66.86	12.16	0.350
With wife	50	66.95	15
Couple and father	183	62.75	15.87
With child/children	19	66.87	14
Household income
Low	22	63.89	13.61	0.822
Moderate	208	64	15.16
High	36	64.44	18.38

Data analysis revealed that patients who had normal BMI, longer history of diabetes, searched information about diabetes, and had more co-morbidity have also scored better in adherence to treatment. The statistical test showed that these relationships were significant (Table 5).

Table 5

Disease-related characteristics and their association with level of adherence to treatment in patients with type 2 diabetes

Variable	n	Mean	SD	P Value
BMI ^*
Underweight	2	66.25	2.47	0.0001
Normal weight	41	72.30	12.89
Pre-obesity	116	64.75	14.80
Obesity class I	107	60.10	15.93
Duration of diabetes
Less than five years	100	52.80	15.69	0.0001
5–10 years	75	70.60	10.07
More than 10 years	91	71.02	11.11
Family history of diabetes type 1or 2
Yes	213	64.18	15.66	0.698
No	53	63.5 3	14.73
History of hospitalization due to complications of diabetes in the past year
Yes	31	67.00	15.57	0.198
No	235	63.66	15.43
Searching for information about diabetes
Yes	170	68.61	13.48	0.0001
No	96	55.99	15.50
Co-morbidities
1–2	129	60.17	16.10	0.0001
3–4	99	66.21	14.38
≥5	38	71.62	11.92

*We have no any subject with obesity class II or III in study sample.

However, despite the observed differences, the relationship between participants’ adherence score and their gender, marital status, area of residency, family history of diabetes, living condition, household income, supplementary insurance, and history of hospitalisation was not significant (Tables 4 & 5).

In this study, we used multivariate logistic regression to determine the influence related variables on patient adherence to treatment. As it has been shown in Table 6, BMI was associated with adherence. Patients with normal BMI were more than triple as likely to be adherent versus obese patients. Also duration of diabetes was associated with adherence to treatment. Patients with less than five years duration of diabetes were 94% less likely to be adherent in comparison to patients with more than 10 years duration of diabetes. Patients who did not search for information about diabetes were 86% less likely to be adherent when compared with patients who did. Patients educational achievement was associated with adherence. Patients with underdiploma and diploma educational achievement were less likely to be adherent compared to academic education, 90% and 80%, respectively. Also, comorbidity was associated with adherence. Patients with 1–2 comorbidity were 69% less likely to be adherent in comparison with patients with more than five comorbidity.

Table 6

Odds ratio resulting from Logistic regression of predictive variables for adherence to treatment in patients with type 2 diabetes

	Odds ratio	95% CI	p
BMI
Underweight vs. obesity	0.73	0.03–20.47	0.85
Normal vs. obesity	3.29	1.43–7.55	0.005
Pre-obesity vs. obesity	1.60	0.89–2.87	0.11
Duration of diabetes
Less than five years vs. More than 10 years	0.06	0.03–0.14	<0.001
5–10 years vs. More than 10 years	0.82	0.4–1.66	0.57
Searching for information about diabetes
No vs. Yes	0.14	0.04–0.56	0.005
Educational achievement
Under diploma vs. Academic education	0.10	0.04–0.26	<0.001
Diploma vs. Academic education	0.20	0.08–0.51	0.001
Co-morbidity
1–2 vs.>5	0.31	0.13–0.74	0.008
3–4 vs.>5	0.52	0.22–1.27	0.15

4. Discussion

Adherence to treatment in chronic diseases is a complex and multivariate phenomenon. Proper adherence plays an important role in controlling diabetes and preventing its complications [28]. There are many tools available to determine adherence to treatment among patients with diabetes, varying from biochemistry tests to specific questionnaires. This later tool, must be consistent with the culture and conditions of the community in which the patients live [8], so we used an Iranian standardised questionnaire to assess the adherence of patients.

According to the results of the present study, most patients had a good status in adherence to treatment. This finding is similar to several studies undertaken in countries such as Ethiopia, South Africa, Lebanon, Kuwait, and India in which most samples had good adherence to treatment [29 , 30–32]. These results may be justified by theories such as Health Belief Model (HBM). Different studies [22 , 32–34] show HBM components such as perceived susceptibility, perceived severity, and perceived benefits positively influence self-care behaviours of patients with diabetes. In response to question 1, most participants in our study responded that they obey the treatment despite having no assistance in surrounding environment. Furthermore, Most subjects of current study pointed out, in response to an item, that they still insist on treatment even in conditions where it disrupts their social activities. Several studies [35 –37] reported the positive effects of Theory of Planned Behaviour on self-care activities in patients with diabetes although this effect may not be profound as a recent meta-analysis showed [38]. In our study a large proportion of subjects declared that they will adhere to treatment when they truly decided to be so.

Nevertheless, the results of the current study are disagree with many other studies in MENA countries [39 –42] where poor adherence of patients to treatment has been reported. The inconsistent findings may be due to several causes such as research designs, study instruments, sampling frames, sample sizes, the use of general measures, and lack of control over confounding factors that could distort the results of the study.

Poor adherence to treatment is very common in patients with diabetes and depends on many factors. These factors can be grouped into four categories, including treatment and disease characteristics, intrapersonal factors, interpersonal factors, and environmental factors [8]. In this study, we also tried to examine the relationship between adherence to treatment among patients with diabetes with a number of measured variables, considering the above classification.

The results of the current study revealed a significant positive relationship between age and adherence. This finding is consistent with the results of several studies [43]. Perhaps one of the reasons for this situation is that as age increases, the person’s perceptual and physical abilities diminish and this can lead to his/her poor adherence [8].

We found that patients with higher educational achievement had better adherence to treatment. This finding is consistent with Al-Haj Mohd [39] and Dinar [44]. The interest of people with higher education in seeking information on how to gain and maintain health is a possible explanation of this finding. The results of the current study showed that retired patients had a better score on adherence to treatment. This finding of our study is not consistent with other studies [45–47] which showed retirement reduced patients’ adherence to treatment. Although, we recommend that the results of our study be viewed with caution, we can also suppose that with the retirement of the individual, they have more freedom to practice in the way of adherence to treatment. According to our results, patients with normal BMI were strongly more likely to be adherent. The relationship between BMI and adherence behaviours has been showed in many studies. However, proposing a cause and effect relation needs further investigations. This study shows the patients who have a longer history of diabetes scored better in adherence. Although this relationship was not significant in Elsous (2017) study [48], we suggest that a person may adhere to treatment regimen as time passes. We found that co-morbidity influenced the likely of patient adherence positively. This finding is inconsistent with many studies [49, 50]. We suggest a part of this inconsistency may be explained by Piette and Kerr [51] notion. They reasoned that when there is no serious conflict between diabetes care recommendation and health related co-morbid guidelines, patients are more likely to have better adherence.

5. Conclusion

With the increasing prevalence of type 2 diabetes and its related costs, more practical studies are needed to determine the adherence of patients to treatment. There are anecdotal results about the degree of adherence to treatment in patients with diabetes, which seem to be related to the assessment tool. Good adherence to treatment has generally been reported in studies using subjective instruments such as self-reported questionnaires. While in a number of studies which used objective tools such as HbA1c poor adherence has been reported. It is recommended that future researchers use a combination of subjective and objective tools to determine the degree to which patients are adherent to treatment.

This study showed factors such as age, educational achievement, occupation, duration of diabetes, BMI, co-morbidities, and health seeking behaviours are related to adherence to treatment among patients with type 2 diabetes. Health authorities may use these findings to promote adherence to treatment in patients with diabetes and provide a basis for more extensive studies in this area.

5.1. Limitations

Although the questionnaire which has been used in this study have adequate psychometric properties and is appropriate for the targeted population, we suggest the comparisons of results of this study with another studies and related interpretations should be considered with cautions as the questionnaire is new and is not widely used in another studies.

Funding

Guilan University of Medical Sciences, Iran, under grant number 96080609, supported this work.

Conflict of interest

The authors declare that they have no conflict of interest.

Footnotes

Appendix

Acknowledgments

The authors would like to thank all those people who took part in this study. We also appreciate Dr. Mahnaz Modanloo for her genius help, providing the questionnaire and its scoring guide.

References

World Health Organization, Fact Sheet. 2017.

Fowler

. Microvascular and Macrovascular Complications of Diabetes. Clinical Diabetes. 2008;26(2)77–82.

International Diabetes Federation, IDF_AnnualReport_2016_Interactive_Final. 2016.

Rahmani

, et al. Investigation of the Prevalence of Obesity in Iran: A Systematic Review and Meta-Analysis Study. Acta Med Iran. 2015;53(10):596–607.

Heidari-Bakavoli

, et al. Prevalence of obesity in Iran and its related socio-economic factors. Mediterranean Journal of Nutrition and Metabolism. 2015;8(2):109–18.

Noshad

, et al. Diabetes Care in Iran: Where We Stand and Where We Are Headed. Ann Glob Health. 2015;81(6):839–50.

International Diabetes Federation, Diabetes Prevalence in Middle East and North Africa region (Estimates for 2017 and 2045). 2017.

World Health Organization, Adherence to long-term therapies - Evidence for action, E. Sabate, Editor. 2003.

Marso

, et al. Liraglutide and cardiovascular outcomes in type 2 diabetes. New England Journal of Medicine. 2016;375(4):311–22.

10.

Neal

, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. New England Journal of Medicine. 2017;377(7):644–57.

11.

Palmer

, et al. Comparison of clinical outcomes and adverse events associated with glucose-lowering drugs in patients with type 2 diabetes: A meta-analysis. Jama. 2016;316(3):313–24.

12.

National Diabetes Clearinghouse. Preventing diabetes complications: Glucose control. 2011.

13.

American Diabetes A. Standards of Medical Care in Diabetes—2013. Diabetes Care. 2013;36(Suppl 1):S11–S66.

14.

Aikens

, Piette

. Longitudinal association between medication adherence and glycaemic control in Type 2 diabetes. Diabet Med. 2013;30(3):338–44.

15.

Forum

IDL.

The diabetes challenge in the Islamic Republic of Iran. 2015.

16.

García-Pérez

L-E

, et al. Adherence to Therapies in Patients with Type 2 Diabetes. Diabetes Therapy. 2013;4(2):175–94.

17.

Lunde

, et al. The Effectiveness of Smartphone Apps for Lifestyle Improvement in Noncommunicable Diseases: Systematic Review and Meta-Analyses. Journal of Medical Internet Research. 2018;20(5):e162.

18.

Sokol

, et al. Impact of medication adherence on hospitalization risk and healthcare cost. Med Care. 2005;43(6):521–30.

19.

Khunti

, et al. Association Between Adherence to Pharmacotherapy and Outcomes in Type 2 Diabetes: A Meta-analysis. Diabetes Care. 2017.

20.

Reyes

, et al. Factors Influencing Diabetes Self-Management Among Medically Underserved Patients With Type II Diabetes. Glob Qual Nurs Res. 2017;4:2333393617713097.

21.

Alrahbi

, Alghenaimi

. Factors influencing diabetes self-management among Omani patients with type-2 diabetes: Patients' perspectives. Journal of Nursing Education and Practice. 2017;7(12):64.

22.

Karimy

, et al. Determinants of adherence to self-care behavior among women with type 2 diabetes: An explanation based on health belief model. Medical Journal of the Islamic Republic of Iran. 2016;30:368.

23.

Asl

, et al. The Effectiveness of "Family-Centered Empowerment Model" on the Treatment Adherence of Patients with Type II Diabetes and Heart Disorder Admitted to AJA Hospitals, During Year 2015. ajaums-mcs. 2017;4(1):58–69.

24.

Seyed Fatemi

, et al. Psychometric properties of the adherence questionnaire in patients with chronic disease: A mix method study. Koomesh Journal. 2018;20(2):179–91.

25.

Rafii

, et al. Compliance to treatment in patients with chronic illness: A concept exploration. Iranian Journal of Nursing and Midwifery Research. 2014;19(2):159–67.

26.

Poshtchaman

, et al. The effect of two ways of using the phone and SMS follow-up care on treatment adherence in Coronary Artery Bypass Graft patients. IJCN. 2014;3(2):6–14.

27.

Poshtchaman

, et al. Assessing patient adherence to treatment after coronary artery bypass graft. Journal of Sabzevar University of Medical Sciences. 2015;22(4):668–75.

28.

Delamater

. Improving Patient Adherence. Clinical Diabetes. 2006;24(2):71.

29.

Abebe

, Berhane

, Worku

. Barriers to diabetes medication adherence in North West Ethiopia. Springerplus. 2014;3(1):195.

30.

Adegbola

, et al. Assessment of self-reported adherence among patients with type 2 diabetes in Matlala District Hospital, Limpopo Province. African Journal of Primary Health Care and Family Medicine. 2016;8(1):1–5.

31.

Halepian

, et al. Adherence to Insulin, Emotional Distress, and Trust in Physician Among Patients with Diabetes: A Cross-Sectional Study. Diabetes Therapy. 2018;9(2):713–26.

32.

Al-Taweel

DM.

Awad

, Johnson

. Evaluation of adherence to international guidelines for treating patients with type 2 diabetes mellitus in Kuwait. Int J Clin Pharm. 2013;35(2):244–50.

33.

Mohebi

, et al. Review the key role of self-efficacy in diabetes care. Journal of Education and Health Promotion. 2013;2.

34.

Dehghani-Tafti

, et al. Determinants of self-care in diabetic patients based on health belief model. Global Journal of Health Science. 2015;7(5):33.

35.

Gatt

, Sammut

. An exploratory study of predictors of self-care behaviour in persons with type 2 diabetes. International Journal of Nursing Studies. 2008;45(10):1525–33.

36.

Plotnikoff

, et al. Physical activity and diabetes: An application of the theory of planned behaviour to explain physical activity for Type 1 and Type 2 diabetes in an adult population sample. Psychology and Health. 2010;25(1):7–23.

37.

Blue

. Does the Theory of Planned Behavior Identify Diabetes-Related Cognitions for Intention to Be Physically Active and Eat a Healthy Diet? Public Health Nursing. 2007;24(2):141–50.

38.

Rich

, et al. Theory of planned behavior and adherence in chronic illness: A meta-analysis. Journal of Behavioral Medicine. 2015;38(4):673–88.

39.

Al-Haj Mohd

, et al. The predictors to medication adherence among adults with diabetes in the United Arab Emirates. J Diabetes Metab Disord. 2015;15:30.

40.

Ashur

, et al. Illness perceptions of Libyans with T2DM and their influence on medication adherence: A study in a diabetes center in Tripoli. Libyan Journal of Medicine. 2015;10(1):29797.

41.

Hegazi

, et al. Epidemiology of and Risk Factors for Type 2 Diabetes in Egypt. Annals of Global Health. 2015;81(6):814–20.

42.

Mahfouz

, Awadalla

. Compliance to diabetes self-management in rural El-Mina, Egypt. Central European Journal of Public Health. 2011;19(1):35.

43.

Rolnick

, et al. Patient characteristics associated with medication adherence. Clin Med Res. 2013;11(2):54–65.

44.

Dinar

, Elkhoudri

, Belahsen

. Effect of education on nutrition and diabetes status in type 2 diabetics in El Jadida province of Morocco. Mediterranean Journal of Nutrition and Metabolism. 2015;8:187–97.

45.

Kivimaki

, et al. Influence of retirement on nonadherence to medication for hypertension and diabetes. Cmaj. 2013;185(17):E784–90.