Barriers to Antidiabetic Medication Adherence among Patients with Diabetes Mellitus in Saudi Arabia: A Systematic Review

This systematic review was conducted with aim of determining different types of barriers that interfere with the adherence to antidiabetic medications among diabetes mellitus patients in Saudi Arabia. A comprehensive systematic literature exploration was performed via databases and search engines as Science Direct, PubMed, Medline, Google Scholar, Scopus, Cochrane Library, and EBSCO. The review includes 20 original research studies on adherence to antidiabetic medications and their associated factors. The review of the studies reveals that the most prominent barriers to antidiabetic medication adherence were forgetfulness, polypharmacy and complexity of regimen, side effects, low perceived self-efficacy of the medications, and feeling better. Likewise, frequent dosing, long history of diabetes, comorbidities, cost of medication or financial issue, disruption of normal routine, carelessness, busy schedule, God centered locus of control, and food habits were among the others. The identified multifaceted barriers can serve as a potential target for interventions to improve adherence to antidiabetic medications and health related quality of life.


Diabetes mellitus (DM) is a chronic metabolic
Although management approaches for diabetes is multifaceted, lifestyle related strategies and pharmacotherapy are the cornerstone [3]. Poor medication adherence, which varies widely from 38% to 93% and is still rising, is one of the most prominent reasons for inadequate glycemic control among diabetic patients [4][5][6]. Poor glycemic control can have many implications, leading to increased morbidity, mortality, hospitalization, cost of therapy, and decreased quality of life (QoL). It may also cause various microvascular (retinopathy, nephropathy and neuropathy) and macrovascular (cardiovascular) complications on long-term [1,7]. This demands a more thorough analysis of various barriers affecting adherence to antidiabetic medications among the people of Saudi Arabia, so that the policymakers could focus on these factors to promote medication adherence and improve health related QoL.
Barriers to medication adherence may be attributed to several factors which are not the same for every population, probably due to cultural, environmental and socioeconomic disparities [8]. Hence, this systematic review was devised with the objective to determine different types of barriers that interfere with adherence to medicines for antidiabetic use among diabetes mellitus patients in Saudi Arabia. The compilation of these factors may help policymakers develop suitable intervention plan to address the issue of non-adherence or poor adherence to antidiabetic medications.

Data and Search Approach Sources
A comprehensive systematic literature exploration was performed through such databases and search engines as Science Direct, PubMed, Medline, Google Scholar, Scopus, Cochrane Library, and EBSCO. The search of literature was executed for the original research articles published up to April 2020. The retrieved articles were evaluated manually for their appropriateness with the objectives of the study. Various search phrases were used such as ''diabetes, diabetic"; ''adherence, compliance"; ''barriers, factors"; and "Saudi Arabia, KSA, Kingdom of Saudi Arabia". To facilitate the search for the relevant articles, Booleans operators such as "AND/OR" were also used without any filter with respect to date and language.

Study Selection
This systematic review includes original research studies based on adherence to antidiabetic medications and their related factors in Saudi Arabia. The review, however, excludes studies related to patients' special situation that affect medication adherence such as mentally disturbed patients, patients with cancer or serious complications and abusers of narcotic drugs. The selection of the articles was performed through PRISMA guidelines (Fig.1).

Data Extraction and Analysis
The data was extracted using self-designed data extraction tool. The information included in the tool includes authors, year of publication, study settings and design, number of participants, type of diabetes (i.e., T1DM or T2DM), average duration of diabetes, medications prescribed to manage diabetes, average value of glycosylated hemoglobin, average age of the participants and their level of education, measure of nonadherence or adherence, and percentage of low or poor adherence. Two independent reviewers conducted the process and consensus was found by debate where there was disagreement.

RESULTS
A total of 4754 articles were discovered from the databases and search engines, and out of which only 34 studies met the criteria to be screened for full text (Fig. 1). However, 14 such full-text articles were excluded with the justification that they did not identify the barriers (n=9), not related to medication adherence (n= 3) and concerned with non-diabetic (n= 2).
Out of 20 studies, three were published in the year 2020, four in 2019, three in 2018, four in 2017, two in 2016, and one in each of 2015, 2013, 2012 and 1999. Nearly one half (45%, n=9) of the studies were conducted in Primary Healthcare Centers (PHC) and University Hospital settings followed by diabetic-care unit or chronic disease centers (20%, n=4). Moreover, Riyadh city was the highly favored place of study (25%, n=5). Most of the studies followed crosssectional study design, but there were few examples of experimental, mixed model and web based studies as well.
The entire studies included a sum of 12862 patients, and most of the participants included in the studies had T2D (70%, n=14), followed by T1 & 2D both (30%, n=6). In 60% (n=12) of the investigations, the information about average duration of diabetes was nonexistent. The information about the medications prescribed was not mentioned in 35% of the diabetic patients, even though adherence was investigated. Conversely, oral antidiabetic mediation (OADM) along with insulin were prescribed in 35% of the patients. The average age of the patients included in the study was 40 years and above.
Two studies were published with no information about literacy, however, the average literacy rate was 25.35% (n=18). A variety of tools were used to measure adherence to antidiabetic medications including self-administered questionnaire (n=9), the Morisky Medication Adherence Scale (MMAS) (n=5), and others (n=6). The data about percentage of low or poor adherence was not illustrated in four studies, but the average value of low or poor adherence was 31.45% (n=16).  Numerous barriers were identified from the studies reviewed and they have been categorized as: patient-related, disease-related, medication-related, provider-related, and societal-related factors (Table 2). However, the most prominent barriers identified were forgetfulness, polypharmacy and complexity of regimen, side effects; low perceived self-efficacy of the medications, feeling better. Likewise, frequent dosing, long history of diabetes, comorbidities, cost of medications or financial issue, disruption of normal routine, carelessness, busy schedule, god centered locus of control, and food habits were among the others.

Diabetes is a rising public health issue in Saudi
Arabia and other parts of the world. Diabetes as an ailment, lasts for several years and even lifelong, and its management requires persistent rigorous approaches including life-style changes, appropriate medications and most importantly optimal antidiabetic medication adherence. However, the review of the studies above shows up various barriers that hinder adherence to antidiabetic medications. Thus, this review focuses on summarizing those barriers and to suggest the policymakers to address them to the maximal extent possible.
Most of the studies were conducted at PHCs and University Hospital settings preferably for convenience of data collection. Many patients obviously choose to visit these facilities to avail public services or less cost effective therapy as the studies point to the exorbitant cost of diabetes treatment 10 times higher than nondiabetic patients [27]. Although most of the studies were quantitative, only one survey was mixed model which necessitates more qualitative or mixed model researches to explore more barriers to antidiabetic medication adherence.
Most (70%) of the studies were found to include type 2 diabetes patients. This is attributed to a greater prevalence of type 2 diabetes compared with type 1 diabetes. The studies also report that the incidence of type 1 diabetes is rising every year in Saudi Arabia and this needs attention in the Kingdom as well [28]. Sixty percent (n=12) of the studies explained the poor medication adherence among the patients with longer duration of diabetes, but they lacked the information about the average duration of diabetes. Interestingly, a Brazilian study found the reverse trend [29]. The contrast associations may be a result of diversity in attitude among the people of different locations [23,30]. The average age of the patients included in the study was 40 years and above. This is often due to the reason that higher fraction of studies included in this review focused on type 2 diabetes which typically occurs in older patients [31,32]. The studies would have informed more understanding if they had included detailed information about medications prescribed. The studies highlight the importance of education and awareness in almost one-fourth of the patients who could benefit from counseling customized to their level of understanding. Also, images, photographs and local language can be utilized amid counselling to have optimal grasp of the information [33]. The studies indicate low or poor adherence to medications in 31.45% of the cases which is less than that reported in the gulf region [34].
The studies have described mixed responses as barriers to antidiabetic medication adherence in terms of gender, marital status and educational status. While there are varied gender findings, it is more persuasive that, relative to males, females are more adherent to antidiabetic medications. This is because females spend most of their time at home and can take advantage of taking their medications as prescribed [35]. Similarly, some researches have shown that married couples are more compliant than single or unmarried couples, although the reverse has been found in other studies. It is more logical to find that married patients particularly the older adhere more to medications because a partner and family can love, care for, and sustain patients [15]. Some investigators have found education as a stronger determinant of medication adherence, whereas others are of the opposite opinion. It is more convincing, however, that literacy facilitates comprehension, enable more complex thinking abilities and enhance patient compliance [36]. Furthermore, educated individuals are more conscious about health benefits of physical activity, and are more compliant with therapy and guidance on physical activity [37].
The studies have identified forgetfulness, polypharmacy and complexity of regimen, side effects (mainly fatigue, dizziness), low perceived self-efficacy of the medications as the common barriers to medication adherence and identification of most of these barriers concurs with other studies [38][39][40][41]. The issue of forgetfulness can be addressed through use of cell phone reminders, involving the children and the family individuals to remind about the medications [42][43][44][45]. Complexity of medication regimen also play significant role in adherence. Higher the complexity of medication regimen, lower or poorer the adherence [46]. Similarly, medications induced side effects and low perceived efficacy also impede the adherence [47,48]. Injectable preparations such as insulin itself is a barrier to medication adherence owing to fear of injection, size of the needle, and even pain associated with it [49]. Therefore, this study recommends critical role of educating the patients and providing counseling ideally in local language and dialect to minimize side effects and to advance secure and viable use of drugs and devices. Similarly, feeling better, frequent dosing, long history of diabetes, comorbidities, cost of medications or financial issue, disruption of normal routine, carelessness, busy schedule, God centered locus of control, and food habits were among the other barriers to medication adherence. These factors highlight the need of coordinated counseling interventions to amplify adherence and therapeutic outcomes.
This review is not free from limitations. In most of the studies, the tools used to measure medication adherence were developed by the researchers with insufficient evidence of reliability and validity. Methodological consistency of the studies included could not be evaluated. Furthermore, the term 'medication adherence' was not clearly defined.
The following recommendations are suggested to incorporate qualitative or mixed model approaches in future researches so as to explore a range of beliefs and barriers towards medication adherence. There is also a need to include type 1 diabetes in future research activities.

CONCLUSION
Barriers to antidiabetic medication adherence are multifactorial. The common barriers ascertained in this review were forgetfulness, complexity of medication regimen, side effects, and low perceived self-efficacy of the medications. Similarly, feeling better, frequent dosing, long history of diabetes, comorbidities, cost of medications or financial issue, disruption of normal routine, carelessness, busy schedule, God centered locus of control, and food habits were among the other barriers to adherence. Religious and social factors as barriers are underreported and require more attention in further studies. The barriers identified and discussed in this review can serve as a potential target for intervention to augment medication adherence and health related quality of life.

CONSENT
It's not applicable.