Translation, cultural adaptation and validation of the General Medication Adherence Scale (GMAS) in moroccan patients with type-2 diabetes

Demographic transitions and lifestyle changes have led to increasing chronic diseases, more so in low- and middle-income countries. These chronic diseases not only have unfavorable outcomes on patients but also increase health care costs [1]. The World Health Organization (WHO) defines chronic diseases as any disease that is characterized by its long-standing and slow evolution, usually resulting in death, apparent disability, or even unhealthy complications [2]. Multimorbidity is especially related to functional incapacity, as well as to apparent frailty, especially in elderly individuals [3]. Some examples of chronic illness that may result in functional decline, debility and/or frailty include cancers, cardiovascular diseases, respiratory disorders, and diabetes [2]. Type-2 diabetes is one of the most complex diseases and is characterized by hyper- or hypoglycemia, accompanied by a significant change in insulin sensitivity or secretion [2]. Although diabetes mellitus is not contagious, it is a public health problem that continues to grow due to its prevalence and related complications [4]. In addition, the health costs attributed to diabetes are constantly rising, reaching 629 million US dollars by 2045 [5], thereby implicating diabetes as a major public health dilemma, which is forecast of being the 7th highest cause of death in the world by 2030 [6]. It is a chronic epidemic condition affecting several countries. It is a major public health problem in China, and its prevalence reached 11.2% among adults in 2017, increasing from 9.7% to 2007 [7]. This is due to a lack of physical exercise, overweight status and even the adoption of an inappropriate diet [8]. A 2016 study from China (N = 15,404) and a 2019 study from Sudan and India (N = 55,639) reported that 80% of patients with diabetes were unable to reach their glycemic goals [9]. In Morocco, the prevalence of diabetes has increased over the years, presenting a major public health concern. According to the Ministry of Health, the national estimated prevalence of diabetes was 10.6% in 2018. In addition, the cost of care with medication nonadherence exceeds $100 billion annually [10]. These widespread chronic diseases have impacted the socioeconomic burden of counties as well as individuals and their families [11]. Type-2 diabetes, which is often diagnosed incidentally during a routine check-up, accounts for 90% of the world’s diabetes population [12]. The condition has serious micro- and macrovascular complications that may develop before the clinical diagnosis of the disease itself. The fundamental goal of self-management of diabetes is controlling blood sugar, maintaining normal or near-normal A1C, lifestyle modifications such as exercise and diet, and adhering to medication regimens to prevent or delay complications. The implementation of these health recommendations is described as therapeutic adherence. [13]. However, nonadherence remains a real problem in patients with chronic conditions, especially those with diabetes mellitus (DM); it deteriorates the patient’s health condition, increases the risk of treatment failure and hospital visits, and finally creates a financial burden to the patient and the health care system. To measure patients’ adherence to their treatments, there are several validated tools to measure medication compliance outside Morocco [14]. Among these tools, we cite The General Medication Adherence Scale (GMAS), which has already been validated in several countries, such as Saudi Arabia [15, 16]. To our knowledge, we do not have such tools validated in our Moroccan context, hence the importance of this work, which comes to fill this gap. Therefore, the objective of our study was to cross-culturally adapt and validate the GMAS for patients with diabetes in Morocco.

Diabetes is a chronic condition requiring both routine and complex individual measures [4]. Efficient therapeutic compliance involves developing knowledge and skills through regular monitoring of blood glucose levels to prevent long-term complications, as well as adopting a valuable strategy for improving the health behaviors of these patients with diabetes [4]. Characterization of medication adherence using a self-report tool is the most common, effective and cost-effective way of assessing medication adherence [22]. Various self-assessment tools have been developed, including the Shea Scale [23], the Adherence to Refills and Medications Scale (ARMS) [24], the Morisky Medication Adherence Scale (MMAS) [25] and others. However, many studies have reported that no single tool qualifies as a standard for estimating adherence [22]. GMAS was designed with the shortcomings of the above instruments in mind. The objective of our study was the cross-cultural adaptation and validation of the GMAS tool in the Moroccan context after it was used by Naqvi et al. in Pakistan [17], an English-speaking population (January 2018), Saudi Arabia (April 2018, with the English language) [26] and elsewhere in the Arabic language in December 2019 [16]. The GMAS tool has been translated and cross-culturally validated in the Moroccan context in the present study. Theoretically, the number needed to successfully validate the scale should be in the range of 55–110 patients [27, 28]; however, the present study was able to exceed this figure by collecting data from 284 patients. This sample size was larger than that used in the study conducted in Saudi Arabia [26] and similarly larger than that employed in another study that was able to validate the eight-item Morisky Medication Adherence Scale (MMAS-8) in Saudi Arabia [29]. This aspect could be qualified as a strong point of our study. The factorial validity was estimated using a CFA by calculating suitability indices such as NFI, CFI, TLI, IFI and the KMO measure. These fit indices in our study had a value of 1, and they were in a logical range (> 0.95). The KMO value was 0.775 (> 0.7). The study conducted in Pakistan revealed that these indices, NFI, TLI, and CFI, had the following values: 0.93, 0.93, and 0.97, respectively, with a KMO value of 0.832. For the study conducted in Saudi Arabia in English, the indices had the following values: 0.93, 0.99, and 0.99, respectively, with a KMO value of 0.705. For the assessment of GMAS in Saudi Arabia in Arabic, they had the following values: 0.960, 0.954, and 0.979, respectively, with a KMO value of 0.870. All these above reported values confirmed the goodness of fit of this three-factor model GMAS tool. Convergent validity is established if the value that concerns the average factor loadings is ≥ 0.7 [30]. Our study revealed an average factor of 0.749 for the three constructs, although in the GMAS tool utilized in Pakistan, this loading had values of 0.70, 0.73, and 0.76, respectively, for the three constructs. However, the English GMAS versions used in Saudi Arabia were 0.75, 0.70, and 0.72 for the three constructs, respectively, and the factor loading of the GMAS in Arabic was 0.725 for all constructs in Saudi Arabia. The construct validity of the tool was estimated via the correlation of the adherence score with the pill burden: the number of medications prescribed by the treating physician. To assess the correlation, Spearman’s σ correlation coefficients were calculated, and the validity was established in the case of a correlation coefficient σ ≥ 0.3, with a p value < 0.05 [31‐33]. In our study, σ had a value of 0.83, with p < 0.001, while σ for the GMAS in Arabic had a value of 0.388 with p < 0.001 in Saudi Arabia. The reliability of the GMAS tool was assessed through Cronbach’s α. In addition, internal consistency was assessed via the item-total correlation (ITC) and ICC. Our study revealed a Cronbach’s α of 0.804 for the three constructs, which was less than that of the following studies: the Pakistani study, which had a value of 0.819; the Sudanese study [34], which had a value of 0.834; the Saudi study (in Arabic), which had a value of 0.865; and the Vietnam study [9], which had a value of 0.879. However, the Cronbach’s α found in our study is higher than that of the study conducted in Saudi Arabia in English, which had a value of 0.740. The present study encountered a number of limitations that need to be specified: the sensitivity, specificity, and accuracy of the scale were not established, and these three validation modalities would have contributed to the strength of the measurement tool. However, the other validity measures that were developed, combined with the acceptable reliability that was demonstrated, reflect the scale’s suitability for assessing adherence in the target population.

The Moroccan Arabic-dialect version of the GMAS tool was translated and validated in Moroccan patients with type-2 diabetes and was able to satisfy the validity criteria for the majority of psychometric parameters. As a result, health care professionals caring for patients with diabetes can use this tool to measure adherence. This is important because physicians, nurse practitioners, physician assistants, and even pharmacists manage diabetes patients.