Noncommunicable diseases (NCDs), also known as chronic diseases, have a long duration and are influenced by a combination of genetic, physiological, environmental and behavioural factors. In 2021, the WHO released data showing that chronic diseases cause 41 million deaths annually, accounting for 71% of all deaths globally[1]. In China, chronic diseases accounted for 88.5% of total deaths in 2019, and the situation is increasingly becoming concerning[2]. Patients with severe chronic disease inflict a heavy economic and emotional burden on their families[3]. So effective prevention and control measures should be taken to prevent, manage and control the development of chronic diseases.
Better medication adherence can be called as the key point between the doctors' health management measures and patients' disease control. The use of medication per prescription requirements is still the primary means of managing and controlling chronic diseases, and patients' medication adherence should be given more attention[4]. Medication adherence means that patients take the correct dose of medication at the correct time[5]. However, previous studies have shown that the current state of patient medication adherence remains a substantial global challenge, with nearly 50% of patients failing to use their medication as prescribed. This leads to additional deaths, increased hospitalisation rates, and increased financial burdens on healthcare services[6].
With the continuous development of the Internet, various modern health management measures are emerging. In the Covid-19 era particularly, chronic disease management has been able to take advantage of new systems in telemedicine, mHealth, smartphone applications and other intelligent tools[7, 8]. Medication monitoring devices with pill counts, electronic monitoring devices, big data, and more, offer new perspectives on monitoring patients' medication adherence[9]. Although these measures have achieved good results, they are not entirely effective for clinical applications as they are laborious, costly, and even invasive[10]. A more comprehensive, convenient, efficient, and accurate assessment of patients' medication adherence behaviours is required so that patients with poor adherence can be identified and intervened with early.
Current scales are a good tool for assessing patients' medication adherence. The scales are time- and labour-saving and do not require much financial support. However, most scholars believe that no scale can be used as the gold standard to measure medication adherence[11].Among the medication adherence evaluation scales developed for chronic patients, the series developed by Morisky's team is the most well-known and widely used. Morisky first developed the original version of Medication Adherence Questionnaire (MAQ) from research on hypertension patients in 1986[12]. It has only four entries and is simple and easy to handle, making it suitable for primary screening. Considering the influences of the surrounding environment, Morisky amended the MAQ Scale in 2008 and developed the 8-item Morisky Medication Adherence Scale (MMAS-8)[13]. It has excellent psychometric characteristics and good applicability in diabetes[14], hypertension[15], mental disorders[16], and other diseases. The scale has been translated into many languages and applied worldwide. However, it does not consider the patients' economic level. In low- and middle-income countries, the economic situation is an important factor affecting patients' medication adherence[17]. The Brief Medication Questionnaire (BMQ) can evaluate different noncompliance behaviours. However, the evaluation system is relatively complicated, and patients need to recall the specific medication name, frequency and medications taken or missed. This evaluation typically requires a lot of time[18]. The Hill-Bone Impairment Scale [19], developed in 2000, has 14 entries that evaluate drug use, sodium intake, and appointment plans. It is usable for patients with low literacy but is limited in its development. Only patients with hypertension can use it, but it is sensitive to cultural differences. The self-efficacy for appropriate medication use scale(SEAMS) with 13 questions was developed in 2007 by Risser et al.[20], and evaluated mainly on self-efficacy, rather than Medication behaviours. Although SEAMS is an excellent self-report tool with good psychological measurement characteristics, but the limitation is that it is difficult to calculate the score and is time-consuming[21].
The General Medication Adherence Scale (GMAS) was initially developed by Naqvi scholars in the Urdu language in 2018 and was validated in Pakistan[22]. In subsequent years it was translated into English[23], Chinese[24], Nepalese[25], Vietnamese[26], Arabic[27], and others. This was done to verify the scale in different countries and regions and for specificity evaluation in chronic diseases, such as rheumatoid arthritis[28] and type II diabetes[29]. In these studies, classical test theory was used to verify the scale from the perspective of reliability and validity. Its effectiveness and psychological measurement characteristics in the population were confirmed. Most of the studies reported relatively good medication adherence in the patient population. According to the selection of health Measurement Instruments guidelines, the GMAS scale achieved better results in measured attributes than previous scales[30]. Although the scale has been developed later and competes with modern intelligence technology, the GMAS can still be seen as a reliable tool.
The Rasch model is a latent trait model proposed by Georg Rasch, a Danish mathematician and statistician[31]. It can measure latent variables that cannot be directly observed through individual performance on social survey questions. Moreover, the Rasch model has a solid theoretical basis and rich mathematical logic expression. This fundamentally improves the objectivity of the investigation and analysis of the research object. However, this idealised mathematical model also requires that the empirical data meet the a priori conditions of the model: that the source is credible and valid; that other external attributes do not confound the internal attributes; and that it meets the requirements of theory-driven research and data-adapted models[32]. The GMAS was conceived using two main concepts. These were the concept of medication adherence—as described by the taxonomy of Vrijens et al.[33] to assist the structural design of the scale and affordability of drugs as a potential research tools structure. As a scale similar to the measurement of medication psychological problems, GMAS is more likely to be suitable for evaluation with the measurement method of modern psychological questionnaire. The Rasch analysis has been widely used in the evaluation of various psychological, medical and nursing scales, and its two-factor model is more conducive to the analysis of the performance and characteristics of the scales.The Rasch analysis can be used to further verify the GMAS scale from the aspects of difficulty and discrimination.
Based on the above characteristics, the research group plans to fit the Rasch model to GMAS data. The aim is to further verify or improve the GMAS, consider the scope of application, and provide a theoretical basis for subsequent application in nursing practice.