Results
The included studies are grouped according to the following categories: Socioeconomic, Treatment, Condition, Personal, and Healthcare-related factors, based on the WHO dimensions of patients’ adherence to treatment.

Socioeconomic factors

Socioeconomic factors can be divided into those factors related to social or environmental variables, economic factors, and those related to the lifestyle of patients. Among the studies analysed, we identified four studies that showed a significant effect of social or environmental factors (i.e., social interaction and support networks) on treatment adherence (Siregar & Andayani, 2020; Zullig et al., 2015; Reddy et al., 2017; Park et al., 2015). Concerning the set of economic factors, several studies reported a significant association between adherence to treatment and financial status (Wooldrich et al., 2015; Crowley et al., 2012; McAlister et al., 2019; Shankari, 2020), education and literacy (Siregar and Andayani, 2020; Crowley et al., 2012; Al-Haj Mohd et al., 2016), employment (Sieben et al., 2021; Zullig et al., 2015; Crowley et al., 2012), and living condition of patients (Meggetto et al., 2017; Tola et al., 2016; Wooldrich et al., 2015). This SR also identified scientific evidence on the effect of patients’ lifestyle on treatment adherence. The lifestyle factors with a reported significant effect are substance use and abuse (Meggetto et al., 2017; Sieben et al., 2021; Llorca et al., 2021) and physical activity (Nascimento et al., 2016; Shankari, 2020). Among the studies reviewed, no reference was made to the study of the effect of the social situation of the patient in adherence to treatment. For full reference to the data extracted, see Table 3 and Supplementary Material; Table S3.
–Table 3 should come around here –

Factors related to the Healthcare System

The healthcare system-related factors were divided into two sets of factors: those concerning the relationship between the patient and the healthcare professional (HCP), and those directly related to the healthcare system. In this SR, ample evidence showed that provision of patient education, training, and follow-up of the patient by the HCP significantly increased adherence (Tola et al., 2016; Bonetti et al., 2018; Hohmann et al., 2014; Hovland et al., 2020; Kamal et al., 2015; Ababneh et al., 2019; Wan, 2016; Asgari et al., 2021; Alfian et al., 2020; Jahn et al., 2014; Wang et al, 2020). Moreover, the patients’ trust in their HCP (McAlister, 2019) and HCPs’ time available for consultation (Colvin et al., 2018) were also found to have a significant effect on patients’ adherence to treatment. When considering only the healthcare system-related factors, it was found that both the provision of feedback and training to the HCP and the support of the community influence patients’ adherence to treatment (Grigoryan et al., 2012; Mitchell et al., 2015). Among the studies reviewed, no reference was made to the study of the effect of the “quality and cost of health services”, “Provider continuity”, “Regulation process”, or “Drug supply” in adherence to treatment. For a complete reference to the data extracted, see Table 4 and Supplementary Material; Table S3.
–Table 4 should come around here –

Disease-related factors

The third dimension of adherence considered in this SR concerned disease-related factors. Two studies found evidence for the effect of progress, duration, and severity of the disease, and its symptomatology as an influencer of adherence (Nieuwkerk et al., 2012; Al-Haj Mohd et al., 2016). Furthermore, several studies identified the existence of comorbidities as a factor significantly affecting adherence to treatment (Grigoryan et al., 2012; Park et al., 2015; Shankari, 2020). In addition, the level of disability caused by the condition at the physical, psychological, social, and vocational levels has also been found to play a significant role in the level of patients’ adherence to treatment, according to three articles (Crowley et al., 2012; Nascimento et al., 2016; Laba et al., 2018). For full reference to the data extracted, see Table 5 and Supplementary Material; Table S3.
–Table 5 should come around here –

Treatment-related factors

Several adherence factors associated with patients’ treatment have also been identified as modifiers of adherence. These treatment-related factors can be further categorized as factors related to the treatment regimen, the effects of the treatment, and the treatment properties. Regarding the treatment regimen, ten articles in this SR reported scientific evidence that complexity and duration of the treatment have a causal effect on patients’ adherence levels (Flicoteaux et al., 2017; Gillespie et al., 2014; Kuypers et al., 2013; Suffoletto et al., 2012; Wooldrich et al., 2015; Sieben et al., 2021; Matsumura et al., 2013; Llorca et al., 2021; Calvo-Arbeloa, 2019; Kamal et al., 2015). Similarly, another study identified how the treatment properties, specifically the formulation and physical properties of the medication, had a significant effect on the patients’ adherence levels (Hohmann, 2014). Focusing on the treatment effects, only one article found evidence that the appearance of beneficial effects or side effects and experience of failures in previous treatments to influence adherence (Shankari et al., 2020; Meggetto et al., 2017). Among the studies reviewed, no reference was made to the study of the effect of the “Interference in the routine of the patient” or “Cost of treatment” in adherence to treatment. For full reference to the data extracted, see Table 6 and Supplementary Material; Table S3.
–Table 6 should come around here –

Patient-related factors

The final dimension in the WHO framework is patient-related factors, which was further divided into three sets of factors: unalterable characteristics, cognitive and psychological factors, and behavioural factors. Regarding the first factor, ample studies in this SR identified demographics to play a significant role in adherence to treatment (Suffoletto, 2012; Crowley et al, 2014; Sieben et al., 2021; Mohan et al., 2014; Crowley, 2012; Grigoryan et al., 2012; Al-Haj Mohd et al., 2016; Matsumura et al., 2013; Llorca et al., 2021; Shankari et al., 2020; Laba et al., 2018; Calvo-Arbeloa, 2019), while only one study showed this significance for experience with treatment and treatment setting (Beckers, 2013). Also, the physical characteristics of the patients were found to be predictors for adherence to treatment in this SR (Meggetto et al., 2017; Crowley, 2012). Among the cognitive and psychological factors that were studied, health literacy (Crowley, 2012), perceptions, beliefs, and concerns of the patients regarding their condition (Crowley et al., 2014; Nieuwkerk et al., 2012; Crowley, 2012), patients’ knowledge about their treatment (Beckers, 2014; Al-Haj Mohd et al., 2016; Shankari et al., 2020), and patients’ knowledge about their disease (Beckers, 2013; Siregar & Anadayani, 2020; Nieuwkerk et al., 2012; Llorca et al., 2021) were reported as predictors for adherence to treatment. Finally, some behavioral factors were found to have an effect on adherence, such as the lifestyle of the patient (Gillespie et al., 2014; Mugo et al., 2014; Llorca et al., 2021; Shankari et al., 2020) and the planning abilities and self-efficacy, which were found in three studies (Kuypers et al., 2013; Crowley et al., 2014; Llorca et al., 2021). The table below shows the results of the quantification of the factors affecting adherence based on the SR of experimental studies. For full reference to the data extracted, see Table 7 and Supplementary Material; Table S3.
–Table 7 should come around here –

Inclusion of covariates

In total, 7 of the 47 studies analysed for this SR reported having controlled the effect of any covariate. From those studies, the table below presents which factors these studies analysed as covariates. As Table 6 shows, demographic factors are the most used as covariates in the reviewed studies. These are followed by factors related to the characteristics of the treatment or disease, or the economic situation of the patient. Other factors are rarely analysed as covariates in the reviewed studies.
–Table 8 should come around here –