From the perspective of healthcare providers, therapeutic compliance is a major clinical issue in RA patients as it impacts disease’s outcomes [5, 17, 18]. Inadequate medication adherence (which includes three major components: persistence, initiation adherence and execution adherence) causes an increased financial burden for society as it has been associated to excess emergency care visits and hospitalizations, higher treatment costs and loss of productivity. Furthermore, as a result of undetected or unreported therapeutic non-compliance, physicians may change the regimen, which may increase the cost or complexity of the treatment and eventually the incidence of adverse events. One logical target in trying to complete the riddle of therapeutic non-compliance in RA would be to identify most common associated factors from the patient’s perspective and to identify predictors as we did in the present study developed in a well characterized population of Mexican Mestizo early RA patients.
We found that 2 out of 3 patients were classified as NP. In the literature, different studies have targeted adherence to DMARDs [5, 6, 9, 10, 13, 15, 16, 24] and shown that the extent to which patients adhere to DMARD therapy varies between underuse and overuse. Such variations may be explained by differences in sample’s size, methods capturing medication adherence, variable disease duration, follow-up, disease activity and therapeutic modalities, although studies confirm that adherence in RA is suboptimal. An important point to consider is that medication adherence is a dynamic feature, not stable over time. We found that 25% of our patients were consistently NP similar to have been reported in longitudinal studies performed in other RA populations [5, 14, 25].
Factors identified from studies and reviews with poor compliance may be grouped into several categories, and divided into 5 domains according to the World Health Organization: namely patient-centered factors, therapy-related factors, healthcare systems factors, social and economic factors, and disease factors [20, 26]. In our study, NP patients were directed to select at least one factor from a list; factors were included in the list as all of them have been shown to impact compliance in different populations [17, 20, 26]. Most frequent patient’s motivations for NP were forgetfulness, lack of financial resources and lack of availability at the drug store. Forgetfulness is a widely reported factor that causes non-compliance with both, medication and clinical appointments in different populations, including Mexican patients with type2-diabetes [27, 28]. Meal frequency has been shown to be an effective tool to remind the patient to take his medications in Japanese patients  and this strategy could be intensified in order to improve compliance. Interestingly, in our study “forgetfulness” was highly correlated to the motivation “Because timing/s when my medication is prescribed is different from mealtime/s “(Rho: 0.92, p = 0.001, data not shown). Also, written instructions are better than oral advice for reminding patients to take their medications and we recommend it implementation for every patient, and at every appointment.
Patient’s perception of at least “some difficulty to find arthritis medication” (previously reported as lack of availability at the drug store) and that “arthritis medication is expensive” (lack of financial resources) were among the most frequent motivations for NP; when both of them were selected at the first evaluation of patient’s compliance, they predicted NP (in addition to a higher education level) and also recurrent- NP. Cost is a crucial issue in patient’s compliance especially for patients with chronic diseases [28, 30]. A number of studies have shown that patients who had no insurance cover or who had low incomes (as our population of patients who had to pay for their medication) are more likely to be non-compliant when compared to patients with health insurance or relatively high incomes [27, 31–34]. In RA patients, inadequate or nonexistent reimbursement by health insurance plans has shown to negatively affect adherence to biologics . Also, among the identified health care systems factors that contribute to poor compliance are lack of availability and accessibility to healthcare ; a significant percentage of our patients selected lack of availability of the medication which is related to the former. Finally, regarding education level, intuitively it may be expected that patients with higher educational level should have better knowledge about the disease and therapy benefits and accordingly better compliance. García-González et al.  found an association between lower education and lower adherence in 102 ethnically diverse patients from Texas, among whom 72 had RA. However, similar to our findings other researchers have shown that non-RA patients with lower educations levels have better compliance [35, 36]. One possible explanation may be that patients with lower education level might have more trust in physician’s advice. DiMatteo proposed that even highly educated patients may not understand their conditions and the benefits of being compliant .
In our study, most frequent patient’s motivations for NP were “unintentional” motivations. As opposed to intentional motivations, they reflect a person’s ability and skill at medicine taking, including forgetting, poor manual dexterity, losing medicines or not being able to afford them. Meanwhile, intentional non-adherence is a behavior driven by the decision not to take medicines [11, 25, 38]; drivers of this decision have been suggested to be based on patient’s beliefs about its illness and its treatment, which can be further categorized as perceived benefits and perceived concerns . Neame et al. reported that most people with RA had positive beliefs regarding the necessity of their medication but levels of concern were also high and were positively associated to poor compliance as 91% of non-adherent RA patients had at least one concern about potential adverse events . Besides the valuable conceptual distinction between intentional and non-intentional motivations of non-adherence in RA patients, a practical approach to poor patient’s medication behavior will be to identify individual main drivers of poor compliance and tailor the content of the adherence-improving-intervention to the individual patient’s motivation of non-adherence . This comprehensive strategy may be more effective that traditional adherence intervention programs in RA that have shown inconsistent and limited effects [41–43]; furthermore, it can be applied in our population as 79% of our patients selected non-intentional motivations and among them, 66% selected exclusively non-intentional motivations during their follow-up meanwhile 28% selected both, intentional and non-intentional motivations (data not shown). This finding is in agreement with the fact that there were no major differences between active and inactive patients with NP.
Some limitations of the study need to be addressed. We did not use a well-validated questionnaire scale to assess persistence; we applied a short-patient-oriented questionnaire, locally designed which has shown adequate internal consistency, high sensitivity and satisfactory specificity to detect persistence on DMARDs . We did analyze neither the construct of adherence nor major factors associated. This study was done in an inception cohort of early RA patients, with particular socio-demographic characteristics, ethnicity, treatment and health system and our results may not be generalized to RA populations with different characteristics .
Compliance with medication is a dynamic process and fluctuates over time; with a more extended length of follow-up, patients formerly classified as persistent with therapy may become non-persistent. We limited the study of factors associated to medication persistence, to patient’s motivations; ultimately, it is the patient who decides whether or not to take his/her medication as prescribed, and non adherent patient’s opinion are essential in order to design effective interventions. Finally, we investigate a limited number of patient’s motivations for non-persistence with medication although they were selected based on the existing literature.