4. Discussion
The present study has highlighted that 70% of the patients on secondary prevention therapy for CVD have an inadequate adherence in the National Institute of Cardiology ”Ignacio Chávez” in Mexico City. In agreement with this finding, a population-based study in Taiwan reported statin discontinuation of 68% at five years.7 These results contrast with previously reported in the United States, where adherence ranged between 50%8 and 87%9within the first year of treatment initiation, declining up to 45%7 in the second. Although our study did not intend to assess adherence over time, our adherence rate may be one of the lowest reported, reflecting the discrepancies observed across countries and ethnic groups.
According to the WHO, nonadherence arises from the interplay between patient-related barriers and socioeconomic, health care system, disease, and therapy-related factors.10 In patients under statin therapy, adverse effects were the leading cause for statin discontinuation in a previous study, reporting that 25% of statin users believed statins cause memory loss,11 a rare and inconsistent finding in the literature.12,13Furthermore, Ahmed ST et al. reported that physicians preferred to avoid explaining all potential adverse effects to patients as it could influence their medication-taking behavior.14Conversely, Rodriguez F et al. observed that patient preference was a higher determinant for statin withdrawal than adverse effects.9 This result does parallel with ours, leading to the suggestion that suboptimal health literacy may account for a significant proportion of nonadherence, reflecting an inadequate doctor-patient relationship. In this regard, a study found statin safety and risks concerns were the main reason for withdrawal, and those nonadherent were more skeptical of having a higher risk of myocardial infarction and stroke with high cholesterol levels than their adherent counterparts.11 Thus, effective communication and education on the natural history of disease and treatment characteristics might be needed to clarify patients’ concerns and demystify statin impact on CVD. Moreover, clinicians should promote adherence by encouraging a higher patient-physician involvement, potentially ameliorating therapy misconceptions, and decreasing patient misinterpretations about their condition, henceforth decreasing nonadherence. In this framework, a significant trend toward statin treatment reinitiation willingness (59.7%) was observed in those patients who discontinued statins after dialogue with their doctor.11
In the current study, forgetfulness accounted for the most prevalent cause of nonadherence (53.5%), likely exposing underlying prescription complexities beyond medication-taking behavior such as multi-drug regimen, different dosing times, blood pressure, and glucose self-tracking, and lifestyle modifications.15 In this sense, polypill implementation could offer potential advantages over conventional pharmacotherapy as it provides an easy-to-remember monotherapy regimen. Multiple studies have reported the cost-effectiveness of a polypill strategy for secondary prevention of CVD, observing an increase in adherence rate, reduction in CVD burden,16-18 and dose titration.19Nonetheless, their clinical employment in diverse populations might be hampered by healthcare systems and socioeconomic disparities across countries. Although some randomized clinical trials in developing countries 17,20 and underserved populations in the United States16 demonstrated its cost-efficacy, drug prices need to be reduced to improve access, availability, and affordability in these countries.21,22 Hence, polypill strategy might be challenging in low- and middle-income countries owing to higher out-of-pocket costs than conventional pills. In this context, limited access to care, medication costs, and low socioeconomic status might account for poor adherence in low- and middle-income countries, as observed in this study, where 24% of the patients reported drug discontinuation for economic factors. Likewise, although forgetfulness might play a role in some patients, particularly in those with mild cognitive impairment and aging patients, underlying causes such as personal indifference to well-being, low self-esteem, suboptimal health literacy, age misconception, absence of family support, polypharmacy, and depression should also be considered as they may contribute to this phenomenon as well. Early decline and suboptimal adherence in such patients could be prevented by frequent communication and assessment of patients’ environment and psychosocial and cognitive status, particularly in patients with clinical and subclinical depression.
Even though physician-related barriers were the less prevalent (1.2%) in our study, lipid-lowering guidelines’ misunderstanding might also account for poor adherence, leading to suboptimal risk reduction in those who may benefit from high-intensity therapy.23 A previous study observed higher adherence patterns in patients taking moderate-intensity statin therapy than patients on low-intensity or high-intensity statin regimens.9 While, Khunti K et al. found that patients receiving high-intensity therapy were the more adherent group.24According to a study, stable LDL-C levels or at goal was the most prevalent cause reported by 42% of physicians for treating patients with a lower statin intensity than recommended.25 Consistently, statin therapy discontinuation by physicians was observed in a large proportion of patients with adverse effects after the index event, receiving a lipid-lowering drug alternative in only 2.6% of the cases.25 Moreover, high to moderate adherence has shown to lead to more significant reductions in total cholesterol and LDL-C levels than statin discontinuation and/ or withdrawal.26 These findings could explain our observations as higher cholesterol, triglycerides, and hemoglobin A1c levels were found in those patients with inadequate adherence. In such patients, physicians might seek to maintain a high to moderate statin regimen in order to reduce overall atherosclerotic cardiovascular disease (ASCVD) risk and improve medication-taking behavior, as patients seem to be more conscious of the importance of medication adherence and disease control.
Furthermore, the number of consultations and the level of care might also impact secondary prevention therapy adherence. In a study performed at the Veterans Affairs system, a higher adherence was observed in those with regular primary care physician (PCP) control, reporting adherence improvement in those patients with at least one visit compared to those without any consultation.27 Rehman H et al. found that patients under cardiology control had 9% higher odds of being more adherent in comparison with those under primary care management.28 This finding seems to differ markedly from ours as the National Institute of Cardiology ”Ignacio Chávez” is one of the highest accredited healthcare organizations in Cardiology in Mexico. However, it is worth mentioning that most of the patients do not have any further medical control and depend solely on their annual Cardiology consultation at the institute. In this understanding, the low adherence found in the institute could portray the country’s health care scenario. Therefore, it could be hypothesized that adherence to statins and renin-angiotensin-aldosterone system blockers, and potentially other medications, might be more severe in other country institutions.
Further analysis observed that low school level and optimal medical therapy (OMT) without revascularization were independent predictors for nonadherence. In such patients, an invasive approach by revascularization or coronary artery bypass graft surgery (CABG) may account for a proper adherence pattern, likely behaving as a placebo. Likewise, these patients may be more knowledgeable about their disease and nonadherence potential consequences as opposed to those under OMT alone, exhibiting more adequate adherence patterns. This discrepancy could be attributed to patient misconception on their condition and medication unawareness due to chronic diseases’ asymptomatic and slowly progressive nature and the lack of short-term clinical evidence of medication administration benefits. From this perspective, nonadherence may arise from ineffective implementation tools of intensive pharmacologic and lifestyle intervention, disclosing current approach failure and the need for strategies reassessment to impel more valuable and efficient enforcement methods. In this sense, whether the initial strategy should be conservative or interventional is still a matter of open-ended debate. Although multiple trials have been conducted to compare the effectiveness of OMT with and without percutaneous coronary intervention (PCI) on cardiovascular mortality, nonfatal MI, and all-cause mortality reduction, none has shown a difference between groups.29,30 In contrast with these conclusions, observational studies have reported lower mortality and cardiac deaths in the PCI arm.31 This disagreement could be attributed to the closed control observed in randomized clinical trials, potentially conferring a predictive model with ideal characteristics which might not represent a real-world setting. Therefore, when superimposed these observations with our findings, it could be hypothesized that observational studies demonstrated better outcomes with invasive strategies by better reflecting real-world variabilities such as medication adherence. Henceforth in order for the OMT to achieve the results observed in clinical trials, it might be necessary to generate effective patient-based adherence strategies.
On the other hand, low school level was the most important predictor for nonadherence. The odds of nonadherence were 2.37 times higher in those patients with elementary school or less, finding similar results to previously documented in a former study.23 This finding could potentially explain adherence rate discrepancies observed with high-income countries, and it could be extrapolated to other low- and middle-income countries as well, reflecting the complex interrelation of economic burden and medication-taking behavior. Hence, medication nonadherence may compel all institutions to develop population-based adherence instruments to address the problem in an integral manner since the spectrum of adherence seems to overcome cultural and ethnic barriers. Besides, adherence might differ among institutions. Measuring adherence data as a quantitative analysis could probably be of limited value. Thus, a new method should be implemented to measure this complex parameter, in which case screening tools could be used to identify high-risk persons on the basis of all five nonadherence-related factors and clinical prediction algorithms. Nonetheless, more data is needed to evaluate these hypotheses, and a properly designed multicenter registry study must be conducted to complement these results with a representative sample of Mexico.
In this area of high unmet medical need, these findings should raise concern among physicians as nonadherence can lead to suboptimal risk reduction in high-risk populations, negatively impacting not only patients’ quality of life and survival rate but also increasing direct and indirect health care system costs. Consistently, these findings could provide incremental value to promote interdisciplinary units and develop patient-centered health policies to improve patients’ quality of life and survival rate in high-risk populations to attenuate the CVD burden. Thus, new and targeted specific population-based strategies are needed to improve the current adherence situation in the country.