The gap between efficacy shown in clinical trials and effectiveness realized in routine care is widening, driven in large part by medication nonadherence. Contemporary therapies produce larger clinical benefits than older agents, yet a substantial proportion of patients discontinue, underuse, or never fill prescriptions. When adherence rates remain roughly static while treatment effects increase, the benefit forgone through nonadherence grows accordingly.
Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) illustrate this problem clearly. A synthesis of real-world evidence reported that between 20% and 50% of patients discontinue GLP-1 RAs within the first year. Many patients also use doses lower than those studied in randomized trials, and weight-loss observed in practice is consistently lower than trial outcomes. Importantly, patients with high adherence approach trial-level weight loss.
Drivers of early discontinuation include gastrointestinal side effects, cost, insurance barriers, and supply shortages. Clinical consequences depend on the indication: for weight management alone, stopping typically leads to weight regain; for patients receiving GLP-1 RAs for cardiorenal benefit or kidney outcomes, discontinuation forfeits ongoing reductions in morbidity and mortality because those benefits require continued exposure rather than producing durable protection after withdrawal.
Adherence problems extend beyond weight-loss therapies. A systematic review and meta-analysis of nearly 595,000 patients with atrial fibrillation found that only about two-thirds maintained good adherence to direct oral anticoagulants (DOACs). Nonadherence was associated with a 39% increase in the risk of stroke. This quantifies the gap from prescribed therapy to delivered benefit even for medications with clear value and simple dosing regimens, reframing nonadherence as a measurable contributor to preventable morbidity and mortality.
The two-dose recombinant zoster vaccine (RZV) also shows adherence challenges. A nationwide US study of more than 726,000 adults found adherence to the recommended 2–6 month schedule was 72%. Completion rates were lower among racial and ethnic minorities, younger adults, and people with lower household income. Emerging longitudinal analyses link herpes zoster vaccination with reduced dementia risk and suggest stronger protection among those who received both doses versus one; if confirmed, failure to complete the series could forgo not only protection against shingles but a possible reduction in dementia risk.
The central claim is that the opportunity cost of nonadherence—benefit forgone when effective therapies go untaken—has risen because modern treatments are more effective than prior options. Examples cited include semaglutide producing substantially greater weight loss than older agents, DOACs reducing stroke or systemic embolism relative to warfarin, and RZV offering markedly greater zoster risk reduction than its predecessor. These improved treatment effects amplify the clinical and public health consequences when patients do not adhere in real-world settings.
Nonadherence is heterogeneous. It ranges from cost-driven gaps and insurance barriers to unintentional lapses in complex regimens. Not all discontinuation is inappropriate; stopping therapy after informed shared decision-making about adverse effects is distinct from nonadherence. Research and interventions should differentiate these scenarios rather than conflating all discontinuation as failure.
Traditional interventions—patient education, simpler dosing schedules, and improved clinician–patient communication—remain necessary but frequently insufficient. A Cochrane review of 182 randomized trials found that even the most effective adherence strategies, typically bundles combining education, counseling, and ongoing support, yielded only modest improvements in adherence and clinical outcomes.
At the point of care, clinicians can treat adherence as a decision point: discuss the cardiorenal benefits lost if a GLP-1 RA is stopped, verify anticoagulant persistence at each renewal, and counsel patients on completing the RZV series. However, individual-level efforts have limits in the face of structural barriers.
A system-level reconceptualization treats adherence as a determinant of therapeutic value and targets the contexts in which prescriptions are written and filled. Potential system-level strategies described include:
A cited randomized trial after myocardial infarction eliminated copayments for cardiovascular medications for nearly 6,000 patients, improving adherence by 4%–6% and reducing first major vascular events without increasing total health spending—an example of system-level change yielding both adherence and outcome gains.
Nonadherence is difficult to detect in routine practice; clinicians tend to underestimate it. Common measurements—pharmacy refill records, self-report, and EHR documentation—each provide only partial visibility. The article notes that artificial intelligence may eventually help identify nonadherence more reliably, but rigorous evidence that AI tools improve clinically meaningful endpoints remains limited.
The limiting factor in realizing the full value of modern therapeutics is no longer solely pharmacology; it is also human behavior shaped by health systems. The author argues for investments in system-level solutions—coverage and funding design, pharmacist-led programs, health information technology, and implementation research—to close the widening adherence gap. Without such investments, health systems will continue to pay for the promise of transformative therapies while realizing only a fraction of their benefit.