Why Young Adults Suddenly Stockpile Unused Pills: Huge Study Reveals Age-Specific Drug Risks

Every year, patients around the world walk out of pharmacies with bags full of medication they will never take. In Japan alone, the value of leftover drugs—pills dispensed but never consumed—is estimated at roughly 50 billion yen annually, a figure that represents both wasted medical resources and a quiet warning sign about the state of patient adherence. Now, one of the largest real-world investigations of its kind has mapped exactly who is most likely to end up with surplus medication, and the answer upends a common assumption: the biggest danger is not old age, but a surprisingly fragile moment in the lives of young adults.

The study, published in the Journal of Pharmacy and Pharmaceutical Sciences, analyzed records from 36,340 outpatients treated at a single Japanese hospital between 2016 and 2021. Rather than relying on patient questionnaires, which are notoriously unreliable when it comes to adherence, the researchers tapped into something far more objective: a protocol-based pharmacotherapy management system, known as PBPM, in which community pharmacists physically count and report leftover medications back to prescribing physicians. Under this standardized leftover drug adjustment protocol, pharmacists interview patients during dispensing, verify unused pills, and send objective reports that allow doctors to fine-tune prescription durations. Any patient with at least one such report during the five-year window was classified as having leftover drugs.

The headline number is deceptively modest: 9.6 percent of patients—3,472 individuals—had leftover drugs reported at least once. But the distribution across the lifespan is anything but uniform. The prevalence climbed steadily from just 0.4 percent in children under ten to a peak of 15.9 percent among patients in their eighties, before dipping back to 10.5 percent in those aged ninety and older. Patients in the leftover group were, on average, nearly seventeen years older than those without leftovers, and they carried roughly twice as many medications, received prescriptions lasting nearly twice as long, and visited significantly more prescribing departments.

What makes the study genuinely novel is its decade-by-decade dissection of risk. Using multivariate logistic regression with interaction terms, the team demonstrated that age does not merely add risk on its own—it fundamentally rewrites which prescription characteristics matter most. The statistical model, which achieved an area under the curve of 0.858, revealed significant interactions between age and three of the four primary factors examined: drug count, prescription duration, and the number of prescribing departments. Sex, by contrast, dropped out as an independent predictor once these prescription complexities were accounted for, suggesting that the fundamental drivers of leftover drug risk are structural features of the prescription itself rather than demographic attributes.

The most striking finding concerns young adults. For patients in their twenties, prescriptions lasting sixty days or longer carried an adjusted odds ratio of 102.15 compared with prescriptions under thirty days—a number so extreme that the researchers themselves caution it comes with wide confidence intervals and sparse event counts. Yet the pattern held across the twenties, thirties, and forties: long-term prescriptions were the single most potent risk factor for younger adults, dwarfing the influence of polypharmacy or multi-department care in these age groups. The researchers interpret this through the lens of episodic non-adherence. Among younger patients, single occurrences of leftover drugs predominated—nearly seventy percent of cases in the teenage group involved just one report—consistent with situation-dependent lapses, such as lifestyle disruptions or treatment changes, rather than chronic mismanagement.

Older patients tell a very different story. For those in their sixties through eighties, the dominant risks were polypharmacy—defined as five or more concurrent medications—and visits to three or more prescribing departments. The predicted probability models revealed a synergistic amplification: when patients on five or more drugs also received prescriptions lasting sixty days or longer, the probability of leftover drugs remained above 0.3 from the thirties through the eighties, eventually approaching 0.6 when multi-department care was added. This cumulative burden structure, the authors argue, reflects the reality that medication-related risks in older adults rarely stem from a single factor but emerge from the interaction of multiple components of pharmacotherapy.

Amid this complexity, one finding cut across every age group: keeping the drug count below five was consistently associated with a lower likelihood of leftover drugs, even when prescription duration or department count was unfavorable. In profiles where all factors sat at baseline levels—fewer than five drugs, prescriptions under thirty days, and a single prescribing department—the predicted probability of leftover drugs stayed below 0.05 across all ages. This suggests that deprescribing, the deliberate process of reducing inappropriate polypharmacy guided by tools such as the Beers Criteria and the STOPP/START criteria, may be the single most transferable strategy across the lifespan.

The middle decades present their own vulnerability. For patients in their forties and fifties, the sensitivity of risk to an increasing drug count was more pronounced than in any other age band. The researchers speculate that this cohort, often juggling employment and family responsibilities, is uniquely exposed when polypharmacy collides with extended prescription durations. For them, suppressing the medication count below five may serve as a practical buffer between the demands of work and the burden of treatment.

The study’s methodology deserves attention because it sidesteps a chronic weakness in adherence research. Self-reported surveys tend to overestimate compliance, and pharmacy refill records can conflate adherence with access issues. By contrast, the PBPM data capture what community pharmacists actually observed and reported—clinical facts negotiated between pharmacist and physician. The approach is not without blind spots: patients who concealed their leftovers, discarded them at home, or had prescriptions adjusted during the medical exam never entered the leftover group. The authors also acknowledge that leftover drugs do not exclusively signal non-adherence; dose adjustments, hospitalizations, treatment completion, and deprescribing can all produce surplus medication, making the indicator multifactorial rather than a direct measure of non-adherence.

Even so, the practical implications are hard to ignore. For younger adults, the data argue for starting with shorter prescriptions to gauge management capability, then using refills and pharmacist follow-up—potentially supported by information and communication technology—to maintain contact. For older adults, the priority shifts toward simplifying regimens and building information-sharing systems between medical institutions, since fragmented care across departments can breed duplicate prescriptions and inappropriate additions. Japan’s medical fee system already requires pharmacies to confirm leftover drugs and adjust prescriptions accordingly, and pharmacists there are legally obligated to monitor medication status throughout treatment. Whether the specific magnitude of risk seen in young Japanese adults—particularly that eye-watering odds ratio above 100—translates to other healthcare systems remains an open question, and the authors call for multicenter prospective studies combined with qualitative research into patient behavior. But the core message travels well: the recipe for leftover medication changes dramatically with age, and one-size-fits-all adherence interventions are doomed to miss half their audience. The future of medication management, this study suggests, belongs to strategies tailored to each decade of a patient’s life—with the number of pills in the regimen as the common denominator of safety.

Subject of Research: Age-stratified risk factors for leftover medications identified through community pharmacy pharmacotherapy management data

Article Title: Age-related risk factors for leftover drugs: a lifespan analysis in community pharmacies

Article References: Hirai, T., Hanaoka, S., Terakado, Y., Seki, T., Hayashi, H., & Watanabe, F. (2026). Age-related risk factors for leftover drugs: a lifespan analysis in community pharmacies. Journal of Pharmacy & Pharmaceutical Sciences, 29, Article 16879. https://doi.org/10.3389/jpps.2026.16879

Image Credits: AI Generated

DOI: 10.3389/jpps.2026.16879

Keywords: medication adherence, leftover drugs, polypharmacy, community pharmacy, prescription duration, pharmacists, older adults, deprescribing, medication safety, PBPM, healthcare costs, Japan