Overprescription of Medications Continues To Plague Older Adults: Study

Machine learning models can predict potentially dangerous drug combinations, leading to less overprescribing by doctors, according to new research.

A new study warns that older patients often take multiple medications at the same time, frequently resulting in hospital emergency situations that could have been avoided.

According to findings published in the June 2025 issue of the medical journal Age and Aging, U.K. researchers found that the cumulative effects of polypharmacy is the greatest predictor that an older patient will need emergency hospitalization.

Polypharmacy is the practice of using multiple medications, often excessive or unnecessary, at the same time. The practice is common among patients older than 65 years. These patients are often given sedative and anticholinergic drugs, or drugs used to treat various chronic conditions like dementia, depression and chronic obstructive pulmonary disease (COPD).

When older adults take multiple medications to manage chronic conditions like diabetes and hypertension, they often face increased risks, including side effects, hospitalizations and a reduced quality of life.

Researchers from the University of Bath in the U.K. analyzed data from nearly 86,900 adults over the age of 65 in the U.K. Biobank, using machine learning to predict 30-day emergency hospitalization risks.

Led by Dr. Robert T. Olender, the team tested three predictive models—Random Forest, XGBoost and Logistic Regression—drawing on factors such as demographics, geriatric syndromes, existing health conditions, and a polypharmacy score, known as the Drug Burden Index, to assess potentially harmful combinations of medications.

Overall, the team found that the models could accurately forecast emergency hospitalizations with an average accuracy of 75%.

The models indicated inappropriate polypharmacy was a major risk factor of emergency hospital admissions among adults aged 65 and over. Other predictors included impaired mobility, a history of fractures and falls, smoking and excessive alcohol consumption.

This study is the first of its kind to use machine learning and data driven methods to examine how patients who use multiple drugs at the same time may face greater risks of side effects, especially emergency hospitalization.

Researchers hope that they can use the machine learning model to develop an app or other digital tool for doctors to use to help predict which patients face the greatest risk of emergency hospitalization due to drug overprescribing and other factors.

The research could help identify patients who are at risk of suffering side effects from medications early, when intervention can help prevent serious outcomes, they concluded.

Overprescription Concerns

This is not the first study to highlight how common polypharmacy is and the risks it poses to patients. A study published in 2018 concluded nearly half of patients prescribed narcotic painkillers don’t need them for pain relief. Instead, patients are given powerful opioid drugs that are highly addictive, when their pain can be treated with safer drugs instead.

Similarly, many patients are not only overprescribed painkillers, but also antibiotics. A study published by researchers from the U.S. Centers for Disease Control and Prevention (CDC) concluded half of all hospital patients are given antibiotics unnecessarily.

Another study published by researchers from the University of Colorado School of Medicine concluded children are often given antibiotics to treat pneumonia when they are seen by doctors in the ER. Children who do not have confirmed pneumonia diagnoses are commonly prescribed full rounds of antibiotics.

Researchers warn that inappropriate prescribing of antibiotics is commonly linked to the increase in antibiotic-resistant bacteria, or superbugs, that cannot be treated with standard types of antibiotics.