Swiss Study Maps Opioid Risks in Hospitalized Adults

A comprehensive observational cohort study published in The Lancet Regional Health – Europe has identified significant patterns of opioid use among hospitalized adults in Switzerland, highlighting a substantial subset of patients who meet specific criteria for deprescribing. Between 2020 and 2024, researchers analyzed data from three Swiss tertiary referral centers to characterize inpatient opioid exposure, map treatment trajectories, and identify high-risk patient groups. The study found that the majority of opioid administrations occurred in patients without cancer diagnoses, with a notable proportion of these non-cancer cases exhibiting stable dose trajectories that persisted until discharge. This finding suggests a potential gap in clinical oversight, as these patients may be at heightened risk for persistent opioid use and associated adverse outcomes following hospitalization.

The research team, led by Marcel Rainer and colleagues, conducted a descriptive, multicenter, longitudinal study using retrospective data from electronic health records. The study included all adult inpatient cases aged 18 years and older who received at least one recorded systemic opioid administration on medical or surgical wards. After applying exclusion criteria for missing data, critical care-only anesthesiologic use, and clinically implausible records, 68,129 cases were included in the final analysis. This large sample size allowed for robust estimation of opioid exposure patterns and the identification of clinically meaningful subgroups across the participating hospitals.

Opioid exposure was quantified by converting all administered opioids into oral morphine milligram equivalents (MMEs) using standardized conversion factors from the 2022 US Centers for Disease Control and Prevention clinical practice guideline. Daily MMEs were calculated for each hospital day to track dose dynamics over time. The study defined sustained therapy as the administration of opioids for at least three consecutive days. To identify patients at risk for inappropriate use, the researchers established predefined deprescribing indicators, which included a high opioid dose exceeding 90 MMEs on the final treatment day, the presence of high-risk comorbidities such as psychiatric or substance use disorders, or the concurrent use of central nervous system (CNS) depressants like benzodiazepines and gabapentinoids.

Baseline characteristics of the cohort revealed that the median age was 63 years, and 54.9% of cases were female. Most patients, 78.6%, had no documented cancer diagnosis during their hospitalization. The most common admission diagnoses were musculoskeletal disorders, accounting for 22.9% of cases, and injury- and poisoning-related conditions, representing 19.4%. Opioid initiation was most frequently managed by orthopedics and general surgery specialties. Overall, 47.0% of all cases received opioids for at least three days, while 11.4% of the total cohort met the predefined criteria for opioid deprescribing on the last day of hospitalization.

To further characterize risk, the researchers applied k-means clustering to the daily MME trajectories of non-cancer cases over the first seven days of hospitalization. This analysis identified three distinct treatment patterns. Cluster 1, comprising 18,828 cases, showed the highest relative MME on day one followed by a rapid decline. Cluster 3, with 15,438 cases, peaked on day two before a steady reduction. Cluster 2 was the largest group, consisting of 19,292 cases, and was characterized by relatively stable MME levels with only marginal dose reductions across the observation period. This stable trajectory in Cluster 2 was clinically significant, as it indicated a lack of dose de-escalation during the hospital stay.

Cluster 2 contained the highest proportion of patients meeting the criteria for deprescribing, with 33.4% of its members identified as high-risk. These patients were older, with a median age of 70 years, and had a higher comorbidity burden compared to the other clusters. Among all non-cancer cases, 7,791 patients met the eligibility criteria for opioid deprescribing. Notably, 83% of these high-risk patients were identified within Cluster 2, suggesting that stable-dose trajectories are a key marker for missed deprescribing opportunities. This subgroup tended to be older and more often female, and was most commonly admitted for injuries, arthritic diseases, and spinal disorders.

The study detailed the specific indicators present in this high-risk group. Among the 6,439 high-risk cases in Cluster 2, 18.1% were discharged on high-dose opioid therapy exceeding 90 MMEs. Concurrent maintenance therapy with CNS depressants was observed in 42.6% of these patients, while 66.4% presented with high-risk comorbidities. A small fraction, 1.6%, met all three predefined deprescribing indicators simultaneously. The distribution of these indicators varied by diagnosis, with osteopathies and surgical or medical complications showing the highest proportion of cases meeting all three criteria. This granular data provides actionable targets for clinical intervention and opioid stewardship programs.

The findings align with broader trends in opioid prescribing, where non-cancer pain management has become a significant driver of opioid exposure. In Switzerland, opioid sales and poisoning calls have risen substantially over the past two decades, predominantly driven by oxycodone formulations. The study highlights that hospital-initiated opioid use is a critical transition point, as a significant number of opioid-naïve patients begin treatment during or immediately after hospitalization. Without structured monitoring and deprescribing protocols, these patients may progress to long-term use, contributing to the burden of opioid-related harm in the community.

The authors emphasize that hospitals represent an important opportunity for primary prevention through regular reassessments of opioid therapy. The stable dose trajectories observed in the high-risk group may reflect institutional inertia, where standardized dosing practices are driven by enhanced recovery protocols rather than individualized clinical decision-making. To address this, the study proposes embedding monitoring systems into routine clinical information infrastructure. Such systems could facilitate the real-time identification of high-risk trajectories and support coordinated, individualized deprescribing efforts by care teams, including geriatricians, pharmacologists, and pharmacists.

Despite the study’s strengths, including its large sample size and use of precise medication administration records, several limitations were acknowledged. The descriptive nature of the study meant that no formal statistical inference was performed, and causal associations could not be established. Indications for opioid therapy were inferred from admission diagnoses, which may not fully capture the clinical context. Additionally, the deprescribing criteria did not account for pain intensity scores, relying instead on MME thresholds as a marker for post-discharge risk. Cancer stratification based on ICD-10 codes may have misclassified some patients, and off-label use or participation in opioid agonist therapy programs could not be reliably identified.

The study concludes that systematic identification of patients with sustained opioid therapy and high-risk indicators is essential for improving patient safety. By targeting specific patient subgroups, such as those with musculoskeletal disorders and stable dose trajectories, hospitals can implement tailored reduction strategies. These efforts should bridge acute care with primary care and home care networks to ensure safe post-discharge dose reductions. Future research should focus on quantifying downstream opioid-related harms associated with sustained therapy after hospital discharge to further validate the clinical impact of these deprescribing interventions.

Subject of Research: Opioid treatment patterns and deprescribing indicators among hospitalized adults in Switzerland

Article Title: Opioid treatment patterns, trajectories, and deprescribing indicators among hospitalised adults in Switzerland, 2020–2024: a multicentre observational cohort study

Article References: Rainer, M., Gasparovic, L., Noseda, R., Colombo, A., Hug, B., Ceschi, A., & Burden, A. M. (2027). Opioid treatment patterns, trajectories, and deprescribing indicators among hospitalised adults in Switzerland, 2020–2024: a multicentre observational cohort study. The Lancet Regional Health – Europe, 72, Article 101905. https://doi.org/10.1016/j.lanepe.2026.101905

Image Credits: AI Generated

DOI: 10.1016/j.lanepe.2026.101905

Keywords: opioid stewardship, hospitalized adults, Switzerland, deprescribing, morphine milligram equivalents, observational cohort, non-cancer pain, clinical risk, medication safety, trajectory analysis, Opioid, treatment