New persistent opioid use among ICU survivors after discharge: incidence, predictors, and nationwide cohort analysis

Critical Care volume 29, Article number: 469 (2025) Published: 03 November 2025

Background

Long-term opioid dependence after critical illness is an emerging concern, yet the incidence and predictors of persistent opioid use among intensive care unit (ICU) survivors remain incompletely characterized. We aimed to estimate the six-month incidence of new persistent opioid use in opioid-naïve ICU survivors and to identify associated risk factors.

Methods

We conducted a retrospective, nationwide cohort study using South Korea’s National Health Insurance Service database. Adults admitted to any ICU between January 1, 2020, and December 31, 2022, were included if they survived to hospital discharge and remained alive for at least six months, with no opioid prescription in the 12 months preceding admission. New persistent opioid use was defined as at least one outpatient opioid prescription within 90 days after discharge and at least one additional prescription between 91 and 180 days. We performed multivariable logistic regression to identify independent predictors.

Results

Among 567,260 opioid-naïve ICU survivors, 23,945 (4.2%) developed new persistent opioid use within six months. Across the cohort, 22,643 (4.0%) received less-potent opioids (tramadol, dihydrocodeine) and 1,643 (0.3%) received potent opioids (morphine, fentanyl, oxycodone, hydromorphone, methadone). Independent predictors included older age (odds ratio [OR] 1.01 per year; 95% confidence interval [CI], 1.01–1.02; P < 0.001), female sex (OR 1.13; 95% CI, 1.09–1.16; P < 0.001), socioeconomic disadvantage (Medical Aid, OR 1.30; 95% CI, 1.23–1.38; P < 0.001), malignancy (OR 1.05; 95% CI, 1.01–1.09; P = 0.017), metastatic tumor (OR 1.24; 95% CI, 1.15–1.35; P < 0.001), extracorporeal membrane oxygenation (OR 1.80; 95% CI, 1.75–1.89; P < 0.001), and continuous renal replacement therapy (OR 1.24; 95% CI, 1.11–1.37; P < 0.001). The strongest predictor was an early opioid prescription within 30 days of discharge (OR 19.7; 95% CI, 19.1–20.3; P < 0.001). Potency-specific analysis showed potent opioid persistence was largely driven by cancer, while less-potent use was shaped more by demographic and socioeconomic factors.

Conclusions

Approximately one in 25 ICU survivors developed new persistent opioid use by six months. Early post-discharge opioid prescription was the dominant risk factor. Risk profiles differed by opioid potency, underscoring the need for early tapering strategies, multimodal non-opioid analgesia, and stewardship programs tailored to patient subgroups and opioid type.