OPIOIDS AND PSYCHOTROPIC MEDICATIONS CO-PRESCRIBING FOR CHRONIC NON-CANCER PAIN PATIENTS IN TAIWAN BEFORE AND AFTER MEDICATION GUIDELINE CHANGES: AN INTERRUPTED TIME SERIES ANALYSIS
Author(s)
SHIH WEI TAI, MS1, CHING-CHING CLAIRE LIN, PhD2.
1Ph.D. Program of Interdisciplinary Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan, 2Institute of Health Policy and Management College of Public Health, National Taiwan University, Taipei, Taiwan.
1Ph.D. Program of Interdisciplinary Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan, 2Institute of Health Policy and Management College of Public Health, National Taiwan University, Taipei, Taiwan.
OBJECTIVES: Chronic non-cancer pain (CNCP) patients on long-term opioid therapy (LTOT) are frequently co-prescribed psychotropics, heightening overdose risk. On 4 December 2018, the Taiwan FDA revised the LTOT guideline to deter routine opioids-benzodiazepines (BZDs) co-prescribing. We examined the potential substitution.
METHODS: Using Taiwan's National Health Insurance Research Database from October 2017 to December 2019, we identified adult CNCP patients receiving LTOT, then calculated monthly proportions of opioid co-prescribing with BZDs and non-BZDs psychotropics (antidepressants, anxiolytics, sedative-hypnotics, antiepileptics, antipsychotics, muscle relaxants), requiring at least 1 day of overlapping. Interrupted time series analysis (ITSA) employed segmented generalized linear models (Gaussian family, identity link) with Newey-West robust standard errors to estimate pre-guideline trends (β1), immediate changes in January 2019 (β2), and post-guideline slope changes (β3).
RESULTS: Before and after guidelines revision, opioids with any of the psychotropics co-prescribing increased from 86.89% to 87.26%. Opioids with BZDs co-prescribing rose from 54.64% to 55.21%, with pre-guideline slope β1=0.0025/month (95% CI -0.0006, 0.0056), immediate change β2=-0.0386 (95% CI -0.0619, -0.0152), and post-guideline slope change β3=0.0026/month (95% CI -0.0009, 0.0061). Opioids co-prescribing showed relative increases for antidepressants, non-BZD sedative-hypnotics, and antiepileptics. Spillover effects occurred in non-BZDs anxiolytics,sedative-hypnotics, antidepressants, and antipsychotics, with smaller immediate declines but upward post-guideline trends.
CONCLUSIONS: The TFDA revision produced a transient reduction in opioids with BZDs co-prescribing, followed by substitution toward other psychotropics, indicating potential spillover effects. Durable safety gains require enhanced guideline dissemination, clinical decision support systems alerts, pharmacist review interventions, and academic detailing to sustain safer prescribing for CNCP patients on LTOT.
METHODS: Using Taiwan's National Health Insurance Research Database from October 2017 to December 2019, we identified adult CNCP patients receiving LTOT, then calculated monthly proportions of opioid co-prescribing with BZDs and non-BZDs psychotropics (antidepressants, anxiolytics, sedative-hypnotics, antiepileptics, antipsychotics, muscle relaxants), requiring at least 1 day of overlapping. Interrupted time series analysis (ITSA) employed segmented generalized linear models (Gaussian family, identity link) with Newey-West robust standard errors to estimate pre-guideline trends (β1), immediate changes in January 2019 (β2), and post-guideline slope changes (β3).
RESULTS: Before and after guidelines revision, opioids with any of the psychotropics co-prescribing increased from 86.89% to 87.26%. Opioids with BZDs co-prescribing rose from 54.64% to 55.21%, with pre-guideline slope β1=0.0025/month (95% CI -0.0006, 0.0056), immediate change β2=-0.0386 (95% CI -0.0619, -0.0152), and post-guideline slope change β3=0.0026/month (95% CI -0.0009, 0.0061). Opioids co-prescribing showed relative increases for antidepressants, non-BZD sedative-hypnotics, and antiepileptics. Spillover effects occurred in non-BZDs anxiolytics,sedative-hypnotics, antidepressants, and antipsychotics, with smaller immediate declines but upward post-guideline trends.
CONCLUSIONS: The TFDA revision produced a transient reduction in opioids with BZDs co-prescribing, followed by substitution toward other psychotropics, indicating potential spillover effects. Durable safety gains require enhanced guideline dissemination, clinical decision support systems alerts, pharmacist review interventions, and academic detailing to sustain safer prescribing for CNCP patients on LTOT.
Conference/Value in Health Info
2026-09, ISPOR Asia Pacific 2026, Bangkok, Thailand
Value in Health, Volume 55, Issue S1
Code
EPH17
Topic
Epidemiology & Public Health
Topic Subcategory
Safety & Pharmacoepidemiology
Disease
SDC: Musculoskeletal Disorders (Arthritis, Bone Disorders, Osteoporosis, Other Musculoskeletal)