Panel Data Opioid Utilization Analysis of Advanced Stage Non-Small Cell Lung Cancer Patients in Taiwan Cancer Centers

Objectives The aim of this study was to explore the association between opioid utilization and hospital characteristics (caseload, geographic variation, and palliative care service) among advanced non-small cell lung cancer (NSCLC) patients in Taiwan cancer centers.aterials and Methods We exploited a retrospective cohort, consisting of all newly diagnosed advanced NSCLC (stage IIIB and stage IV) patients from Taiwan Cancer Center Data Base (TCDB) from Jan 2004 to June 2004, and excluding those who died after Jan 1st 2005 or had no definite data of diagnosis date or initial anti-cancer treatment. We then retrieved the details of opioid utilization of the last six months of each patient by linking this cohort with claimed National Health Insurance Data (NHID). The dosage of each kind of opioid analgesics was transformed to “oral morphine equiavalents” (OMEQ). The definitions of “opioid utilization” were: the experience of taking opioid analgesics (yes or no), the average OMEQ per person per day, and the trend of OMEQ utilization in the last 26 weeks before death.esults Of the 1042 advanced NSCLC patients enrolled in this study, 714 (68.5%) were males and 328 (31.5%) were females. Most of them were above 65 years old (71.7%) and with the diagnosis of stage IV cancer (77.7%). More than a half of them (56.4%) did not receive anti-cancer treatment at the time of diagnosis. Most of these patients (80.7%) had experiences in taking opioid analgesics. However, the proportion of taking weak opioid analgesics was only 44.2%. The average OMEQ per person per day was 32mg, and strong or non-parenteral opiod analgesics accounted for most of the dosage. The utilization of transdermal Fentanyl was quite high, accounting for one-third of total dosage and half of non-parenteral dosage. Noteworthily, the utilization of parenteral Meperidine was still high, accounting for one-tenth of total dosage and one-third of parenteral dosage. We then used a mixed model with patient and patient-time interaction fitted as random. After adjusting all the confounders (sex, age, stage, and initial anti-cancer treatment), we found that the trend of OMEQ utilization in the last 26 weeks before death was higher when the hospital offered palliative care service. (p=0.034)onclusions Among the advanced NSCLC patients in Taiwan cancer centers, the trend of OMEQ utilization in the last 26 weeks before death was correlated with palliative care service, but not correlated to caseload or geographic variation. The causal relationship needs further investigation.