S92 Racial and Ethnic Disparities in Opioid Prescriptions in Benign and Malignant Pancreatic Disease in the United States, 2006-2015

Abstract

Introduction: Racial and ethnic disparities in pain management are well established, but not for patients with pancreatic disease. We sought to evaluate racial-ethnic disparities in opioid prescriptions for patients with acute and chronic pancreatitis and pancreatic cancer. Methods: Data from the National Ambulatory Medical Care Survey (NAMCS) were used to examine opioid prescriptions recorded during ambulatory visits by adult pancreatic disease patients between 2006 and 2015. Diagnoses were determined by ICD-9 codes 557.0 and 577.1 for acute and chronic pancreatitis and 157 for pancreatic cancer. Visits by patients with other painful conditions associated with opioid use were excluded. Bivariate analysis was used to detect differences in opioid prescriptions by patient race, ethnicity, and sex. Data weights for national-level estimates were not applied in accordance with NAMCS guidelines requiring a relative standard error of 30% or less. Results: We identified 421 outpatient visits by adults with pancreatic disease. Eighteen patients with comorbid painful conditions (lumbago, myalgia, generalized pain, ankylosing spondylitis, osteoarthritis, multiple sclerosis, and mononeuritis) were excluded. A total of 207 pancreatitis and 196 pancreatic cancer patient visits were identified, representing 9.8 million visits, but weights were repealed for analysis. No sex differences in opioid prescriptions were found among pancreatitis (p=0.78) or pancreatic cancer patient visits (p=0.57). Opioids were prescribed at 58% of non-Hispanic Black, 37% non-Hispanic White, and 19% Hispanic pancreatitis (P=0.05) patient visits (Table). Opioid prescriptions were less common in visits by Hispanic compared to non-Hispanic pancreatitis patients (OR 0.35, 95% CI 0.14-0.91, p=0.03) and there was a trend toward higher rates in non-Hispanic Black versus non-Hispanic White patient visits. No racial-ethnic differences in opioid prescriptions were seen among pancreatic cancer patient visits. Conclusion: While racial-ethnic disparities in opioid prescriptions were observed in pancreatitis patient visits, none were seen in pancreatic cancer patient visits. These findings suggest racial and ethnic bias in opioid prescription practices for patients with painful but benign pancreatic disease, but there may be a lower threshold for opioid provision in the treatment of malignant and terminal pancreatic disease. Table 1. - Demographic characteristics and likelihood of opioid use among visits by adult pancreatitis and pancreatic cancer patients in NAMCS, 2006-2015 No opioid Opioid Prescription OR (95% CI) p-value Pancreatitis N=132 (64%) N=75 (36%) Race/Ethnicity, % A) Non-Hispanic 106 (61) 69 (39) Reference Reference Hispanic 26 (81) 6 (19) 0.35 (0.14-0.91) .03 B) White non-Hispanic 87 (63) 52 (37) Reference Reference Black non-Hispanic 10 (42) 14 (58) 2.34 (0.97-5.65) .06 Hispanic 26 (81) 6 (19) 0.39 (0.15-1.00) .05 Other non-Hispanic 9 (75) 3 (25) 0.56 (0.42-0.84) .40 Sex, % Female 59 (65) 32 (35) Reference Reference Male 73 (63) 43 (37) 1.09 (0.61-1.92) .78 Pancreatic Cancer N=126 (64%) N=70 (36%) Race/Ethnicity, % White non-Hispanic 97(64) 54(36) Reference Reference Black non-Hispanic 17(65) 9(35) 0.95 (0.40-2.28) .91 Hispanic 8(62) 5(38) 1.12 (0.35-3.60) .85 Other non-Hispanic 4(67) 2(33) 0.90 (0.16-5.06) .90 Sex, % Female 63(62) 38(38) Reference Reference Male 63(66) 32(34) 0.84 (0.47-1.51) .57