762 - Heading West: Regional Trends in Maternal Opioid Use and Neonatal Abstinence Syndrome

Poster Number: 762
Publication Number: 762.12

Erwin T. Cabacungan, Medical College of Wisconsin, Wauwatosa, WI, United States; Kelsey S. Ryan, Medical College of Wisconsin, Wauwatosa, WI, United States

Background: Maternal opioid use (OU) and its associated neonatal abstinence syndrome (NAS) are significant public health problems leading to severe morbidities and prolonged hospital stays. The population burden of opioid use and NAS varies geographically, with the Southern and Northeastern United States historically demonstrating the highest prevalence prior to 2016. We hypothesize that OU and NAS regional prevalence are linked and may change over time.

Objective: To determine and compare the yearly regional OU and NAS prevalence and resource utilization using a national hospital discharge database.

Design/Methods: We conducted a retrospective cohort study of hospital discharge data for mothers with OU and infants with NAS using the 2016-2020 Healthcare Cost Utilization Project National Inpatient Sample (HCUP-NIS). Weighted analyses produced nationally representative estimates. We compared the four US census regions: Northeast (NE), Midwest (MW), South (S), and West (W). Cochrane-Armitage test and Correlation Coefficient (r) compared yearly prevalence over time. Generalized linear model with Poisson distribution compared length of stay (LOS) and γ distribution compared Total Charges (TCH). Analysis was adjusted for covariates, patient and hospital characteristics, and year.

Results: HCUP-NIS database included 174,776,205 weighted discharge records from US hospitals from 2016-2020, which included 158,075 maternity discharges affected by opioid use and 116,755 infant discharges affected by NAS (Figure 1). There were significant regional variations in the prevalence of both OU and NAS (Figure 2). The NE consistently had the highest rates of OU and NAS. Yearly OU and NAS increased the most in the W. There was a strong positive relationship between the yearly increase in OU and NAS in the W (r = 0.95), a fairly strong positive relationship in the S (r = 0.76), and a fairly strong negative relationship in the M (r = -0.70). Median LOS and TCH for NAS decreased in all four regions (Table 1). Compared to the NE, the S had significantly higher LOS for NAS. Across all four regions, median LOS for maternal OU did not decrease, and median TCH increased. Compared to the NE, the W demonstrated the most change in TCH for both OU and NAS.

Conclusion(s): Prevalence and healthcare utilization for OU and NAS demonstrate the most acceleration in the W, with the least change in the NE from 2016-2020. This regional variability can inform resource allocation and reveal opportunities for standardization.