National Trends in Hepatitis C Infection by Opioid Use Disorder Status Among Pregnant Women at Delivery Hospitalization — United States, 2000–2015

Jean Y. Ko, PhD; Sarah C. Haight, MPH; Sarah F. Schillie, MD; Michele K. Bohm, MPH; Patricia M. Dietz, DrPH

Disclosures

Morbidity and Mortality Weekly Report. 2019;68(39):833-838. 

In This Article

Abstract and Introduction

Introduction

Hepatitis C virus (HCV) is transmitted primarily through parenteral exposures to infectious blood or body fluids that contain blood (e.g., via injection drug use, needle stick injuries).[1] In the last 10 years, increases in HCV infection in the general U.S. population[1] and among pregnant women[2] are attributed to a surge in injection drug use associated with the opioid crisis. Opioid use disorders among pregnant women have increased,[3] and approximately 68% of pregnant women with HCV infection have opioid use disorder.[4] National trends in HCV infection among pregnant women by opioid use disorder status have not been reported to date. CDC analyzed hospital discharge data from the 2000–2015 Healthcare Cost and Utilization Project (HCUP) to determine whether HCV infection trends differ by opioid use disorder status at delivery. During this period, the national rate of HCV infection among women giving birth increased >400%, from 0.8 to 4.1 per 1,000 deliveries. Among women with opioid use disorder, rates of HCV infection increased 148%, from 87.4 to 216.9 per 1,000 deliveries, and among those without opioid use disorder, rates increased 271%, although the rates in this group were much lower, increasing from 0.7 to 2.6 per 1,000 deliveries. These findings align with prior ecological data linking hepatitis C increases with the opioid crisis.[2] Treatment of opioid use disorder should include screening and referral for related conditions such as HCV infection.

To evaluate HCV infection prevalence at hospital delivery among women with and without opioid use disorder, data from HCUP's National Inpatient Sample (NIS, 2000–2015) (https://www.hcup-us.ahrq.gov/) were analyzed. The fourth quarter of 2015 and more recent data were excluded because of the transition to the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) during that period. The NIS is the largest publicly available all-payer inpatient health care database in the United States, yielding national estimates representing approximately 35 million hospitalizations. Discharges for in-hospital deliveries were identified using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnostic and procedure codes pertaining to obstetric delivery.[5]

HCV infection was identified from ICD-9-CM codes 070.41, 070.44, 070.51, 070.54, 070.70, 070.71, and V02.62; and opioid use disorder was identified from codes for opioid dependence and nondependent abuse (304.00–304.03, 304.70–304.73, and 305.50–305.53), aligning with Diagnostic and Statistical Manual of Mental Disorders, 5th Edition criteria*.[6] Deliveries were categorized by maternal diagnoses: HCV infection only, opioid use disorder only, both HCV infection and opioid use disorder, or neither. Demographic variables of interest included age, payer source, race/ethnicity, median income quartiles for residency ZIP code, and hospital geographic region.

Survey-specific analysis techniques accounted for clustering, stratification, and weighting. National annual prevalence rates of opioid use disorder and HCV infection per 1,000 delivery hospitalizations during 2000–2015 and 95% confidence intervals (CIs) were calculated using SAS (version 9.4; SAS Institute). HCV infection rates were calculated by opioid use disorder status. Joinpoint regression was used to model the average percentage change in HCV infection and opioid use disorder rates over time and their statistical significance. The program identifies points (joinpoints) where the slope of the trend significantly changes and calculates the average percentage change in the rate during the years between joinpoints. Using 2015 data, distribution of diagnoses by payer source, race/ethnicity, median income for residency ZIP code, and hospital region were calculated. Polytomous logistic regression models were used to calculate unadjusted odds ratios (ORs) and 95% CIs comparing the likelihood of each delivery hospitalization having one or both diagnoses versus neither by sociodemographic characteristics. Statistical significance was set at p<0.05.

During 2000–2015, the rate of HCV infection increased from 0.8 (95% CI = 0.7–0.9) to 4.1 (95% CI = 3.7–4.4) per 1,000 deliveries. Rates significantly increased from 2000 to 2004 (15.7%; p<0.001), 2004 to 2010 (6.1%; p<0.001), and 2010 to 2015 (14.9%; p<0.001). Among deliveries with opioid use disorder diagnoses, the rate of maternal HCV infection increased from 87.4 (95% CI = 56.3–118.5) to 216.9 (95% CI = 197.9–235.9) per 1,000 deliveries (Figure). The rate significantly increased during 2000–2004 (17.2%; p<0.001), remained statistically unchanged during 2004–2011 (−2.4%; p = 0.1), and significantly increased during 2011–2015 (7.9%; p<0.001). Among deliveries without opioid use disorder diagnoses, the rate of HCV infection increased from 0.7 (95% CI = 0.6–0.8) to 2.6 (95% CI = 2.4–2.9) per 1,000 deliveries during 2000–2015. The rate remained statistically unchanged during 2000–2002 (21.1%; p = 0.1), and significantly increased during 2002–2011 (5.5%; p<0.001) and 2011–2015 (15.0%; p<0.001).

Figure.

National prevalence* of maternal hepatitis C virus (HCV) infection per 1,000 delivery hospitalizations, by opioid use disorder (OUD) status, 2000–2015
* Prevalence numerator consisted of HCV infection International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes (070.41, 070.44, 070.51, 070.54, 070.70, 070.71, and V02.62), and denominator consisted of delivery hospitalizations discharges with and without opioid type dependence and nondependent opioid abuse based on ICD-9-CM codes (304.00–304.03, 304.70–304.73, and 305.50–305.53).
Rates are for 2000 through the third quarter of 2015.

In 2015, all three groups (those with HCV infection only, opioid use disorder only, and both HCV infection and opioid use disorder) shared similar risk factors (Table 1). Compared with women aged ≥35 years, those aged 25–34 years were more likely to have a diagnosis of HCV infection (OR = 1.2, 95% CI = 1.0–1.4), opioid use disorder (OR = 1.8, 95% CI = 1.6–2.0), or both (OR = 1.8, 95% CI: 1.4–2.3) at delivery (Table 2). Women with publicly billed deliveries (Medicaid or Medicare) were the most likely to have a diagnosis of HCV infection (OR = 5.5, 95% CI = 4.7–6.4), opioid use disorder (OR = 6.4, 95% CI = 5.8–7.2), or both (OR = 9.9, 95% CI = 7.8–12.6) at delivery, compared with privately billed deliveries. Compared with non-Hispanic black women, Native American women were the most likely to have a diagnosis of HCV infection (OR = 5.0, 95% CI = 2.9–8.7) or opioid use disorder (OR = 5.9, 95% CI = 4.0–8.8) at delivery, and non-Hispanic white women were the most likely to have a diagnosis of both (OR = 10.9, 95% CI = 6.3–18.6) at delivery. Women from areas with median income of <$42,000 were the most likely to receive a diagnosis of HCV infection (OR = 2.5, 95% CI = 2.0–3.0), opioid use disorder (OR = 2.0, 95% CI = 1.7–2.3), or both (OR = 2.5, 95% CI = 1.8–3.4) at delivery, compared with those from areas with median income ≥$68,000. Compared with U.S. residents of the Western census region (the referent group), residents of the South were the most likely to receive a diagnosis of HCV infection (OR = 1.9, 95% CI = 1.5–2.3) at delivery. Women living in the Northeast were the most likely to receive a diagnosis of opioid use disorder (OR = 2.0, 95% CI = 1.6–2.4) or both HCV infection and opioid use disorder (OR = 4.8, 95% CI = 3.1–7.5) at delivery.

* ICD-9-CM codes related to opioid dependence and nondependent abuse, in remission, were included in this analysis because both early remission and opioid use disorder could have occurred during pregnancy.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....