Key Points
Question
What were the prevalence and maternal and perinatal outcomes of hepatitis C (HCV)-positive pregnancies during the opioid epidemic?
Findings
This cross-sectional study of more than 70 million births or spontaneous abortions showed the prevalence of HCV-positive pregnancies in the US increased 16-fold between 1998 and 2018. Maternal HCV infection was associated with increased odds of preterm labor, poor fetal growth, or fetal distress.
Meaning
The data from this study suggest that universal HCV screening with each pregnancy may be useful, as the Centers for Disease Control and Prevention propose, but research is needed on the subsequent needs for appropriate specialist care for affected mothers and newborns.
Abstract
Importance
Injection drug use is the primary risk factor for hepatitis C virus (HCV) infection in adults. More than one-third of newly reported HCV cases occur in women, particularly among persons aged 20 to 39 years. However, nationally representative data on HCV during pregnancy are limited.
Objective
To estimate the temporal trend of HCV-positive pregnancies during the opioid epidemic and identify HCV-associated maternal and perinatal outcomes.
Design, Setting, and Participants
A cross-sectional study was performed with data from the US, from calendar year 1998 through 2018. Data analysis was conducted from November 14, 2021, to May 14, 2023. Participants included women during in-hospital childbirth or spontaneous abortion in the National Inpatient Sample of the Healthcare Cost and Utilization Project.
Exposure
Maternal HCV infection.
Main Outcomes and Measures
The main outcome was the temporal trend, measured as change in the annual prevalence, in the prevalence of HCV positivity among pregnant women since the start of the opioid epidemic in the late 1990s. Secondary outcomes were the associations shown as relative odds between maternal HCV infection and maternal and perinatal adverse events.
Results
During the study period, more than 70 million hospital admissions resulted in childbirth or spontaneous abortion. Among them, 137 259 (0.20%; 95% CI, 0.19%-0.21%) involved mothers with HCV; these individuals were more often White (77.4%; 95% CI, 76.1%-78.6%), low-income (40.0%; 95% CI, 38.6%-41.5%), and likely to have histories of tobacco (41.7%; 95% CI, 40.6%-42.9%), alcohol (1.8%; 95% CI, 1.6%-2.0%), and opioid (28.9%; 95% CI, 27.3%-30.6%) use compared with HCV-negative mothers. The median age of women with HCV was 28.0 (IQR, 24.3-32.2) years, and the median age of HCV-negative women was 27.2 (IQR, 22.7-31.8) years. The prevalence of HCV-positive pregnancies increased 16-fold during the study period, reaching 5.3 (95% CI, 4.9-5.7) cases per 1000 pregnancies in 2018. Age-specific prevalence increases ranged from 3-fold (age, 41-50 years) to 31-fold (age, 21-30 years). Higher odds of cesarean delivery, preterm labor, poor fetal growth, or fetal distress were associated with HCV-positivity during pregnancy. However, no significant differences were observed in gestational diabetes, preeclampsia, eclampsia, or stillbirths.
Conclusions and Relevance
In this cross-sectional study, the prevalence of HCV-positive pregnancies increased markedly, and maternal HCV infection was associated with increased risks for adverse perinatal outcomes. These data may support recent recommendations for universal HCV screening with each pregnancy.
This cross-sectional study examines the prevalence of hepatitis C virus in pregnant women from 1998 to 2018, with a focus on changes since the start of the opioid epidemic in the late 1990s.
Introduction
Hepatitis C virus (HCV) is the most commonly reported chronic bloodborne infection in the US,1 and injection drug use is the primary risk factor for disease transmission among adults.2 The US saw a tripling of acute HCV cases between 2010 and 2015,3 coinciding with increased heroin (ie, second wave) and synthetic opioid (ie, third wave) overdose deaths during the opioid epidemic.4 More than one-third of newly reported HCV cases were among women, and the highest incidence consistently occurred among persons aged 20 to 39 years.5 From 2014 to 2017, HCV cases among pregnant women in the US exceeded that of hepatitis B and syphilis combined, despite HCV being the only infection among the 3 not to receive universal screening during prenatal care.6 Antepartum opioid use disorder has also more than quadrupled since the start of the opioid epidemic in the late 1990s.7 Maternal HCV infection poses potential risks for the fetus, including preterm birth and neonatal death.6 Mother-to-child transmission of HCV during pregnancy occurs in approximately 6% of children delivered by women with HCV viremia, and the risk doubles in the setting of poorly controlled HIV coinfection.8 Consequently, multiple US medical and public health organizations—including the American College of Obstetricians and Gynecologists—have recently recommended universal HCV screening during each pregnancy.9,10
Until lately, data on HCV infection during pregnancy have been scarce, with conflicting reports on the role of HCV in pregnancy and perinatal outcomes. Several newer publications have sought to address the literature gap. Collectively, they suggested an increasing US prevalence of HCV infections during pregnancy over the past 2 decades, although most of these studies reported only on a segment of the period.6,11,12,13,14 A few studies also reported adverse perinatal events associated with HCV, although they may not have fully accounted for the potential confounding effects of substance use, which is another known risk for poor pregnancy outcomes.6,14
Given increasing HCV infections among women of childbearing age,5 the risk of vertical HCV transmission,8 and potential pregnancy-related adverse events,6,14 clinicians and policy makers need a heightened understanding of HCV infection during pregnancies and its implications on outcomes. Our study aimed to construct the temporal trend of HCV-positive pregnancies in the US since the start of the opioid epidemic in the late 1990s and identify adverse maternal and perinatal results associated with HCV infection.
Methods
Our report follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. The Johns Hopkins Institutional Review Board reviewed and acknowledged our study design as not human participant research. Consequently, the need for informed consent did not apply.
The National Inpatient Sample (NIS) is the largest publicly available all-payer inpatient care database in the US and a part of the Healthcare Cost and Utilization Project by the Agency for Healthcare Research and Quality.15 The database contains discharge data, including patient demographic data, diagnoses, procedures, insurance payers, hospital characteristics, total charges, and lengths of stay. Before 2012, the NIS was a stratified random sample of approximately 20% of nonfederal, nonrehabilitation hospitals. Starting in 2012, it has approximated a 20% stratified sample of all discharges from participating nonfederal hospitals in the US, excluding rehabilitation and long-term acute care hospitals. Nationwide participation in the NIS has increased over time. As of 2017, 47 states and the District of Columbia contribute to the NIS. Their data are weighted to allow the calculation of national estimates.
We used the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Clinical Modification (ICD-10-CM), and International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Procedure Coding System (ICD-10-PCS) codes in the first 40 billing positions to capture all hospital admissions from calendar year 1998 through 2018 that culminated either in childbirth (vaginal and cesarean delivery), stillbirth, or spontaneous abortion (defined as the loss of the fetus before 20 weeks of pregnancy) (eTable 1 in Supplement 1). All included women were aged 18 to 50 years during the pregnancy. Within the cohort, we identified hospitalizations with known diagnoses of HCV positivity. As ICD-9-CM and ICD-10-CM codes do not specify the absence of a diagnosis, the remainder of hospitalizations presumably were in 1 of 3 categories: (1) HCV-negative, (2) HCV-positive but were unscreened or undetected, or (3) known to be HCV-positive but were uncoded at hospital discharge. For simplicity of nomenclature, we refer to this collective group as HCV-negative.
Outcomes
Our study examined adverse clinical outcomes that broadly focused on maternal and perinatal events. Maternal outcomes were anemia complicating pregnancy, gestational diabetes, hypertension complicating pregnancy (excluding preeclampsia), preeclampsia or eclampsia, and thyroid dysfunction complicating pregnancy. Perinatal events comprised cesarean delivery, stillbirth, spontaneous abortion, preterm labor (ie, delivery after 20 weeks and before 37 weeks of pregnancy), poor fetal growth, fetal distress, and premature rupture of membranes.
Covariates
Our models were adjusted for potential confounders (ie, covariates simultaneously associated with HCV infection and maternal and/or perinatal outcomes), which we determined after reviewing a published study design on viral hepatitis-related pregnancy outcomes.16 Maternal covariates included age by decade, race and ethnicity, substances used (10 categories including tobacco, alcohol, opioids, and cocaine), HIV, hemodialysis, diabetes, thyroid disorders, hypertension, anemia, primary payer, and median household income for the patient’s zip code. The Healthcare Cost and Utilization Project and its partner organizations defined the racial and ethnic categories for the hospital-reported information in our analysis. We collapsed smaller categories into “other” due to their limited sample sizes. Our analysis adjusted for race and ethnicity as potential confounders, given their possible associations with both the exposure (HCV infection) and the outcome (maternal or perinatal events). In addition, we accounted for medical comorbidities via the validated Charlson-Deyo Comorbidity Index,17 using adapted ICD-9-CM and ICD-10-CM coding algorithms.18 We summarized medical comorbidities into 3 severity categories per the Charlson-Deyo Comorbidity Index: 0, 1 to 2, or more than 2. Hospital-related covariates were bed size, location and teaching status, and geographic region.
Statistical Analysis
Data analysis was conducted from November 14, 2021, to May 14, 2023. We stratified, clustered, and weighted our analyses to navigate the sampling design of the NIS. Additionally, we applied the suggested procedures for multiyear analysis to account for the 2012 NIS sampling redesign.19 Categorical variables were compared using Rao-Scott χ2 tests for complex surveys.20 We performed multivariable logistic regression models to evaluate the relative odds of maternal or perinatal events as a function of HCV infection during pregnancy. Our collective content knowledge informed the model variable selection—no statistical selection procedures aided the process. We used variance inflation factors to identify the presence of possible collinearity among independent variables of the multivariable regression models. All statistical tests were 2-sided and unpaired, and analyses were performed using SAS, version 9.4 (SAS Institute LLC) with α = .05 used as the threshold for significance.
Results
From 1998 to 2018, more than 70 million hospital admissions in the cohort resulted in vaginal delivery, cesarean delivery, stillbirth, or spontaneous abortion. Of these, 137 259 (0.20%; 95% CI, 0.19%-0.21%) involved HCV-positive mothers (eTable 2 in Supplement 1).
Patient characteristics stratified by HCV status are listed in Table 1. Compared with the HCV-negative group, women in the HCV-positive group were slightly older (median, 27.2 [IQR, 22.7-31.8] years vs 28.0 [IQR, 24.3-32.2] years), more often White (77.4%; 95% CI, 76.1%-78.6% vs 53.9%; 95% CI, 52.6%-55.1%), and generally had a lower socioeconomic status (40.0%; 95% CI, 38.6%-41.5% vs 26.9%; 95% CI, 25.8%-27.9% resided in zip codes with the lowest quartile median household incomes and 74.5%; 95% CI, 73.4%-75.6% vs 39.3%; 95% CI, 38.3%-40.3% received Medicaid). Women who were HCV-positive additionally had more severe Charlson-Deyo Comorbidity Index scores (89.5%; 95% CI, 89.0%-90.0% vs 96.2%; 95% CI, 96.1%-96.3% scored 0) and a higher prevalence of anemia (13.2%; 95% CI, 12.7%-13.8% vs 8.8%; 95% CI, 8.5%-9.1%), but the 2 cohorts did not have clinically meaningful differences in the baseline prevalence of diabetes, HIV, and thyroid dysfunction. Overall, HCV-positive women were significantly more likely to have a history of tobacco (41.7%; 95% CI, 40.6%-42.9% vs 4.0%; 95% CI, 3.8%-4.2%), alcohol (1.8%; 95% CI, 1.6%-2.0% vs 0.1%; 95% CI, 0.11%-0.12%), opioid (28.9%; 95% CI, 27.3%-30.6% vs 0.3%; 95% CI, 0.25%-0.29%), and cocaine (6.9%; 95% CI, 6.4%-7.4% vs 0.3%; 95% CI, 0.26%-0.30%) use.
Table 1. Characteristics of Pregnant Women and Hospitals in the US from 1998 to 2018a,b.
| Characteristic | HCV status, No. (%) | |
|---|---|---|
| HCV-negative (n = 69 901 008) | HCV-positive (n = 137 259) | |
| Age, y | ||
| 18-20 | 7 392 264 (10.6) | 5825 (4.2) |
| 21-30 | 37 894 073 (54.2) | 78 684 (57.3) |
| 31-40 | 23 321 958 (33.4) | 49 253 (35.9) |
| 41-50 | 1 292 712 (1.8) | 3497 (2.5) |
| Race and ethnicity | ||
| Black | 7 747 775 (13.8) | 8559 (7.1) |
| White | 30 239 399 (53.9) | 93 414 (77.4) |
| Otherc | 18 163 428 (32.3) | 18 739 (15.5) |
| Charlson-Deyo Comorbidity Index | ||
| 0 | 67 271 073 (96.2) | 122 813 (89.5) |
| 1-2 | 2 596 628 (3.7) | 13 694 (10.0) |
| ≥3 | 33 307 (0.05) | 752 (0.5) |
| Diabetes | 498 078 (0.7) | 1497 (1.1) |
| HIV | 16 762 (0.02) | 585 (0.4) |
| Hypertension | 454 605 (0.7) | 1322 (1.0) |
| Thyroid dysfunction | 1 477 110 (2.1) | 3709 (2.7) |
| Anemia | 6 136 463 (8.8) | 18 166 (13.2) |
| Substance use | ||
| Tobacco | 2 814 721 (4.0) | 57 263 (41.7) |
| Alcohol | 81 829 (0.1) | 2437 (1.8) |
| Opioids | 188 609 (0.3) | 39 700 (28.9) |
| Cannabis | 419 522 (0.6) | 8209 (6.0) |
| Sedatives/hypnotics | 16 624 (0.02) | 1269 (0.9) |
| Cocaine | 193 489 (0.3) | 9459 (6.9) |
| Other stimulants | 110 977 (0.2) | 6312 (4.6) |
| Hallucinogens | 1869 (0.003) | 70 (0.05) |
| Inhalants | 80 (0.0001) | 15 (0.01) |
| Other psychoactives | 18 572 (0.03) | 3378 (2.5) |
| Primary payer | ||
| Medicare | 387 062 (0.6) | 3494 (2.6) |
| Medicaid | 27 404 834 (39.3) | 102 013 (74.5) |
| Private insurance | 37 742 800 (54.1) | 23 580 (17.2) |
| Other | 4 222 637 (6.1) | 7879 (5.8) |
| Median household income for zip code | ||
| Quartile 1/lowest | 13 383 777 (26.9) | 49 972 (40.0) |
| Quartile 2 | 12 519 516 (25.1) | 34 835 (27.9) |
| Quartile 3 | 12 274 811 (24.7) | 24 997 (20.0) |
| Quartile 4/highest | 11 606 748 (23.3) | 15 034 (12.0) |
| Hospital size | ||
| Small | 8 587 316 (12.3) | 17 811 (13.0) |
| Medium | 19 092 676 (27.4) | 35 901 (26.3) |
| Large | 41 974 613 (60.3) | 82 849 (60.7) |
| Hospital location/teaching status | ||
| Rural | 8 084 866 (11.6) | 19 154 (14.0) |
| Urban nonteaching | 26 899 867 (38.6) | 33 952 (24.9) |
| Urban teaching | 34 669 873 (49.8) | 83 454 (61.1) |
| Hospital region | ||
| Northeast | 11 928 810 (17.1) | 31 439 (22.9) |
| Midwest | 15 196 028 (21.7) | 24 276 (17.7) |
| South | 26 055 748 (37.3) | 57 158 (41.6) |
| West | 16 720 422 (23.9) | 24 385 (17.8) |
Abbreviation: HCV, hepatitis C virus.
Deliveries and spontaneous abortions included.
Rao-Scott χ2 test was used to compare proportions between HCV status; all differences were significant at P < .001.
Included Asian or Pacific Islander, Hispanic, Native American, and other categories as defined by the Healthcare Cost and Utilization Project; data collapsed due to small sample sizes.
Regarding hospital characteristics, the distribution of admitting hospital sizes was similar between the HCV-positive and HCV-negative cohorts. However, women who were HCV-positive more often presented in rural (14.0%; 95% CI, 12.9%-15.2% vs 11.6%; 95% CI, 10.9%-12.3%) and urban teaching (61.1%; 95% CI, 58.9%-63.4% vs 49.8%; 95% CI, 47.9%-51.6%) hospitals.
The prevalence of maternal HCV infection increased 16-fold, from 0.34 (95% CI, 0.26-0.41) cases per 1000 pregnancies in 1998 to 5.3 (95% CI, 4.9-5.7) cases per 1000 pregnancies in 2018. Over the 21-year study period, the proportion of HCV-positive pregnancies increased in all age groups: 22-fold among women aged 18 to 20 years, 31-fold among women aged 21 to 30 years, 8-fold among women aged 31 to 40 years, and 3-fold among those aged 41 to 50 years (Figure). These increases were concurrent with increasing proportions of pregnancies in women with opioid use histories (Figure; eTable 3 in Supplement 1). The group aged 21 to 30 years experienced an accelerated increase in HCV-positive pregnancies after the start of the second wave of the opioid epidemic in 2010.
Figure. Prevalence of Hepatitis C-Positive or Opioid Use Status Among Pregnancies in the US, Across the First 3 Waves of the Opioid Epidemic.
Multivariable Analysis
After adjusting for maternal- and hospital-related covariates, maternal HCV infection was associated with higher odds of gestational hypertension (adjusted odds ratio [AOR], 1.08; 95% CI, 1.03-1.14). We found no association between maternal HCV infection and gestational anemia, gestational diabetes, preeclampsia, eclampsia, or thyroid dysfunction complicating pregnancy (Table 2).
Table 2. Frequencies and AORs of Maternal and Perinatal Outcomes Between HCV-Positive and HCV-Negative Groups.
| Variable | HCV-negative (n = 69 901 008) | HCV-positive (n = 137 259) | AOR (95% CI)a |
|---|---|---|---|
| Maternal outcomes | |||
| Gestational anemia | 6 525 131 (9.3) | 18 046 (13.1) | 0.99 (0.89-1.10) |
| Gestational diabetes | 4 363 717 (6.2) | 9049 (6.6) | 1.04 (0.98-1.10) |
| Gestational hypertension | 3 697 210 (5.3) | 10 149 (7.4) | 1.08 (1.03-1.14) |
| Preeclampsia/eclampsia | 2 847 323 (4.1) | 6781 (4.9) | 1.00 (0.94-1.06) |
| Thyroid dysfunction complicating pregnancy | 1 006 368 (1.4) | 4564 (3.3) | 1.07 (0.98-1.17) |
| Perinatal outcomes | |||
| Cesarean delivery | 21 133 625 (30.2) | 53 047 (38.6) | 1.19 (1.15-1.22) |
| Stillbirth | 479 088 (0.7) | 1339 (1.0) | 0.95 (0.87-1.03) |
| Spontaneous abortion | 1 029 547 (1.5) | 2263 (1.6) | 0.88 (0.82-0.95) |
| Preterm labor | 4 755 044 (6.8) | 15 892 (11.6) | 1.10 (1.05-1.14) |
| Poor fetal growth | 1 423 590 (2.0) | 7858 (5.7) | 1.29 (1.21-1.37) |
| Fetal distress | 10 891 929 (15.6) | 27 853 (20.3) | 1.11 (1.08-1.15) |
| Premature rupture of membranes | 3 119 340 (4.5) | 8618 (6.3) | 1.04 (0.99-1.10) |
Abbreviations: AOR, adjusted odds ratio; HCV, hepatitis C virus.
Adjusted for maternal age (by decade), race and ethnicity, substance use (by individual categories: tobacco, alcohol, opioids, cannabis, sedatives/hypnotics, cocaine, other stimulants, hallucinogens, inhalants, other psychoactives), HIV, hemodialysis, diabetes, thyroid disorders, hypertension, anemia, Charlson-Deyo Comorbidity Index, primary payer, median household income for mother’s zip code, hospital bed size, hospital location and teaching status, and geographic region.
Concerning perinatal outcomes, maternal HCV infection was associated with higher odds of cesarean delivery (AOR 1.19; 95% CI, 1.15-1.22), preterm labor (AOR 1.10; 95% CI, 1.05-1.14), poor fetal growth (AOR 1.29; 95% CI, 1.21-1.37), and fetal distress (AOR 1.11; 95% CI, 1.08-1.15). Conversely, maternal HCV was associated with lower odds of spontaneous abortion (AOR 0.88; 95% CI, 0.82-0.95). We noted no statistically significant differences in stillbirths or premature rupture of membranes.
Discussion
Our 21-year cross-sectional study observed a significant increase in the prevalence of HCV-positive pregnancies since the start of the US opioid epidemic in the late 1990s. Between 1998 and 2018, the greatest relative increase occurred in women aged 21 to 30 years, followed by those aged 18 to 20 years. The prevalence of HCV-positive pregnancies increased steeply in women aged 21 to 30 years during the second wave of the opioid epidemic between 2010 and 2013 (Figure), when heroin overdose–related deaths first became dominant, and continued through the third wave.4 Our study findings note the increase of a health care epidemic affecting women in their reproductive years in the shadow of the US opioid crisis. Overall, our data are in keeping with those of the Centers for Disease Control and Prevention, which have shown increasing cases of HCV since 2010.3
In our nationally representative cohort, the overall prevalence of HCV-positive pregnancies in 2018 was 5.3 cases per 1000 pregnancies. The estimate was similar to that of a recent study using the US Standard Certificate of Live Birth from 2016 to 2020.21 However, the earliest recommendation for universal HCV screening in pregnancy did not arrive until 2018, so our findings may have underestimated the true prevalence of HCV-positive pregnancies in the US during the study period.9,22 Our data source also could not differentiate between active HCV viremia vs mere HCV seropositivity, which remains long after the viremia resolves. Given an approximate 38% spontaneous viral clearance among women acutely exposed to HCV and 3.7 million births in the US in 2021,23,24 we conservatively estimate approximately 12 000 pregnancies yearly with HCV viremia. Thus, at a 6% risk of vertical transmission,8 approximately 725 infants are born with HCV infection in the US each year.
Hepatitis C virus infection has health associations that extend beyond the liver. In this study cohort, HCV-positive women scored higher on the baseline Charlson-Deyo Comorbidity Index than their HCV-negative peers. We also noted that maternal HCV infection was associated with higher odds of gestational hypertension, cesarean delivery, preterm labor, poor fetal growth, and fetal distress. In contrast, maternal HCV infection was associated with lower odds of spontaneous abortion, although potential explanatory mechanisms were elusive, as early pregnancy losses most often result from genetic anomalies.25 A notable historical consideration was the use of elective cesarean delivery aimed to circumvent mother-to-child HCV transmissions in the 1990s and early 2000s due to conflicting available data.26,27 The American College of Obstetricians and Gynecologists formally recommended against elective cesarean delivery for maternal HCV in 2007,28 so a time-dependent association was possible. Nonetheless, an analysis of nationally representative data from 2012 to 2018 noted increased odds of cesarean delivery in the HCV-positive group compared with HCV-negative controls.12 Overall, our study findings were consistent with many, albeit not all, prior studies on associations between HCV infection and maternal and perinatal outcomes. For instance, 2 meta-analyses of studies on maternal HCV infection, preterm birth, and poor fetal growth calculated pooled estimates that qualitatively concurred with our findings, but individual studies included in the meta-analyses did not always reach the same conclusions.29,30 Elucidating the mechanisms of perinatal HCV outcomes was ultimately outside the scope of our study; further physiologic-based investigations are needed.
Our study complemented other publications on HCV infection during pregnancy. While many existing studies showed epidemiologic snapshots of shorter lengths after the early 2000s,6,11,12 our analysis of a nationally representative data set provides a single, unified trend estimate over 2 decades to encompass the start of the US opioid epidemic. We also featured more granularity than earlier studies when representing substance use.14 Instead of a generic composite clinical factor (eg, drug use), our analysis specifically captured and adjusted for 10 individual categories of substance use (eg, alcohol, opioids, and cocaine), several of which may have higher magnitudes of association with adverse pregnancy-related outcomes. Thus, we expect our investigation of HCV-associated adverse maternal and perinatal outcomes to have less residual confounding from substance use disorders.
Our findings on the increasing prevalence of HCV-positive pregnancies and associated adverse outcomes support the recent recommendations for universal HCV screening during each pregnancy.9,10 Universal screening is undoubtedly more comprehensive than previous strategies of targeted screening.22 One cohort study reported that universal HCV screening during pregnancy identified 31% more expectant mothers with HCV seropositivity than a risk-based screening protocol.31 Yet, universal HCV screening of pregnant women also may remain cost-effective even when the disease prevalence is as low as 0.07%.32 Despite these advantages, universal screening is only one step in the HCV care cascade. Given the health-related implications and vertical transmission risk, connecting to appropriate specialist care is paramount for both mother and child. However, previous data report that postpartum engagement with HCV care occurs in only about a quarter of the cases.31,33
Our study findings suggest that multidisciplinary collaborations in caring for women with HCV may help enhance engagement with specialist care. In the cohort studied herein, alcohol, tobacco, and drug use and limited income were more common among HCV-positive women. Substance use and financial insecurity have been associated with higher barriers to health care access,34,35 increasing the risk of medical complications. Obstetric and primary care clinics can aim to establish a more durable and comprehensive care model through collaborations with hepatology or infectious disease, addiction medicine, social work, case management, and psychiatry services. A recently published perinatal care model from a safety-net hospital combining a linkage protocol and a multidisciplinary colocated clinic resulted in increased HCV treatment initiations (adjusted rate ratio, 3.36; 95% CI, 1.57-7.17).36 Further investigations into the optimal implementation of maternal-infant care linkage programs can help advance the mission of HCV eradication.
Strengths and Limitations
A strength of our study is its large sample size, which allowed for sufficient statistical precision to analyze relatively infrequent events during pregnancy. The sample size enabled us to account for multiple potential confounders with our multivariable regression models. For instance, our model was able to condition on 10 individual substance categories. Nonetheless, the large sample size could also increase statistically significant findings with unclear implications in the clinical setting; therefore, informed interpretations and comparisons with other published studies are still necessary. Additionally, the NIS is nationally representative, attenuating regional practice pattern variations. We also queried and collated 21 years of data from the NIS within the same analysis to generate long-term trend estimates and create a summarizing visualization.
We acknowledge several limitations. First, ICD-9-CM and ICD-10-CM codes often do not explicitly capture the absence of a diagnosis (eg, HCV-negative) and thus can introduce misclassification bias. For example, some pregnant women with HCV may have been unscreened or miscoded at hospital discharge. Thus, the true prevalence of HCV-positive pregnancies in the US may be higher than reported herein, and our data may have underestimated the associations between maternal HCV and maternal and perinatal adverse events. However, a study of National Center for Health Statistics birth certificate data from 2011 to 2016 suggested that HCV screening practices improved over time,37 although it could also partly explain the increasing trend we observed. Second, the NIS does not collect data on HCV viral titers, restricting us from further defining HCV-positive as active HCV viremia or nonviremic HCV seropositivity. Consequently, our prevalence estimates likely contained a mix of both scenarios; we also could not analyze the differential outcomes of viral titer levels or resolved infection in perinatal events. Third, the NIS does not capture several potentially relevant background characteristics, such as prenatal care or health literacy, so some unmeasured or residual confounding could have influenced our statistical inferences. Fourth, the NIS reports only hospitalization data; it does not capture adverse outpatient events occurring earlier in pregnancy or during home births. However, out-of-hospital births account for fewer than 2% of all childbirths in the US.38 Fifth, the NIS underwent a sampling redesign that resulted in a one-time disruption of the 2012 data.19 Despite the revised trend weights used in our analysis,39 the sampling redesign likely still contributed to the unexpected single-year drop in the prevalence of HCV-positive pregnancies in 2012 among women aged 41 to 50 years. In comparison, the other 3 age cohorts were substantially larger and more robust against aberrant statistical processes. Sixth, the NIS does not link mother-to-infant medical records, so we could not assess the long-term outcomes of children born to HCV-positive mothers.
Conclusions
This cross-sectional study noted the prevalence of maternal HCV infection has increased markedly since the start of the opioid epidemic in the US, reaching 5.3 cases per 1000 pregnancies in 2018. The most substantial increases during the opioid epidemic were noted in women aged 18 to 30 years. Additionally, HCV infection is associated with a heightened risk of adverse maternal and perinatal outcomes. Overall, our data support the recommendations for universal HCV screening with each pregnancy proposed by the Centers for Disease Control and Prevention and American College of Obstetricians and Gynecologists. Perinatal care and delivery may be the initial health care exposure for many women. These touchpoints represent an opportunity for health care professionals to identify HCV infection and link women and their children to appropriate specialist care.
eTable 1. International Classification of Diseases (ICD) Codes Used in the Analysis, Excluding Codes for the Charlson-Deyo Comorbidity Index
eTable 2. Frequencies of Hepatitis C–Positive Pregnancies in the US by Year, From 1998 to 2018
eTable 3. Pregnancies in the US With and Without Opioid Use by Year, From 1998 to 2018
Data Sharing Statement
References
- 1.Centers for Disease Control and Prevention. Blood safety. Diseases and organisms. May 7, 2019. Accessed June 7, 2022. https://www.cdc.gov/bloodsafety/bbp/diseases-organisms.html
- 2.Centers for Disease Control and Prevention. Viral hepatitis. Hepatitis C questions and answers for health professionals. March 30, 2022. Accessed June 7, 2022. https://www.cdc.gov/hepatitis/hcv/hcvfaq.htm
- 3.Centers for Disease Control and Prevention. Viral hepatitis. and TB Prevention. 2017 Surveillance. Accessed June 7, 2022. https://www.cdc.gov/hepatitis/statistics/2015surveillance/index.htm
- 4.Centers for Disease Control and Prevention. Understanding the opioid overdose epidemic. June 1, 2022. Accessed June 7, 2022. https://www.cdc.gov/opioids/basics/epidemic.html
- 5.Ryerson AB, Schillie S, Barker LK, Kupronis BA, Wester C. Vital signs: newly reported acute and chronic hepatitis C cases—United States, 2009-2018. MMWR Morb Mortal Wkly Rep. 2020;69(14):399-404. doi: 10.15585/mmwr.mm6914a2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rossi RM, Wolfe C, Brokamp R, et al. Reported prevalence of maternal hepatitis C virus infection in the United States. Obstet Gynecol. 2020;135(2):387-395. doi: 10.1097/AOG.0000000000003644 [DOI] [PubMed] [Google Scholar]
- 7.Haight SC, Ko JY, Tong VT, Bohm MK, Callaghan WM. Opioid use disorder documented at delivery hospitalization—United States, 1999-2014. MMWR Morb Mortal Wkly Rep. 2018;67(31):845-849. doi: 10.15585/mmwr.mm6731a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Benova L, Mohamoud YA, Calvert C, Abu-Raddad LJ. Vertical transmission of hepatitis C virus: systematic review and meta-analysis. Clin Infect Dis. 2014;59(6):765-773. doi: 10.1093/cid/ciu447 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Schillie S, Wester C, Osborne M, Wesolowski L, Ryerson AB. CDC recommendations for hepatitis C screening among adults—United States, 2020. MMWR Recomm Rep. 2020;69(2):1-17. doi: 10.15585/mmwr.rr6902a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Routine hepatitis C virus screening in pregnant individuals. American College of Obstetricians and Gynecologists. May 2021. Accessed February 19, 2022. https://www.acog.org/en/clinical/clinical-guidance/practice-advisory/articles/2021/05/routine-hepatitis-c-virus-screening-in-pregnant-individuals
- 11.Patrick SW, Dupont WD, McNeer E, et al. Association of individual and community factors with hepatitis C infections among pregnant people and newborns. JAMA Health Forum. 2021;2(10):e213470. doi: 10.1001/jamahealthforum.2021.3470 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chen B, Wang Y, Lange M, Kushner T. Hepatitis C is associated with more adverse pregnancy outcomes than hepatitis B: a 7-year national inpatient sample study. Hepatol Commun. 2022;6(9):2465-2473. doi: 10.1002/hep4.2002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ko JY, Haight SC, Schillie SF, Bohm MK, Dietz PM. National trends in hepatitis C infection by opioid use disorder status among pregnant women at delivery hospitalization—United States, 2000-2015. MMWR Morb Mortal Wkly Rep. 2019;68(39):833-838. doi: 10.15585/mmwr.mm6839a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Arditi B, Emont J, Friedman AM, D’Alton ME, Wen T. Deliveries among patients with maternal hepatitis C virus infection in the United States, 2000-2019. Obstet Gynecol. 2023;141(4):828-836. doi: 10.1097/AOG.0000000000005119 [DOI] [PubMed] [Google Scholar]
- 15.Agency for Healthcare Research and Quality. NIS database documentation. Accessed June 7, 2022. https://www.hcup-us.ahrq.gov/db/nation/nis/nisdbdocumentation.jsp
- 16.Reddick KLB, Jhaveri R, Gandhi M, James AH, Swamy GK. Pregnancy outcomes associated with viral hepatitis. J Viral Hepat. 2011;18(7):e394-e398. doi: 10.1111/j.1365-2893.2011.01436.x [DOI] [PubMed] [Google Scholar]
- 17.Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-619. doi: 10.1016/0895-4356(92)90133-8 [DOI] [PubMed] [Google Scholar]
- 18.Sundararajan V, Henderson T, Perry C, Muggivan A, Quan H, Ghali WA. New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality. J Clin Epidemiol. 2004;57(12):1288-1294. doi: 10.1016/j.jclinepi.2004.03.012 [DOI] [PubMed] [Google Scholar]
- 19.Agency for Healthcare Research and Quality. Houchens R, Ross D, Elixhauser A, Jiang J. Nationwide inpatient sample redesign final report. US; 2014. Accessed June 7, 2022. https://www.hcup-us.ahrq.gov/tech_assist/trends/508/508course_2019.jsp#changes
- 20.Rao JNK, Scott AJ. The analysis of categorical data from complex sample surveys: chi-squared tests for goodness of fit and independence in two-way tables. J Am Stat Assoc. 1981;76(374):221-230. doi: 10.1080/01621459.1981.10477633 [DOI] [Google Scholar]
- 21.Ahrens KA, Rossen LM, Burgess AR, Palmsten K, Ziller EC. Maternal hepatitis C prevalence and trends by county, US: 2016-2020. Paediatr Perinat Epidemiol. 2023;37(2):134-142. doi: 10.1111/ppe.12934 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Carey I, Christiana M, Marie-Ange M, et al. Universal versus targeted screening for HCV infection in pregnancy in a diverse, multi-ethnic population: universal screening is more comprehensive. J Viral Hepat. 2022;29(12):1079-1088. doi: 10.1111/jvh.13752 [DOI] [PubMed] [Google Scholar]
- 23.Hamilton BE, Martin JA, Osterman MJ. Births: provisional data for 2021. National Center for Health Statistics. May 2022:1-12. Accessed September 20, 2022. https://www.cdc.gov/nchs/data/vsrr/vsrr020.pdf
- 24.Grebely J, Page K, Sacks-Davis R, et al. ; InC3 Study Group . The effects of female sex, viral genotype, and IL28B genotype on spontaneous clearance of acute hepatitis C virus infection. Hepatology. 2014;59(1):109-120. doi: 10.1002/hep.26639 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Blue NR, Page JM, Silver RM. Genetic abnormalities and pregnancy loss. Semin Perinatol. 2019;43(2):66-73. doi: 10.1053/j.semperi.2018.12.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gibb DM, Goodall RL, Dunn DT, et al. Mother-to-child transmission of hepatitis C virus: evidence for preventable peripartum transmission. Lancet. 2000;356(9233):904-907. doi: 10.1016/S0140-6736(00)02681-7 [DOI] [PubMed] [Google Scholar]
- 27.European Paediatric Hepatitis C Virus Network. Effects of mode of delivery and infant feeding on the risk of mother-to-child transmission of hepatitis C virus. BJOG. 2001;108(4):371-377. [PubMed] [Google Scholar]
- 28.American College of Obstetricians and Gynecologists . ACOG Practice Bulletin No. 86: viral hepatitis in pregnancy. Obstet Gynecol. 2007;110(4):941-956. doi: 10.1097/01.AOG.0000263930.28382.2a [DOI] [PubMed] [Google Scholar]
- 29.Huang QT, Huang Q, Zhong M, et al. Chronic hepatitis C virus infection is associated with increased risk of preterm birth: a meta-analysis of observational studies. J Viral Hepat. 2015;22(12):1033-1042. doi: 10.1111/jvh.12430 [DOI] [PubMed] [Google Scholar]
- 30.Huang QT, Hang LL, Zhong M, Gao YF, Luo ML, Yu YH. Maternal HCV infection is associated with intrauterine fetal growth disturbance: meta-analysis of observational studies. Medicine (Baltimore). 2016;95(35):e4777. doi: 10.1097/MD.0000000000004777 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bushman ET, Subramani L, Sanjanwala A, et al. Pragmatic experience with risk-based versus universal hepatitis C screening in pregnancy: detection of infection and postpartum linkage to care. Am J Perinatol. 2021;38(11):1109-1116. doi: 10.1055/s-0041-1728827 [DOI] [PubMed] [Google Scholar]
- 32.Chaillon A, Rand EB, Reau N, Martin NK. Cost-effectiveness of universal hepatitis C virus screening of pregnant women in the United States. Clin Infect Dis. 2019;69(11):1888-1895. doi: 10.1093/cid/ciz063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Epstein RL, Sabharwal V, Wachman EM, et al. Perinatal transmission of hepatitis C virus: defining the cascade of care. J Pediatr. 2018;203:34-40.e1. doi: 10.1016/j.jpeds.2018.07.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Devoe JE, Baez A, Angier H, Krois L, Edlund C, Carney PA. Insurance + access not equal to health care: typology of barriers to health care access for low-income families. Ann Fam Med. 2007;5(6):511-518. doi: 10.1370/afm.748 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Matsuzaki M, Vu QM, Gwadz M, et al. Perceived access and barriers to care among illicit drug users and hazardous drinkers: findings from the Seek, Test, Treat, and Retain data harmonization initiative (STTR). BMC Public Health. 2018;18(1):366. doi: 10.1186/s12889-018-5291-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Epstein RL, Moloney C, Garfinkel J, et al. Enhancing linkage to hepatitis C virus treatment following pregnancy in women identified during perinatal care. Hepatol Commun. 2021;5(9):1543-1554. doi: 10.1002/hep4.1748 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Schillie SF, Canary L, Koneru A, et al. Hepatitis C virus in women of childbearing age, pregnant women, and children. Am J Prev Med. 2018;55(5):633-641. doi: 10.1016/j.amepre.2018.05.029 [DOI] [PubMed] [Google Scholar]
- 38.MacDorman MF, Declercq E. Trends and state variations in out-of-hospital births in the United States, 2004-2017. Birth. 2019;46(2):279-288. doi: 10.1111/birt.12411 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Agency for Healthcare Research and Quality Healthcare Cost and Utilization Project (HCUP) . HCUP NIS trend weights. Published October 2021. Accessed July 13, 2022. https://www.hcup-us.ahrq.gov/db/nation/nis/trendwghts.jsp
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. International Classification of Diseases (ICD) Codes Used in the Analysis, Excluding Codes for the Charlson-Deyo Comorbidity Index
eTable 2. Frequencies of Hepatitis C–Positive Pregnancies in the US by Year, From 1998 to 2018
eTable 3. Pregnancies in the US With and Without Opioid Use by Year, From 1998 to 2018
Data Sharing Statement