CDC Recommendations for Hepatitis C Testing …

Methods

The Recommendations Steering Committee, composed of CDC staff with expertise in viral hepatitis, obstetrics, pediatrics, infectious diseases, and policy (Supplementary Appendix 1, https://stacks.cdc.gov/view/cdc/134020), met regularly to oversee the development of the recommendations. The steering committee designed a comprehensive systematic review of the literature to guide the decision for these recommendations. The purpose of the review was to examine the benefits and harms of different testing strategies for identifying children with perinatally acquired HCV infection.

The following research question guided the development of the recommendations:

• Among children perinatally exposed to HCV, does NAT for HCV RNA at age 2–6 months^* compared with HCV antibody testing with reflex RNA testing (i.e., NAT for HCV RNA after a reactive anti-HCV test) at age ≥18 months, increase the diagnosis of current HCV infections, increase linkage to care and treatment, and decrease cirrhosis and deaths attributable to HCV infection?

This question was further broken down into five questions guiding a chain of indirect evidence:

• Compared with HCV antibody testing at age ≥18 months, how would NAT for HCV RNA at age 2–6 months affect the number of children identified with perinatally acquired HCV infection?
• How many additional children with perinatally acquired HCV infection would be identified by testing with NAT for HCV RNA at age 2–6 months?
• How many additional children with perinatally acquired HCV infection would be linked to care by testing with NAT for HCV RNA at age 2–6 months?
• Do desirable effects (i.e., benefits) of testing for HCV infection outweigh undesirable effects (harms)?
• What is the effect of diagnosis at age 2–6 months with NAT for HCV RNA on cirrhosis and deaths attributable to HCV infection?

Key questions (KQs) were developed for each of the five questions from the chain (Supplementary Table 1, https://stacks.cdc.gov/view/cdc/133599):

• K.Q.1.a. What is the prevalence of HCV infection among pregnant persons in the United States?
• K.Q.1.b. What proportion of pregnant persons are tested for HCV infection in the United States?
• K.Q.1.c. What proportion of children perinatally exposed to HCV become infected?
• K.Q.2.a. What is the diagnostic accuracy of HCV antibody testing and NAT for HCV RNA among perinatally exposed children?
• K.Q.2.b. What proportion of children perinatally exposed to HCV are tested for HCV infection?
• K.Q.3.a. What proportion of children with confirmed HCV infection are linked to care?
• K.Q.4.a. What are the benefits of HCV testing among perinatally exposed children?
• K.Q.4.b. What are the harms of HCV testing among perinatally exposed children?
• K.Q.5.a. What is the effect of hepatitis C diagnosis in childhood on related morbidity and mortality (including cirrhosis, HCC, and death)?
• K.Q.5.b. What is the effect of DAA treatment in childhood on hepatitis C–related morbidity (including cirrhosis and HCC)?

Literature Review

A systematic review of the literature was conducted to examine available evidence on HCV infection prevalence among pregnant persons and perinatally exposed children, loss of follow-up among perinatally exposed children, and the benefits and harms of testing perinatally exposed children.

A search for English language, peer-reviewed journal articles published in Medline (Ovid), Embase (Ovid), Cochrane Library, CINAHL (EBSCO), and Scopus was performed (Supplementary Table 2, https://stacks.cdc.gov/view/cdc/133599). The search included articles published during January 1, 2001–June 8, 2021. The 20-year period was selected because of the expected scarcity of data among this population. Duplicates were identified using EndNote software (version 20; Clarivate), which automated the “find duplicates” function with preference set to match on title, author, and year. Duplicates were removed from the EndNote library.

All references from the initial search were uploaded into DistillerSR software (version 2.35; Evidence Partners) for further review by the Recommendations Work Group (Supplementary Appendix 1, https://stacks.cdc.gov/view/cdc/134020). Two independent reviewers checked all titles and abstracts for relevance to the research question (AS, JB, or NN and LP). Titles determined to be non-English language articles or not relevant to the study question were not included in the abstract review. All articles determined to be relevant in the title review and articles with conflicting results in the title review were included in the abstract review. Similarly, all abstracts determined to be relevant to the research question and articles with conflicting inclusion results in the abstract review were included for full text review.

Included articles were separated into three categories for the full text review: 1) U.S. articles only discussing HCV in pregnancy (without data on perinatal HCV), 2) U.S. articles that included data regarding perinatal HCV testing, and 3) international articles that potentially included harms of perinatal HCV testing. All full text articles were independently reviewed by two reviewers (AS, EC, JB, LC, or MF and LP). Relevant data were abstracted independently and compared. All differences in abstracted data were discussed by the two reviewers until they reached agreement.

Articles were excluded if an English language version could not be found, were not related to HCV infection, treatment, or testing; were international articles not specific to perinatal HCV transmission and testing; were case reports, case series, opinion articles, editorials, review articles, or conference abstracts; contained only modeled data or only animal data; included adults aged ≥18 years (unless also included perinatally infected or pregnant persons); reported HCV infection among children not acquired perinatally; or only reported on medications not recommended for use among children. Data on testing rates and prevalence of HCV infection during pregnancy that had been considered in the development of the 2020 report on CDC recommendations for HCV screening among adults were also included in this literature summary (12). On completion of the formal literature review, reference lists from all U.S. and international review articles were reviewed to identify additional articles for full text review.

After data abstraction, all included articles related to pregnancy and perinatal HCV testing rates, incidence, linkage to care, and outcomes underwent review to assess the quality of the evidence using the National Institutes of Health (NIH) Study Quality Assessment Tools, which were developed specific to certain study designs and focus on concepts key to a study’s internal validity (Supplementary Table 3, https://stacks.cdc.gov/view/cdc/133599). Articles were first categorized by study design, and each criterion and overall rating was independently scored by two reviewers (AS or EC and LP) using the descriptions available at https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools. The two reviewers discussed and reached agreement on giving the articles an overall rating of good, fair, or poor. All articles, regardless of their rating, were included in the overall analysis. Because articles describing harms of perinatal HCV screening varied in the relation between the study design and the harm mentioned, the quality of U.S. and international articles on the harms of perinatal HCV testing was scored based on two standardized measures developed by the Recommendations Steering Committee: 1) level of confidence (high, moderate, or low), indicating how the specific harm was measured in the study, and 2) outcome prioritization (critical, important, not important), indicating the relevance of the harm as it related to the study question. Although these determinations were based on expert opinion and judgment, two reviewers (AS, EC, or MF and LP) independently scored each study with a harm; discrepancies in the level of confidence or the outcome prioritization were discussed by the two reviewers until they reached agreement. The quality assessment of the cost-effectiveness study was evaluated by two independent reviewers (TA and LP) using the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) checklist (80).

To identify recently published studies through December 31, 2022, a supplemental literature search was conducted on January 17, 2023, using a search strategy identical to the original search (Supplementary Table 4, https://stacks.cdc.gov/view/cdc/133599). Titles and abstracts were independently reviewed by NN or MW and LP. All studies with conflicting agreement on inclusion proceeded for full text reviews, which were independently reviewed by AS or EC and LP. Evidence quality reviews for all included articles were conducted using the same methods as for the original reviews by AS or EC and LP. Abstracted data and evidence assessments from full text reviews were added to the original review.

In developing recommendations, the Recommendations Steering Committee considered the results of the literature review and findings from the effectiveness and cost-effectiveness modeling study (see Evidence Summary). Furthermore, the steering committee had biweekly meetings to discuss implementation feasibility, public health implications, and equitable access to testing.

CDC determined that these recommendations included influential scientific information with a clear and substantial impact on important public policies and private sector decisions. As required by the Information Quality Act (81), peer review by external specialists not involved in the development of the recommendations was conducted. CDC solicited nominations for peer reviewers from the American Academy of Pediatrics (AAP), NASPGHAN, AASLD, the American Academy of Family Physicians (AAFP), and ACOG. Six peer reviewers from the listed organizations with expertise in pediatrics, infectious diseases, hepatology, and obstetrics reviewed the recommendations and provided structured peer reviews (Supplementary Appendix 1, https://stacks.cdc.gov/view/cdc/134020). Representatives from professional societies, providers, advocacy groups, and public health professionals have communicated the need for clear recommendations through conferences, journal commentary, and other types of communications. Buy-in from relevant parties was obtained at inception of recommendation development in April 2021, and methods for developing the recommendations were summarized in a presentation only (i.e., no vote was conducted) at the April 2022 meeting of the CDC/Health Resources and Services Administration Advisory Committee on HIV, Viral Hepatitis and STD Prevention and Treatment. Opportunity for reaction and feedback to the draft recommendation was provided through a public comment period (November 22, 2022–January 27, 2023) and an informational webinar open to the public, academia, advocacy groups, and partner organizations. CDC received 22 public comments on the draft document from the public, providers, advocacy groups, industry, medical professional associations, think tanks, and one public health department. Peer reviewer and public comments were considered by the work group, and edits made in response were documented (Supplementary Appendix 2 and Supplementary Appendix 3, https://stacks.cdc.gov/view/cdc/134020).

Summary of the Literature

The initial literature search yielded 3,802 articles. All titles were screened, 1,241 (32.6%) abstracts were reviewed, and 201 (5.3%) full texts were reviewed for possible inclusion. A total of 35 articles (0.9%) from the initial literature review had data available to abstract. An additional six articles were identified from references and were included for data abstraction. After review, 41 articles were included.

The supplementary literature search yielded an additional 459 articles. All titles were screened, 160 abstracts (34.9%) were reviewed, and 23 (5.0%) full texts were reviewed for possible inclusion. A total of 11 articles (2.4%) from the supplementary literature review had data to abstract. An additional four articles were identified from references and were included for abstraction. The supplementary review added 15 articles to the original 41 for a total of 56 articles included.

Sixteen articles included data related to HCV testing rates during pregnancy, 12 of which were reviewed in the 2020 CDC adult HCV screening recommendations (Supplementary Table 5, https://stacks.cdc.gov/view/cdc/133599). The median percentage of pregnant persons tested for HCV infection was 47.6% (range: 0.7%–98.4%) (Table 1). Thirty-five articles included data regarding anti-HCV positivity or hepatitis C diagnoses during pregnancy, 26 of which were reviewed in the 2020 CDC adult HCV screening recommendations. The median prevalence of anti-HCV positivity or diagnosis was 1.1% (range: 0.1%–70.8%). Eleven articles included data regarding HCV RNA positivity in pregnancy among those who were anti-HCV positive, of which four were reviewed in the 2020 CDC adult HCV screening recommendations. The median prevalence of HCV RNA positivity was 68.2% (range: 29.6%–81.3%).

Two articles presented data on the proportion of perinatally exposed children who were referred for HCV testing (Supplementary Table 6, https://stacks.cdc.gov/view/cdc/133599). The median percentage of children referred for testing was 16.7% (range: 1.9%–31.4%) (Table 1). Twelve articles presented data related to the proportion of perinatally exposed children tested for HCV infection, either with an anti-HCV test or NAT for HCV RNA. The median percentage tested for HCV infection was 30.1% (range: 8.6%–53.1%). Thirteen articles presented data regarding the rate of perinatal transmission. The median rate was 4.7% (range: 0.0%–11.1%). One article presented information regarding linkages to care among perinatally infected children and indicated that five of five (100%) were linked to care. Seven studies included data related to DAA treatment among perinatally HCV infected children aged 3–17 years (Supplementary Table 7, https://stacks.cdc.gov/view/cdc/133599). In these studies, the median percentage of children with chronic HCV infection who achieved sustained virologic response 12 weeks posttreatment was 98.1% (range: 96%–100%). Assessment of the quality of evidence was performed for each of the included articles using the NIH Study Quality Assessment tool, and the results ranged from fair to good (Supplementary Table 8, https://stacks.cdc.gov/view/cdc/133599).

Both U.S. and international articles were evaluated for harms associated with testing perinatally exposed children for HCV infection; six U.S. articles and 24 international articles were included (Supplementary Table 9, https://stacks.cdc.gov/view/cdc/133599). Among 30 studies that described potential harms, the most commonly reported harms were related to interpretation of test results, including intermittent or transient viremia (13 studies), false-positive antibody results (one study), and false-negative antibody results (two studies). Other harms included the cost of testing (four studies); stigma (four studies); guilt, stress, and concern about a child’s health, school, employment, and future marriage (one study); wait time for screening at 18 months (one study); misclassification of vertical transmission (one study); parental refusal of testing (one study); absence of approved treatment (one study); involvement of social services (one study); distance to follow up for families living far away (one study); time to go to the laboratory and wait for the test (one study); lack of testing and transportation availability (one study); pressure on clinic staff members to order and explain test results (one study); and delay in infection status or uncertain prognosis clarification contributing to parental confusion and poor clinic attendance (one study). All studies with harms were evaluated for the level of confidence and outcome prioritization (Supplementary Table 9, https://stacks.cdc.gov/view/cdc/133599). After careful evaluation and review, the Recommendations Steering Committee concluded that the benefits of testing outweighed the identified and potential harms of testing.

The data from the literature review are subject to at least four limitations. First, although data were abstracted in a consistent method, there was heterogeneity in the studies included (e.g., certain studies did not differentiate between the type of diagnostic test used [i.e., anti-HCV or NAT for RNA] or the age of the infant or child at the time the test was performed). In addition, studies might have defined maternal HCV infection as a single positive anti-HCV test, which might not represent current infection. Second, approximately 20 years of data and knowledge about perinatal HCV infection from 2001 to 2022 were included, including exposure and transmission, diagnosis, HCV RNA test performance, treatment, and outcomes. However, the data and knowledge about perinatal HCV infection have progressed substantially over this time. As a result, certain studies might have included previous definitions of HCV infection or less sensitive testing methods. However, because of the scarcity of data on perinatal HCV infection, including more rather than less data was essential. Third, with the exception of articles that described potential harms, articles with no U.S. data were excluded for all reviews to make the recommendations generalizable to the U.S. population on the basis of similar populations, medical care, treatment guidelines, and outcomes. However, this step might have excluded certain potentially relevant international studies. Finally, data from the systematic review of the literature were limited by the designs of the individual studies and the quality of the evidence. For this reason, standardized quality assessment tools created by NIH for observational cohort and cross-sectional studies and for before-after (prepost) studies with no control group (https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools) were used to assess the quality of the evidence from included studies.

Cost-Effectiveness Considerations

Evaluating the cost-effectiveness of public health interventions is critical to developing recommendations. No cost-effectiveness studies comparing testing approaches for perinatal HCV infections were identified during the development of these recommendations. Therefore, a CDC-conducted novel analysis evaluating the optimal testing strategy for perinatally exposed children guided these recommendations (82). Through CDC’s National Center for HIV, Viral Hepatitis, STD, and TB Prevention’s Epidemiologic and Economic Modeling Agreement, a mathematic modeling study was conducted using an economic analysis framework to compare the current strategy of anti-HCV testing with reflex to NAT for HCV RNA starting at age 18 months with a proposed strategy of a single NAT for HCV RNA at age 2–6 months. Also included were considerations for universal testing strategies for both options (i.e., all infants regardless of maternal HCV status). Inputs and estimates for the study were guided by published literature. For rate of hepatitis C screening during pregnancy, a mean estimate of 44.7% was used with sensitivity analyses to account for expected increases in screening during pregnancy because universal screening recommendations are becoming more widely implemented. In addition, decreased loss to follow-up was accounted for with the proposed strategy because more infants are expected to attend 2–6-month well-child visits than 18-month well-child visits (83). Costs and health outcomes of the various strategies were modeled and incorporated rates of spontaneous clearance of infection. Outcomes included diagnosed infections, treated or cured infections, HCC, liver transplants, and liver-related deaths.

The modeling study found that, compared with the baseline strategy of testing exposed children aged 18 months with an anti-HCV test and reflex to NAT for HCV RNA, an increased number of infants received diagnoses and had improved health outcomes when NAT for HCV RNA was performed at age 2–6 months among exposed children. Universal screening with both anti-HCV testing with reflex to NAT for HCV RNA at 18 months and NAT for HCV RNA at age 2–6 months also improved infant diagnoses and health outcomes. However, testing known exposed infants at age 2–6 months was the only strategy that was cost-saving compared with the baseline strategy, with a population level cost savings of $469,671 per year, assuming 3.6 million births with 0.64% of births occurring among persons with HCV infection. Although the universal testing strategies resulted in an increase in quality-adjusted life years (QALYs) compared with the baseline strategy, the strategies resulted in increased total costs ranging from $38 million for the universal anti-HCV test with reflex to NAT for the HCV RNA strategy at 18 months to $129 million for the universal NAT for HCV RNA at age 2–6 months (incremental cost-effectiveness ratios [ICERs] of 26,105 and 35,887 per QALY gained, respectively). Compared with NAT for HCV RNA among exposed children at age 2–6 months, universal NAT for HCV RNA screening becomes decreasingly cost-effective as the proportion of pregnant persons screened for HCV increases (i.e., because universal screening allows the HCV infection status of every pregnant person to be known, universal testing of infants would result in diminishing return). The strategy of testing known perinatally exposed infants with a NAT for HCV RNA at age 2–6 months was the only perinatal hepatitis C testing strategy that was both cost-saving and resulted in better health outcomes. Limitations of the study included a paucity of data on testing strategies for infants and children exposed to perinatal HCV, attributing the cost and benefits of testing only to the exposed child, assuming perinatal HCV not diagnosed in childhood would not be diagnosed and treated later in life and assuming that there was no difference in linkage to treatment or costs before age 3 years with the different testing strategies. A health economist determined the study met the overall standards for quality of the CHEERS checklist (Supplementary Table 10, https://stacks.cdc.gov/view/cdc/133599). The steering committee determined that any minor deviations from the checklist were appropriate and did not compromise the quality of the evidence.

Rationale for Recommendations

Rates of HCV infections during pregnancy have been increasing (6,7), corresponding with the ongoing opioid crisis (11). Although perinatal HCV transmission occurs in up to 7% of perinatally exposed children (13,14), approximately 70% of children aged ≥18 months are not being tested with the current strategy of anti-HCV testing, leading to loss to follow-up (Table 1). With the availability of highly sensitive and specific NATs for HCV RNA detection starting at age 2 months (79) and more children attending well-child visits at age 2–6 months compared with those aged ≥18 months (17,84), a NAT for HCV RNA at age 2–6 months is both cost-effective and cost-saving and will identify more children with perinatal HCV transmission who are eligible for curative treatment beginning at age 3 years (82).

HCV Testing Strategy

HCV testing of children exposed perinatally identifies children who are at risk for developing complications from chronic HCV infection. Standard clinical practice for testing with a NAT for HCV RNA is done at or after age 2 months, and testing with anti-HCV is done at or after age 18 months because of the persistence of antibody passively transferred from the infected birthing parent to the infant (37,85–87). The older, less sensitive NAT for HCV RNA used in certain studies of perinatally exposed infants was associated with potential false-negative results indicating intermittent viremia (37,88,89); however, currently used tests are highly sensitive with a lower limit of detection of 15.0 IU/mL or less for genotype 1a (90). One NAT for HCV RNA at age 2–6 months is sufficient to determine perinatal infection. On the basis of data related to the sensitivity, specificity, and PPVs and NPVs, a detectable HCV RNA result confirms perinatal HCV transmission and an undetectable HCV RNA rules out perinatal HCV transmission (79,85,88,91). If anti-HCV testing is done at or after age 18 months, a NAT for HCV RNA on all reactive specimens will identify current HCV infection. Children with nonreactive anti-HCV tests at age ≥18 months and children with undetectable HCV RNA test results after a reactive anti-HCV test at age ≥18 months do not need further follow-up.