BACKGROUND: The California Perinatal Hepatitis B Prevention Program (PHBPP) identifies and provides case management of hepatitis B virus (HBV)-infected pregnant women and their infants. The goal is to enroll all pregnant, HBV-infected women into PHBPP to ensure that their infants receive timely administration of post-exposure prophylaxis. In California, prenatal HBsAg screening is mandatory, maximizing the probability that HBV-infected pregnant women are reported to PHBPP. However, barriers to care, testing and reporting may remain, thus compromising the program’s effectiveness. We aimed to evaluate the completeness of PHBPP surveillance and examine risk factors associated with underreporting.
METHODS: A retrospective capture-recapture analysis was used to identify unreported PHBPP-eligible women. Eligible women were identified by a probabilistic matching algorithm using name, date of birth, and zip code of residence. Women with chronic hepatitis B reported to the California Chronic Hepatitis B Registry (CHBR) were matched to women with a live birth in 2013 or 2014 per California birth certificate data. Identified matches were compared to women reported to PHBPP in 2013 and 2014 using the same matching algorithm. A Lincoln-Petersen estimate was used to approximate the total number of HBV-infected pregnant women in California. Potential risk factors for underreporting were selected a priori, and Chi square analysis (p≤0.2) was used to determine the variables included in the final model. Multivariate logistic regression was used to assess the relationship between underreporting and predictors of interest.
RESULTS: 5,832 women were reported to PHBPP from 2013-2014. 5,442 PHBPP-eligible women were identified in the CHBR, of whom 1,269 (23%) were not reported to the program. Factors associated with underreporting include mother’s age ≥ 45 years (vs. ≤ 20 years) (P=0.012); living in a rural county with < 88.4 people per square mile (vs. a county with ≥ 2,000 people per square mile) (P<0.0001); and having been born in the U.S. (P<0.0001). 712 women were estimated to be missing from both data sources, resulting in an overall estimate of 7,813 HBV-infected pregnant women in California from 2013-2014.
CONCLUSIONS: The California PHBPP currently enrolls approximately 75% of HBV-infected pregnant women, short of the goal of enrolling all eligible women. However, data were missing for pregnancies resulting in miscarriages and fetal deaths, as these pregnancies are not issued birth certificates. Therefore, additional program-eligible women might be missing from these analyses, suggesting that the underreporting rate may be an underestimate. Further research is needed to improve the detection and reporting of PHBPP-eligible women.