Comparative Effectiveness of Treatment Options for Genital Herpes Infection to Reduce Adverse Pregnancy Outcomes

Overview

Preterm delivery (PTD), together with low birthweight (LBW), is the leading cause of infant death and illness, affecting 500,000 births with annual medical costs of more than $26 billion in the U.S. each year. Identifying changeable risk factors to reduce PTD is considered a top research priority. Recent research has shown genital herpes infection (HSV) is associated with increased risks of PTD and LBW. More importantly, treating this infection, including infection with no symptoms, using readily available antiviral medications can be effective in removing the risk due to HSV. Thus, early identification and treatment of HSV in pregnant women could be an effective way to prevent PTD and LBW. Currently, many pregnant women with HSV infection, especially those with no symptoms, choose not to treat due to (a) a lack of demonstrated benefit of treatment and (b) general hesitance to use medications during pregnancy due to safety concerns for the fetus. Thus, emerging evidence of an increased risk of PTD/LBW associated with HSV infection, if untreated, and treatment effectiveness by anti-herpes medications has significantly changed current treatment paradigms among pregnant women. This evidence also provides new hope that effectively treating HSV infection among pregnant women, especially before the 3rd trimester, could lead to a new method to reduce PTD and LBW and reduce racial/ethnic disparities in these risks due to high rates of the infection in minority groups. To further examine the effectiveness of treating HSV in pregnant women to reduce adverse pregnancy outcomes, the investigators propose to conduct a prospective cohort study with a two-stage design combining the large pregnant women population (N=90,000) in Stage I identified through Kaiser Permanente Northern California (KPNC) electronic medical records (EMRs), with a Stage II sample to collect detailed information on additional factors that might muddle our understanding of this issue. This study will address the following: (1) Does treating HSV infection in pregnant women reduce the risk of PTD or LBW? (2) Does timing of the treatment during pregnancy influence treatment effectiveness? (3) Do other factors influence treatment effectiveness? and (4) Does HSV infection in pregnancy, if untreated, increase the risk of PTD and LBW, compared to no infection? Answers to these questions will be valuable to pregnant women and clinicians, and directly address their concerns when making treatment decisions

Preterm delivery (PTD), along with low birthweight (LBW), is the leading cause of perinatal mortality and morbidity. In the U.S., 12% of livebirths are PTDs, resulting in more than $26 billion in medical costs annually. The impact on infant health and staggering costs makes PTD one of the top research priorities of PCORI, AHRQ, the Institute of Medicine (IOM) and the World Health Organization (WHO), due to a lack of effective interventions to reduce PTD. Genital herpes infection is prevalent, with a recent WHO estimation of 500 million people worldwide infected. Treating pregnant women with genital herpes infection, especially before the 3rd trimester, has been shown to reduce the risk of PTD and LBW, thus it can be an effective intervention to reduce PTD/LBW. However, the effectiveness and benefit of treating genital herpes to reduce PTD and LBW needs to be further demonstrated in order to be incorporated into the treatment decision making process. Currently, many pregnant women choose not to treat genital herpes due to a general aversion to taking medications during pregnancy for the safety of their fetuses, and a lack of demonstrated evidence of benefits. Paradoxically, the choice of no treatment for genital herpes may adversely impact fetal health, leading to PTD and LBW. Given that pregnant women frequently prefer no treatment, studies are urgently needed to establish the risk-benefit profile between treatment and no treatment for genital herpes infection in the context of improving fetal health, including the timing of treatment (before the 3rd trimester). This study is designed to provide clear evidence of treatment effectiveness in real-world clinical practice, and risk-benefit profiles to inform both treatment decisions by pregnant women and clinicians. Study Aims: This proposed comparative effectiveness study will address the following questions: 1. Does treating genital herpes infection in pregnant women reduce the risk of adverse pregnancy outcomes including PTD or LBW? (treated vs. untreated) 2. Does the timing of the treatment during pregnancy influence the treatment effectiveness on reducing adverse pregnancy outcomes (PTD and LBW)? (head-to-head comparison of treatment timing: before the 3rd trimester vs. during the 3rd trimester). 3. Do other treatment metrics, including treatment duration, dosage, and compliance, impact treatment effectiveness in reducing the risk of PTD and LBW? 4. Does treatment effectiveness vary depending on the type (or severity) of underlying genital herpes infection? (e.g., treating symptomatic genital herpes infection vs. treating latent/asymptomatic genital herpes) 5. Does genital herpes infection in pregnancy, if untreated, increase the risk of PTD and LBW, compared to no genital herpes infection? (untreated vs. controls without genital herpes) In addition, this study is especially relevant in addressing racial disparities, given that minority pregnant women have higher rates of both genital herpes infection and PTD: 3 times the infection rate and 150% higher PTD rate among African-Americans compared to Whites. Thus, demonstrating the effectiveness of treating genital herpes in reducing PTD could lead to a reduction in the existing racial disparity in PTD rates. Study Description Overall study design: The investigators will conduct a prospective cohort study with a two-stage design based on more than 90,000 pregnant KPNC members in real-world clinical practice. Due to the increased fetal risk of untreated genital herpes infection, randomizing pregnant women with the infection into treated and untreated groups presents ethical problems, thus is not feasible. Our innovative two-stage prospective cohort design, leveraging our large membership and comprehensive electronic medical record (EMR) data, is a robust alternative option for examining the comparative effectiveness of treating genital herpes infection in pregnant women to reduce PTD and LBW. Comparators: Three comparisons will be made: 1. When assessing treatment effectiveness, women with the infection who choose not to receive treatment will serve as the comparator (untreated). This comparator is a frequently preferred treatment option chosen by pregnant women due to their reluctance to use medications during pregnancy, based on their predominant concerns for the safety of their developing fetus as well as a lack of evidence that treating genital herpes infection is beneficial to their fetus. This comparator will also make the comparison groups more comparable by controlling for confounding by indication. 2. When assessing the timing of treatment effectiveness (before vs. after the start of the 3rd trimester), those who receive treatment during the 3rd trimester will be used as the comparator. Using this comparator will allow a head-to-head comparison between the timing of the treatment. 3. When assessing the effect of choosing not to treat during pregnancy, women without an underlying genital herpes infection or receipt of any treatment will serve as the comparator (normal controls). This comparison will provide evidence of the increased risk of PTD and LBW if genital herpes infection is not treated during pregnancy. Our comparators will allow us to control for confounding by indication (genital herpes, its type and severity). Our EMR contains extensive questions on risk factors, including lifestyle factors, for all 90,000 mother-infant dyads. Through the unique two-stage study design, investigators will collect additional information, through interviews, on a subsample of women that will further allow controlling for additional confounders. Multiple statistical methodologies, in accordance with PCORI's methodology standards, will be employed in the analytic plan (e.g., propensity scores, instrumental variable methods) to ensure compatibility between comparison groups.