PharmacoEconomics (2014) 32:467–478 DOI 10.1007/s40273-014-0133-2
ORIGINAL RESEARCH ARTICLE
Cost and Cost Effectiveness of Vaginal Progesterone Gel in Reducing Preterm Birth: An Economic Analysis of the PREGNANT Trial Laura T. Pizzi • Neil S. Seligman • Jason K. Baxter Eric Jutkowitz • Vincenzo Berghella
•
Published online: 9 April 2014 Ó Springer International Publishing Switzerland 2014
Abstract Background Preterm birth (PTB) is a costly public health problem in the USA. The PREGNANT trial tested the efficacy of vaginal progesterone (VP) 8 % gel in reducing the likelihood of PTB among women with a short cervix. Objective We calculated the costs and cost effectiveness of VP gel versus placebo using decision analytic models informed by PREGNANT patient-level data. Methods PREGNANT enrolled 459 pregnant women with a cervical length of 10–20 mm and randomized them to either VP 8 % gel or placebo. We used a cost model to estimate the total cost of treatment per mother and a cost-effectiveness model to estimate the cost per PTB averted with VP gel versus L. T. Pizzi Thomas Jefferson University, 901 Walnut Street, Suite 911, Philadelphia, PA 19107, USA L. T. Pizzi (&) Department of Pharmacy Practice, Jefferson School of Pharmacy, 901 Walnut Street, Suite 901, Philadelphia, PA 19107, USA e-mail: [email protected] N. S. Seligman Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Rochester Medical Center, School of Medicine and Dentistry, 601 Elmwood Ave, Box 668, Rochester, NY 14642, USA J. K. Baxter V. Berghella Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Jefferson Medical College of Thomas Jefferson University, 834 Chestnut Street, Suite 400, Philadelphia, PA 19107, USA E. Jutkowitz Division of Health Policy and Management, University of Minnesota, 420 Delaware Street SE, MMC 729, Minneapolis, MN 55455, USA
placebo. Patient-level trial data informed model inputs and included PTB rates in low- and high-risk women in each study group at\28 weeks gestation, 28–31, 32–36, and C37 weeks. Cost assumptions were based on 2010 US healthcare services reimbursements. The cost model was validated against patient-level data. Sensitivity analyses were used to test the robustness of the cost-effectiveness model. Results The estimated cost per mother was $US23,079 for VP gel and $US36,436 for placebo. The cost-effectiveness model showed savings of $US24,071 per PTB averted with VP gel. VP gel realized cost savings and cost effectiveness in 79 % of simulations. Conclusion Based on findings from PREGNANT, VP gel was associated with cost savings and cost effectiveness compared with placebo. Future trials designed to include cost metrics are needed to better understand the value of VP. Key Points for Decision Makers This is the first analysis that employs clinical trial data to model the cost and cost effectiveness of vaginal progesterone when used to reduce preterm birth. In the base-case scenario, vaginal progesterone gel was cost saving and cost effective based on the public health definition of preterm birth (\37 weeks gestation). Vaginal progesterone gel treatment does not appear to benefit all women. The probabilistic sensitivity analysis suggests that the treatment is not cost effective in approximately 20 % of simulated cases. Treatment with vaginal progesterone gel does not clinically benefit high-risk women based on our model’s definition of high risk (i.e., having a history of preterm birth).
468
1 Introduction Preterm birth (PTB), defined as births before 37 weeks gestation [1], accounts for 75 % of overall neonatal mortality (death within the first 28 days of life) [2]. The rate of PTB in the USA peaked in 2006 at 12.8 %, remaining higher than that of other developed nations [3]. Among surviving preterm babies, short- and long-term morbidity is common. Short-term consequences of prematurity include respiratory distress syndrome, bronchopulmonary dysplasia, retinopathy of prematurity, sepsis, intraventricular hemorrhage, and necrotizing enterocolitis. Long-term morbidity includes neurologic impairments in children, higher rates of rehospitalization during the first 2 years of life, asthma or reactive airway disease, cerebral palsy, mental retardation, and behavioral disorders [4]. It comes as no surprise that the high morbidity associated with PTB necessitates significant healthcare utilization. PTB is a costly public health issue in the USA, resulting in annual societal costs of $US26 billion ($US51,600 per preterm infant) [5]. The cost of initial hospitalization can exceed $US200,000 for neonates born \28 weeks, compared with the cost of term birth, which can be \$US2,000 [6]. However, the long-term costs of chronic medical conditions (e.g., respiratory disease) and educational (e.g., early intervention services) impairments of children born preterm, as well as long-term costs due to lost productivity of the parents and intangible effects (e.g., stress and anxiety due to caregiver burdens), are not as well defined. It is well established that prior PTB and short cervical length (B25 mm) are the most significant risks for delivering preterm [7–12]. Transvaginal ultrasound (TVU) is the standard diagnostic test by which cervical length is assessed. Identification of a short cervix affords an opportunity to treat with a progestogen or cervical cerclage, depending on obstetrical history. The Society for Maternal Fetal Medicine (SMFM) and the American College of Obstetricians and Gynecologists (ACOG) recently published evidence-based clinical practice guidelines [1, 13]. The recommendation for women with singleton pregnancies, no prior history of PTB, and a short cervical length (B20 mm at 24 weeks gestation) is treatment with vaginal progesterone (VP) dosed as either the 90-mg gel or a 200-mg vaginal suppository [13]. The recommendation for women with a history of PTB is treatment with progesterone (commonly given as 17-alphahydroxy progesterone caproate [17P]) starting at 16–20 weeks (SMFM recommendation; ACOG states 16–24 weeks), weekly for 36 weeks [1, 13]. Women with a cervical length B25 mm before 24 weeks should receive cervical cerclage in addition to treatment with 17P [13]. A physician’s physical examination may warrant cerclage irrespective of obstetric history or progestogen use.
L. T. Pizzi et al.
The available and emerging evidence on progestogens has led to controversy regarding whether all women with a singleton gestation (‘universal screening’), or only those with particular risk factors for PTB, should be screened for a short cervix via TVU. Supporters of universal screening argue that TVU is a safe, well tolerated, and effective means of identifying women at increased risk of PTB, and there are evidence-based interventions for these women. Opponents of universal screening argue that it is inefficient to screen women without a prior PTB to detect the few that would be eligible for treatment (the number needed to screen to prevent one PTB is estimated to be between 387 and 604) [13]. Other reasons for opposing universal screening include (i) existing trials have not randomized to TVU universal screening versus no such screening (while the efficacy of VP has been assessed, the efficacy of screening has not); (ii) the ‘real world’ performance of TVU screening is unknown; (iii) assessment of cervical length by TVU requires expertise; and (iv) TVU may not be available in some geographic areas [13]. SMFM and ACOG have recently stated that there is currently insufficient evidence to support universal screening; however, it is reasonable for physicians to determine when it is necessary and to expect screening to be reimbursed by thirdparty payers [1, 13]. VP gel, the recommended treatment for women with a short cervix without a history of PTB, has been tested in various formulations. In 2003, Fonseca et al. [14] found that daily self-administered VP 100-mg (given as a compounded formulation) reduced the incidence of PTB \37 weeks of gestation by 14.7 % in Brazilian women with high-risk singleton gestations, compared with the placebo group. In a later study, Fonseca et al. [15] performed a multicenter, international study of women carrying singletons or twins, identified those with a short cervix, and found that a 200-mg VP suppository given each night decreased the incidence of PTB \34 weeks to 19.2 %, compared with 34.4 % in the placebo group. More recently, a daily, self-administered vaginal gel preparation was tested in a multi-country trial by Hassan et al. [16] and demonstrated a 45 % reduction in PTB \33 weeks versus placebo. A new drug application (NDA) for the progesterone gel formulation tested in PREGNANT (ProchieveÒ) was submitted to the US FDA in April 2011. Although efficacy findings from PREGNANT demonstrated a statistically significant reduction in PTB in the study sample as a whole [16], the FDA expressed concerns about the lack of statistical significance in PTB in the subsample of US women and declined approval [17]. This paper reports findings of a cost-effectiveness analysis of PREGNANT. Though the VP product included in the study was declined US approval, this is the first known
Economic Analysis of PREGNANT
economic analysis of a progestogen in PTB based on a decision analysis of primary patient-level trial data. Given the public health importance of PTB and the need for more cost evidence regarding progestogen use, we believe that our analysis will be valuable to clinicians and other healthcare decision makers involved in delivering care to pregnant women. Our primary objective was to determine the total cost per mother treated with VP gel in PREGNANT. We also reported the cost effectiveness of VP gel in terms of cost per PTB averted. In addition, we determined the PTB threshold at which VP gel costs would equal costs of those in the placebo group (i.e., the point at which a decision would be cost neutral).
2 Methods The perspective taken in our economic analysis is that of a US healthcare payer. Our time horizon initiated upon a diagnosis of short cervix and ended upon the neonate’s discharge from the hospital or death. The project was approved by the investigators’ Institutional Review Board and initiated upon completion of PREGNANT (but before the FDA decision was rendered). The PREGNANT study was a randomized, multicenter, clinical trial that investigated the safety and effectiveness of using a VP gel to decrease the incidence of PTB among women with a short cervix, with or without a history of PTB [16]. Inclusion criteria were singleton pregnancy, a maternal cervical length between 10 and 20 mm as measured by TVU performed between 19 ? 0 and 23 ? 6 weeks, and the absence of preterm labor symptoms. Exclusion criteria for trial participation included planned cerclage, cervical dilation, progesterone allergy, treatment with any progestogen (either current or within the previous 4 weeks), inability to fully participate in the study due to chronic medical conditions, vaginal bleeding, or other fetal complications. The PREGNANT intent-to-treat (ITT) population consisted of 459 women, representing 44 sites in ten countries. Of this population, 235 women were randomized to receive ProchieveÒ 8 % VP gel, and 224 were randomized to receive placebo. Treatment began between 20 and 23 ? 6 weeks and continued until one of the following occurred: 36 ? 6 gestational weeks, rupture of membranes, or delivery. The primary clinical measure for the trial was prevention of PTB before 33 weeks. 2.1 Cost Model Our primary objective was to develop a decision analytic model to estimate the total costs of administering 8 % VP gel to pregnant women with a short cervix (10–20 mm),
469
using data from the PREGNANT trial. The decision analysis incorporated both efficacy and safety findings from the trial. The framework of the model was constructed using TreeAge Pro 2011 (TreeAge, Williamstown, MA, USA), and is presented in Fig. 1. The tree depicts the clinical course of women, from enrollment in PREGNANT through delivery of the neonate. The branches of the tree illustrate the two study arms: VP 8 % and placebo. Each arm was further branched to account for a maternal risk of PTB, defined based on a history of PTB: women with prior PTB were defined as ‘high risk’ (n = 38/235 [16.2 %] in the VP gel arm; n = 33/224 [14.7 %] in the placebo arm); women without prior PTB were defined as ‘low risk’ (n = 137/235 [83.8 %] in the VP gel arm; n = 191/224 [85.3 %] in the placebo arm). The output of this tree is the total cost per case (maternal ? neonatal costs) in the PREGNANT trial for those who received VP gel or placebo. 2.2 Cost-Effectiveness Model We then fit the cost model into a cost-effectiveness format to report results in terms of cost per PTB averted. The clinical and cost inputs of this model were identical to those in the primary model but were expressed as cost per PTB averted in PREGNANT. The purpose of this secondary model was to further inform healthcare payers by expressing our findings in terms of cost per meaningful outcome, as well as to enable a more robust sensitivity analyses. The main outcome measure for this model was the incremental cost-effectiveness ratio (ICER) calculated as: ICER ¼
Total costsVPgel cohort Total costsplacebo cohort # PTB averted
In this formula, PTB averted was calculated as the number of PTBs in the VP gel cohort minus the number of PTBs in the placebo cohort, thus representing the efficacy difference achieved by treating women with VP gel versus placebo. The PTBs averted with VP gel could also be viewed as the babies born full-term (FTBs) gained in this group, since, by our definition, births could only be preterm or term, with \37 weeks being the cutoff applied to define PTBs, and a PTB averted must be an FTB gained. 2.3 Model Inputs 2.3.1 Clinical Inputs Clinical inputs for the models were obtained by analyzing patient-level PREGNANT trial data. Our modeling sample was defined as all participants who were included in the ITT analyses (n = 459 total, of whom 235 received VP gel and 224 received placebo). The clinical inputs resulting
470
Variable Name p_HxPTB_Plac p_HxPTB_VP p_PTB28_PlacHR p_PTB28_PlacLR p_PTB28_VPHR p_PTB28_VPLR p_PTB31_PlacHR p_PTB31_PlacLR p_PTB31_VPHR p_PTB31_VPLR p_PTBPlacHR p_PTBPlacLR p_PTBVPHR p_PTBVPLR
L. T. Pizzi et al.
Description Probability of a history of preterm birth in the placebo arm Probability of a history of preterm birth in the VP gel arm Probability of a preterm birth
ORIGINAL RESEARCH ARTICLE
Cost and Cost Effectiveness of Vaginal Progesterone Gel in Reducing Preterm Birth: An Economic Analysis of the PREGNANT Trial Laura T. Pizzi • Neil S. Seligman • Jason K. Baxter Eric Jutkowitz • Vincenzo Berghella
•
Published online: 9 April 2014 Ó Springer International Publishing Switzerland 2014
Abstract Background Preterm birth (PTB) is a costly public health problem in the USA. The PREGNANT trial tested the efficacy of vaginal progesterone (VP) 8 % gel in reducing the likelihood of PTB among women with a short cervix. Objective We calculated the costs and cost effectiveness of VP gel versus placebo using decision analytic models informed by PREGNANT patient-level data. Methods PREGNANT enrolled 459 pregnant women with a cervical length of 10–20 mm and randomized them to either VP 8 % gel or placebo. We used a cost model to estimate the total cost of treatment per mother and a cost-effectiveness model to estimate the cost per PTB averted with VP gel versus L. T. Pizzi Thomas Jefferson University, 901 Walnut Street, Suite 911, Philadelphia, PA 19107, USA L. T. Pizzi (&) Department of Pharmacy Practice, Jefferson School of Pharmacy, 901 Walnut Street, Suite 901, Philadelphia, PA 19107, USA e-mail: [email protected] N. S. Seligman Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, University of Rochester Medical Center, School of Medicine and Dentistry, 601 Elmwood Ave, Box 668, Rochester, NY 14642, USA J. K. Baxter V. Berghella Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Jefferson Medical College of Thomas Jefferson University, 834 Chestnut Street, Suite 400, Philadelphia, PA 19107, USA E. Jutkowitz Division of Health Policy and Management, University of Minnesota, 420 Delaware Street SE, MMC 729, Minneapolis, MN 55455, USA
placebo. Patient-level trial data informed model inputs and included PTB rates in low- and high-risk women in each study group at\28 weeks gestation, 28–31, 32–36, and C37 weeks. Cost assumptions were based on 2010 US healthcare services reimbursements. The cost model was validated against patient-level data. Sensitivity analyses were used to test the robustness of the cost-effectiveness model. Results The estimated cost per mother was $US23,079 for VP gel and $US36,436 for placebo. The cost-effectiveness model showed savings of $US24,071 per PTB averted with VP gel. VP gel realized cost savings and cost effectiveness in 79 % of simulations. Conclusion Based on findings from PREGNANT, VP gel was associated with cost savings and cost effectiveness compared with placebo. Future trials designed to include cost metrics are needed to better understand the value of VP. Key Points for Decision Makers This is the first analysis that employs clinical trial data to model the cost and cost effectiveness of vaginal progesterone when used to reduce preterm birth. In the base-case scenario, vaginal progesterone gel was cost saving and cost effective based on the public health definition of preterm birth (\37 weeks gestation). Vaginal progesterone gel treatment does not appear to benefit all women. The probabilistic sensitivity analysis suggests that the treatment is not cost effective in approximately 20 % of simulated cases. Treatment with vaginal progesterone gel does not clinically benefit high-risk women based on our model’s definition of high risk (i.e., having a history of preterm birth).
468
1 Introduction Preterm birth (PTB), defined as births before 37 weeks gestation [1], accounts for 75 % of overall neonatal mortality (death within the first 28 days of life) [2]. The rate of PTB in the USA peaked in 2006 at 12.8 %, remaining higher than that of other developed nations [3]. Among surviving preterm babies, short- and long-term morbidity is common. Short-term consequences of prematurity include respiratory distress syndrome, bronchopulmonary dysplasia, retinopathy of prematurity, sepsis, intraventricular hemorrhage, and necrotizing enterocolitis. Long-term morbidity includes neurologic impairments in children, higher rates of rehospitalization during the first 2 years of life, asthma or reactive airway disease, cerebral palsy, mental retardation, and behavioral disorders [4]. It comes as no surprise that the high morbidity associated with PTB necessitates significant healthcare utilization. PTB is a costly public health issue in the USA, resulting in annual societal costs of $US26 billion ($US51,600 per preterm infant) [5]. The cost of initial hospitalization can exceed $US200,000 for neonates born \28 weeks, compared with the cost of term birth, which can be \$US2,000 [6]. However, the long-term costs of chronic medical conditions (e.g., respiratory disease) and educational (e.g., early intervention services) impairments of children born preterm, as well as long-term costs due to lost productivity of the parents and intangible effects (e.g., stress and anxiety due to caregiver burdens), are not as well defined. It is well established that prior PTB and short cervical length (B25 mm) are the most significant risks for delivering preterm [7–12]. Transvaginal ultrasound (TVU) is the standard diagnostic test by which cervical length is assessed. Identification of a short cervix affords an opportunity to treat with a progestogen or cervical cerclage, depending on obstetrical history. The Society for Maternal Fetal Medicine (SMFM) and the American College of Obstetricians and Gynecologists (ACOG) recently published evidence-based clinical practice guidelines [1, 13]. The recommendation for women with singleton pregnancies, no prior history of PTB, and a short cervical length (B20 mm at 24 weeks gestation) is treatment with vaginal progesterone (VP) dosed as either the 90-mg gel or a 200-mg vaginal suppository [13]. The recommendation for women with a history of PTB is treatment with progesterone (commonly given as 17-alphahydroxy progesterone caproate [17P]) starting at 16–20 weeks (SMFM recommendation; ACOG states 16–24 weeks), weekly for 36 weeks [1, 13]. Women with a cervical length B25 mm before 24 weeks should receive cervical cerclage in addition to treatment with 17P [13]. A physician’s physical examination may warrant cerclage irrespective of obstetric history or progestogen use.
L. T. Pizzi et al.
The available and emerging evidence on progestogens has led to controversy regarding whether all women with a singleton gestation (‘universal screening’), or only those with particular risk factors for PTB, should be screened for a short cervix via TVU. Supporters of universal screening argue that TVU is a safe, well tolerated, and effective means of identifying women at increased risk of PTB, and there are evidence-based interventions for these women. Opponents of universal screening argue that it is inefficient to screen women without a prior PTB to detect the few that would be eligible for treatment (the number needed to screen to prevent one PTB is estimated to be between 387 and 604) [13]. Other reasons for opposing universal screening include (i) existing trials have not randomized to TVU universal screening versus no such screening (while the efficacy of VP has been assessed, the efficacy of screening has not); (ii) the ‘real world’ performance of TVU screening is unknown; (iii) assessment of cervical length by TVU requires expertise; and (iv) TVU may not be available in some geographic areas [13]. SMFM and ACOG have recently stated that there is currently insufficient evidence to support universal screening; however, it is reasonable for physicians to determine when it is necessary and to expect screening to be reimbursed by thirdparty payers [1, 13]. VP gel, the recommended treatment for women with a short cervix without a history of PTB, has been tested in various formulations. In 2003, Fonseca et al. [14] found that daily self-administered VP 100-mg (given as a compounded formulation) reduced the incidence of PTB \37 weeks of gestation by 14.7 % in Brazilian women with high-risk singleton gestations, compared with the placebo group. In a later study, Fonseca et al. [15] performed a multicenter, international study of women carrying singletons or twins, identified those with a short cervix, and found that a 200-mg VP suppository given each night decreased the incidence of PTB \34 weeks to 19.2 %, compared with 34.4 % in the placebo group. More recently, a daily, self-administered vaginal gel preparation was tested in a multi-country trial by Hassan et al. [16] and demonstrated a 45 % reduction in PTB \33 weeks versus placebo. A new drug application (NDA) for the progesterone gel formulation tested in PREGNANT (ProchieveÒ) was submitted to the US FDA in April 2011. Although efficacy findings from PREGNANT demonstrated a statistically significant reduction in PTB in the study sample as a whole [16], the FDA expressed concerns about the lack of statistical significance in PTB in the subsample of US women and declined approval [17]. This paper reports findings of a cost-effectiveness analysis of PREGNANT. Though the VP product included in the study was declined US approval, this is the first known
Economic Analysis of PREGNANT
economic analysis of a progestogen in PTB based on a decision analysis of primary patient-level trial data. Given the public health importance of PTB and the need for more cost evidence regarding progestogen use, we believe that our analysis will be valuable to clinicians and other healthcare decision makers involved in delivering care to pregnant women. Our primary objective was to determine the total cost per mother treated with VP gel in PREGNANT. We also reported the cost effectiveness of VP gel in terms of cost per PTB averted. In addition, we determined the PTB threshold at which VP gel costs would equal costs of those in the placebo group (i.e., the point at which a decision would be cost neutral).
2 Methods The perspective taken in our economic analysis is that of a US healthcare payer. Our time horizon initiated upon a diagnosis of short cervix and ended upon the neonate’s discharge from the hospital or death. The project was approved by the investigators’ Institutional Review Board and initiated upon completion of PREGNANT (but before the FDA decision was rendered). The PREGNANT study was a randomized, multicenter, clinical trial that investigated the safety and effectiveness of using a VP gel to decrease the incidence of PTB among women with a short cervix, with or without a history of PTB [16]. Inclusion criteria were singleton pregnancy, a maternal cervical length between 10 and 20 mm as measured by TVU performed between 19 ? 0 and 23 ? 6 weeks, and the absence of preterm labor symptoms. Exclusion criteria for trial participation included planned cerclage, cervical dilation, progesterone allergy, treatment with any progestogen (either current or within the previous 4 weeks), inability to fully participate in the study due to chronic medical conditions, vaginal bleeding, or other fetal complications. The PREGNANT intent-to-treat (ITT) population consisted of 459 women, representing 44 sites in ten countries. Of this population, 235 women were randomized to receive ProchieveÒ 8 % VP gel, and 224 were randomized to receive placebo. Treatment began between 20 and 23 ? 6 weeks and continued until one of the following occurred: 36 ? 6 gestational weeks, rupture of membranes, or delivery. The primary clinical measure for the trial was prevention of PTB before 33 weeks. 2.1 Cost Model Our primary objective was to develop a decision analytic model to estimate the total costs of administering 8 % VP gel to pregnant women with a short cervix (10–20 mm),
469
using data from the PREGNANT trial. The decision analysis incorporated both efficacy and safety findings from the trial. The framework of the model was constructed using TreeAge Pro 2011 (TreeAge, Williamstown, MA, USA), and is presented in Fig. 1. The tree depicts the clinical course of women, from enrollment in PREGNANT through delivery of the neonate. The branches of the tree illustrate the two study arms: VP 8 % and placebo. Each arm was further branched to account for a maternal risk of PTB, defined based on a history of PTB: women with prior PTB were defined as ‘high risk’ (n = 38/235 [16.2 %] in the VP gel arm; n = 33/224 [14.7 %] in the placebo arm); women without prior PTB were defined as ‘low risk’ (n = 137/235 [83.8 %] in the VP gel arm; n = 191/224 [85.3 %] in the placebo arm). The output of this tree is the total cost per case (maternal ? neonatal costs) in the PREGNANT trial for those who received VP gel or placebo. 2.2 Cost-Effectiveness Model We then fit the cost model into a cost-effectiveness format to report results in terms of cost per PTB averted. The clinical and cost inputs of this model were identical to those in the primary model but were expressed as cost per PTB averted in PREGNANT. The purpose of this secondary model was to further inform healthcare payers by expressing our findings in terms of cost per meaningful outcome, as well as to enable a more robust sensitivity analyses. The main outcome measure for this model was the incremental cost-effectiveness ratio (ICER) calculated as: ICER ¼
Total costsVPgel cohort Total costsplacebo cohort # PTB averted
In this formula, PTB averted was calculated as the number of PTBs in the VP gel cohort minus the number of PTBs in the placebo cohort, thus representing the efficacy difference achieved by treating women with VP gel versus placebo. The PTBs averted with VP gel could also be viewed as the babies born full-term (FTBs) gained in this group, since, by our definition, births could only be preterm or term, with \37 weeks being the cutoff applied to define PTBs, and a PTB averted must be an FTB gained. 2.3 Model Inputs 2.3.1 Clinical Inputs Clinical inputs for the models were obtained by analyzing patient-level PREGNANT trial data. Our modeling sample was defined as all participants who were included in the ITT analyses (n = 459 total, of whom 235 received VP gel and 224 received placebo). The clinical inputs resulting
470
Variable Name p_HxPTB_Plac p_HxPTB_VP p_PTB28_PlacHR p_PTB28_PlacLR p_PTB28_VPHR p_PTB28_VPLR p_PTB31_PlacHR p_PTB31_PlacLR p_PTB31_VPHR p_PTB31_VPLR p_PTBPlacHR p_PTBPlacLR p_PTBVPHR p_PTBVPLR
L. T. Pizzi et al.
Description Probability of a history of preterm birth in the placebo arm Probability of a history of preterm birth in the VP gel arm Probability of a preterm birth
