Oral misoprostol tablets (25 µg) for induction of labor: a targeted literature review and cost analysis

Abstract

Background

Various methods exist for the induction of labor (IOL), and there is limited consensus as to optimal methods. Off-label misoprostol is recommended by the World Health Organization (WHO) for IOL but preparing it into doses suitable for IOL lacks precision, with potential adverse outcomes if dosing is inaccurate. This study explores potential outcomes and costs associated with increased uptake of a low-dose (25 µg) oral misoprostol formulation (Angusta; Norgine BV, Amsterdam) approved for IOL, in France, Belgium, and the Netherlands.

Methods

A literature review was undertaken to derive probabilities of delivery outcomes (vaginal, instrumental, and cesarean sections) for IOL methods, from published meta-analyses. Outcomes for oral misoprostol tablets (25 µg) were unavailable in the meta-analyses, so were estimated using data from two published retrospective cohort studies. A model was developed to predict the frequency of IOL outcomes and associated costs at the national level, across multiple scenarios. Scenarios were tested using a moderate, medium, and high increase in oral misoprostol tablet (25 µg) uptake. Market shares, costs, and induction rates were defined for each country using multiple data sources.

Results

Increased uptake of oral misoprostol tablets (25 µg) was estimated to be associated with a slightly increased rate of routine vaginal deliveries, and concurrent decreases in instrumental vaginal deliveries and cesarean sections. Since routine vaginal deliveries are less costly than other delivery outcomes, increased uptake of oral misoprostol tablets (25 µg) within the IOL market has the potential to be cost-saving. These trends were predicted using 25 µg oral misoprostol tablet outcomes informed by both retrospective studies.

Conclusion

Preliminary outcomes suggest that oral misoprostol tablets at 25 µg per dose may improve outcomes in IOL and be cost-saving. Further study is required to validate these findings and assess the comparative efficacy of IOL methods, including oral misoprostol tablets (25 µg).

Keywords:

• Induction of labor
• misoprostol
• oral misoprostol
• obstetrics
• costs

JEL CLASSIFICATION CODES:

• I00
• I
• A10
• A1
• A

Introduction

Induction of labor (IOL) is recommended where there is a clear medical indication and the expected benefits outweigh its potential harms¹. The benefits of IOL (where clinically indicated) in pregnancies at or beyond term include improvement of neonatal and maternal outcomes and reducing the risk of cesarean sections². Globally, rates of IOL have increased over time. Across European countries, induction rates differ. In France, 22.0% of all births involve IOL, whereas in Belgium the proportion is 28.5%, and in the Netherlands, 21.3%^3–10.

Guidelines on IOL exist internationally, nationally, and at the local hospital level^1,11–15. A recent systematic review of guidelines found consistency in guidance for inducing labor for prolonged pregnancy and premature rupture of membranes, but inconsistencies in the timing of IOL for other indications¹⁶. There are also large disparities in IOL rates between hospitals, based on local guidelines and practices.

Multiple methods exist for inducing labor, including pharmacological, mechanical, and alternative interventions. Options recommended by the World Health Organization (WHO) are low dose oral or vaginal misoprostol, low dose vaginal prostaglandins, and balloon catheter; oxytocin is an alternative when prostaglandins (including misoprostol) are not available¹⁷. There is considerable variation between countries in choice of method, due to differing opinions nationally and locally^18–21.

Misoprostol is a synthetic prostaglandin E1 analogue²². It is sold under the brand name Cytotec (Pfizer Inc, Groton, CT) and other brand names, and is indicated for the treatment of gastric and duodenal ulcers²³. Misoprostol is among the options recommended by the WHO and has commonly been used off-label for IOL; however, off-label use is associated with various concerns. Cytotec is formulated as 200 µg tablets, but the dosing required for IOL is either 25 or 50 µg per dose. There is no single protocol for the preparation of the correct dose, and practitioners use differing preparation methods and dosing regimens²⁴. Cutting up the tablet is difficult to do precisely and may lead to dosing inaccuracy^25,26. Further, compounded Cytotec deteriorates rapidly and therefore has a short shelf life²⁷. Correct misoprostol dosing is vital to ensure efficacy and safety and minimize adverse events²⁸. Too high a dose can increase the risk of uterine hyperstimulation²⁹, too low a dose can lead to ineffective induction³⁰. Because of the risk of negative outcomes associated with off-label use, Cytotec has been withdrawn for all indications in France due to safety concerns^23,31, and is subject to warnings and restrictions in other countries^32,33.

Oral misoprostol tablets (25 µg) (Angusta, Norgine BV, Amsterdam) are designed specifically for IOL, providing a fixed ready-to-use dose of misoprostol requiring no additional preparation time in the pharmacy^28,34. They are the only oral formulation of misoprostol approved for IOL and were developed following requests from the Danish and French authorities for a controlled low dose form of misoprostol. Evidence from real-world studies shows favorable obstetric outcomes for oral misoprostol tablets (25 µg) in various IOL protocols^30,34–36.

Comparing different IOL methods in terms of both clinical outcomes and costs is important for clinical and payer decision-making. However, there are significant gaps in clinical evidence³⁷, and considerable uncertainty has been found when ranking methods in terms of clinical outcomes³⁸. Moreover, limited trial sizes have rendered analyses of adverse outcomes, such as uterine rupture imprecise³⁷. The aim of IOL is to achieve vaginal delivery, without the use of instruments or cesarean section. Instrumental vaginal delivery and cesarean section are suboptimal outcomes for IOL and are also associated with greater costs than routine vaginal delivery^39,40. Therefore, interventions that improve these outcomes also have the potential to save costs.

The objective of this study is to analyze the outcomes and associated costs for the different IOL methods available in three countries (the Netherlands, Belgium, France), and to explore the effects of increased use of oral misoprostol tablets (25 µg) at a population level, assuming displacement of various other interventions (based on country-specific market shares). This is performed by a cost-calculator model using a national healthcare insurance perspective, with clinical outcomes informed by a targeted literature review (TLR). These three countries were selected based on the broad similarity and comparability of their healthcare systems and associated costs.

Methods

Targeted literature review

A TLR was conducted to assess the efficacy of IOL interventions and to inform the cost calculator model. Searches of databases (PubMed, Embase) were conducted to identify systematic literature reviews (SLRs) comparing mechanical or pharmacological interventions for IOL. The Cochrane Pregnancy and Childbirth Group website was also searched⁴¹. Findings were grouped by study type (network meta-analysis [NMA] or meta-analysis [MA]) and by the number of comparisons in MA studies (>2 or ≤2).

Eligibility criteria were applied following the PICOS framework (population, intervention, comparison, outcome, and study design), in line with Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols (PRISMA-P) guidance⁴². In brief, only SLR studies written in English and published from 2015 to June 2020 (the date of the search) were included; the date limit was applied to obtain recent information relevant to current practice. Full details of the PICOS criteria, search strategy, and PRISMA diagram are given in the Supplementary Material (Tables S1–S4 and Figure S1). The population of interest was women with a viable fetus undergoing third trimester IOL, and interventions of interest were pharmacological or mechanical interventions for IOL (i.e. not alternative interventions, such as acupuncture), and not methods for augmentation of labor, such as oxytocin. Ideally, data would have been limited to women with an unripe cervix, to ensure that only IOL, rather than labor augmentation, was evaluated. However, the evidence available specifically for this subpopulation is limited and therefore the population was not restricted based on cervical ripeness. An overall population of women undergoing IOL was evaluated; no specific subpopulations were assessed and studies assessing a specific subpopulation (e.g. nulliparous women only, or women with a previous cesarean section) were excluded due to the differences in risk and outcomes in these specific subpopulations.

Review outputs and calculation of probabilities

A total of 369 studies were identified of which 33 met the inclusion criteria. Three were NMA and the remainder were MA. Of the MA, five evaluated two or more interventions (Supplementary Material, Figure S1). None of the identified studies evaluated oral misoprostol tablets (25 µg). Three studies have been published which describe delivery outcomes with oral misoprostol tablets (25 µg) (Table 1), however, only two are of interest^35,36. The third evaluated outcomes in nulliparous women only so did not meet the inclusion criteria³⁰. One of the studies evaluated two treatment regimens, one of which did not reflect the recommended dosing in the Summary of Product Characteristics (SmPC),³⁴ and was therefore not incorporated³⁵. The sensitivity of the model to outcome probabilities reported within the two included studies for oral misoprostol tablets (25 µg) (Bendix et al.³⁵ and Helmig et al.³⁶) was assessed in scenario analyses.

Table 1. Population and dosing in oral misoprostol tablets (25 µg) studies.

Study	Population	N	Oral misoprostol tablets dosing and regimen studied	Incorporated
Helmig et al.^36	Nulliparous or multiparous women, singleton pregnancies, at ≥37 weeks gestation.	978	25 µg every 2 h, up to 6 doses on day one, 8 doses day two, 3 doses day three.	Yes
Eriksson et al.^30	Nulliparous women, singleton pregnancies, at ≥37 weeks gestation	193	25 µg every 2 h	No (nulliparous subpopulation only)
Bendix et al.^35	Nulliparous or multiparous women, singleton or multiple pregnancies, predominantly at ≥37 weeks gestation	418	25 µg every 2 h	Yes
		398	50 µg every 6 h	No (dosing regimen did not reflect SmPC^34)

Absolute probability values for each outcome and intervention of interest were calculated (Table 2). These data were assessed for clinical plausibility by expert clinicians (RH, FP, and RD) and were deemed plausible.

Table 2. Absolute probability values for each intervention and outcomes.

Study type	Study details	Specific intervention	Probability of outcome
			Any vaginal delivery	Caesarean section	Instrumental vaginal delivery
Observational study**	Oral misoprostol tablets (25 µg × 2 h), Bendix et al.^35	Oral misoprostol tablets (25 µg)	84.0%^Δ	16.0%^Δ	10.0%^Δ
	Oral misoprostol tablets (25 µg × 2 h), Helmig et al.^36	Oral misoprostol tablets (25 µg)	85.1%^Δ	14.9%^Δ	9.6%^Δ
TLR findings (NMA and MA studies)	Double-balloon or Cook's catheter	Cook balloon—vaginal	77.2%*	22.8%*	17.1%^†
	Oral misoprostol tablet (<50 µg if described)	Cytotec off-label—oral (<50 µg)	77.2%*	22.8%*	9.7%^†
	Oral misoprostol tablet (≥50 µg if described)	Cytotec off-label—oral (≥50 µg)	83.9%*	16.1%*	9.3%^§
	Vaginal misoprostol (<50 µg if described)	Cytotec off-label—vaginal (<50 µg)	84.3%*	15.7%*	16.1%^†
	Vaginal misoprostol (≥50 µg if described)	Cytotec off-label—vaginal (≥50 µg)	83.7%*	16.3%*	21.1%^§
	Foley catheter	Foley catheter—vaginal	83.2%*	16.8%*	14.8%^†
	Intracervical PGE2	Prepidil—endocervical gel	81.9%*	18.1%*	10.8%^†
	Vaginal PGE2 Pessary (normal release)	PROPESS/CERVIDIL—vaginal	82.1%*	17.9%*	11.7%^†
	Vaginal PGE2 Gel	Prostin E2—vaginal gel	78.3%*	21.7%*	11.7%^†
	Vaginal PGE2 Tablet	Prostin E2—vaginal tablet	82.6%*	17.4%*	11.7%^†

Abbreviations. PGE2, prostaglandin E2.

Doses (where reported) describe the maximum single dose; other assumptions are detailed within the text.

^ΔAbsolute probabilities sourced from Bendix et al.³⁵ and Helmig et al.³⁶.

*Absolute probabilities calculated from Alfirevic et al.²⁹, comparator (placebo) probability from Alfirevic et al.⁴⁸.

^†Absolute probabilities reported by de Vaan et al.⁴⁹.

^§Absolute probabilities reported by Vogel et al.⁴⁵.

**As data for oral misoprostol tablets (25 µg) is from observational studies, and data for other interventions is from NMA/MA findings, only naïve comparisons between oral misoprostol tablets (25 µg) and other interventions are possible.

Comparators were selectively drawn from the TLR outputs based on the interventions used within each of the scope countries. Outcomes of interest were cesarean section rates (from which total vaginal delivery rates could be calculated), and instrumental vaginal delivery rates, as these were the only outcomes routinely published across all oral misoprostol tablets (25 µg) and comparator studies.

Ideally, a single study that included all interventions and associated probabilities for the outcomes of interest would have been used to inform the cost calculator inputs, but none of the identified studies fulfilled these criteria. To minimize variability in values due to different study methods, the number of studies used was kept to a minimum. In total, four were selected based on the following order of priorities (Supplementary Material, Table S5):

1. Absolute probabilities directly reported in NMA studies

2. Odds ratios reported in NMA studies, where absolute probability could be calculated

3. Absolute probabilities directly sourced from the minimum number of MA studies to fulfill all the interventions

To capture the absolute probability of outcomes, probabilities had to be calculated from odds ratios vs. placebo. Given that: $$\mathit{Odds\; ratio}=\mathrm{ }\frac{\mathit{Odds}\left(\mathit{Intervention}\right)}{\mathit{Odds}\left(\mathit{Comparator}\right)}$$ and: $$\mathit{Odds}\left(\mathit{Intervention}\right)=\frac{P\left(\mathit{Intervention}\right)}{1-P\left(\mathit{Intervention}\right)};$$

The probability of an outcome with the intervention can be calculated as: $$P\left(\mathit{Intervention}\right)=\frac{\left(\mathit{Odds\; ratio\; x\; Odds}\left(\mathit{Comparator}\right)\right)}{1+\mathrm{ }\left(\mathit{Odds\; ratio\; x\; Odds}\right(\mathit{Comparator}\left)\right)}$$

Where P( ) is the probability with either intervention or comparator. Therefore, knowing the probability of an outcome with the comparator, the probability of an outcome with the intervention of interest may be calculated using: $$P\left(\mathit{Intervention}\right)=\frac{\left(\mathit{Odds\; ratio\; x}\frac{P\left(\mathit{Comparator}\right)}{1-P\left(\mathit{Comparator}\right)}\right)}{1+\mathrm{ }\left(\mathit{Odds\; ratio\; x}\frac{P\left(\mathit{Comparator}\right)}{1-P\left(\mathit{Comparator}\right)}\right)}$$

To capture the absolute probability of outcomes, from risk ratios, probabilities were calculated using the following equation: $$\mathit{Risk\; ratio}=\mathrm{ }\frac{P\left(\mathit{Intervention}\right)}{P\left(\mathit{Comparator}\right)}$$

Where P() is the probability with either intervention or comparator

Cost calculator model

A cost calculator (Figure 1) was developed in Microsoft Excel to calculate the costs and number of events associated with IOL for each country. This used a one-year time horizon without discounting rates (due to 1-year horizon). The cost calculator evaluated core obstetric outcomes (vaginal delivery and cesarean section rates) for various interventions, and market shares for these interventions. The model was developed from a national healthcare insurance perspective, focusing on diagnosis-related group (DRG) costs for the outcomes of interest. DRG costs are fixed fees that encompass all the costs in the relevant episode, regardless of the length of stay and other individual variations in a patient’s care. The cost calculator was quality checked before analysis of results (calculations, and face validity of outputs).

Figure 1. Conceptual illustration of the cost calculator model.

Model inputs

Inputs into the model comprised absolute probabilities of outcomes for each IOL intervention, and country-specific data on the number of interventions, unit costs for outcomes, and market shares for each intervention (Table 3) Drug costs are inherently included within these unit costs for France and the Netherlands (based on their payment system). In Belgium these are not included as they are paid for by the individual undergoing IOL (and therefore are not explicitly included given the perspective used), except for a small proportion of the cost of Prostin E2 (Pfizer Ltd.); this cost was considered negligible and was not included. Assumptions were made on country-specific details, for example on the route of Prostin E2 and Cytotec delivery, and exclusion of oxytocin (based on the assumption that studies evaluated induction rather than augmentation of labor). Model inputs and assumptions were discussed with expert clinicians (RH, FP, and RD) and were deemed clinically plausible.

Table 3. Summary of country-specific inputs.

		Belgium	France	Netherlands
Unit costs	Standard vaginal delivery	€2,836.00	€2,564.00	€2,360.00
	Instrumental vaginal delivery	€3,282.00	€2,564.00	€2,885.00
	Caesarean section	€4,245.00	€4,352.00	€4,440.00
	Source	tct.fgov.be; national data bank^39	ScanSante^50	www.opendisdata.nl/, Dutch healthcare authority data^40
Number of annual inductions	Total births	116,681	740,000	172,273
	Proportion inductions	28.5%	22.0%	21.3%
	Total eligible inductions	21,100	100,773	23,349
	Source	Statbel 2019^6	INSEE^4	Perined 2018^5
		Perinatale activiteiten in Vlanderen 2019^8	Enquête périnatale, 2016^10
		Santé périnatale en Wallonie 2019^7
		Santé périnatale Région bruxelloise 2019^9
Market shares of relevant methods for induction of labor	Misoprostol (25 µg)—oral tablet	0.0%	29.0%	0.0%
	Prostin E2—vaginal tablet	62.0%*	0.0%	0.0%
	Prostin E2—vaginal gel	0.0%	15.3%*	3.4%*
	Prepidil—endocervical gel	8.0%	0.0%	1.6%
	PROPESS—vaginal insert	12.0%	48.8%	0.5%
	Cytotec—oral or vaginal tablet	4.0%**	0.0%	10.9%**
	Foley balloon catheter	14.0%^+	3.5%^+	83.5%^+
	Cook balloon catheter	0.0%^+	3.5%^+	0.0%^+
	Source	Norgine Market research^51; IQVIA data MAT June 2019	Blanc-Petitjean et al.^52, clinical expert opinion	Perined 2018^5

All values in the table were verified by clinical experts. Market share values may not sum to 100% within the table due to rounding. Only methods for IOL and cervical ripening were incorporated, i.e. not oxytocin which is used for the augmentation of labor only.

⁺Market share data for mechanical methods is limited. For Belgium and the Netherlands, all mechanical methods were assumed to be Foley catheter, as no data is available for the Cook balloon catheter. For France, half were assumed to be Foley catheter, and half Cook catheter.

*Prostin was assumed to be equivalent to vaginal tablets for Belgium, vaginal gel for France and the Netherlands.

**Cytotec was assumed to be 100% vaginal for Belgium and 90% vaginal and 10% oral route for the Netherlands, with a 50:50 split between doses (<50 µg and ≥ 50 µg).

Scenario testing in cost calculator

Efficacy, market share, and cost data reflecting current practice at the time of writing were incorporated into the cost calculator. For each country, scenarios were evaluated using hypothetical moderate, medium, and high increases in oral misoprostol tablet (25 µg) market share compared to current values. For each country and set of market share scenarios, two different datasets from retrospective observational cohort studies were used to inform outcome probabilities with oral misoprostol tablets (25 µg) (Table 2). The model was run twice, once for the outcomes reported in each retrospective study.

Results

Market share scenarios

The market share of oral misoprostol tablets (25 µg) differs between countries. In Belgium and the Netherlands, oral misoprostol tablets (25 µg) had not been introduced at the time that market share data was obtained, so began from a base of zero. Under the moderate, medium, and high increases, its estimated market share reached 27.0, 40.7, and 60.0%, respectively in Belgium (Supplementary Material, Table S6) and 19.9, 34.9, and 57.7%, respectively in the Netherlands (Supplementary Material, Table S7). In France, from a baseline market share of 29.0%, the share rose to 36.1, 43.2, and 50.3% under the respective scenarios (Supplementary Material, Table S8). Other interventions were displaced accordingly.

Outcomes assessment

Based on data for oral misoprostol tablets (25 µg) informed by Bendix et al.³⁵, increased use of this formulation was predicted to result in a reduction in instrumental vaginal deliveries and cesarean sections in each country, with a corresponding increase in routine vaginal deliveries (Figure 2). The same trends were predicted based on the data from Helmig et al.³⁶, again driven by predicted reduced cesarean section and instrumental vaginal delivery rates with oral misoprostol tablets (25 µg) (Figure 3). This result was driven by the low cesarean section and instrumental vaginal delivery rate reported for oral misoprostol tablets (25 µg) in the original datasets. For Belgium and the Netherlands, there was a greater predicted reduction in the number of instrumental deliveries than cesarean sections, whereas in France the reverse was true (i.e. a greater reduction in cesarean sections).

Figure 2. Differences in annual costs and outcomes of induction of labor under scenarios of moderate, medium, and high uptake of oral misoprostol tablets (25 µg), based on data from Bendix et al.³⁵, 25 µg per dose (N = Belgium 21,100; France 110,773; Netherlands 23,349).

Figure 3. Differences in annual costs and outcomes of induction of labor under scenarios of moderate, medium and high uptake of oral misoprostol tablets (25 µg), based on data from Helmig et al.³⁶, 25 µg per dose (N = Belgium 21,100; France 110,773; Netherlands 23,349).

Impact on costs

Since both instrumental vaginal deliveries and cesarean sections have a greater unit cost than routine vaginal deliveries, this translated into predicted cost savings. In all three countries, increasing the use of oral misoprostol tablets (25 µg) was predicted to be associated with a net annual cost saving with both of the relevant datasets (Figure 4). As oral misoprostol tablets (25 µg) use increased, so did cost savings.

Figure 4. Estimated change in total annual cost, relative to current market shares, under scenarios of moderate, medium and high uptake of oral misoprostol tablets (25 µg). Negative values denote cost saving (N = Belgium 21,100; France 110,773; Netherlands 23,349).

Discussion

Our study describes potential clinical and cost advantages from increased use of oral misoprostol tablets (25 µg) for IOL in France, Belgium, and the Netherlands. The study focuses on core delivery outcomes for IOL, namely routine vaginal delivery, instrumental delivery, and cesarean section. Comparing outcomes from the Bendix et al.³⁵ and the Helmig et al.³⁶ observational studies of oral misoprostol tablets (25 µg) in real-world practice with evidence from published meta-analyses, our study suggests that an increased population-level uptake of oral misoprostol tablets at 25 µg per dose (dosing every 2 h) may be associated with improved outcomes compared with other IOL methods, namely slightly lower rates of instrumental vaginal delivery and cesarean section. This trend was predicted in all three countries. However, the differences estimated are small and their significance is uncertain, as the analysis was descriptive and generation of confidence intervals and significance testing were not undertaken. Clinical experience as the formulation is used more widely could be evaluated to confirm whether improvement in outcomes and the associated cost savings are realized in real-world practice.

Since instrumental vaginal delivery and cesarean section are more expensive than routine vaginal delivery^39,40, increased use of oral misoprostol tablets (25 µg) has the potential to be cost-saving if the estimated differences in outcomes were confirmed in practice. This was predicted in all three of the scope countries, based on the data available. In France, savings were predicted even though instrumental and standard vaginal delivery are assigned the same cost by the French healthcare system. The cost savings predicted may be conservative, as the study did not take into account additional cost savings associated with reducing primary cesarean section rates, such as the consequent reduction in repeat caesareans for subsequent pregnancies⁴³ and the cost of neonatal complications. Relatively small improvements in primary and repeat cesarean rates have been found to be associated with substantial reductions in hospital costs⁴⁴.

There is a paucity of data explicitly comparing outcomes between different IOL methods. Few head-to-head comparisons between interventions are available in the literature, and therefore alternative methods had to be used to estimate absolute probability values for outcomes with each intervention. To make the estimates as robust as possible given the available data, NMA studies were used in preference to MA studies, and the number of studies used was kept to the minimum necessary to obtain all the data. However, since oral misoprostol tablets (25 µg) have only been approved for use since 2017, it is not explicitly included in any of the identified NMA and MA studies that were used to evaluate efficacy and outcomes for the other methods. Therefore, a naïve comparison had to be undertaken between outcomes in the oral misoprostol tablet (25 µg) studies and calculated absolute probabilities from the literature. This is a limitation of the study since there is no adjustment to control for heterogeneity between populations across studies. A further limitation is that the analysis was descriptive only, meaning that the degree of uncertainty around the point estimates, or the statistical significance of the differences calculated, are not known. Additionally, this paucity of data meant that outcomes, such as neonatal intensive care admissions could not be evaluated, as this was not sufficiently captured in the NMA and MA studies.

An additional limitation of the data is the lack of information on whether women in the studies have an unripe cervix. Our study focuses on cervical ripening and IOL, and as such would ideally only incorporate data from women with an unripe cervix in whom methods, such as oxytocin, are used for the augmentation of labor as opposed to cervical ripening, would be excluded. One SLR attempted to evaluate a low Bishop score population, aligning to an unripe cervix, but inconsistencies within the data meant this was not possible²⁹. Other authors have commented that no or very few relevant clinical trials specifically recruited women with an unripe cervix^38,45. Therefore, although the data included in our study does not specifically represent the subpopulation of women with an unripe cervix, it would not be feasible to limit the analysis to this group, and the majority of women in studies of IOL receiving the interventions of interest are likely to have an unripe cervix at baseline.

There are various avenues for future research building on the analyses presented herein. For example, incorporating neonatal and post-natal maternal outcomes, and assessment of individual subpopulations (such as nulliparous women). Full cost-effectiveness analyses would generate additional information; these may be undertaken in the future depending on the reimbursement requirements for each country. Using a full cost-effectiveness analysis, further impacts on costs, labor outcomes, and quality of life may be evaluated at a patient level. However, the paucity of high-quality comparative effectiveness evidence noted above is likely to result in uncertainty. A comprehensive cost-effectiveness evaluation comparing 19 different IOL interventions (not including oral misoprostol tablets [25 µg]) in the UK reported “considerable uncertainty around effect estimates” and “a high degree of uncertainty in [our] estimates of cost-effectiveness”²⁹.

There is a need for a low-dose oral misoprostol formulation, such as oral misoprostol tablets (25 µg), that is specifically designed and approved for IOL, to extend the options available to women and their physicians and obviate the risks associated with off-label use of Cytotec. A prospective qualitative study in Germany reported that the most common reasons for not using misoprostol were concern over the lack of a license (69%) and uncertainty of the legal situation (27%)⁴⁶. Furthermore, gaps in providing information on the choice of method for inducing labor, the associated risks, and reduced ability to make informed choices, are reported as reducing maternal satisfaction with IOL⁴⁷. The introduction of an approved oral formulation provides additional choices and may improve women’s satisfaction with their treatment.

The preliminary outcomes reported here suggest that increased use of oral misoprostol tablets (25 µg) within IOL could potentially increase the proportion of routine vaginal deliveries, with concurrent decreases in instrumental vaginal deliveries and cesarean sections, both of which may be associated with negative implications for both maternal and fetal health. If confirmed in clinical practice, this predicted improvement in obstetric outcomes has the potential to be cost-saving vs. current IOL practice, based on the increased cost associated with instrumental vaginal deliveries and cesarean sections, compared with routine vaginal deliveries. Further study is required to validate these findings and assess the comparative efficacy of IOL methods, including oral misoprostol tablets (25 µg).

Transparency

Declaration of funding

This study was supported by Norgine Ltd.

Declaration of financial/other relationships

A.C.P. is an employee of Norgine Ltd. K.P. is an employee of Health Economics and Outcomes Research Ltd., which received fees from Norgine in relation to this study and development of the manuscript. R.H. received consulting fees from Norgine Ltd. for this article and reports no other potential conflict of interest relevant to this work. F.P. received consulting fees from Norgine Ltd. for this article, has received consulting fees from Roche Diagnostics France, and reports no other potential conflict of interest relevant to this work. R.D. received consulting fees from Norgine Ltd. for this article, has received research grants from Ferring and consulting fees from Metagenics, and reports no other potential conflict of interest relevant to this work.

Peer reviewers on this manuscript have received an honorarium from JME for their review work but have no other relevant financial relationships to disclose.

Author contributions

All authors: study conception and design. KP: analysis of data. All authors: interpretation of data. K.P. and A.C.P.: drafting of the manuscript. All authors: critical revision of the manuscript and final approval of the manuscript. All authors agree to be accountable for all aspects of the work.

Acknowledgements

Editorial assistance was provided by Jo Whelan of Health Economics and Outcomes Research Ltd., funded by Norgine.