Outpatient versus inpatient induction of labour for improving birth outcomes

  • Conclusions changed
  • Review
  • Intervention

Authors

  • Anthony J Kelly,

    Corresponding author
    1. Brighton and Sussex University Hospitals NHS Trust, Department of Obstetrics and Gynaecology, Brighton, UK
    • Anthony J Kelly, Department of Obstetrics and Gynaecology, Brighton and Sussex University Hospitals NHS Trust, Royal Sussex County Hospital, Eastern Road, Brighton, BN2 5BE, UK. tony.kelly@bsuh.nhs.uk.

    Search for more papers by this author
  • Zarko Alfirevic,

    1. The University of Liverpool, Department of Women's and Children's Health, Liverpool, UK
    Search for more papers by this author
  • Arpita Ghosh

    1. Brighton and Sussex University Hospitals NHS Trust, Department of Obstetrics and Gynaecology, Brighton, UK
    Search for more papers by this author

Abstract

Background

More than 20% of women undergo induction of labour in some countries. The different methods used to induce labour have been the focus of previous reviews, but the setting in which induction takes place (hospital versus outpatient settings) may have implications for maternal satisfaction and costs. It is not known whether some methods of induction that are effective and safe in hospital are suitable in outpatient settings.

Objectives

To assess the effects on outcomes for mothers and babies of induction of labour for women managed as outpatients versus inpatients.

Search methods

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (30 June 2013).

Selection criteria

Published and unpublished randomised and quasi-randomised trials in which inpatient and outpatient methods of cervical ripening or induction of labour have been compared.

Data collection and analysis

Two review authors independently assessed trial reports for inclusion. Two review authors carried out data extraction and assessment of risk of bias independently.

Main results

We included four trials, with a combined total of 1439 women in the review; each trial examined a different method of induction and we were unable to pool the results from trials.

1. Vaginal PGE2 (two studies including 1028 women). There were no differences between women managed as outpatients versus inpatients for most review outcomes. There was no evidence of a difference between the likelihood of women requiring instrumental delivery in either setting (risk ratio (RR) 1.29; 95% confidence interval (CI) 0.79 to 2.13). The overall length of hospital stay was similar in the two groups.

2. Controlled release PGE2 10 mg (one study including 300 women). There was no evidence of differences between groups for most review outcomes, including success of induction. During the induction period itself, women in the outpatient group were more likely to report high levels of satisfaction with their care (satisfaction rated seven or more on a nine-point scale, RR 1.42; 95% CI 1.11 to 1.81), but satisfaction scores measured postnatally were similar in the two groups.

3. Foley catheter (one study including 111 women). There was no evidence of differences between groups for caesarean section rates, total induction time and the numbers of babies admitted to neonatal intensive care.

Authors' conclusions

The data available to evaluate the efficacy or potential hazards of outpatient induction are limited. It is, therefore, not yet possible to determine whether induction of labour is effective and safe in outpatient settings.

Résumé scientifique

Induction du travail en ambulatoire ou en milieu hospitalier pour améliorer les accouchements

Contexte

Dans certains pays, le travail est déclenché pour plus de 20% des femmes. Les différentes méthodes utilisées pour déclencher le travail ont fait l'objet de revues antérieures, mais le cadre dans lequel le déclenchement se fait (hôpital vs ambulatoire) peut avoir des conséquences en termes de satisfaction de la mère et de coûts. On ne sait pas si certaines méthodes de déclenchement, qui sont efficaces et sûres en milieu hospitalier, sont adaptées en ambulatoire.

Objectifs

Évaluer les effets sur les résultats cliniques de la mère et de l'enfant du déclenchement du travail, chez les femmes en hospitalisation ou en ambulatoire.

Stratégie de recherche documentaire

Nous avons effectué des recherches dans le registre des essais cliniques du groupe Cochrane sur la grossesse et la naissance (30 juin 2013).

Critères de sélection

Essais randomisés et quasi-randomisés, publiés et non-publiés, dans lesquels ont été comparées des méthodes de maturation cervicale et de déclenchement du travail en ambulatoire et en milieu hospitalier.

Recueil et analyse des données

Deux auteurs de la revue ont évalué les rapports des essais à l'inclusion de façon indépendante. Deux auteurs de la revue ont entrepris l'extraction des données et l'évaluation du risque de biais, de manière indépendante.

Résultats principaux

Quatre essais ont été inclus dans la revue, soit un total de 1439 femmes; chaque essai a examiné une méthode différente de déclenchement du travail et il nous a été impossible de grouper les résultats des essais.

1. La PGE2 vaginale (deux études incluant 1028 femmes). Pour la plupart des résultats cliniques de la revue, il n'y avait aucune différence entre les patientes prises en charge en ambulatoire ou en milieu hospitalier. Il n'y avait aucune preuve d'une différence entre la probabilité de nécessiter un accouchement instrumental dans n'importe quel contexte (risque relatif (RR) 1,29; intervalle de confiance (IC) à 95% 0,79 à 2,13). La durée globale d'hospitalisation était similaire dans les deux groupes.

2. PGE2 10 mg à libération contrôlée (une étude avec 300 femmes). Aucune différence n'a été mise en évidence entre les groupes pour la plupart des résultats cliniques de la revue, y compris le succès du déclenchement. Pendant la période du déclenchement lui-même, le niveau de satisfaction des soins était plus élevé pour les femmes en ambulatoire (note de 7 ou plus sur une échelle de satisfaction à 9 points, RR 1,42; IC à 95%, de 1,11 à 1,81), mais les scores de satisfaction post-natale étaient similaires dans les deux groupes.

3. Le cathéter de Foley (une étude avec 111 femmes). Aucune différence n'a été mise en évidence entre les groupes en termes de taux de césariennes, de durée totale du déclenchement et du nombre de bébés admis en soins intensifs néonataux.

Conclusions des auteurs

Les données disponibles permettant d'évaluer l'efficacité ou les risques potentiels du déclenchement en ambulatoire sont limitées. Par conséquent, il n'est pas encore possible de déterminer si le déclenchement du travail est efficace et sans risque lorsqu' il est pratiqué en ambulatoire.

Plain language summary

Outpatient versus inpatient induction of labour

Up to a quarter of pregnant women may need their labour started artificially, or induced, with the use of medication or by other means. With most methods of induction it takes some time for labour to actually start. This means that it may be more convenient to women, and cheaper for health service providers, if they are cared for in outpatient settings, such as in their own homes. Women who are at low obstetric or medical risk could be assessed in hospital, given the induction agent and then return home with clear instructions. The use of outpatient induction of labour attempts to balance possible improvements in maternal satisfaction, convenience, reduced length of stay in hospital and lower cost with the safety of both the mother and baby.

Four randomised controlled trials with a combined total of 1439 women assessed the effects of induction of labour for women managed as outpatients versus inpatients. Out of a total of four studies, the induction agents differed in two studies, whereas the remaining two studies evaluated the same induction agent (vaginal PGE2). The limited information from these trials did not support more successful induction within 24 hours, shorter length of stay in hospital or differences in need for further induction or the mode of giving birth. The information available was limited and it is, therefore, not yet possible to determine whether induction of labour is effective and safe in outpatient settings.

Résumé simplifié

Induction du travail en milieu hospitalier ou en ambulatoire

Jusqu' à un quart des femmes enceintes peuvent avoir besoin d'un déclenchement artificiel du travail, par des médicaments ou par d'autres moyens. Avec la plupart des méthodes de déclenchement, il faut du temps avant que le travail commence. Cela signifie qu' il pourrait être plus pratique pour les femmes, et plus économique pour les prestataires de services de santé, d'être prises en charge en milieu ambulatoire, par exemple au domicile. Les femmes qui présentent un faible risque au plan obstétrique ou médical pourraient être examinées à l'hôpital, y recevoir l'agent d'induction, puis retourner chez elles avec des instructions claires. Le déclenchement du travail en ambulatoire vise à mesurer les améliorations possibles en termes de satisfaction de la mère, l'aspect pratique, la durée réduite de séjour à l'hôpital et la réduction des coûts, tout en garantissant la sécurité de la mère et de l'enfant.

Quatre essais contrôlés randomisés, 1439 femmes au total, ont évalué les effets du déclenchement du travail chez des femmes prises en charge en ambulatoire comparés à celles hospitalisées. Sur un total de quatre études, les agents de déclenchement différaient dans deux études, tandis que les deux autres études évaluaient les mêmes agents d'induction (PGE2 vaginale). Les données limitées de ces essais n'indiquaient pas un meilleur taux de réussite pour le déclenchement dans les 24 heures, ne diminuent pas la durée d'hospitalisation et il n'y avaient pas de différences quant à la nécessité d'un second déclenchement ou au mode d'accouchement. Les données disponibles étaient limitées et par conséquent il n'est pas encore possible de déterminer si le déclenchement du travail est efficace et sans risque lorsqu' il est pratiqué en ambulatoire.

Notes de traduction

Traduit par: French Cochrane Centre 14th January, 2014
Traduction financée par: Ministère du Travail, de l'Emploi et de la Santé Français

Background

Multiple interventions are available for cervical ripening and induction of labour at term (Hofmeyr 2009).

The method used, and indications for induction of labour, have been the main focus of previous trials, systematic reviews and guidelines, but there has been less published work focusing on the place of induction of labour. This is an important issue, as the number of women undergoing labour induction seems to be increasing; in England as a whole rates are above 20%, and some units report more than 25% of women being induced (NHS 2007).

The ideal agent for induction of labour would achieve cervical ripening followed by ‘spontaneous’ onset of labour without causing uterine hyperstimulation (Calder 1998). Currently, most commonly used induction agents result in significant uterine activity, requiring close monitoring of mother and baby within a hospital environment. Some induction methods, such as intravenous syntocinon infusions, will only be suitable for use in an inpatient setting. Induction of labour in an outpatient setting is therefore, restricted to low-risk circumstances when cervical ripening and labour induction is carried out without an ongoing requirement for continuous or frequent maternal or fetal monitoring. The use of outpatient induction of labour attempts to balance potential improvements in maternal satisfaction, convenience, reduced length of hospitalisation and lower cost, against those of safety (both maternal and fetal).  For the purpose of this review, outpatient settings include home and all other facilities (healthcare or otherwise) that are physically distant from the actual place of birth and require transport to hospital in case of complications. In most instances an outpatient induction will include an initial hospital assessment, including administration of the ripening or induction agent.

Description of the condition

Induction of labour is performed for a number of reasons towards the end of pregnancy. The vast majority occur after the expected date of delivery (mainly after 41 weeks) and as such are performed for post maturity.

Description of the intervention

Induction of labour is commonly performed in hospital settings using a range of interventions. Commonly this involves the use of vaginal prostaglandin agents. In recent years there has been increased interest in the use of outpatient induction of labour. In this setting, women attend the hospital to receive the induction agent and then return home afterwards. They then return to the hospital when they start to contract regularly, or if they require further interventions.

How the intervention might work

It is hoped that outpatient induction of labour will be as safe and efficient as the current standard of hospital-based induction of labour. Outpatient induction of labour may well be associated with greater satisfaction and financial savings when compared to hospital-based policies.

Why it is important to do this review

There is an increasing interest in the use of ambulatory models of care within medicine and specifically in obstetric practice. This review attempts to address safety and efficacy issues regarding the comparison of outpatient induction of labour to inpatient practices.

Objectives

  1. The primary objective of this review is to assess the effects on maternal and neonatal outcomes of cervical ripening or third trimester induction of labour for women managed as outpatients compared with inpatient management.

  2. A secondary objective of the review is to determine whether the effects on maternal and neonatal outcomes are influenced by predefined clinical subgroups including the effect of parity, membrane status (intact or ruptured) and cervical status (unfavourable, favourable or undefined).

  3. This review does not attempt to compare the relative effects of different methods of induction of labour on maternal and neonatal outcomes within an outpatient setting. This is the topic of a separate review (Kelly 2009a).

Methods

Criteria for considering studies for this review

Types of studies

All published and unpublished randomised trials, in which inpatient and outpatient methods of cervical ripening or induction of labour are compared. The trials include some form of random allocation to either group; and they report one or more of the pre-stated outcomes. In updates of the review, we plan to include cluster-randomised trials if they are otherwise eligible.

Types of participants

Pregnant women with a viable fetus suitable for cervical ripening or induction of labour at or near term (greater than 35 weeks) in an outpatient setting.

Types of interventions

Outpatient cervical ripening or induction of labour with pharmacological agents or mechanical methods. Outpatient ripening is defined as any cervical ripening or induction of labour intervention (with the exception of membrane sweeping) that can be carried out at home or within community healthcare settings. It also includes a package of care initially provided in hospital (fetal monitoring, drug administration) after which the patient is allowed home until a later review or until admission in labour.

Types of outcome measures

Clinically relevant outcomes for trials of methods of cervical ripening and labour induction have been prespecified by two authors of labour induction reviews (Justus Hofmeyr and Zarko Alfirevic) (Hofmeyr 2000). Most of these outcomes relevant to both inpatient and outpatient settings are used in this review.

In addition, an attempt has been made to use relevant outcome measures to quantify any cost effectiveness benefits of outpatient ripening.

Primary outcomes
  1. Failure to achieve spontaneous vaginal delivery.

  2. Additional induction agents required.

  3. Length of hospital stay.

  4. Use of emergency services.

  5. Mother not satisfied.

  6. Caregiver not satisfied.

  7. Serious neonatal morbidity or perinatal death (e.g. seizures, birth asphyxia defined by trialists, neonatal encephalopathy, disability in childhood).

  8. Serious maternal morbidity or death (e.g. uterine rupture, admission to intensive care unit, septicaemia).

Secondary outcomes

Outcomes related to measures of effectiveness, complications and satisfaction.

Measures of effectiveness
  1. Vaginal delivery not achieved within 24/48/72 hours.

  2. Randomisation to delivery interval.

  3. Oxytocin augmentation.

  4. Pain relief requirements (epidural, opioids).

Complications
  1. Uterine hyperstimulation.

  2. Instrumental vaginal delivery.

  3. Caesarean section.

  4. Apgar score less than seven at five minutes.

  5. Neonatal intensive care unit admission.

  6. Perinatal death.

  7. Uterine rupture.

  8. Postpartum haemorrhage (as defined by the trial authors).

  9. Serious maternal complications (e.g. intensive care unit admission, septicaemia).

Where formal economic evaluation is lacking, we will attempt to describe potential cost savings and the impact of interventions used within an outpatient setting. Where possible, these estimates will involve using some measures of effectiveness and complications in combination with estimates of healthcare provision.

Detailed definitions for outcomes
  • Perinatal and maternal morbidity and mortality are composite outcomes. This is not an ideal solution because some components are clearly less severe than others. It is possible for one intervention to cause more deaths but less severe morbidity. However, in the context of labour induction at term, this is unlikely. All these events will be rare, and a modest change in their incidence will be easier to detect if composite outcomes are presented. The incidence of individual components will be explored as secondary outcomes (see above).

  • 'Uterine rupture' includes all clinically significant ruptures of unscarred or scarred uteri. Trivial scar dehiscence noted incidentally at the time of surgery is excluded.

  • The terminology of uterine hyperstimulation is problematic (Curtis 1987). In the reviews, the term 'uterine hyperstimulation' is defined as uterine tachysystole (more than five contractions per 10 minutes for at least 20 minutes) and uterine hypersystole/hypertonus (a contraction lasting at least two minutes).

  • 'Uterine hyperstimulation with fetal heart rate (FHR) changes' is usually defined as uterine hyperstimulation syndrome (tachysystole or hypersystole with FHR changes such as persistent decelerations, tachycardia or decreased short-term variability). However, due to varied reporting, there is the possibility of subjective bias in the interpretation of these outcomes. Also, it is not always clear from the trials if these outcomes are reported in a mutually exclusive manner. More importantly, continuous monitoring is unlikely in an outpatient setting. Therefore, there is a high risk of biased reporting of uterine hyperstimulation (with or without FHR changes). It is possible that bias will favour the outpatient setting (i.e. by failure to recognise mild forms of hyperstimulation without continuous monitoring). On the other hand, clinicians who favour inpatient induction may, in the absence of continuous monitoring, label any maternal description of painful, frequent uterine contractions as hyperstimulation. Therefore, in the absence of blinding, hyperstimulation and other 'soft' outcomes should be interpreted with extreme caution.

While we sought data on all of the outcomes listed above, we have documented only those with data in the analysis tables. We have included outcomes in the analysis if reasonable measures were taken to minimise observer bias, and data were available according to original treatment allocation.

Search methods for identification of studies

Electronic searches

We contacted the Trials Search Co-ordinator to search the Cochrane Pregnancy and Childbirth Group’s Trials Register (30 June 2013)

The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co-ordinator and contains trials identified from: 

  1. monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);

  2. weekly searches of MEDLINE;

  3. weekly searches of Embase;

  4. handsearches of 30 journals and the proceedings of major conferences;

  5. weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.

Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group

Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co-ordinator searches the register for each review using the topic list rather than keywords.  

We did not apply any language restrictions.

Data collection and analysis

For methods used in previous versions of this review, see Appendix 1.

For the 2013 update, the following methods were used for assessing the reports that were identified as a result of the updated search.

Selection of studies

Two review authors independently assessed for inclusion all the potential studies identified as a result of the search strategy. We resolved any disagreement through discussion or, if required, we consulted the third review author.

Data extraction and management

We designed a form to extract data. For eligible studies, two review authors extracted the data using the agreed form. We resolved discrepancies through discussion or, if required, we consulted the third review author. Data were entered into Review Manager software (RevMan 2012) and checked for accuracy.

When information regarding any of the above was unclear, we planned to contact authors of the original reports to provide further details.

Assessment of risk of bias in included studies

Two review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Any disagreement was resolved by discussion or by involving a third assessor.

(1) Random sequence generation (checking for possible selection bias)

We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.

We assessed the method as:

  • low risk of bias (any truly random process, e.g. random number table; computer random number generator);

  • high risk of bias (any non-random process, e.g. odd or even date of birth; hospital or clinic record number);

  • unclear risk of bias.

(2) Allocation concealment (checking for possible selection bias)

We described for each included study the method used to conceal allocation to interventions prior to assignment and assessed whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.

We assessed the methods as:

  • low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);

  • high risk of bias (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth);

  • unclear risk of bias.

(3.1) Blinding of participants and personnel (checking for possible performance bias)

We described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered that studies were at low risk of bias if they were blinded, or if we judged that the lack of blinding unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes.

We assessed the methods as:

  • low, high or unclear risk of bias for participants;

  • low, high or unclear risk of bias for personnel.

(3.2) Blinding of outcome assessment (checking for possible detection bias)

We described for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed blinding separately for different outcomes or classes of outcomes.

We assessed methods used to blind outcome assessment as:

  • low, high or unclear risk of bias.

(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)

We described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported and the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported, or could be supplied by the trial authors, we planned to re-include missing data in the analyses which we undertook.

We assessed methods as:

  • low risk of bias (e.g. no missing outcome data; missing outcome data balanced across groups);

  • high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; ‘as treated’ analysis done with substantial departure of intervention received from that assigned at randomisation);

  • unclear risk of bias.

(5) Selective reporting (checking for reporting bias)

We described for each included study how we investigated the possibility of selective outcome reporting bias and what we found.

We assessed the methods as:

  • low risk of bias (where it is clear that all of the study’s pre-specified outcomes and all expected outcomes of interest to the review have been reported);

  • high risk of bias (where not all the study’s pre-specified outcomes have been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported);

  • unclear risk of bias.

(6) Other bias (checking for bias due to problems not covered by (1) to (5) above)

We described for each included study any important concerns we had about other possible sources of bias.

(7) Overall risk of bias

We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Cochrane Handbook (Higgins 2011). With reference to (1) to (6) above, we planned to assess the likely magnitude and direction of the bias and whether we considered it is likely to impact on the findings. In future updates, we will explore the impact of the level of bias through undertaking sensitivity analyses - see Sensitivity analysis.

Measures of treatment effect

Dichotomous data

For dichotomous data, we presented results as summary risk ratio with 95% confidence intervals.

Continuous data

For continuous data, we used the mean difference if outcomes were measured in the same way between trials. In future updates, if appropriate, we will use the standardised mean difference to combine trials that measure the same outcome, but use different methods.

Unit of analysis issues

Cluster-randomised trials

We did not identify any cluster-randomised trials for inclusion in this review, but we may include trials of this type in future updates. If we do, we plan to include cluster-randomised trials in the analyses along with individually-randomised trials. Their sample sizes will be adjusted using the methods described in Gates 2005 using an estimate of the intracluster correlation co-efficient (ICC) derived from the trial (if possible), or from another source. If ICCs from other sources are used, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify both cluster-randomised trials and individually-randomised trials, we plan to synthesise the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely. We will also acknowledged heterogeneity in the randomisation unit and perform a sensitivity analysis to investigate the effects of the randomisation unit.

Cross-over trials

Cross-over trials were not eligible for inclusion in this review.

Dealing with missing data

For included studies, levels of attrition were noted. In future updates, if more eligible studies are included, the impact of including studies with high levels of missing data in the overall assessment of treatment effect will be explored by using sensitivity analysis.

For all outcomes, analyses were carried out, as far as possible, on an intention-to-treat basis, i.e. we attempted to include all participants randomised to each group in the analyses. The denominator for each outcome in each trial was the number randomised minus any participants whose outcomes were known to be missing.

Assessment of heterogeneity

We assessed statistical heterogeneity in each meta-analysis using the Tau², I² and Chi² statistics. We regarded heterogeneity as substantial if an I² was greater than 30% and either a Tau² was greater than zero, or there was a low P value (less than 0.10) in the Chi² test for heterogeneity. If we identified substantial heterogeneity (above 30%), we planned to explore it by pre-specified subgroup analysis.

Assessment of reporting biases

In future updates, if there are 10 or more studies in the meta-analysis we will investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually. If asymmetry is suggested by a visual assessment, we will perform exploratory analyses to investigate it.

Data synthesis

We carried out statistical analysis using the Review Manager software (RevMan 2012). We used fixed-effect meta-analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar.

If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if substantial statistical heterogeneity was detected, we used random-effects meta-analysis to produce an overall summary if an average treatment effect across trials was considered clinically meaningful. The random-effects summary was treated as the average of the range of possible treatment effects and we planned to discuss the clinical implications of treatment effects differing between trials. If the average treatment effect was not clinically meaningful, we did not combine trials. If we used random-effects analyses, the results were presented as the average treatment effect with 95% confidence intervals, and the estimates of Tau² and I².

Subgroup analysis and investigation of heterogeneity

In future updates, if we identify substantial heterogeneity, we will investigate it using subgroup analyses and sensitivity analyses. We will consider whether an overall summary is meaningful, and if it is, use random-effects analysis to produce it.

We plan to carry out the following subgroup analyses:

  • nulliparous women versus multiparous women;

  • membrane status (intact versus ruptured);

  • cervical status (unfavourable versus favourable or undefined);

  • induction indication, i.e. postdates (41 weeks or greater).

Subgroup analyses will be restricted to the review's primary outcomes.

We will assess subgroup differences by interaction tests available within RevMan (RevMan 2012). We will report the results of subgroup analyses quoting the Chi² statistic and P value, and the interaction test I² value.

Sensitivity analysis

In future updates, if more studies are included, we will carry out sensitivity analyses to explore the effect of trial quality assessed by concealment of allocation, high attrition rates, or both, with poor quality studies being excluded from the analyses in order to assess whether this makes any difference to the overall result.

Results

Description of studies

Results of the search

The search strategy identified six studies for possible inclusion in the review. Four studies were included in the review with a total of 1439 participants (Biem 2003; Ryan 1998; Sciscione 2001; Wilkinson 2012).

Included studies

Sciscione 2001 was carried out in the USA, Wilkinson 2012 was conducted in Australia and the remaining two studies in Canada (Biem 2003; Ryan 1998).

The interventions examined in the four studies all involved initial treatment and monitoring in hospital, with subsequent discharge home for the outpatient groups.

Ryan 1998 and Wilkinson 2012 used vaginal prostaglandin E2 (PGE2) preparation for cervical ripening/induction agents, whereas Biem 2003 used controlled release PGE2 and Sciscione 2001 used Foley catheters.

(1) Evaluation of vaginal PGE2 in an outpatient versus inpatient setting

In the Ryan 1998 study and Wilkinson 2012 study, the induction agent was vaginal PGE2; little information was provided on eligibility criteria in both studies.

(2) Evaluation of controlled release PGE2 in an outpatient versus inpatient setting

In the Biem 2003 study, women received controlled release PGE2 10 mg. The trial authors set out detailed inclusion criteria, including low obstetric risk and access to reliable transportation. Women randomised to the outpatient group were provided with instructions on when to seek help, had regular telephone contacts by a nurse, and were asked to return to the hospital within 24 hours or less.

(3) Evaluation of the Foley catheter in an outpatient versus inpatient setting

The study by Sciscione 2001 examined the use of a Foley catheter to induce labour. In order to enter the trial, women had to be of low obstetric and medical risk; they were given detailed information on when to seek help and had 24-hour telephone access to a doctor.

Details of the three interventions and full inclusion and exclusion criteria are set out in the Characteristics of included studies table.

Excluded studies

Two studies were excluded from the review (Henry 2011; Rijnders 2007). One study was excluded because they compared two different methods of induction of labour in outpatient and inpatient setting Henry 2011. One study was excluded as there was no published or unpublished information available regarding the result of the study (Rijnders 2007).

Risk of bias in included studies

See Figure 1 and Figure 2 for a summary of 'Risk of bias' assessments.

Figure 1.

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 2.

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Allocation

In the Ryan 1998 and Wilkinson 2012 studies, the results were reported in conference abstracts, and very little information was provided on study methods. In the two other trials, the allocation sequence was computer-generated (Biem 2003; Sciscione 2001). In the Biem 2003 study, allocation concealment was by using sequentially numbered, sealed opaque envelopes opened immediately after the insertion of the induction agent, while Sciscione 2001 describes the use of sequentially numbered envelopes.

Blinding

With interventions where management in different settings are compared, it is not feasible to blind study participants to group allocation, and in the four included studies blinding of the outcome assessors was not attempted. The lack of blinding introduces the potential for bias in these trials and this should be kept in mind when interpreting the results.

Incomplete outcome data

Attrition for outcomes measured in labour were low in all four studies. In one study (Wilkinson 2012), only 51% patients were analysed post randomisation, but all patients were subsequently analysed within allocated groups and there were no drop outs. High levels of attrition were reported in the Sciscione 2001 study for patient satisfaction outcomes measured in the postnatal period.

Selective reporting

There was no evidence of selective reporting in one of the studies (Biem 2003) and in the other studies it was unclear because of the limited information reported in the abstracts (Ryan 1998; Sciscione 2001; Wilkinson 2012), or results being difficult to interpret (Sciscione 2001).

Other potential sources of bias

There was no evidence of other sources of bias in one of the studies (Biem 2003) and in the other studies it was unclear (Ryan 1998; Sciscione 2001; Wilkinson 2012).

Effects of interventions

In view of the fact that the three studies used different methods to induce labour, we have not pooled results in a meta-analysis. Two of the included studies used vaginal prostaglandin E2, but the delivery mechanism was different in each study. A related review (Kelly 2003) provides evidence of different treatment effects depending on the vehicle used to deliver the PGE2.

In the text below, and in the data tables and forest plots, we have described the results for each type of induction method separately; we have calculated no overall effects.

(1) Evaluation of vaginal PGE2 in an inpatient versus outpatient setting (two studies, 1028 women)

Primary outcomes

One study (Wilkinson 2012) reported no evidence of a significant difference in the number of women achieving spontaneous birth, however the time taken to achieve this is not mentioned. There was no evidence of a difference between groups for the numbers of women requiring a further induction agent (oxytocin) (risk ratio (RR) 1.00, 95% confidence interval (CI) 0.87 to 1.14, two studies, 626 women) (Analysis 1.1).

The total mean length of hospital stay (hours) was very similar for those receiving PGE2 and managed as out- and inpatients. While the length of time from admission to delivery was reduced in the outpatient group, this was offset by an increased length of postpartum hospital stay for this group (Analysis 1.2; Analysis 1.3; Analysis 1.4; one study, 201 women).

Serious neonatal morbidity/ death was reported along with development of hypoxaemic ischaemic encephalopathy in one study (Wilkinson 2012). The results were not shown to be statistically significant between the two groups. Eleven serious outcomes were noted in the inpatient group in comparison to 16 in the outpatient group (7.4% versus 5.2%, risk ratio (RR) 1.42; 95% confidence interval (CI) 0.68 to 2.99 (Analysis 1.5).

In the Ryan 1998 study no information was collected on the numbers of women achieving delivery within 24 hours of insertion of the vaginal prostaglandin gel, nor on other primary review outcomes including serious maternal or neonatal morbidity or mortality, maternal satisfaction, or use of emergency services. In the Wilkinson 2012 study no information was collected on length of hospital stay and use of emergency services. Patient preference and satisfaction were studied but not reported.

Secondary outcomes

There was no evidence of a difference between the rates of instrumental delivery if they were in the outpatient group compared with those cared for in hospital (24.8% versus 20.3%, average RR 1.29; 95% CI 0.79 to 2.13, two studies, 626 women; Tau² = 0.08; I² = 60%) (Analysis 1.7). Substantial heterogeneity was evident in the analysis (I² = 60%) and so the analysis was performed using a random-effects model. The rates of instrumental delivery within the outpatient arm of the study by Ryan 1998 are much higher than seen in other studies that have examined the use of vaginal PGE2, and this increase is difficult to explain but may be responsible for the high level of heterogeneity seen.There was no evidence of a difference between groups in terms of the numbers of women undergoing caesarean section (Analysis 1.8).

There was no evidence of a difference between groups for the numbers of women receiving epidural analgesia (Analysis 1.6), or babies with Apgar scores less than seven at five minutes (Analysis 1.9), or admission to neonatal intensive care (Analysis 1.10).

The trial authors concluded that "cost-savings for outpatients were $585", but it was not clear how this figure was calculated.

In the Wilkinson 2012 study no information was available for the number of women receiving epidural analgesia, Apgar scores less than seven at five minutes, admission to neonatal intensive care or serious maternal complications. Neither study reported on vaginal delivery not achieved within 24 hours.

(2) Evaluation of controlled release PGE2 in outpatient and inpatient settings (one study, 300 women)

Primary outcomes

In this study (Biem 2003), information was collected on most of the primary outcomes of the review.

For failure of induction, figures included both those women failing to deliver and those not in spontaneous labour after 24 hours. There was no evidence of a difference between the groups (receiving controlled release prostaglandin) managed at home or as inpatients (22.8% versus 28.7%, RR 0.80; 95% CI 0.54 to 1.17) for vaginal delivery not achieved within 24 hours (Analysis 2.1).

There was no evidence of differences between groups randomised to outpatient versus inpatient settings for use of additional induction agents (Analysis 2.2).

The total length of stay was similar for those managed as outpatients and inpatients for women receiving controlled release prostaglandin (median stay in the outpatient group was reported as 73.6 hours versus 75.6 hours for the inpatient group). The length of stay was defined as the time from insertion of CR-PGE2 to discharge for the inpatient group, and from admission to discharge for the outpatient group. There was no information on costs or cost savings.

This was the only one of the four included studies that provided information on maternal satisfaction for both the outpatient and inpatient groups. Satisfaction and anxiety levels were measured four-hourly during the first 12 hours of the induction period (on a scale from zero to nine, women keyed their rating into a telephone keypad to reduce interviewer bias) and mean figures were calculated for each woman. Overall satisfaction was also measured on the day after delivery.

During the induction period, compared with women in the inpatient group, women in the outpatient group rated their satisfaction higher (mean difference (MD) 0.90; 95% CI 0.34 to 1.46) (Analysis 2.3) and were more likely to report high levels of satisfaction with their care (defined as the numbers of women rating satisfaction higher than seven on a nine-point scale, 55.7% versus 39.3%, RR 1.42; 95% CI 1.11 to 1.81) (Analysis 2.4). Overall satisfaction with labour and delivery for the two groups (measured postnatally) was similar (Analysis 2.5).

Three mothers were described as experiencing "major delivery complications". One woman had a caesarean section, and because of bleeding, was transfused four units of blood (outpatient group). One women in the inpatient group had an emergency caesarean section and a uterine rupture necessitating a hysterectomy, a second women in the inpatient group required hysterectomy for postpartum haemorrhage.

Secondary outcomes

The median time from induction to delivery was reported as being similar in the two groups: 21.4 hours in the outpatient group and 20.7 hours for the inpatients.

Biem 2003 reported relatively high numbers of women with uterine hyperstimulation (with or without fetal heart rate changes), but there were no differences between groups. This outcome must be interpreted with caution in view of problems with definition and diagnosis (Analysis 2.7).

There was no evidence of differences between groups randomised to outpatient versus inpatient settings for mode of delivery (Analysis 2.8; Analysis 2.9).

Outcomes for babies were similar in the two groups. There was no evidence of a difference between groups for admission to neonatal intensive care or non-reassuring fetal heart rate patterns, (Analysis 2.10; Analysis 2.11).

Other outcomes - not prespecified

In this study, anxiety was measured during the initial 12 hours of the induction period; there were no significant differences between the outpatient and inpatient groups (Analysis 2.12).

(3) Evaluation of the Foley catheter in outpatient versus inpatient settings (one study, 111 women)

Primary outcomes

In this study (Sciscione 2001), no information was provided on the numbers of women failing to deliver within 24 hours.

Information on patient satisfaction was only provided for the outpatient group and 94.6% (35 of the 37 respondents) reported that they would recommend outpatient induction to someone else. Most of these women (33 of the 37) had been able to remain in their own homes overnight. The response rate for these outcomes measured in the postnatal period was relatively low (37 of the 61 women in the outpatient group responded), so these satisfaction outcomes should be interpreted with caution.

The authors reported that "on average, the outpatient group avoided 9.6 hours of time in the hospital", but separate figures were not provided on length of hospital stay for those receiving care in the two different settings.

Secondary outcomes

There was no evidence of a difference between groups for the number of women receiving epidural analgesia (Analysis 3.1) or the numbers of women delivering by caesarean section (Analysis 3.2).

There was no evidence of any difference between groups in the numbers of babies admitted to special care (Analysis 3.3). There was no evidence of a difference between groups for the total induction time (Analysis 3.4).

There was little information on outcomes for babies.

Discussion

Summary of main results

There are limited data currently available to evaluate the efficacy or potential hazards of outpatient induction. We found four trials involving 1439 women. Two trials used vaginal PGE2 preparation for cervical ripening/induction agents, one used controlled release PGE2 and one used a Foley catheter.

There were very few differences between groups for most of the outcomes measured in this review. On the basis of the available data, it is not possible to determine whether these interventions are effective and safe within an outpatient setting.

Overall completeness and applicability of evidence

The outcomes chosen for the review reflect measures of both efficacy and harm (both direct, e.g. hyperstimulation and indirect, e.g. use of emergency services).

The studies included in the review did not have the statistical power to detect differences between the randomised groups for most outcomes, and more information is required to assess the safety and effectiveness of methods of induction of labour compared between the inpatient and outpatient setting. In one of the included trials, three women experienced serious labour complications, and the author of this study calls for larger studies in different settings "to compare the frequency of uncommon adverse events in labour and delivery" (Biem 2003).

Interpreting some of the results from the included studies was not simple. Outcome data using time intervals when examining induction of labour are often complicated. There are a variety of start and end points used. The data within these trials are recorded using a variety of methods, which makes comparing findings from studies difficult.

As patient convenience is often cited as a reason for carrying out care in outpatient settings, it is surprising that there is so little information on this. In fact, only one study collected information on maternal satisfaction from women in both arms of the trial. One study looked at maternal anxiety, and this may be useful additional outcome for future updates. One study states that the patient preference and satisfaction were assessed but the results were not reported or published.

Cost savings are also frequently mentioned as a reason for providing less inpatient care. Again, although three studies provided some information on length of hospital stay, this did not easily translate into cost data. Without a full breakdown of health service utilisation, it is not possible to impute costs.

The included studies had strict eligibility criteria and it is likely that outpatient cervical ripening and induction is only suitable for selected groups of women. The criteria cited within these studies reflect suitable 'low-risk' groups.

We were unable to pool results; it would be helpful to have more trials examining the interventions considered in this review where women are cared for in different settings. Without these direct comparisons between settings, we cannot make the assumption that a method that is known to be safe and effective in one setting would have the same effect in a different one.

Quality of the evidence

The included trials were of varying quality. There was limited information regarding randomisation and concealment in two of the included studies, and ideally, these processes should be explicit to avoid the introduction of bias. When comparing the same intervention within two settings, it is not necessary to blind participants and patients to the actual method of induction of labour, but where possible, researchers assessing outcomes should be blinded to the setting.

All of the studies were underpowered for the outcomes assessed. One trial recruited significant number of women at the point of randomisation however, only half of the total number entered the study and were analysed when evaluating the results. It is unclear how this may have effected the overall quality of this trial and the results of this study should therefore be interpreted with caution.

For two included studies in this review, we have only published abstracts available to analyse (Ryan 1998; Wilkinson 2012).

Authors' conclusions

Implications for practice

This review highlights the small volume of available evidence relating to cervical ripening and induction of labour in an outpatient setting. Conclusions regarding the efficacy or hazards of cervical ripening and induction on labour in an outpatient setting cannot be drawn from the available evidence.

Implications for research

The recent national UK guideline by The National Institute for Clinical Excellence on intrapartum care (NICE 2008) highlighted the need for more research into the safety and efficacy of outpatient ripening. Further trials are required to assess both efficacy and the potential hazards of initiating labour away from a hospital setting, and researchers are guided to consider the use of outcomes similar to those developed within this review.

Acknowledgements

We would like to acknowledge the contribution of Therese Dowswell, Research Associate, Cochrane Pregnancy and Childbirth Group, as an author on previous versions of this review Kelly 2009b . Therese was supported by a grant from the National Institute for Health Research (NIHR), UK NIHR NHS Cochrane Collaboration Programme Grant Scheme award for NHS prioritised, centrally-managed, pregnancy and childbirth systematic reviews: CPGS02.

The National Institute for Health Research (NIHR) is the largest single funder of the Cochrane Pregnancy and Childbirth Group. The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR, NHS or the Department of Health.

Data and analyses

Download statistical data

Comparison 1. Outpatient versus inpatient induction with vaginal PGE2
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Additional induction agent required (oxytocin)2626Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.87, 1.14]
2 Total length of hospital stay1201Mean Difference (IV, Fixed, 95% CI)0.20 [-8.86, 9.26]
3 Length of stay: admission to delivery1201Mean Difference (IV, Fixed, 95% CI)-9.60 [-13.02, -6.18]
4 Postnatal hospital stay1201Mean Difference (IV, Fixed, 95% CI)7.60 [-0.12, 15.32]
5 Serious neonatal morbidity or death1425Risk Ratio (M-H, Fixed, 95% CI)1.42 [0.68, 2.99]
6 Women receiving epidural analgesia1201Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.88, 1.08]
7 Instrumental vaginal delivery2626Risk Ratio (M-H, Random, 95% CI)1.29 [0.79, 2.13]
8 Caesarean section rate2626Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.73, 1.29]
9 Apgar score less than 7 at 5 minutes1201Risk Ratio (M-H, Fixed, 95% CI)2.23 [0.21, 24.22]
10 Neonatal intensive care unit admission1201Risk Ratio (M-H, Fixed, 95% CI)1.30 [0.45, 3.74]
Analysis 1.1.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 1 Additional induction agent required (oxytocin).

Analysis 1.2.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 2 Total length of hospital stay.

Analysis 1.3.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 3 Length of stay: admission to delivery.

Analysis 1.4.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 4 Postnatal hospital stay.

Analysis 1.5.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 5 Serious neonatal morbidity or death.

Analysis 1.6.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 6 Women receiving epidural analgesia.

Analysis 1.7.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 7 Instrumental vaginal delivery.

Analysis 1.8.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 8 Caesarean section rate.

Analysis 1.9.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 9 Apgar score less than 7 at 5 minutes.

Analysis 1.10.

Comparison 1 Outpatient versus inpatient induction with vaginal PGE2, Outcome 10 Neonatal intensive care unit admission.

Comparison 2. Outpatient versus inpatient induction with controlled release PGE
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Vaginal delivery or labour not achieved within 24 hours1299Risk Ratio (M-H, Fixed, 95% CI)0.80 [0.54, 1.17]
2 Oxytocin required1299Risk Ratio (M-H, Fixed, 95% CI)0.76 [0.46, 1.27]
3 Mean satisfaction score measured during induction1299Mean Difference (IV, Fixed, 95% CI)0.90 [0.34, 1.46]
4 High level of maternal satisfaction measured during induction period1299Risk Ratio (M-H, Fixed, 95% CI)1.42 [1.11, 1.81]
5 Overall satisfaction score with labour and delivery measured in the postnatal period1299Mean Difference (IV, Fixed, 95% CI)0.20 [-0.31, 0.71]
6 Number receiving epidural1299Risk Ratio (M-H, Fixed, 95% CI)1.02 [0.91, 1.16]
7 Uterine hyperstimulation (with or without fetal heart rate changes)1299Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.51, 1.98]
8 Instrumental vaginal delivery1299Risk Ratio (M-H, Fixed, 95% CI)1.34 [0.83, 2.17]
9 Caesarean section rate1299Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.64, 1.42]
10 Neonatal intensive care admission1299Risk Ratio (M-H, Fixed, 95% CI)1.38 [0.57, 3.34]
11 Nonreassuring fetal heart rate1299Risk Ratio (M-H, Fixed, 95% CI)1.48 [0.80, 2.73]
12 Anxiety score measured during induction period (not pre-specified outcome)1299Mean Difference (IV, Fixed, 95% CI)-0.20 [-0.60, 0.20]
Analysis 2.1.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 1 Vaginal delivery or labour not achieved within 24 hours.

Analysis 2.2.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 2 Oxytocin required.

Analysis 2.3.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 3 Mean satisfaction score measured during induction.

Analysis 2.4.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 4 High level of maternal satisfaction measured during induction period.

Analysis 2.5.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 5 Overall satisfaction score with labour and delivery measured in the postnatal period.

Analysis 2.6.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 6 Number receiving epidural.

Analysis 2.7.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 7 Uterine hyperstimulation (with or without fetal heart rate changes).

Analysis 2.8.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 8 Instrumental vaginal delivery.

Analysis 2.9.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 9 Caesarean section rate.

Analysis 2.10.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 10 Neonatal intensive care admission.

Analysis 2.11.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 11 Nonreassuring fetal heart rate.

Analysis 2.12.

Comparison 2 Outpatient versus inpatient induction with controlled release PGE, Outcome 12 Anxiety score measured during induction period (not pre-specified outcome).

Comparison 3. Outpatient versus inpatient induction with Foley catheter
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Women receiving epidural analgesia1111Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.95, 1.06]
2 Caesarean section rate1111Risk Ratio (M-H, Fixed, 95% CI)0.67 [0.41, 1.10]
3 Neonatal intensive care unit admission1111Risk Ratio (M-H, Fixed, 95% CI)0.20 [0.02, 1.78]
4 Total induction time (insertion of catheter to delivery)1111Mean Difference (IV, Fixed, 95% CI)2.0 [-292.61, 296.61]
Analysis 3.1.

Comparison 3 Outpatient versus inpatient induction with Foley catheter, Outcome 1 Women receiving epidural analgesia.

Analysis 3.2.

Comparison 3 Outpatient versus inpatient induction with Foley catheter, Outcome 2 Caesarean section rate.

Analysis 3.3.

Comparison 3 Outpatient versus inpatient induction with Foley catheter, Outcome 3 Neonatal intensive care unit admission.

Analysis 3.4.

Comparison 3 Outpatient versus inpatient induction with Foley catheter, Outcome 4 Total induction time (insertion of catheter to delivery).

Appendices

Appendix 1. Methods used in previous versions of this review

Data collection and analysis

The review authors worked independently to assess trials for inclusion and for methodological quality. We resolved differences in interpretation by discussion.

Selection of studies

Two review authors (A Kelly and T Dowswell) independently assessed for inclusion all the potential studies identified as a result of the search strategy. We resolved any disagreement through discussion, or if required we consulted the third author.

Data extraction and management

We designed a form to extract data. At least two review authors extracted the data using the agreed form. We resolved discrepancies through discussion. We entered data into Review Manager software (RevMan 2008) and checked for accuracy.When information on study design or outcomes was unclear, we attempted to contact authors of the original reports to provide further details.

Assessment of risk of bias in included studies

Two review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008). We resolved any disagreement by discussion or by involving a third assessor.

(1) Sequence generation (checking for possible selection bias)

We have described for each included study the methods used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.

We assessed the methods as:

  • adequate (e.g. random number table; computer random number generator);

  • inadequate (odd or even date of birth; hospital or clinic record number);

  • or,unclear.

(2) Allocation concealment (checking for possible selection bias)

We have described for each included study the method used to conceal the allocation sequence in sufficient detail and determine whether intervention allocation could have been foreseen in advance of, or during, recruitment.

We assessed the methods as:

  • adequate (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);

  • inadequate (open random allocation; unsealed or non opaque envelopes;

  • alternation; date of birth);unclear.

(3) Blinding (checking for possible performance bias)

When comparing induction of labour in different settings blinding women and clinical staff is not usually possible (although it may be feasible to blind outcome assessors for some outcomes). We did not formally assess blinding (or lack of it). Studies have been judged as being at lower risk of bias where we considered that lack of blinding was unlikely to affect results.

(4) Incomplete outcome data (checking for possible attrition bias through withdrawals, dropouts, protocol deviations)

We have described for each included study the completeness of data for each main outcome, including attrition and exclusions from the analysis. We have noted whether attrition and exclusions were reported, the numbers (compared with the total randomised participants), reasons for attrition/exclusion where reported, and any re-inclusions in analyses which we have undertaken.

We have assessed the methods as:

  • adequate (e.g. where there are no missing data or relatively low levels of attrition (less than 20%) and reasons for missing data are balanced across groups);

  • inadequate (e.g. where there are higher levels of missing data or where missing data are not balanced across groups);

  • unclear (e.g. where there is insufficient reporting of attrition or exclusions to permit a judgement to be made).

(5) Selective reporting bias

We have described for each included study how the possibility of selective outcome reporting bias was examined by us and what we found.

We have assessed the methods as:

  • adequate (where it is clear that all of the study’s prespecified outcomes and all expected outcomes of interest to the review have been reported);

  • inadequate (where not all the study’s prespecified outcomes have been reported; one or more reported primary outcomes were not prespecified; outcomes of interest are reported incompletely and so cannot be used; or, study fails to include results of a key outcome that would have been expected to have been reported);

  • unclear.

(6) Other sources of bias

We have noted for each included study any important concerns we have about other possible sources of bias. For example, any potential sources of bias associated with a particular study design.

Measures of treatment effect

In future updates of the review, if more than one study examines the same intervention, we will pool the results from studies and carry out meta-analyses using the methods set out in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008).

For dichotomous data, we will present results as summary risk ratio with 95% confidence intervals.

For continuous data, we will use the mean difference if outcomes were measured in the same way between trials. We will use the standardised mean difference to combine trials that measure the same outcome, but use different methods. Again, we will report 95% confidence intervals.

We will assess heterogeneity by visual inspection of the outcomes, and by applying tests of heterogeneity between trials using the I² statistic. If we identify heterogeneity among the trials, we will explore it by prespecified subgroup analysis and perform sensitivity analyses.

As more data become available, we plan to conduct subgroup analyses using the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008). Data permitting, we would perform subgroup analyses by: nulliparous women; induction indication, i.e. postdates (41 weeks or greater).

Sensitivity analyses

In updates of the review, if results from more than one trial are pooled we will carry out sensitivity analyses to explore the effect of trial quality for important outcomes in the review. Where there is risk of bias associated with a particular aspect of study quality (e.g. inadequate allocation concealment or high levels of missing data), we will explore this by sensitivity analyses.

Analysis of cluster-randomised trials

If we identify cluster-randomised trials that are eligible for inclusion, we will include such trials in analyses along with individually-randomised trials. Their sample sizes will be adjusted using the methods described in Gates 2005 using an estimate of the intracluster correlation co-efficient (ICC) derived from the trial (if possible), or from another source. If ICCs from other sources are used, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify both cluster-randomised trials and individually-randomised trials, we plan to synthesise the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely. We will also acknowledge heterogeneity in the randomisation unit and perform a separate meta-analysis. Therefore, we will perform the meta-analysis in two parts as well.

What's new

DateEventDescription
14 August 2013New citation required and conclusions have changedWith the addition of one new trial of vaginal PGE2, there is now no evidence of a difference between the likelihood of women requiring instrumental delivery in either setting. In the previous version of this review, women in the outpatient group were more likely to have instrumental deliveries.
30 June 2013New search has been performedSearch updated. Eight new trial reports identified. One new trial (five reports) included (Wilkinson 2012) and one trial (three reports) excluded (Henry 2011). We have also excluded a trial previously awaiting classification (Rijnders 2007) and moved one previously ongoing study report (Turnbull 2009) to the included section under Wilkinson 2012 .

Contributions of authors

For the 2013 update, Anthony Kelly and Arpita Ghosh prepared the text and Zarko Alfirevic commented on drafts.

Declarations of interest

None known.

Sources of support

Internal sources

  • The University of Liverpool, UK.

External sources

  • No sources of support supplied

Differences between protocol and review

Methods have been updated to the current standard methods for the Cochrane Pregnancy and Childbirth Group (2013). Anxiety was included as an additional outcome.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Biem 2003

MethodsRCT.
Participants

300 women randomised (1 woman withdrew).

Inclusion criteria: singleton, term pregnancy, cephalic presentation, intact membranes, Bishop's score 6 or less, parity 5 or less, unscarred uterus, normal nonstress test, reliable transportation from home.

Exclusion criteria: congenital anomaly, dead fetus, IUGR, hypertension, abnormal placenta, poly- or oligohydramnios.

Interventions

Both groups received vaginal controlled release prostaglandin E2 10 mg. Both groups were monitored in the antenatal ward for 1 hour.

Outpatient group: after initial monitoring women were discharged home to return when in labour or were reviewed after 12 hours (nonstress test). If they were not in labour 24 hours later they returned to hospital for induction of labour as an inpatient. Women were in telephone contact with a nurse every 4 hours and were given detailed instructions on when to seek help. They were asked to remain within easy travelling distance of the hospital.

Inpatient group: women remained on the antenatal ward throughout and managed in a similar way to the outpatient group.

OutcomesSatisfaction with care, length of hospital stay, length of labour, mode of delivery, labour interventions, maternal, fetal and neonatal complications.
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated sequence.
Allocation concealment (selection bias)Low risk"Sequential sealed opaque envelopes" opened immediately after the insertion of the PGE2.
Blinding (performance bias and detection bias)
Women
High riskNot feasible for this intervention.
Blinding (performance bias and detection bias)
Clinical staff
High riskNot feasible for this intervention.
Blinding (performance bias and detection bias)
Outcome assessors
High riskBlinding of the outcome assessors was not attempted.
Incomplete outcome data (attrition bias)
All outcomes
Low riskOnly 1 woman withdrew from the study after randomisation.
Selective reporting (reporting bias)Low riskNone apparent.
Other biasLow riskNone apparent. No baseline imbalance apparent.

Ryan 1998

MethodsRCT.
Participants

201 women.

Inclusion and exclusion criteria were not clear. Women at "term... who met the eligibility criteria to receive PGE2 as an outpatient".

InterventionsUnclear. Information not provided. Inpatient and outpatient management after insertion of PGE2 gel (dose not stated) were compared.
OutcomesLength of hospital stay, mode of delivery, Apgar score and neonatal admission to special care.
NotesAbstract only available, we have attempted to contact the author for more information but to date (November 2008) we have had no response.
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskDescribed as "randomised".
Allocation concealment (selection bias)Unclear riskNo information provided.
Blinding (performance bias and detection bias)
Women
High riskNot feasible.
Blinding (performance bias and detection bias)
Clinical staff
High riskNot feasible.
Blinding (performance bias and detection bias)
Outcome assessors
High riskBlinding of the outcome assessors was not attempted.
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNo loss to follow-up apparent but little information provided.
Selective reporting (reporting bias)Unclear riskAbstract only available.
Other biasUnclear riskAbstract only available.

Sciscione 2001

MethodsRCT.
Participants

111 women.

Inclusion criteria: singleton, term pregnancy, cephalic position, intact membranes with a Bishop's score < 6, with reactive nonstress test, "...attending physician had requested pre-induction cervical ripening using the Foley catheter".

Exclusion criteria: fetal anomaly or dead fetus, hypertension, vaginal bleeding, ruptured membranes, placenta praevia, IUGR, active herpes infection, without access to phone, without reliable transportation or living more than 30 minutes' distance from the hospital.

Interventions

Both groups had a number 16 Foley catheter inserted into the endocervical canal to or past the internal os; the balloon was filled with 30 mL of sterile water, the end of the catheter was taped to the thigh. After placement of the catheter if there was a reactive nonstress test and no signs of uterine hyperstimulation and the amniotic fluid index was > 5th percentile women were randomised.

Outpatient group: women received detailed oral and written guidelines on when to seek advice and then were discharged home. 24 hour phone access to a doctor was provided.They were asked to return for review the next morning for induction of labour with oxytocin.

Inpatient group: women were admitted to the labour ward. They were allowed to ambulate. The catheter was checked every 2-4 hrs and the fetal heart rate was assessed hourly.

OutcomesPrimary outcome: Bishop score.
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random number table.
Allocation concealment (selection bias)Low riskSequentially numbered envelopes.
Blinding (performance bias and detection bias)
Women
High riskBlinding not feasible.
Blinding (performance bias and detection bias)
Clinical staff
High riskBlinding not feasible.
Blinding (performance bias and detection bias)
Outcome assessors
High riskBlinding of the outcome assessors was not attempted.
Incomplete outcome data (attrition bias)
All outcomes
Low riskComplete data for main outcomes. Only the outpatient group was followed up in the postnatal period and there was high attrition (40%) for this longer-term follow-up.
Selective reporting (reporting bias)Unclear riskSome of the results were difficult to interpret.
Other biasUnclear riskNot clear how many of the women approached were eligible for this trial. Not clear how women were managed as regards oxytocin and this may have had an impact on results. Not clear how many women in the outpatient group were surveyed in the postnatal period; figures differ between the main study paper and an abstract reporting survey results.

Wilkinson 2012

  1. a

    IUGR: intrauterine growth restriction
    PGE2: prostaglandin E2
    RCT: randomised controlled trial.

Methods2-centre RCT.
Participants

827 women from 2 tertiary centres were randomised to 411 women in experimental arm (outpatient group) and 416 in control arm (inpatient group). 215 women in experimental group and 210 women in the control group received the intervention. The rest of the women either went into labour spontaneously or did not need the intervention by induction date.

Inclusion and exclusion criteria were not clear. It states, women were low risk, induced for postdates or social reasons. Women with glucose intolerance in pregnancy were excluded.

InterventionsPGE2. Dose not stated. The study compared inpatient to outpatient cervical ripening for induction of labour with PG2.
OutcomesOxytocin use, spontaneous birth, instrumental delivery or caesarean section, maternal, fetal and neonatal complications.
Notes

51% patients were analysed post randomisation.

22% patients receiving PGE2 as an intended outpatient did not go home.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskOnly abstract available for this study. It states women were randomised and stratified for parity but its unclear what method was used to generate random sequence.
Allocation concealment (selection bias)Unclear riskNot stated.
Blinding (performance bias and detection bias)
Women
High riskNot feasible.
Blinding (performance bias and detection bias)
Clinical staff
High riskNot feasible.
Blinding (performance bias and detection bias)
Outcome assessors
Unclear riskNot stated.
Incomplete outcome data (attrition bias)
All outcomes
Unclear risk51% of patients who were randomised were not analysed. This cohort of patients was balanced between the groups and this was judged to be a factor relating to the point of randomisation rather than a flaw in the actual study design.
Selective reporting (reporting bias)Unclear riskUnclear - only abstract available.
Other biasUnclear riskUnclear - only abstract available.

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    PGE2: prostaglandin E2
    RCT: randomised controlled trial.

Henry 2011The study compared 2 different methods of induction in outpatient and inpatient setting. It compared intracervical Foley catheter in the outpatient setting with intravaginal PGE2 gel in the inpatient setting for cervical ripening at term. The study reports clinical effectiveness, safety and cost-effectiveness.
Rijnders 2007It is a multicentre RCT that aims to explore costs and effects of amniotomy at home for induction of post term pregnancy. However, only trial registration form for this study is available for analysis. According to this, the study commenced in 2004 and its anticipated end date was 2007. Litrature search did not reveal any published results for this study and the authors were not contactable via e-mail or phone for any further information.

Ancillary