Background: Operative vaginal birth (OVB) is recognised by the World Health Organization as one of the seven critical components of basic obstetric and neonatal emergency care¹. The global rate of operative vaginal birth has declined significantly over the past two decades². This decline has been attributed to the heightened attention given to its complications and the associated medicolegal concerns³, negative perceptions among service users⁴, and insufficient training in the use of instruments or in complex interventions⁵.
Operative vaginal birth is associated with better maternal and neonatal outcomes than caesarean section at full cervical dilatation when performed by an experienced operator in an appropriate setting⁶. In particular, among women requiring operative vaginal birth, the incidence of postpartum haemorrhage is reduced and the length of hospital stay is shorter. In addition, their newborns are less likely to require admission to the neonatal intensive care unit⁶. Operative vaginal delivery is typically undertaken in urgent clinical scenarios, thereby constraining the time available for adequate preparation and informed consent.⁷. Nevertheless, it is crucial to ensure that women and their companions feel informed and supported throughout this experience⁸,⁹. A recent review suggests that women may perceive operative vaginal birth positively when they receive effective communication from the medical team and feel involved in the decision-making process¹⁰.
In Spain, approximately 14% of births are operative vaginal births, of which around 60% are performed using vacuum extraction¹¹. Proficiency in the application of the vacuum device depends on the level of exposure of junior obstetricians to situations in which they can learn the procedure under the supervision of more experienced practitioners. Obstetric experience is directly related to the number of times operative vaginal birth is performed on real patients. However, this learning approach raises ethical concerns and has driven the development of simulation-based training programmes in several countries. These programmes aim to ensure that obstetric trainees acquire both the technical and non-technical skills necessary to perform vacuum extraction safely and appropriately¹³. Objective: The objective of this study is to evaluate the clinical impact of a structured simulation-based training programme in vacuum-assisted vaginal birth on maternal and neonatal outcomes, using a stepped-wedge interrupted time-series design.
Primary study objectives:
1. To establish the rate of cup detachment (pop-off) associated with vacuum-assisted vaginal birth (Kirkpatrick Model Level III).
2. To determine the incidence of neonatal injuries associated with vacuum-assisted vaginal birth.
3. To determine the incidence of perineal tears associated with vacuum-assisted vaginal birth.
4. To assess course satisfaction and reproducibility using a Likert scale (Kirkpatrick Model Level I).
Secondary study objectives:
1. To determine the proportion of obstetricians who have received formal simulation-based training in vacuum-assisted vaginal birth.
2. To identify the type of vacuum device most commonly used in maternity units in our setting.
3. To establish the rate of sequential instrumentation (Kirkpatrick Model Level III).
4. It would be useful to clarify whether the objectives of the observational study differ before and after the intervention.
5. To determine the percentage of obstetricians who improve their confidence in the technique used, measured using a Likert scale (levels of concern and agreement) (Kirkpatrick Model Level II).
6. To determine the percentage of acquired knowledge retained at least six months after training (Kirkpatrick Model Level III). Methods: Study Design
Study Design
This study is a multicentre, quasi-experimental evaluation of an educational intervention using a stepped-wedge interrupted time-series design. The design was selected to assess the impact of a structured simulation-based training programme in vacuum-assisted vaginal birth, implemented sequentially across participating centres.
Randomisation at the individual clinician or institutional level was not feasible due to the educational and organisational nature of the intervention. The stepped-wedge design allows all centres to ultimately receive the intervention while contributing both pre-intervention (control) and post-intervention (exposed) data, and enables adjustment for secular time trends and between-centre variability.
The study protocol adheres to the STROBE guidelines for observational studies and to methodological recommendations for stepped-wedge and interrupted time-series designs.
Study Setting
The study will be conducted in maternity units across Spain participating in the Registro Español de Morbimortalidad Materna y Perinatal (REMMP), a national registry that systematically collects maternal and neonatal outcome data.
Participating hospitals represent a range of institutional characteristics, including differences in delivery volume, geographic location, and level of obstetric care, thereby enhancing the external validity and generalisability of the findings.
Each centre will implement the intervention according to a predefined staggered schedule established prior to study initiation.
Intervention
The intervention consists of a standardised simulation-based training programme in vacuum-assisted vaginal birth, conceptualised as a complex clinical and educational intervention.
The programme includes:
● Access to the theoretical part of the course was provided online one month in advance.
● A structured face-to-face practical workshop (approximately 1 hour)
● High-fidelity simulation of vacuum-assisted vaginal birth
● Use of a standardised technical skills checklist
● Supervised hands-on practice with expert feedback
● Training of local instructors using a train-the-trainer model to ensure sustainability
● An opinion survey evaluating perceived usefulness and effectiveness of the simulation activity
The training curriculum focuses on:
● Appropriate clinical indications for vacuum-assisted vaginal birth
● Selection and correct placement of the vacuum cup at the flexion point
● Generation and regulation of negative pressure
● Traction technique and coordination with maternal effort
● Recognition of failed attempts and criteria for abandoning the procedure
● Appropriate escalation to alternative delivery methods
To minimise contamination during implementation, no clinical outcome data will be collected during the intervention delivery (washout) period, when varying proportions of clinicians within each centre are undergoing training.
Participants
Study Population
The study population will include women and neonates undergoing an attempted vacuum-assisted vaginal birth at participating centres during the study period.
Inclusion Criteria
● Maternal age ≥18 years
● Vaginal birth at a participating centre
● Attempted operative vaginal birth using a vacuum extractor
● Maternal informed consent
Exclusion Criteria
● Gestational age < 37 weeks
● Multiple pregnancy
● Use of a vacuum device exclusively during caesarean section
Study Periods
For each participating centre, the study timeline will be divided into three phases:
1. Pre-intervention period: six months prior to implementation of the training programme, during which clinical outcome data are collected.
2. Intervention implementation (washout) period: training delivery phase; no clinical outcome data collected to minimise contamination.
3. Post-intervention period: six months following completion of training, during which clinical outcome data are collected.
This structure allows comparison of outcomes before and after exposure to the intervention while controlling for secular trends.
Outcome Measures
Primary Outcome
The primary outcome is failure of vacuum-assisted vaginal birth as the first intended instrument, defined as:
● Use of a second operative instrument, or
● Conversion to caesarean section following attempted vacuum extraction
We focused on the collection of the pop-off rate in order to assess technical failures of the procedure. This outcome was selected due to its clinical relevance and consistency with international literature on operative vaginal birth.
Secondary Outcome
Maternal outcomes
● Sequential instrumentation
● Caesarean section
● Total instrument application time (minutes)
● Number of instrument applications
● Number of traction pulls required
● Episiotomy
● Perineal trauma
● Cervical tear requiring suturing
● Postpartum haemorrhage (estimated blood loss ≥ 1000 mL)
● Use of general anaesthesia or regional anaesthesia
Neonatal outcomes
● Apgar score < 7 at 5 minutes
● Umbilical arterial pH < 7.10
● Shoulder dystocia
● Neonatal scalp injuries (cephalohematoma or subgaleal haematoma)
● Admission to neonatal intensive care unit
● Neonatal hyperbilirubinaemia requiring phototherapy at a mínimum
Covariates
The following maternal, fetal, and procedural variables will be collected to characterise the study population and adjust analyses as appropriate:
● Participant age (years)
● Country of professional training (Spain, Europe, outside Europe)
● Operator seniority (attending physician or resident)
● Operator experience (years of professional experience)
● Training method (simulation or in vivo)
● Indication for operative vaginal birth
● Type of obstetric vacuum device used
● Number of cup detachments (pop-off)
● Presence of caput succedaneum
● Analgesia use
● Fetal head position
Data Collection and Management
Clinical data will be collected retrospectively and prospectively from routinely recorded medical records at each participating centre. All data will be anonymised prior to entry into a secure, password-protected electronic database accessible only to authorised members of the research team.
Data quality checks and consistency reviews will be performed regularly throughout the study period.
Statistical Analysis
Analyses will be conducted on an intention-to-treat basis at the centre level, whereby each hospital will be considered exposed to the intervention from the predefined post-intervention period, regardless of individual clinician attendance.
Statistical Framework and Notation
Let i index individual births (i = 1,…, N), h(i) denote the hospital (cluster), and t(i) the discrete time period (step). Exposure to the intervention is defined at the hospital–time level. Hospitals are considered unexposed during the pre-intervention period and exposed from the post-intervention period onwards.
Primary Analysis
The primary outcome will be analysed using modified Poisson regression with a log link and robust (sandwich) variance estimation to estimate adjusted risk ratios (RRs).
Models will include:
● Fixed effects for hospital to account for clustering
● Fixed effects for time period to adjust for secular trends
Secondary Analyses
Secondary binary outcomes will be analysed using the same modified Poisson framework. Count outcomes will be analysed using Poisson or negative binomial regression as appropriate. Continuous outcomes will be analysed using linear regression or transformed if required to meet model assumptions.
Sensitivity analyses will explore alternative specifications of time (categorical versus continuous) and the impact of delayed or incomplete implementation at individual centres.
All analyses will be performed using R statistical software, with a two-sided significance level of 0.05.
Sample Size Considerations
Formal sample size calculation for stepped-wedge designs is complex due to clustering and repeated measures. Therefore, a pragmatic approach was adopted.
Based on published literature, the baseline failure rate of vacuum-assisted vaginal birth as the first intended instrument is estimated at approximately 20%. The study is designed to detect a clinically meaningful relative reduction to 10–12% following implementation of the training programme.
Assuming an intracluster correlation coefficient between 0.05 and 0.10 and accounting for clustering and temporal effects, a minimum sample size of approximately 3,000 vacuum-assisted vaginal births is considered sufficient to detect the anticipated effect size with adequate statistical power.
Given the national scope of REMMP, the expected number of eligible cases during the study period is anticipated to exceed this threshold.
Ethics Approval
The study involves the analysis of routinely collected anonymised clinical data. Ethical approval was granted by the CEIm Girona (Code: 2024.070) on 05/08/2024. All participants sign a written informed consent form authorising the use and sharing of their data for research purposes. All procedures are conducted in accordance with national data protection regulations and the Declaration of Helsinki.
Multimedia Appendices
● Multimedia Appendix 1: SPIRIT schedule of enrolment, intervention, and assessments
● Multimedia Appendix 2: STROBE Checklist (for Observational Studies)
References
1. Trends in maternal mortality: 2000 to 2017: estimates by WHO, UNICEF, UNFPA, World Bank Group and the United Nations Population Division [en línea]. 2019 [Consultado el 11 de octubre de 2021]. Disponible en: https://www.unfpa.org/featured-publication/trends-maternal-mortality-2000-2017.
2. Murphy DJ. Medico-legal considerations and operative vaginal delivery. Best Pract Res Clin Obstet Gynaecol 2019;56:114-24.
3. Feinmann J. Why do doctors still use forceps when they killed our baby? Daily Mail. 22 de febrero de 2010 [Consultado el 11 de octubre de 2021]. Disponible en: https://www.dailymail.co.uk/health/article-1253013/Forceps-killed-baby-doctors-using-them.html.
4. Gale A, Siassakos D, Attilakos G, et al. Operative vaginal birth: better training for better outcomes. BJOG 2014;121:643-4.
5. Registro Español de Seguridad Obstétrica 2021. Grupo Español de Seguridad Obstétrica.
6. Murphy DJ, Liebling RE, Verity L, et al. Early maternal and neonatal morbidity associated with operative delivery in the second stage of labor: a cohort study. Lancet 2001;358:1203-7.
7. Gale A, Siassakos D, Attilakos G, et al. Operative vaginal birth: better training for better outcomes. BJOG 2014;121:643–4.
8. RCOG. Operative vaginal delivery: green–top guideline No. 26 [en línea]. 2011 [Consultado el 11 de octubre de 2021]. Disponible en: https://www.rcog.org.uk/globalassets/documents/guidelines/gtg_26.pdf.
9. Mobbs N, Williams C, Weeks A. Humanising birth: does the language we use matter? [en línea]. 2018 [Consultado el 11 de octubre de 2021]. Disponible en: http://blogs.bmj.com/bmj/2018/02//humanising-birth-does-the-language-we-use-matter/.
10. NICE. Intrapartum care for healthy women and babies [en línea]. 2014 [Consultado el 11 de octubre de 2021]. Disponible en: https://www.nice.org.uk/guidance/cg190/resources/intrapartum-care-for-healthy-women-and-babies-pdf-35109866447557.
11. Registro Español de Seguridad Obstétrica 2022. Grupo Español de Seguridad Obstétrica.
12. Moreau R, Pham MT, Brun X, Redarce T, Dupuis O. Assessment of forceps use in obstetrics during a simulated childbirth. Int J Med Robot. 2008;4(4):373-80. Disponible en: https://doi.org/10.1002/rcs.222.
13. O'Grady JP, Pope CS, Patel SS. Vacuum extraction in modern obstetric practice: a review and critique. Curr Opin Obstet Gynecol. 2000;12(6):475-80. Disponible en: https://doi.org/10.1097/00001703-200012000-00003.
14. Vacca A. Reducing the risks of a vacuum delivery. Fetal and Maternal Medicine Review 2006; 17:4. DOI: I O./ O I 7/S096553950600 l 823
Appendix 1. SPIRIT Schedule of Enrolment, Intervention, and Assessments
Title: SPIRIT schedule for the stepped-wedge simulation-based training programme in vacuum-assisted vaginal birth
Assumptions used for the schedule:
● Step duration: Step duration varies according to the period
● Total periods per centre: 3 periods (t1-t3)
● Washout period: training implementation; no clinical outcome data collected
● Each centre transitions once from pre-intervention to post-intervention
Activity / Variable Pre-intervention
Clinical
outcome
collection
(t1) Washout
(t 2) Post-intervention
Clinical outcome collection
(t 3)
Time (months) -12 0 6
Centre selection and activation (agreements, audit, standardisation) X
Training of local faculty (train-the-trainer) x
Simulation training implementation (workshop + checklist + debriefing) X
Process measures (number of sessions, proportion of staff
trained) X
Primary outcome (failure of first instrument) X X
Maternal secondary outcomes X X
Neonatal secondary outcomes X X
Educational measures (Likert scale, checklist performance) X
SPIRIT Table. Outcome data are not collected during the washout period to minimise contamination, as varying proportions of clinicians are trained during this phase.
Appendix 2. STROBE Checklist (for Observational Studies)
STROBE Item Recommendation Location in Manuscript
1 Study design indicated in title/abstract Title; Abstract – Methods
2 Scientific background and rationale Abstract – Background; Introduction
3 Specific objectives Abstract – Objective
4 Study design Methods – Study Design
5 Setting, locations, dates Methods – Study Setting; Results (Current Study Status)
6 Participants and eligibility criteria Methods – Participants
7 Variables clearly defined Methods – Outcome Measures; Covariates
8 Data sources and measurement Methods – Data Collection and Management
9 Bias Methods – Study Design; Intervention (washout period)
10 Study size Methods – Sample Size Considerations
11 Quantitative variables Methods – Statistical Analysis
12 Statistical methods Methods – Statistical Analysis
13 Participants flow Multimedia Appendix 2
14 Descriptive data Planned (protocol)
15 Outcome data Planned (protocol)
16 Main results Not applicable (protocol)
17 Other analyses Methods – Secondary Analyses
18 Key results Not applicable (protocol)
19 Limitations Planned for final results paper
20 Interpretation Planned for final results paper
21 Generalisability Introduction; Study Setting
22 Funding Results (Current Study Status) Results: At the time of manuscript submission, ethical approval and funding were obtained (July 2024). Centre recruitment and site agreements are ongoing. Data collection is planned to begin in the fourth quarter of 2024 and is expected to be completed by mid-2026. Data analysis will commence thereafter, with results anticipated for publication in late 2026. Conclusions: This paper describes the protocol for a multicentre study designed to assess the effectiveness of a simulation-based training programme in vacuum-assisted vaginal birth. By providing robust evidence on its impact on operative vaginal birth practice and outcomes, the study aims to address important methodological shortcomings of previous research and to inform the development of standardised training strategies in this field.
