Original Research

Development and validation of a labour epidural analgesia documentation instrument for South African hospitals

Sean Chetty, Elizabeth J. Jacobs, Estie Mostert

Received: 07 Jan. 2026; Accepted: 20 May 2026; Published: 04 Aug. 2026

Copyright: © 2026. The Authors. Licensee: AOSIS.
This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license (https://creativecommons.org/licenses/by/4.0/).

Abstract

Background: Labour epidural analgesia is the gold standard for intrapartum pain relief, yet documentation practices remain inconsistent, particularly in resource-constrained settings. Inadequate documentation creates patient-safety risk and limits audit and governance. No formally validated labour epidural documentation instrument has been described in the published literature.

Aim: This study aimed to develop and content-validate a standardised labour epidural analgesia documentation instrument for the South African health sector.

Setting: Study was conducted in the South African health sector.

Methods: A two-phase instrument development and content validation study were conducted. In the development phase, a concept instrument was generated from a literature review and refined through expert consensus. In the quantification phase, national anaesthesiology experts independently rated items for relevance and importance using a four-point Likert scale. Item-level and scale-level content validity indices and inter-rater agreement values were calculated using predefined thresholds.

Results: The initial 68-item concept instrument was refined to 56 items. Quantitative content validation resulted in a final instrument comprising four domains: pre-procedural assessment, intra-procedural documentation, post-procedural follow-up and chart layout. Most retained items demonstrated high item-level content validity (I-CVI ≥ 0.78), with inter-rater agreement values from 0.80 to 1.00. The layout domain had a borderline scale-level content validity index (S-CVI = 0.79), with all retained items meeting item-level validity criteria.

Conclusion: This study presents a content-validated labour epidural analgesia documentation instrument, defining a consensus-based minimum dataset for clinical practice.

Contribution: The findings of this research provide a foundation for patient-safety initiatives, audit, training and future electronic anaesthesia record development.

Keywords: labour epidural; clinical documentation; content validity; patient safety; obstetric anaesthesia.

Introduction

Neuraxial labour analgesia is widely regarded as the gold standard for intrapartum pain relief, providing effective analgesia while allowing the obstetric anaesthetist to optimise both maternal and foetal well-being.^1,2 Despite its recognised benefits, access to labour epidural analgesia in many South African public-sector hospitals remains limited and is often reserved for higher-risk patients.^3,4,5 In such contexts, the quality of anaesthetic documentation assumes heightened importance, as prolonged care episodes, frequent reassessment and multiple handovers increase vulnerability to communication errors and adverse events.

South African audit data indicate that labour epidural analgesia remains relatively uncommon in public-sector hospitals and is frequently provided in higher-risk clinical contexts.^3,4,6 Where services are available, complication rates are comparable to those reported in high-income settings.^3,4 In such environments, characterised by low procedural volume, complex case-mix, prolonged care episodes and frequent handovers, the quality of clinical documentation assumes particular importance for patient safety, continuity of care and clinical governance.^7,8 Both locally and internationally, anaesthetic record-keeping has been shown to be variable and often suboptimal.^9,10 In a South African audit, only one-third of anaesthetic records met minimum documentation standards defined by the Health Professions Council of South Africa.¹⁰ These findings underscore the need for a standardised and validated documentation framework in obstetric anaesthesia practice.

Against this backdrop, the absence of a standardised and validated labour epidural analgesia record represents a significant gap in obstetric anaesthesia practice within the South African health sector. In settings where labour epidurals are infrequent, frequently performed in higher-risk patients and delivered within complex health-system constraints, robust documentation assumes heightened importance. A validated documentation tool has the potential to support safer clinical care, improve supervision and handover, strengthen clinical governance and enhance medico-legal defensibility.

In the development of healthcare measurement instruments, content validity is recognised as the foundational measurement property, reflecting the extent to which an instrument captures the essential elements of the construct of interest. These methodological principles have been formalised within the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) framework, which provides internationally accepted guidance for evaluating content validity through structured expert assessment.^11,12 Although COSMIN was originally developed for patient-reported outcome measures, its principles regarding expert evaluation of item relevance and comprehensiveness are applicable to structured clinical documentation instruments.

The aim of this study was to develop and content-validate a standardised labour epidural analgesia documentation tool that reflects expert consensus on essential documentation elements and provides a baseline standard for obstetric anaesthesia practice.

Research methods and design

Study design

This study employed a two-stage instrument development and content validation design to develop a standardised labour epidural analgesia documentation tool. The study comprised an initial instrument development phase followed by a quantitative content validity and inter-rater agreement assessment.

Expert selection and participation

Participants were South African specialist anaesthesiologists with recognised expertise in obstetric anaesthesia and active involvement in labour epidural analgesia services. Purposive sampling was employed to recruit experts with substantial clinical exposure to labour epidural practice within the public and/or private sectors.

The development-stage panel comprised of 10 specialist anaesthesiologists from three tertiary academic hospitals in Gauteng province (Chris Hani Baragwanath Academic Hospital, Charlotte Maxeke Johannesburg Academic Hospital and Rahima Moosa Mother and Child Hospital). Eight participants were Grade 1 specialists, one was Grade 2, and one was Grade 3. Six of the 10 (60%) were female. All were actively practising in urban public-sector settings.

For the quantification phase, 24 national experts were invited to participate. Experts were invited from multiple provinces and from both public- and private-sector practice settings to ensure a range of South African obstetric anaesthesia perspectives.

The development-stage and quantification-stage expert panels were recruited independently. Participation in one phase did not guarantee or require participation in the other.

Years of post-specialist experience and demographic variables such as race and ethnicity were not formally recorded, as the primary inclusion criterion was recognised specialist status with active involvement in obstetric anaesthesia practice.

Instrument development stage

Instrument development was guided by established principles for content validity determination as described by Lynn and subsequently refined by Polit and Beck.^13,14,15 These principles are consistent with methodological standards subsequently formalised by the COSMIN initiative, which emphasises expert evaluation of item relevance and comprehensiveness during early-phase instrument development.^12,16 This stage comprised three sequential steps: domain identification, item generation and item formation (Figure 1).

FIGURE 1: Methodological steps showing stages of instrument development.

The initial concept instrument was generated from a comprehensive literature review undertaken to identify domains and candidate items relevant to labour epidural analgesia documentation. This review was conducted to support domain identification and item generation in accordance with established instrument development methodology, rather than as a formal systematic review.

The concept instrument was distributed electronically to development-stage experts prior to a structured peer review meeting. During this stage, items were reviewed for relevance and importance to clinical practice and refined through consensus-based discussion. Items were rated using a four-point Likert scale reflecting perceived relevance and importance:

1 = irrelevant

2 = relevant but unimportant

3 = relevant and important

4 = relevant and essential

Items rated as irrelevant were removed. Additional items were incorporated based on expert recommendations. Items were revised, expanded or repositioned as required to improve clarity and alignment with clinical workflow. The resulting instrument constituted the rated instrument that progressed to quantitative content validation.

Instrument quantification stage

Quantitative content validation was undertaken using a national panel of expert anaesthesiologists drawn from both public and private sectors. Experts independently rated each item in the rated instrument using the same four-point relevance scale applied during the development stage.

Data analysis

Item-level content validity indices (I-CVI) were calculated as the proportion of experts assigning a rating of 3 or 4 to each item. Given a 12-expert panel, an I-CVI threshold of ≥ 0.78 was applied to determine item retention. This threshold is consistent with published recommendations for expert panels of six or more participants.^14,17

To assess agreement beyond chance, modified kappa statistics were calculated for all items in the validation dataset using the method described by Polit et al.^14,17 The probability of chance agreement (Pc) was calculated from the number of experts rating an item as relevant (score 3 or 4), and modified kappa was then calculated as K* = (I-CVI − Pc)/(1 − Pc). Because only fully completed instruments were included in the final analysis, no missing ratings were present. Modified kappa values ≥ 0.80 were interpreted as strong agreement beyond chance. Items not meeting the predefined I-CVI threshold were excluded from the final instrument.

Scale-level content validity was evaluated using the average scale CVI (S-CVI/Ave), with values ≥ 0.90 interpreted as indicative of excellent content validity.^14,15.

To further assess the consistency of expert ratings beyond chance agreement, inter-rater (IR) agreement values were calculated for retained items using the modified kappa statistic. IR agreement values ≥ 0.80 were considered acceptable.

Data were captured and analysed using Microsoft Excel^®.

The methodological approach to content validity was informed by COSMIN guidance for instrument development and content validity studies.^11,12,16 This article was prepared with reference to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) principles, where applicable, to reporting of the study design and participant characteristics.

Ethical considerations

Ethical approval for this study was obtained from the Human Research Ethics Committee (Medical) of the University of the Witwatersrand (M140129). Participation was voluntary for all expert contributors, including members of the peer group involved in instrument development and the national expert panel involved in the validation phase. Potential participants were provided with an information letter describing the purpose and methodology of the study, and consent to participate was implied by participation in the peer group process or completion and return of the study instrument.

Data were collected and analysed anonymously. No personally identifiable information was recorded, and responses were de-identified prior to analysis. Access to study data was restricted to the research team and stored on password-protected devices.

The study involved expert participants providing professional opinion and did not involve patients, clinical interventions or the use of identifiable patient data.

Results

Instrument development stage: Item refinement and reduction

The initial concept instrument (Online Appendix 2), developed from a structured literature review, comprised 68 candidate items relevant to labour epidural analgesia documentation.

The development-stage expert panel comprised 10 specialist anaesthesiologists from three Gauteng tertiary academic hospitals; eight were Grade 1 specialists, one was Grade 2, and one was Grade 3. Six of the ten participants (60%) were female.

Following expert review and consensus-based refinement, 18 items were removed or reclassified, two new items were added, and four items were modified, resulting in a 56-item rated instrument (Online Appendix 1) that progressed to quantitative content validity analysis.

Items were removed predominantly for one of four reasons:

• limited or absent relevance to labour epidural analgesia practice
• duplication of information routinely documented elsewhere in the clinical record
• inclusion of information outside the anaesthetist’s scope of practice or responsibility
• contextual irrelevance within South African public-sector labour ward settings.

Examples of excluded items included isolated obstetric variables, form design characteristics (such as paper type or colour) and patient or ward descriptors that were considered outside the scope of labour epidural anaesthesia. Several items were repositioned within the instrument to align with the clinical workflow. For example, documentation of sensory block level was relocated to the monitoring section to reflect time-dependent reassessment, and items relating to mode of delivery were expanded to capture anaesthetic management should labour progress to caesarean section.

Two additional items were incorporated following expert recommendation to enhance clinical completeness, and four items were revised to improve clarity, scope or interpretability.

All items retained in the rated instrument met the predefined criteria for relevance and were advanced to the quantification stage for formal content validity assessment. A detailed summary of item-level modifications undertaken during the development stage is presented in Table 1.

Instrument quantification stage: Content validity and inter-rater agreement

Twenty-four experts were invited to participate in the quantification phase. Sixteen returned completed instruments (response rate 67%). Of these, 12 were fully completed and met predefined inclusion criteria and were included in the final analysis (50% of those invited). All 12 included experts completed ratings for all items in the 56-item instrument. The validation panel included experts from Gauteng (n = 8), KwaZulu-Natal (n = 2), Western Cape (n = 1) and North West province (n = 1). Eighty-three per cent were working primarily in the public sector, and 67% were male.

Quantitative content validation was performed on the 56-item rated instrument.

Across the instrument, I-CVI values ranged from 0.42 to 1.00, indicating variation in expert agreement across domains. Items failing to meet the predefined threshold were excluded from the final instrument.

Item-level I-CVI values, IR agreement values and decisions to retain in the instrument are presented in Table 2 to Table 5, along with the S-CVI/Ave values for each domain.

Demographic and pre-procedural assessment domain

This domain comprised 21 items at the quantification stage. Of these, 18 items met the predefined I-CVI threshold (≥ 0.78) and were retained in the final instrument.

I-CVI values for retained items in this domain ranged from 0.80 to 1.00, indicating good to excellent content validity. Three items (previous labour epidural, cervical dilatation and parity) with an I-CVI range of 0.40–0.75, failed to meet the predefined I-CVI threshold and were excluded.

Inter-rater agreement values for retained items in this domain ranged from 0.80 to 1.00. Items that failed to meet I-CVI thresholds also demonstrated lower IR agreement, reinforcing the decision to exclude them.

Intra-procedural documentation domain

The intra-procedural domain included 17 items at the quantification stage. Sixteen of 17 items in the intra-procedural domain met the predefined I-CVI threshold, with I-CVI values ranging from 0.83 to 1.00.

Items relating to technical aspects of epidural placement, physiological monitoring and documentation of complications achieved unanimous expert agreement (I-CVI = 1.00). A single item (documentation of position - I-CVI = 0.67) failed to meet the I-CVI threshold and was excluded.

Inter-rater agreement values for retained intra-procedural items ranged from 0.83 to 1.00.

Post-procedural follow-up domain

At the quantification stage, the post-procedural domain comprised 9 items. Eight items met the predefined I-CVI threshold and were retained.

Retained items demonstrated I-CVI values ranging from 0.92 to 1.00, particularly for documentation of analgesic adequacy, complications and epidural catheter removal. The experiential item ‘Would the patient want an epidural again?’ failed to meet the predefined I-CVI threshold (I-CVI = 0.50) and was excluded.

IR agreement values for retained items ranged from 0.92 to 1.00, confirming consistent expert judgement across this domain.

Layout and format domain

The layout and format domain initially comprised 10 items. Four items met content validity criteria and were retained.

Excluded items demonstrated lower I-CVI values (≤ 0.75), indicating lower levels of expert agreement. Items retained within this domain included time-based grids, standardised date and time formats and documentation of test results.

Inter-rater agreement for retained layout items ranged from 0.92 to 1.00, supporting acceptable agreement despite greater variability in expert opinion within this domain.

Scale-level content validity and final instrument composition

The scale-level content validity index met or exceeded commonly cited benchmarks in three of the four domains. For the Layout of Chart domain, expert agreement demonstrated greater variability, reflecting differences in opinion regarding formatting and design features rather than disagreement about clinical content. The resulting S-CVI for this domain was 0.79.

Item-level content validity and inter-rater agreement values for retained layout items all met predefined thresholds (Table 2 to Table 5), and no retained item demonstrated poor agreement.

Following quantitative content validation, 10 additional items were excluded, resulting in a final 46-item labour epidural analgesia documentation instrument. The final instrument is organised into four domains: demographic and pre-procedural assessment, intra-procedural documentation, post-procedural follow-up and essential layout elements. The validated instrument is presented in Table 6.

Discussion

This study demonstrates that South African obstetric anaesthesia experts share a high degree of consensus regarding the safety-critical elements that should be documented during labour epidural analgesia. Through formal content validation, the findings define a consensus-based minimum dataset for labour epidural documentation nationally and address a gap that persists in the contemporary obstetric anaesthesia literature.

Labour epidural analgesia differs from many anaesthetic encounters in that it unfolds over prolonged periods, requires repeated reassessment and involves multiple clinical handovers. In such settings, documentation functions prospectively, supporting continuity of care, supervision, escalation of concerns and shared situational awareness.

The importance of documentation to patient safety in anaesthesia is well established. Poor or incomplete anaesthetic records have been identified as latent system failures that contribute to adverse events and limit the ability to learn from harm.^7,8 In obstetric anaesthesia, audits and confidential enquiries repeatedly highlight inadequate documentation of neuraxial procedures and monitoring as a contributory factor to adverse outcomes.^18,19

In the present study, intra-procedural and post-procedural domains demonstrated consistently high content validity indices and strong inter-rater agreement, indicating robust expert consensus regarding safety-critical elements of labour epidural care. Documentation of procedural details, physiological monitoring, sensory block assessment and complications was universally endorsed, aligning with international neuraxial safety guidance that emphasises meticulous documentation as a core component of safe practice.^20,21

Within the scope of the literature reviewed during instrument development, we did not identify any formally validated labour epidural analgesia documentation instruments reported using structured content validity or psychometric methodologies. Despite widespread recognition of the importance of documentation in obstetric anaesthesia, we did not identify any peer-reviewed studies reporting the development and quantitative content validation of a labour epidural documentation tool using established content validity and IR agreement methodologies.

Instead, the existing literature in this field can be broadly divided into three strands. Firstly, it comprises professional guidelines and consensus statements that outline recommended documentation elements for labour epidural analgesia but do not report formal instrument development or validation.²² Secondly, it consists of audit and service-evaluation studies that identify deficiencies in labour epidural documentation and monitoring practices, without providing validated solutions.¹⁹ Thirdly, it includes educational and simulation-based tools designed to assess technical competence in epidural placement, which are not intended for routine clinical documentation.^23,24

The present study addresses this methodological gap by applying established content validity principles to define and validate a minimum dataset for labour epidural analgesia documentation, grounded in expert consensus and quantitative analysis. In doing so, it distinguishes documentation as a safety-critical clinical process rather than an administrative task, providing a baseline standard for practice, audit and future system development.

Although professional bodies such as the Royal College of Anaesthetists and the Society for Obstetric Anesthesia and Perinatology provide guidance on neuraxial safety practices, no formally validated labour epidural documentation instruments have been endorsed internationally. The present study therefore addresses a methodological and governance gap rather than duplicating existing society guidance.

Lower expert agreement regarding the inclusion of obstetric variables such as parity, cervical dilatation and previous labour epidural experience reflects a boundary-of-practice distinction, rather than a dismissal of clinical relevance. Additional contextual variables, such as provider grade and elective versus emergency categorisation of caesarean section, were likewise not retained following quantitative content validation. Certain safety-relevant clinical variables that some readers may consider important, such as anticoagulant or antiplatelet exposure, baseline neurological symptoms, infection screening or test dose documentation, were also not retained as discrete labour epidural-specific items. Their exclusion reflects the fact that these variables are typically captured within broader anaesthetic pre-assessment and peri-procedural records, and the instrument was designed to avoid unnecessary duplication while defining a labour epidural-specific minimum dataset. It is not a determination that these variables lack clinical importance. In multidisciplinary labour ward environments, anaesthetic documentation must prioritise information that directly informs anaesthetic decision-making, monitoring and follow-up while avoiding unnecessary duplication of data recorded elsewhere. This prioritisation is particularly important in high-volume obstetric units, where excessive documentation risks obscuring safety-critical information. These contextual variables may therefore be incorporated into locally adapted versions of the instrument where service configuration, governance requirements or audit objectives warrant their routine capture.

Although the Layout of Chart domain achieved an S-CVI marginally below commonly cited thresholds for excellent content validity, methodological guidance cautions against rigid application of scale-level cut-offs during early-phase, formative instrument development. COnsensus-based Standards for the selection of health Measurement INstruments principles emphasise that, at this stage, instruments should prioritise comprehensive coverage of clinically relevant content and practical usability, rather than early exclusion of items based solely on aggregate validity indices.²⁵

Consistent with this guidance, layout items demonstrating borderline content validity indices were retained for refinement rather than automatic exclusion. Importantly, all retained layout items met predefined item-level content validity and inter-rater agreement thresholds, supporting their inclusion for further iterative development.¹⁷

The retention of essential layout features, such as time-based grids and standardised date and time formats, aligns with human-factors literature demonstrating that clear temporal structure and standardisation support situational awareness and reduce cognitive load in high-pressure clinical environments.^8,26 The item ‘Test results’ was retained as a deliberately broad documentation field to allow recording of investigations where clinically relevant to the anaesthetic episode. The instrument does not prescribe routine laboratory testing.

Documentation practices in anaesthesia, particularly within public-sector systems, evolve slowly, and many institutions continue to rely on locally developed paper-based records without formal validation. Importantly, the targeted literature review did not identify any validated labour epidural documentation instruments in this field, reinforcing the contemporary relevance of this work.

Moreover, labour epidural analgesia remains relatively uncommon in many South African public hospitals and is frequently reserved for higher-risk patients.¹⁹ In such contexts, the consequences of inadequate documentation are amplified rather than diminished. The validated tool therefore provides a baseline standard against which future audit cycles, training interventions and digital implementations can be evaluated. The present study was designed to define the content of a minimum dataset rather than to prescribe a single final chart format, as operationalisation into a paper or electronic tool will depend on local workflow, governance requirements and service context.

In achieving its stated aim, this study provides a content-validated labour epidural analgesia documentation tool that reflects expert consensus on essential documentation elements and aligns with minimum regulatory expectations. The instrument provides a consensus-based baseline standard aligned with recognised regulatory and professional documentation expectations in South African obstetric anaesthesia practice, serving as a foundation for patient safety, training, audit and future system-level evaluation.

Limitations and future directions

This study has several limitations. Firstly, the data were collected in 2015; however, a targeted literature review did not identify any formally validated labour epidural documentation instruments published since that time, supporting the ongoing relevance of the findings. Secondly, the study focused exclusively on content validity and expert agreement and did not assess usability, clinical workflow integration or impact on documentation completeness or patient outcomes. Accordingly, the study was not intended to produce or validate a universally applicable end-user chart, and local adaptation will be required when operationalising the minimum dataset in specific practice settings. Thirdly, expert input was limited to anaesthesiologists - while appropriate for defining anaesthesia-specific documentation requirements, future work should incorporate multidisciplinary perspectives, including obstetric, midwifery, patients and medico-legal stakeholders. These limitations reflect the study’s intended role as a foundational instrument development rather than a final implementation evaluation. In accordance with COSMIN recommendations, further evaluation of additional measurement properties, including usability, reliability and implementation performance, is planned as a subsequent phase of instrument development.

Conclusion

This study defines, through expert consensus and quantitative content validation, the essential elements of labour epidural analgesia documentation within a South African context. In the absence of internationally validated labour epidural documentation instruments, the findings address a persistent and under-recognised gap in obstetric anaesthesia practice. By reframing documentation as a patient-safety intervention rather than an administrative task, the content-validated tool provides a practical foundation for improving clinical care, training and governance in contemporary obstetric anaesthesia.

Acknowledgements

The authors would like to thank Juan Scribante and Helen Perrie for their valuable support and guidance during the development and execution of this study.

The research presented in this article formed part of Elizabeth J. Jacobs’ postgraduate studies and was originally conducted as part of their MMed (Anaes) thesis titled ‘The Development and validation of an instrument for labour epidural analgesia recordkeeping in hospitals in Southern Gauteng’, submitted to the Department of Anaesthesia, Faculty of Health Sciences, University of the Witwatersrand in 2015, under the supervision of Sean Chetty and Estie Mostert. The original thesis is publicly available at: http://hdl.handle.net/10539/22259. The thesis was submitted in partial fulfilment of the requirements for the MMed (Anaes) degree. Portions of the thesis have been revised, updated and adapted for publication as a journal article.

Portions of the data analysis (calculation of the modified kappa values) were generated with assistance from ChatGPT, Version 5.2. The final code and interpretations were developed and validated by the authors. The content was reviewed and edited by the authors, who take full responsibility for its accuracy.

The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article. The author, Sean Chetty, serve as an editorial board member of this journal. The peer review process for this submission was handled independently, and the author had no involvement in the editorial decision-making process for this article. The authors have no other competing interests to declare.

Sean Chetty: Conceptualisation, Methodology, Supervision, Writing – original draft. Elizabeth J. Jacobs: Conceptualisation, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Writing – review & editing. Estie Mostert: Conceptualisation, Project administration, Supervision, Writing – review & editing. All authors reviewed the article, contributed to the discussion of results, approved the final version for submission and publication and take responsibility for the integrity of its findings.

The authors received no financial support for the research, authorship and/or publication of this article.

Data sharing is not applicable to this article as no new data were created or analysed in this study.

The views and opinions expressed in this article are those of the authors and are the product of professional research. They do not necessarily reflect the official policy or position of any affiliated institution, funder, agency or that of the publisher. The authors are responsible for this article’s results, findings and content.

References