Quality improvement methodology used in enhanced recovery after caesarean delivery implementation studies: a narrative review and author survey

Abstract

Background

Enhanced recovery after caesarean delivery (ERAC) is gaining popularity and has been shown to improve maternal and neonatal outcomes. We aimed to identify the presence and type of quality improvement (QI) methodology used in ERAC studies through an exploratory analysis, including author surveys and literature review.

Methods

We performed a literature search using four databases (MEDLINE through PubMed, Cumulative Index of Nursing and Allied Health Literature, Web of Science and Embase) to identify ERAC studies. Studies were considered if they compared an ‘enhanced’ or ‘fast track’ protocol to a control group and evaluated more than one system or outcome. The adopted QI methodology was evaluated through a standardised questionnaire developed by the authors which was sent to the authors of included studies.

Results

We identified 29 studies. A standardised questionnaire was applied to all included papers to evaluate the presence and type of QI methodology, and 24 authors were approached to complete the survey. We received results from 15 authors, yielding a response rate of 63%. 40% of authors reported use of defined QI methodology, the majority using the Model for Improvement. The QI components most used by those not reporting use of a defined methodology were engagement with multidisciplinary key stakeholders (80%), use of key drivers (87%) and process mapping (60%). Most authors reported use of traditional statistical methodology when analysing results (73%), and 60% reported use of Standards for Quality Improvement Reporting Excellence guidelines. The mean duration of baseline data collection by non-randomised controlled trial studies was 11 months and 9.9 months after implementation.

Conclusion

There is a large variation in the QI practices used in protocol implementation and publication of ERAC studies. The minority of authors report the use of a defined QI methodology and very few report the use of standardised tools in their published works.

PROSPERO registration number

CRD42023399418.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Enhanced recovery after caesarean delivery protocols improve maternal outcomes, but the quality improvement methodologies used for implementation have not been systematically evaluated.

WHAT THIS STUDY ADDS

• Only 8% of published enhanced recovery after caesarean delivery (ERAC) studies reported using formal quality improvement (QI) methodology, yet 40% of surveyed authors actually employed these approaches, revealing significant gaps between practice and reporting.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Standardised QI reporting guidelines are needed for ERAC studies to improve reproducibility and identify effective implementation strategies.

Introduction

Caesarean delivery (CD) remains the most performed inpatient surgical procedure globally. Thus, enhancing postoperative recovery and optimising maternal and neonatal outcomes has the potential to impact millions of people.¹ Enhanced recovery after caesarean delivery (ERAC) protocols have emerged as promising strategies to address these challenges, offering a structured, evidence-based approach to peripartum care.^{2 3} ERAC protocols typically include preoperative, intraoperative and postoperative interventions. The Society for Obstetric Anesthesia and Perinatology² and Enhanced Recovery After Surgery Society⁴,⁶ recommendations encompass: preoperative (counselling, education, patient optimisation, reduced fasting); intraoperative (neuraxial anaesthesia, hypotension management, anti-emetics, euvolemia, antibiotics, uterotonics, multimodal analgesia, breastfeeding promotion); and postoperative interventions (thromboprophylaxis, early feeding and mobilisation, continued multimodal analgesia, rest periods, facilitated discharge). A previous systematic review identified significant heterogeneity, with 40 different interventions used across just 11 studies.⁷ ERAC offers several benefits to institutions and patients, including reduced length of stay and time to ambulation, increased satisfaction among patients, reduced complication rates and improved pain and satisfaction outcomes, without increasing readmission rates. Additionally, ERAC fosters collaboration among diverse multidisciplinary team members, including anaesthesiologists, nurses, obstetricians, paediatricians, lactation specialists and midwives, reflecting its comprehensive approach to holistic patient care.

Due to the limitations of randomised controlled trials in assessing systematic ERAC changes, most studies evaluating the impact of ERAC use a before-and-after study design. Despite this study design being mostly amenable to quality improvement (QI) methodology,⁷,⁹ the QI methodologies employed to implement and evaluate ERAC protocols have not been described. To bridge this gap in knowledge, our study aimed to review published ERAC studies and contact study authors to identify and characterise QI methodologies used. By exploring the range of methodologies, we sought to gain insights into current practices and identify areas for enhancement in future research and QI endeavours.

Methods

A literature search was conducted by a medical librarian (LB) using the following databases: PubMed MEDLINE, Embase, Web of Science and Cumulative Index of Nursing and Allied Health Literature. The search focused on updating the search performed in a previously published systematic review in 2021.⁸ All databases were searched from October 2020 until 8 November 2022. All searches were conducted in English, using the same search strategy previously used (online supplemental appendix). The original search yielded 341 results, which were narrowed to 216 results after manual deduplication using Rayyan systematic review software. After title and abstract review, 15 additional full-text articles were sought for retrieval. All 15 articles were found and used for full-text review in addition to the 9 full text articles from the previously published systematic review. In total, 24 studies were eligible for review. Figure 1 outlines the number of articles included. Studies were considered eligible if they implemented an ‘enhanced’ or ‘fast track’ protocol, comparing ERAC protocols to a control group and evaluated more than one outcome or system. Studies were excluded if they were published as abstracts, letters or other non-full-text publications.

Figure 1. Summary of literature search. CINHAL, Cumulative Index of Nursing and Allied Health Literature.

An updated search was performed in January of 2025 and yielded five more eligible studies.¹⁰,¹⁴ Due to feasibility, the author survey was not performed for these publications, but analysis of reported methodology was again done by author KF and reported in table 1.

Table 1. Questionnaire applied to reviewed literature.

Standardised questionnaire applied to reviewed literature (N=29)	N (%)
Use of defined QI methodology	2 (6.9)
Use of a SMART goal	1 (3.4)
Engagement with a multidisciplinary team	18 (62.1)
Performance of a root cause analysis	0 (0.0)
Use of process mapping	1 (3.4)
Identification of key drivers	0 (0.0)
Use of SQUIRE or other guidelines	1 (3.4)

QI, quality improvement; SMART, Specific, Measurable, Achievable, Relevant, Time-bound; SQUIRE, Standards for Quality Improvement Reporting Excellence.

The study protocol was developed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure transparency and completeness in reporting. The protocol was registered prospectively with PROSPERO, registration # CRD42023399418. Rayyan software was used for efficient management of the literature search process, including screening of titles and abstracts, full-text review and data extraction.

The authors developed a standardised questionnaire (table 1) to ascertain the presence and type of QI methodology reported in each included study. The questionnaire was developed via consensus of author’s expert opinion and the scope was based on a literature review of QI methodologies. The questionnaire was answered by one of the authors (KF) using information provided in the published studies alone. A survey was subsequently emailed to corresponding authors of the included studies to gather additional insights into the methodologies employed in ERAC implementation and evaluation. In the event of non-response, coauthors were contacted via email. Authors were advised that completion of the survey would yield an acknowledgement at the time of publication. Survey questions are outlined in table 2.

Table 2. Questions used in the survey of authors publishing on enhanced recovery after caesarean (ERAC) and the author responses.

Question (N=15)	Response N (%)
Did you use a defined improvement methodology (eg, Lean, Six Sigma, etc) to implement the elements of the enhanced recovery protocol in your study?
Yes	6 (40%)
No	9 (60%)
If yes, please select the following improvement methodology that best describes your implementation practice
Lean	1 (16.7%)
Six-Sigma	1 (16.7%)
Lean Six-Sigma	0 (0%)
Model for Improvement (PDSA or PDCA)	4 (66.7%)

	Yes	No	Not sure	NA
If you did not use a defined improvement methodology, did you use any of the following components when implementing the elements of your enhanced recovery protocol
a) Objective goal setting, eg, using SMART goals?	10 (66.7%)	3 (20%)	0 (0%)	2 (13.3%)
b) Did you engage a group of multidisciplinary key stakeholders in developing your protocol or during the implementation phase?	12 (80.0%)	1 (6.7%)	0 (0%)	2 (13.3%)
c) Did you perform a root cause analysis when analysing your current state?	4 (26.7%)	8 (53.3%)	1 (6.7%)	2 (13.3%)
d) Did you engage in process mapping when analysing your current state?	9 (60.0%)	2 (13.3%)	2 (13.3%)	2 (13.3%)
e) Did you identify key drivers that would need to be achieved to realise successful implementation?	13 (86.7%)	0 (0%)	0 (0%)	2 (13.3%)
How did you test the impact of the changes implemented?
Traditional statistical methodology (eg, before/after t-test)	11 (73.3%)
Time Series Analysis (>2 time points)	1 (6.7%)
Statistical Process Control (SPC)	0 (0%)
Both traditional statistics and SPC	2 (13.3%)
Other	1 (6.7%)
Did you use SQUIRE guidelines when developing your protocol or publishing your work?
Yes	9 (60.0%)
No	5 (33.3%)
Other	1 (6.7%)

	Mean (SD)	Median (IQR)
If your study was not a randomised controlled trial, how long did you collect baseline data before your improvement implementation?	11 months (5.9)	12 months (6–12)
If your study was not a randomised controlled trial, how long did you test your improvement interventions after implementation?	9.9 months (5.5)	2 months (6–12)

PDCA, Plan Do Change Act; PDSA, Plan Do Study Act; SMART, Specific, Measurable, Achievable, Relevant, Time-bound; SQUIRE, Standards for Quality Improvement Reporting Excellence.

Results

Review of published studies

Results of the standardised questionnaire applied to eligible studies can be seen in table 1. Of the 24 ERAC studies included in our analysis, only two (6.9%) reported the adoption of defined QI methodology, namely Lean Six Sigma and Model for Improvement/Plan-Do-Study-Act (PDSA) cycling. Regarding QI components used outside of a defined methodology, one study documented the implementation of Specific-Measurable-Achievable-Relevant-Timebound (SMART) goals. 18 studies (62.1%) reported engagement with multidisciplinary teams. No studies reported the use of root cause analysis or the identification of key drivers. Only one study incorporated process mapping into their implementation strategy. Concerning the duration of improvement interventions, two studies did not specify a timeline, three studies reported a duration of 0–3 months, three studies 3–6 months and twelve studies >6 months. Notably, only one group of authors reported use of the Standards for Quality Improvement Reporting Excellence (SQUIRE) guidelines¹⁵ in their publications.

Survey of authors

Corresponding authors of all 24 initially included studies were approached with the survey. In total, 15 out of 24 groups responded regarding their QI methodology, yielding a response rate of 63%. Of the 15 responders, six required further inquiry to non-corresponding authors to achieve a survey response. Contrary to the results from the standardised questionnaire, 40% of authors reported the use of a defined QI methodology via our survey, with the Model for Improvement being the most used approach. For authors who did not report the use of a defined methodology, the QI components most frequently employed included engagement with multidisciplinary key stakeholders (80%), utilisation of key drivers (87%), use of SMART goals (67%) and process mapping (60%). The majority of authors (73%) reported applying traditional statistical methodology to analyse the impact of ERAC. Additionally, 60% of authors reported adhering to SQUIRE.¹⁵ In the 13 non-randomised controlled trial (RCT) studies with author surveys, the mean duration of baseline data collection was 11 months pre-implementation and 9.9 months post implementation.

Discussion

The main finding of this study is the incomplete reporting of QI methodology in the development, implementation and analysis of published ERAC studies. There was considerable heterogeneity in the implementation processes employed across ERAC studies, with a notable lack of formal description of the QI methodologies used in published manuscripts. This suggests that the best practice methodology for implementing ERAC protocols is currently lacking.

QI methodology in ERAC

The widespread uptake of ERAC protocols has been promoted by governing societies such as the Society for Obstetric Anesthesia and Perinatology² and the Enhanced Recovery After Surgery Society.⁴,⁶ Clinicians may turn to the study protocols and publications of these studies to determine the best means of protocol implementation and definition of implementation success. There is currently no consensus on the optimal QI methodology for use in healthcare, especially concerning ERAC study design and protocol implementation. Common methodologies in perioperative medicine and anaesthesiology include Six Sigma, Lean, Lean-Six Sigma (a combination of the former two), and the Institute for Healthcare Improvement (IHI) Model for Improvement PDSA cycling.^{16 17} A systematic review by Beyranvand et al identified nine different QI models used in surgery, including industry-derived approaches like PDSA and Lean, and surgery-specific frameworks, most commonly applied to improve processes, reduce length of stay and enhance efficiency.¹⁸ While there is no consensus on the optimal QI methodology, our findings confirm that most studies comparing ERAC to control groups lack detailed reporting on the methodologies used for training, education and implementation strategies, as only two studies mentioned the use of any methodology in their publications. In our author survey, 40% reported using a defined QI methodology, with the majority using the Model for Improvement. Among those not using a defined methodology, most authors reported the use of various QI components, including multidisciplinary stakeholder groups (80%) and identifying key drivers for change (87%).

Statistical methodology in ERAC

RCTs are typically designed for single interventions with concealed protocols and may not effectively test bundled interventions like those in enhanced recovery pathways. Many researchers investigating enhanced recovery therefore rely on observational before-and-after, quasi-experimental study design to analyse protocol components, development, implementation and outcome measures.¹⁹

SPC (Statistical Process Control) is particularly valuable in QI interventions because it allows for real-time monitoring and analysis of processes over time.¹⁹ Unlike traditional statistical methods, which often focus on comparing groups or measuring outcomes at specific time points, SPC allows the continuous monitoring of data to detect variations, trends and patterns that may indicate whether a process is in control or out of control, which is very useful after the introduction of a bundled intervention, as in enhanced recovery protocols.²⁰ While the majority (73%) of the studies included in our review employed traditional statistical methods, very few reported supplemental use of SPC (13%) and none reported sole use of SPC. Future guidelines for ERAC protocol implementation and reporting may recommend the use of SPC in such studies due to its relevance and advantages.

Discrepancies between information disseminated in the published manuscripts and the responses obtained directly from authors were evident in this study. Although our survey did not explore reasons for or against using defined methodologies, there appears to be a significant gap in methodology utilisation, individual QI components and result reporting. The significant variability in the published and reported duration of observation, ranging from 6 to 12 months, with no groups reporting observation beyond 1 year after implementation, limits the ability to assess the long-term validity of results reported. With such heterogeneity, it is currently difficult to isolate which of these methodologies or components is most effective.

Reporting discrepancies in ERAC

Despite the existence of established guidelines such as the SQUIRE framework for QI reporting,¹⁵ only one group of the authors acknowledged adherence to these standards in their publications. However, subsequent surveys revealed that 60% of respondents used these guidelines in either their protocol development or publication practices. ERAC studies that incorporate QI methodology or components could benefit from following standardised reporting guidelines, such as SQUIRE, similar to how Consolidated Standards of Reporting Trials (CONSORT) is used for randomised controlled trials. Specialised tools and guidelines tailored to assessing the methodological rigour and quality of QI projects in healthcare are needed.

Limitations

Our study has several limitations. First, due to the novel nature of our query, we were not able to rely on a validated questionnaire when assessing the included studies via a literature analysis and author survey, but rather generated a questionnaire based on expert opinion and literature review. While we faced challenges in attaining full author participation, a survey response rate of 63% is considered good for studies of this nature. Non-response bias and limiting to the English language may, however, limit generalisability of our findings regarding QI practices. After initial analysis and submission, we noted that five more eligible studies had been published that met the inclusion criteria. Due to feasibility, these were not included in our author survey.

In conclusion, ERAC implementation studies demonstrate a lack of standardised QI methodology and heterogeneity in reporting of methodologies used and addressing this gap through consensus-building efforts and standardised reporting guidelines is needed. Additionally, future research endeavours should address the ideal timeframes for assessing the impact of ERAC interventions, and the use of SPC and statistical tests in evaluating outcomes.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.