Enhanced recovery after surgery in colorectal surgery: Do average benefits mask heterogeneous effects in patients underrepresented in trials?

Table 1 Summary of included study characteristics and quality assessment

Study type	Number of studies	Quality assessment (NOS/RoB)	Key findings relevant to this review
Randomized controlled trials	8	Low-to-moderate RoB	ERAS reduces LOS by 2-3 days, complications by 30%-40%
Prospective multicenter cohorts	12	NOS 7-9 (high quality)	Dose-response: 10% compliance increment → approximately 5% complication reduction
Single-center retrospective cohorts	23	NOS 5-7 (moderate quality)	High-risk subgroups have lower compliance and worse outcomes
Registry analyses	7	NOS 7-8 (high quality)	Socioeconomic gradients in compliance and outcomes
Qualitative studies	4	CASP qualitative checklist	Barriers to adherence: Clinician judgement, patient factors, system constraints
Cognitive psychology/safety science	8	N/A (indirect evidence)	Automation bias, Goodhart’s Law - hypothesis-generating only

CASP: Critical Appraisal Skills Program; ERAS: Enhanced recovery after surgery; LOS: Length of stay; N/A: Not applicable; NOS: Newcastle-Ottawa Scale; RoB: Risk of bias.

Table 2 Key limitations and their implications

Limitation	Nature of the constraint	Implication for conclusions
Narrative review design	Non-systematic study selection; author judgement in synthesis	Different reviewers might emphasis different evidence
Indirect evidence for core thesis	No prospective study has tested whether distinguishing deviation types improves outcomes	Central claim is hypothesis-generating, not confirmatory
Post hoc nature of subgroup signals	Signals of heterogeneity derive from unplanned analyses	Cannot distinguish effect modification from confounding
Cross-domain theoretical frameworks	Goodhart’s Law and automation bias untested in ERAS settings	Theoretical discussion is heuristic, not evidentiary
Geographic concentration of equity evidence	Racial and socioeconomic disparities data predominantly from United States	Findings may not generalize to universal coverage health systems
English-language restriction	Grey literature and non-English publications excluded	Language bias and publication bias cannot be quantified
Single-discipline authorship	Authors are clinicians with expertise spanning surgery and ultrasound medicine	Risk of oversimplification in psychology and equity literatures
Temporal scope	Search conducted to July 2026	New evidence may have since emerged
Scope limited to elective colorectal surgery	Findings may not extend to emergency surgery or other specialties	Boundary of applicability is defined but untested
High-income country orientation of recommendations	Stratified audit presupposes electronic audit infrastructure	Applicability to low-resource settings is uncertain

ERAS: Enhanced recovery after surgery.