ASA - Association of Traditional and New Wait Time Benchmarks for Curative-Intent Colorectal Cancer Surgery and Survival: A Population-Based Cohort Study

Back to 2025 Abstracts

Association of Traditional and New Wait Time Benchmarks for Curative-Intent Colorectal Cancer Surgery and Survival: A Population-Based Cohort Study
*Adom Bondzi-Simpson¹, *Rinku Sutradhar³, Frances C. Wright¹, *Tiago Ribeiro¹, *Sheron Perera², *Wing Chan³, *Andrea Covelli¹, *Aisha Lofters⁴, *Rebecca Snyder⁵, *Callisia N. Clarke⁶, Natalie G. Coburn¹, *Julie Hallet¹
¹Department of Surgery, University of Toronto, Toronto, ON, Canada; ²Department of Medicine, Division of Medical Oncology, University of Toronto, Toronto, ON, Canada; ³ICES, Toronto, ON, Canada; ⁴Department of Family and Community Medicine, University of Toronto, Toronto, ON, Canada; ⁵Departments of Surgical Oncology and Health Services Research, MD Anderson, Houston, TX; ⁶Department of Surgery, Medical College of Wisconsin, Milwaukee, WI

Background: Wait time (WT) to surgery is a common quality indicator for colorectal cancer (CRC), using arbitrarily defined benchmarks primarily based on clinical reasoning. However, the definition of target WT and its association with clinically relevant outcomes remains poorly defined. We assessed the association between WT to CRC surgery and overall survival (OS) for curative-intent surgery.

Methods: We conducted a population-based retrospective cohort study of adults undergoing resection for stage I-III CRC over 2007-2020, using health administrative data in Ontario, Canada. The exposure was WT, measured as the time from the decision to operate to surgery (in days). The outcome was OS, measured as time from surgery to death. Restricted cubic spline regression (RCS) examined the relationship between WT and hazards of death to identify meaningful WT thresholds. WT was then categorized as a) a traditional WT benchmark (< 28 days) and b) a new data-informed benchmark defined by RCS. Multivariable Cox proportional hazards explored the association between each WT benchmark and the hazards of death after adjusting for confounders established a priori. Subgroup analysis was performed by cancer stage.

Results: Of 35,533 patients, 27,102 (76.3%) underwent surgery within the traditional WT benchmark. The median WT was 19 days (interquartile range 12-28). RCS revealed an inflection point around 45 days associated with increasing hazards of death, particularly for stages II and III. After adjusting for age, sex, co-morbidity, cancer site, stage, neo-adjuvant or adjuvant therapy, and year of surgery, having surgery within the traditional WT benchmark (< 28 days) was not associated with OS (Hazards Ratio, [HR] 0.97; 95% confidence interval, [CI] 0.92-1.02). Having surgery within the new WT benchmark (< 45 days) was independently associated with superior OS (HR 0.90, 95% CI 0.82-0.99). No significant associations were identified for stage-specific analyses.

Conclusion: In a large population-based cohort of patients undergoing curative-intent resection for stage I-III CRC, undergoing surgery < 45 days of decision to operate was independently associated with superior OS. However, traditional WT of < 28 days was not. These data highlight the importance of using data-driven approaches to define quality indicators linked to clinically significant outcomes. The definition of the WT benchmark may be reconsidered, as it has important implications for patients, health providers, organizations, and health systems in delivering high-quality and high-value cancer care.
Adjusted association of traditional and new data-driven wait time benchmarks with hazards of death.

Cohort	Benchmark	HR	Lower CI	Upper CI	p-value
Entire Cohort	Traditional WT Benchmark (< 28 days)	0.973	0.924	1.024	0.296
	New WT Benchmark (< 45 days)	0.903	0.823	0.99	0.03
Stage I	Traditional WT Benchmark (< 28 days)	1.001	0.891	1.126	0.98
	New WT Benchmark (< 45 days)	0.931	0.763	1.137	0.485
Stage II	Traditional WT Benchmark (< 28 days)	0.944	0.862	1.034	0.212
	New WT Benchmark (< 45 days)	0.949	0.803	1.122	0.541
Stage III	Traditional WT Benchmark (< 28 days)	0.985	0.914	1.06	0.68
	New WT Benchmark (< 45 days)	0.894	0.783	1.022	0.101

Adjusted associations of traditional and new data-driven WT benchmarks and hazards of death. Traditional WT benchmarks based on National Health Services (United Kingdom) Cancer Care Ontario (Canada) benchmark of < 28 days from the decision to operate to the date of the procedure. Multivariable Cox proportional hazards models explored the association of exposures of interest and hazards of death after adjusting for confounders established a priori including age, sex, co-morbidity burden (Elixhauser score), cancer site, stage, neo-adjuvant or adjuvant therapy, and year of surgery. Censoring events in the time-to-event analysis included death, lost to follow-up, and five years post-date of surgery. One model was fit for each outcome/exposure combination, for a total of eight separate models.
Hazard rate of > 1 equals an increased hazard of death when compared to the reference group. In all cases the reference group was compared to patients that did not meet benchmarks for wait times.
Acronyms: CI = confidence interval, HR = hazard ratio, WT = Wait Time.

Restricted cubic spline regression showing univariable association of increasing wait times to surgery and hazards of death.
Figure demonstrating restricted cubic spline regression data (with 3 prespecified knots) for entire cohort (A) and stratified by stage of colorectal cancer. (B) stage I, (C) stage II, (D) stage III. Hazard ratios (HRs) are shown on a log scale; shaded areas indicate 95% Confidence Intervals (CIs).

Back to 2025 Abstracts