By Any Other Name: Bowel Dysfunction After Proctectomy for Cancer and Its Predictive Factors in Administrative Databases

Introduction

Diagnosis, outcomes, and costs of care associated with bowel dysfunction after proctectomy for cancer remain underexplored in population-based studies. The lack of administrative coding for bowel dysfunction or low anterior resection syndrome has historically limited secondary data set outcomes analysis. The purpose of this study was to identify a bowel dysfunction phenotype in administrative claims data and characterize its prevalence, predictive factors, and costs.

Materials and methods

Patients were identified with employer-sponsored commercial insurance (MarketScan research databases) undergoing proctectomy for cancer for a retrospective cohort study. Bowel dysfunction was defined as any patient with diagnostic codes for diarrhea, constipation, incontinence, pelvic floor diagnostic testing, or rehabilitative procedures that occurred in the 18 mo to follow surgery. We performed Poisson regression to identify statistically significant covariates of bowel dysfunction occurrence following low anterior resection. A secondary comparative analysis was also performed of total costs of healthcare utilization following gastrointestinal continuity.

Results

6426 proctectomy patients were identified, out of which 2131 had surgery for cancer. 847 patients undergoing proctectomy for cancer (39.7%) experienced bowel dysfunction during 18 mo of follow-up. The most common diagnoses were constipation (53.5%) and diarrhea (40.3%). Diagnostic procedures and rehabilitative procedures were performed in only 29.8% of those with symptoms. Neoadjuvant chemotherapy administration with radiation (incidence rate ratio = 1.23, 95% CI: 1.01-1.51) and without (incidence rate ratio = 1.20, 95% CI: 1.01-1.42) remained associated with postoperative bowel dysfunction when controlling for other factors. Chemoradiation therapy alone was not associated with bowel dysfunction. The median total follow-up costs with bowel dysfunction were $30,769 greater (P < 0.001).

Conclusions

More than one-third of patients have symptomatic bowel dysfunction within 18 mo after restored continuity, with multiagent chemotherapy being the strongest independent predictor. Bowel dysfunction is associated with more than twice healthcare costs postop.