Abstract

Objectives

The objective of this study was to identify the association between healthcare fragmentation and survival for patients with colorectal cancer in Colombia.

Methods

A retrospective cohort study was performed using administrative databases, with an electronic algorithm to identify patients with colorectal cancer based on codes. The patients were enrolled between January 1, 2013, and December 31, 2016. The exposure variable was fragmentation, which was measured based on the number of different healthcare institutions that treated a patient during the first year after diagnosis. Matching was performed using propensity scores to control for confounding, and the hazard ratio for exposure to higher fragmentation was calculated for the matched sample.

Results

A total of 5036 patients with colorectal cancer were identified, 2525 (49.88%) of whom were women. The mean number of network healthcare institutions for the total sample was 5.71 (SD 1.98). The patients in the quartile with higher fragmentation had the highest mortality rate, 35.67 (95% CI 33.63-38.06) per 100 patients. The comparison of higher and lower quartiles of fragmentation resulted in an incidence rate ratio of 1.23 (95% CI 1.04-1.45; P = .02). Of the 5036 patients, 422 (8.38%) were classified as the exposed cohort (higher fragmentation). The total matched sample consisted of 844 subjects, and an HR of 1.26 (95%CI; 1.05-1.51) was estimated.

Conclusions

Exposure to more highly fragmented healthcare networks decreases overall 4-year survival for patients with colorectal cancer in Colombia.