How do integrated health care systems address racial and ethnic disparities in colon cancer?

Abstract

Purpose: Colorectal cancer (CRC) disparities have persisted over the last two decades. CRC is a complex disease requiring multidisciplinary care from specialists who may be geographically separated. Few studies have assessed the association between integrated health care system (IHS) CRC care quality, survival, and disparities. The purpose of this study was to determine if exposure to an IHS positively affects quality of care, risk of mortality, and disparities.

Patients and methods: This retrospective secondary-data analysis study, using the California Cancer Registry linked to state discharge abstracts of patients treated for colon cancer (2001 to 2006), compared the rates of National Comprehensive Cancer Network (NCCN) guideline-based care, the hazard of mortality, and racial/ethnic disparities in an IHS versus other settings.

Results: More than 30,000 patient records were evaluated. The IHS had overall higher rates of adherence to NCCN guidelines. Propensity score-matched Cox models showed an independent and protective association between care in the IHS and survival (hazard ratio [HR], 0.87; 95% CI, 0.85 to 0.90). This advantage persisted across stage groups. Black race was associated with increased hazard of mortality in all other settings (HR, 1.15; 95% CI, 1.04 to 1.27); however, there was no disparity within the IHS for any minority group (P > .11 for all groups) when compared with white race.

Conclusion: The IHS delivered higher rates of evidence-based care and was associated with lower 5-year mortality. Racial/ethnic disparities in survival were absent in the IHS. Integrated systems may serve as the cornerstone for developing accountable care organizations poised to improve cancer outcomes and eliminate disparities under health care reform.

Fig 1.

Percent delivery of evidence-based care comparing integrated health system (IHS) setting against other settings, stratified by race/ethnicity, for stage I to III colon cancer from 2001 to 2006. Surgery indicates resection for stage I to III disease; 12 lymph nodes indicates examination of ≥ 12 lymph nodes after resection; chemotherapy indicates American Joint Committee on Cancer stage–appropriate chemotherapy. API, Asian/Pacific Islander. (*) P < .05 for comparisons between racial groups within each setting.

Fig 2.

Propensity score–matched Kaplan-Meier curves (upper panel) and Cox models (lower panel) comparing survival in integrated health system (IHS) setting versus other settings. HR, hazard ratio; ref, referent.

Fig 3.

Kaplan-Meier curves showing survival by race, comparing outcomes stratified by health care setting. Survival after care in (A) integrated health system setting and (B) all other settings. API, Asian/Pacific Islander. (*) P < .05 and (†) P < .001 for minority groups compared with white patients after Dunnett adjustment.