Skip to main content
. 2009 May 26;27(24):3945–3950. doi: 10.1200/JCO.2008.20.8546

Table 2.

Influence of Patient and Hospital Variables on 5-Year Overall Survival Among Black and White Patients

Factor Breast Cancer
Colon Cancer
5-Year Mortality, Black v White
% of Excess Mortality Explained* 5-Year Mortality, Black v White
% of Excess Mortality Explained*
Hazard Ratio 95% CI Hazard Ratio 95% CI
Unadjusted 1.42 1.32 to 1.52 1.14 1.08 to 1.20
Adjusted individually for:
    Patient factors 1.25 1.16 to 1.34 40.48 1.13 1.07 to 1.19 7.14
    SES 1.28 1.18 to 1.38 33.33 1.08 1.02 to 1.14 42.86
    Hospital volume 1.44 1.34 to 1.55 NA 1.16 1.10 to 1.23 NA
    Hospital fixed effects§ 1.31 1.21 to 1.42 26.19 1.08 1.01 to 1.16 42.86
Adjusted sequentially for:
    Patient factors 1.25 1.16 to 1.34 40.48 1.13 1.07 to 1.19 7.14
    Patient factors + SES 1.16 1.08 to 1.26 36.00 1.08 1.02 to 1.14 38.46
    Patient factors + hospital fixed effects 1.16 1.07 to 1.26 36.00 1.06 1.00 to 1.14 53.84
    Patient factors + hospital fixed effects + SES 1.13 1.03 to 1.23 48.00 1.03 0.96 to 1.11 76.92

Abbreviations: SES, socioeconomic status; NA, not applicable.

*

Proportion of excess mortality is relative to unadjusted model for individual factors in the univariate analysis and is relative to the risk-adjusted model (adjusted for patient factors) for the sequential multivariate analysis.

Patient factors include age, sex, comorbidities, and cancer stage.

Hospital volume adjustment also included adjustment for hospital teaching status and cancer center designation, the effects of which were negligible.

§

Estimates of hazard ratios and 95% CIs were obtained from model including hospital fixed effects.