Treatment differences between blacks and whites with colorectal cancer

Med Care. 1996 Sep;34(9):970-84. doi: 10.1097/00005650-199609000-00008.


Objectives: The authors examine interracial variations in treatment for over 20,000 patients hospitalized with colorectal cancer in a national sample of hospitals.

Methods: To reduce clinical heterogeneity that could explain differences in treatment, hospitalizations were classified into relatively homogeneous subgroups based on diagnoses indicating primary colorectal tumor, oncologic sequelae, and metastasis. Procedures were classified into clinically relevant treatment types. Multivariate techniques controlled for differences in patient demographics, insurance status, other clinical factors, and provider characteristics.

Results: Blacks were more likely than whites to be hospitalized with oncologic sequelae, diagnoses indicating advanced disease, which may capture the effects of unmanaged or poorly managed cancer. Inpatient mortality was equivalent only for the most severely ill. Otherwise, the odds of inpatient mortality were 59% to 98% higher for blacks than whites. Treatment, in terms of procedure type, was equivalent only for the sickest patients. Among the less severely ill, blacks were less likely than whites to receive major therapeutic procedures.

Conclusions: Multiple findings suggest that blacks with colorectal cancer were hospitalized with more severe conditions and treated less aggressively than whites. In an era of health-care reform, such differences, which are net of insurance effects, may require more than universal insurance coverage to be overcome.

MeSH terms

  • Adolescent
  • Adult
  • African Americans*
  • Aged
  • Colorectal Neoplasms / ethnology*
  • Colorectal Neoplasms / therapy*
  • European Continental Ancestry Group*
  • Female
  • Health Services Accessibility / statistics & numerical data*
  • Health Services Research
  • Hospital Mortality
  • Hospitalization / statistics & numerical data*
  • Humans
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Outcome Assessment, Health Care
  • Practice Patterns, Physicians' / statistics & numerical data*
  • United States / epidemiology