Comparison of interview-based and medical-record based indices of comorbidity among breast cancer patients

OBJETIVES. To compare patient interview-based and medical-record based meas ures of comorbidity and their relation to primary tumor therapy, all cause mortality, self-reported upper body function, and overall physical function . METHODS. Three-hundred and three breast cancer patients (greater than or eq ual to 55 years) who were diagnosed in 1 of 5 Boston hospitals were enrolle d. Patient interviews and medical record abstracts provided the information necessary to construct the Charlson index, Satariano index, and a new inte rview-based index of cardiopulmonary comorbidity. Those indices were used a lone and in combination to predict the patient outcomes. RESULTS. The indices of comorbidity corresponded well with one another. No index of comorbidity predicted mortality or receipt of definitive primary t herapy. The new interview-based index of cardiopulmonary comorbidity was a better predictor of upper body function and overall physical function than was the interview-based or medical record-based Charlson or Satariano indic es of comorbidity. CONCLUSION. Older breast cancer patients are able to provide information ab out their diseases and related symptoms that correlates well with medical r ecord-based measures of comorbidity and displays similar patterns of predic tive power. A new self-reported measure of cardiopulmonary comorbidity perf orms better than the medical record-based measures for predicting patient r elated functional outcomes. Key words: epidemiologic factors-comorbidity; b reast neoplasms.