BACKGROUND. Although a number of studies have used Medicare claims data to
study trends and variations in breast cancer treatment, the accuracy and co
mpleteness of information on surgical treatment for breast cancer in the Me
dicare data have not been validated.
OBJECTIVES. This study assessed the accuracy and completeness of Medicare c
laims data for breast cancer surgery to determine whether Medicare claims c
an serve as a source of data to augment information collected by cancer reg
istries.
METHODS. We used the Surveillance, Epidemiology and End Results (SEER) Canc
er Registry-Medicare data and compared Medicare claims on surgery with the
surgery recorded by the SEER registries for 23,709 women diagnosed with bre
ast cancer at greater than or equal to 65 years of age from 1991 through 19
93.
RESULTS. More than 95% of women having mastectomies according to the Medica
re data were confirmed by SEER. For breast-conserving surgery, 91% of cases
were confirmed by SEER. The Medicare physician services claims and inpatie
nt claims were approximately equal in accuracy on type of surgery. The Medi
care outpatient claims were less accurate for breast-conserving surgery. In
terms of completeness, when the 3 claims sources were combined, 94% of pat
ients receiving breast cancer surgery according to SEER were identified by
Medicare.
CONCLUSIONS. The combined Medicare claims database, which includes the inpa
tient, outpatient, and physician service claims, provides valid information
on surgical treatment among women known to have breast cancer. The claims
are a rich source of data to augment the information collected by tumor reg
istries and provide information that can be used to follow long-term outcom
es of Medicare beneficiaries.