Rc. Burack et al., Follow-up among women with an abnormal mammogram in an HMO: Is it complete, timely, and efficient?, AM J M CARE, 6(10), 2000, pp. 1102-1113
Citations number
25
Categorie Soggetti
Public Health & Health Care Science","Health Care Sciences & Services
Objective: To describe the extent to which women with seriously abnormal ma
mmograms complete indicated follow-up, the timeliness of this follow-up, an
d variations in the pattern of use of diagnostic procedures.
Study Design: Retrospective chart review.
Patients and Methods: Ninety-two women enrolled in a single urban health ma
intenance organization (HMO) with an abnormal index mammogram (mass or susp
icious calcifications) during 1995 or 1996 were identified by review of all
HMO mammography reports. Data were abstracted from medical records concern
ing all clinical services received over the 11 months after the date of the
abnormal mammogram. Procedure costs were estimated based on 1997 Medicare
relative-value units. Logistic regression and a multivariate accelerated fa
ilure-time model were used to evaluate the association between predictor va
riables and the occurrence and timing of completion of follow-up.
Results: Follow-up was not completed by 31 (34%) of the 92 study women and
was delayed beyond 60 days for another 32 (35%). In adjusted analysis, fact
ors associated with completion within 60 days included age less than 50 yea
rs and inclusion of a specific follow-up recommendation in the mammogram re
port. Completion by the end of the study (a minimum of 11 months after the
index mammogram) was associated only with the presence of a specific follow
-up recommendation. The follow-up process (ie, the diagnostic procedures us
ed) was highly variable but almost always included surgical evaluation. The
average cost among those completing follow-up was about $1900 (in 1997 dol
lars).
Conclusions: Incomplete follow-up after a potentially seriously abnormal ma
mmogram constitutes an important barrier to breast cancer control efforts i
n the study HMO, but its explanation remains incompletely understood. The f
ollow-up process itself is highly variable, and improvement in its efficien
cy and timely completion will require a better understanding of its determi
nants.