• Poster

The cost of hospital care for relapsed breast cancer in the UK based on a large patient-level resource utilisation dataset.


Wolowacz, S. E., Roskell, N. S., Christie, A., Kerr, G., & Cameron, D. (2005, November). The cost of hospital care for relapsed breast cancer in the UK based on a large patient-level resource utilisation dataset.. Presented at ISPOR 8th Annual European Congress, Florence, Italy.


The development of new adjuvant chemotherapy
regimens for early breast cancer has resulted in clinically
important reductions in locoregional and distant
(metastatic) relapse in these patients (Martin et al.,
2005). In addition to improvements in survival and
health-related quality of life, these agents may be
expected to reduce the cost associated with management
of relapse.
• In this study, we estimate the total expected lifetime cost
of hospital treatments for relapsed early breast cancer
patients in the UK by type of first relapse (locoregional
or distant).
• These estimates may be used to inform economic
analyses of adjuvant chemotherapy regimens for early
breast cancer.
• Data were analysed for 571 node-positive early breast
cancer patients treated at the Western General Hospital,
Edinburgh between 1992 and 1998, of whom 180
experienced disease recurrence up to 2004.
• Data collected included patient characteristics, dates of
relapse by type and detailed resource use data over a 5-
year period post-relapse.
• Missing data (primarily due to incomplete follow-up of
surviving patients) were imputed using a set of prespecified
• Unit costs from national sources (primarily NHS Returns,
NHS Reference Costs and the British National Formulary)
were applied to the patient-level resource use data.
• Costs were estimated by bootstrapping (1000
simulations; with replacement).
• Resource use data post-relapse were available for all 180
patients who experienced disease relapse.
• In 145 patients the first relapse was distant. In the
remaining 35 the first relapse was locoregional; 22 of
these went on to develop distant disease.
• Complete data were available for 83% with first distant
recurrence and 71% of the group with first locoregional
recurrence as they died within follow-up.
• Descriptive statistics for the population are provided in
Table 1. Bootstrap mean costs post-relapse (and 95%
confidence intervals) are presented in (Tables 2 and 3).
• Chemotherapy and in-patient stay were the largest cost
• Costs were similar between the two groups across
resource categories with the exception of surgery where
there was a statistically significant difference between
groups (p