Analytical approach:
This study updated a published Markov model reported by Mansel, et al. in 2007 (see 'Other Publications of Related Interest' below for bibliographic details). This Markov model had a 25-year time horizon. The data were collected from 2000 until the patient’s death or until January 2007, when the study ended. The perspective was not explicitly stated.
Effectiveness data:
: This study did not evaluate any clinical data, but updated the estimated costs within an existing Markov model (Mansel, et al. 2007). The disease-free survival at 10 years, for patients who did not relapse compared with those who did, was estimated by applying Weibull curves to the data from the 'Arimidex', Tamoxifen, Alone or in Combination (ATAC) trial reported in Mansel, et al.
Monetary benefit and utility valuations:
Not relevant.
Measure of benefit:
The measure of benefit was life-years gained (LYG).
Cost data:
The initial costs were taken from the original model reported in Mansel, et al. These included the costs of drugs and bone density scans. The hospital and community costs associated with managing patients with relapsed breast cancer were estimated with data from the Bedford Breast Unit database of 232 relapsed women. The community costs included general practitioner (GP) home visits, hospice nights and visits, palliative community visits, etc. The hospital costs included those of surgical interventions, blood transfusions, hospitalisation, out-patient visits and procedures, including electrocardiogram, radiotherapy and emergency admissions, and some hospital services including transport. The resource quantities were from hospital, GP, pharmacy, district nurse, and hospice records. The unit costs were from official national sources. The price year was not explicitly reported.
Analysis of uncertainty:
No analysis of uncertainty was reported.