Liana Fraenkel and colleagues at Yale, using Adaptive Conjoint Analysis, found that patient preferences for osteoporosis medications were strongly influenced by the route of administration, reflecting a high trade-off of fracture risk reduction with this attribute. - Vertebral fracture ARR (absolute risk reduction)
- Hip fracture ARR (absolute risk reduction)
- Frequency of Medication(1/365; 4/365; 52/365; 365/365)
- Route of Medication (oral>infusion>injection)
- Gastrointestinal SE (side effects)
- Infusion SE (side effects)
- Oral bisphosphonates taken once a week
- Intravenous bisphosphonates administered every 3 months
- Intravenous bisphosphonates administered once a year
- Subcutaneous injection of rhPTH daily (recombinant human ParaThyroid Hormone)
Fraenkel et al do not report the ARRs provided to respondents, but say they assumed (in their base case) a 50% Relative RR of both VF and HF from all bisphophonates irrspective of ROA. rhPTA was assumed to offer a 65%RRR of VF and a 50% RRR for HF. No data are reported for frequency of Side Effects. Fraenkel et al did not split Frequency and Route of Administration or the two side effects, as we have done. The average weights for their sample were ROA 29.3%; VFARR 24.2%; HFARR 24.3%; Side Effects 22.2%.[ We have split both firstand last 50-50 for this starter Annalisa.] Standard Deviations are not reported but were clearly large, given later results presented, indicating that mean weights will have little value in the context of individual patient decision making As implied by  Ratings and Weightings. The rank order reflects that of the average patient in the study by Fraenkel et al. The % of patients who preferred each method was OB 40%; IVB 3monthly 6%; IV B annually 44%; rhPTH 6%; undecided 10%. Slide 3 shows the very high weight on VFARR necessary to make rhPTH the best option given our ratings Liana Fraenkel, Barbara Gulanski, Dick Wittink (2006) 'Patient treatment preferences for osteoporosis' Arthritis & Rheumatism 55 (5) pp 729-735. author email: liana.fraenkel@yale.edu The purpose of presenting an Annalisa version of a study conducted using another MCDA method (such as Adaptive Conjoint Analysis) is to summarise and present the analysis in a simple and attractive one-screen format for a large audience - and to permit the interested viewer to download the .alt file and interact with the Weightings and Ratings to determine the effect of changes on the Scores. Alternatively the .alt file may be used as a 'starter' for a new analysis. No opinion is expressed or implied in regard to the merits of the study on any criterion.
download alt file here
|
