作者: Antonella Zambon , Gianluca Baio , Giampiero Mazzaglia , Luca Merlino , Giovanni Corrao
DOI: 10.1002/PDS.1530
关键词:
摘要: Background Data from a cohort of women treated with bisphosphonates were used to illustrate that multi-state models may be the useful tools analysis where two causes treatment failure are ultimate outcome interest, and sequence recurrent episodes starting discontinuing is also concern.Methods All 11863 resident in Italian Region Lombardy, aged 45 years or over who received for first time during 2003 entered into study followed up until December 2005. Multi-state in-treatment treatment-free periods as transient states, fractures gastrointestinal-related events competing failure, fitted data. The effect several covariates on transition rates between states was estimated.Results One half less than 27% period observation. Negative prognostic factors discontinuation younger age use alendronate at once-daily dosing, low refill compliance follow-up. risk fracture higher older experienced before entry cohort, those switched drugs, had compliance, corticosteroids follow-up.Conclusions pharmacoepidemiology provides non-biased estimates predicting discontinuation, restarting failures. Our analyses emphasize importance maximising order reduce risks both fracture. Copyright (c) 2008 John Wiley & Sons, Ltd.