Maturitas 57 (2007) 315–324 Persistence with oral and transdermal hormone replacement therapy and hospitalisation for cardiovascular outcomes Giovanni Corrao a,∗ , Antonella Zambon a , Federica Nicotra a , Carla Fornari b , Carlo La Vecchia c,d , Mario Mezzanzanica e , Rossella E. Nappi f,g , Luca Merlino h , Giancarlo Cesana b a Department of Statistics, Unit of Biostatistics, University of Milan-Bicocca, Milan, Italy b Research Center on Chronic Degenerative Diseases, University of Milan-Bicocca, Milan, Italy c Institute of Pharmacological Researches “Mario Negri”, Milan, Italy d Institute of Medical Statistics and Biometry, University of Milan, Milan, Italy e Department of Statistics, Unit of Informatics, University of Milan-Bicocca, Milan, Italy f Research Center for Reproductive Medicine, Department of Morphological, Eidological & Clinical Sciences, University of Pavia, Pavia, Italy g Endocrinology & Internal Medicine Unit, Section of Gynaecological Endocrinology, IRCCS Maugeri Foundation, University of Pavia, Pavia, Italy h Operative Unit of Territorial Health Services, Region Lombardia, Milan, Italy Received 16 January 2007; received in revised form 8 March 2007; accepted 12 March 2007 Abstract Background: The effect of persistence with transdermal and oral administrations of hormone replacement therapy (HRT) on the risk of hospitalisation for cardiovascular disease, and the role of income as potential confounder, were explored in a large population-based cohort study. Methods: Seventy-eight thousand eight hundred and seventy-five women resident in the Italian Lombardy Region aged 45–65 years who received at least one HRT prescription during 1998–2000 were followed until December 2003. The 828 cohort members who experienced at least one hospitalisation for a circulatory system disease were identified from the Regional hospital discharge database. The Regional prescription drug database was used to assess cumulative persistence with hormone treatment during follow-up. Data on individual taxable income was also obtained for women resident in the city of Milan. A proportional hazards model was fitted to estimate the association between cumulative time-dependent persistence with HRT and cardiovascular risk. Results: Compared with women who took HRT for less than 6 months, those exposed for more than 3 years to HRT as a whole, and to transdermal and oral HRT, respectively, showed hazard ratios of 0.65 (95% confidence interval: 0.45, 0.92), 0.53 (0.34, 0.82), and 1.15 (0.47, 2.79). CVD reducing potential of HRT disappeared when estimates were adjusted for income being hazard ratio associated with long-term use 0.94 (0.52, 1.71). ∗ Corresponding author at: Dipartimento di Statistica, Universit` a degli Studi di Milano-Bicocca, Via Bicocca degli Arcimboldi 8, Edificio U7, 20126 Milano, Italy. Tel.: +39 02 64485854; fax: +39 02 64485899. E-mail address: giovanni.corrao@unimib.it (G. Corrao). 0378-5122/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.maturitas.2007.03.001