This Article Full Text (PDF) References Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Nijhuis, R. L. Articles by Hunink, M. G. M. Search for Related Content PubMed PubMed Citation Articles by Nijhuis, R. L. Articles by Hunink, M. G. M. Social Bookmarking                         What's this?

Apparent and Internal Validity of a Monte Carlo–Markov Model for Cardiovascular Disease in a Cohort Follow-up Study

Theo Stijnen, PhD

Department of Epidemiology and Biostatistics, Erasmus Medical Center, Rotterdam, the Netherlands

Anna Peeters, PhD

Department of Public Health, Erasmus Medical Center, Rotterdam, the Netherlands

Jacqueline C.M. Witteman, PhD

Albert Hofman, MD, PhD

Department of Epidemiology and Biostatistics, Erasmus Medical Center, Rotterdam, the Netherlands

M. G. Myriam Hunink, MD, PhD

Department of Radiology, Erasmus Medical Center, Rotterdam, the Netherlands; Departments of Health Policy & Management, Harvard School of Public Health, Boston, MA; Dept. of Epidemiology & Biostatistics, Erasmus Medical Center Rotterdam, P.O. Box 1738, 3000 DR Rotterdam, the Netherlands; phone: +31 10 408 7391; fax: +31 10 408 9382 m.hunink{at}erasmusmc.nl

Objective. To determine the apparent and internal validity of the Rotterdam Ischemic heart disease & Stroke Computer (RISC) model, a Monte Carlo–Markov model, designed to evaluate the impact of cardiovascular disease (CVD) risk factors and their modification on life expectancy (LE) and cardiovascular disease–free LE (DFLE) in a general population (hereinafter, these will be referred to together as (DF)LE). Methods. The model is based on data from the Rotterdam Study, a cohort follow-up study of 6871 subjects aged 55 years and older who visited the research center for risk factor assessment at baseline (1990–1993) and completed a follow-up visit 7 years later (original cohort). The transition probabilities and risk factor trends used in the RISC model were based on data from 3501 subjects (the study cohort). To validate the RISC model, the number of simulated CVD events during 7 years’ follow-up were compared with the observed number of events in the study cohort and the original cohort, respectively, and simulated (DF)LEs were compared with the (DF)LEs calculated from multistate life tables. Results .Both in the study cohort and in the original cohort, the simulated distribution of CVD events was consistent with the observed number of events (CVD deaths: 7.1% v. 6.6% and 7.4% v. 7.6%, respectively; non-CVD deaths: 11.2% v. 11.5% and 12.9% v. 13.0%, respectively). The distribution of (DF)LEs estimated with the RISC model consistently encompassed the (DF)LEs calculated with multistate life tables. Conclusions. The simulated events and (DF)LE estimates from the RISC model are consistent with observed data from a cohort follow-up study

Key Words: validation • apparent and internal • Monte Carlo–Markov model • cardiovascular disease • cohort follow-up study

Medical Decision Making, Vol. 26, No. 2, 134-144 (2006)
DOI: 10.1177/0272989X05284103


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?