Repeated measures and risk factor trajectories of C-reactive protein and body mass index in relation to cardiovascular disease
Dr. Mark O’Doherty (CoEfPHNI)
11.30, Tuesday 9 July. The Clubhouse, level 4
Data on health of ageing populations are essential for health, social and economic research and management. Such data should both be valid and precise. Furthermore, they should be based on repeated measurements over time, to identify and quantify changes in health-related parameters, their determinants and confounders. CHANCES (Consortium on Heath and Ageing Network of Cohorts in Europe and the United States) is a collaborative large scale project funded by the European Commission within the Seventh Framework Programme, which aims at combining and integrating ongoing studies in order to produce evidence on ageing-related health characteristics and determinants in Europe.
Both C-reactive protein (CRP) and obesity have been shown to be related to cardiovascular disease outcomes. However, much of this epidemiologic evidence is usually based on these risk variables being measured once at ‘baseline.’ True estimations using only ‘baseline’ measurements may not be accurate because many physiologic variables are not stable over time, and may have diurnal, seasonal, or long-term variation, which may have considerable impact on the accuracy of risk prediction.
More so, missing data is a common problem in repeated analysis, and in multivariate analysis of large sample surveys even a small proportion of missing data on many variables quickly adds up to a large number of cases being deleted. Journal editors and reviewers are increasingly strict about missing data problems and solutions, and therefore appropriate techniques need to be employed where possible to avoid missing data becoming an issue.
To assess this, we aim to use cohorts within CHANCES which have the appropriate repeated measures available, to clarify the relationship between CRP and obesity with incidence of cardiovascular outcomes and total mortality.
Both C-reactive protein (CRP) and obesity have been shown to be related to cardiovascular disease outcomes. However, much of this epidemiologic evidence is usually based on these risk variables being measured once at ‘baseline.’ True estimations using only ‘baseline’ measurements may not be accurate because many physiologic variables are not stable over time, and may have diurnal, seasonal, or long-term variation, which may have considerable impact on the accuracy of risk prediction.
More so, missing data is a common problem in repeated analysis, and in multivariate analysis of large sample surveys even a small proportion of missing data on many variables quickly adds up to a large number of cases being deleted. Journal editors and reviewers are increasingly strict about missing data problems and solutions, and therefore appropriate techniques need to be employed where possible to avoid missing data becoming an issue.
To assess this, we aim to use cohorts within CHANCES which have the appropriate repeated measures available, to clarify the relationship between CRP and obesity with incidence of cardiovascular outcomes and total mortality.