Underestimation of risk associations due to regression dilution in long-term follow-up of prospective studies.
ABSTRACT In prospective studies, disease rates during follow-up are typically analyzed with respect to the values of factors measured during an initial baseline survey. However, because of "regression dilution," this generally tends to underestimate the real associations of disease rates with the "usual" levels of such risk factors during some particular exposure period. The "regression dilution ratio" describes the ratio of the steepness of the uncorrected association to that of the real association. To assess the relevance of the usual value of a risk factor during particular exposure periods (e.g., first, second, and third decades) to disease risks, regression dilution ratios can be derived by relating baseline measurements of the risk factor to replicate measurements from a reasonably representative sample of study participants after an interval equivalent to about the midpoint of each exposure period (e.g., at 5, 15, and 25 years, respectively). This report illustrates the impact of this time interval on the magnitude of the regression dilution ratios for blood pressure and blood cholesterol. The analyses were based on biennial remeasurements over 30 years for participants in the Framingham Study (Framingham, Massachusetts) and a 26-year resurvey for a sample of men in the Whitehall Study (London, England). They show that uncorrected associations of disease risk with baseline measurements underestimate the strength of the real associations with usual levels of these risk factors during the first decade of exposure by about one-third, the second decade by about one-half, and the third decade by about two-thirds. Hence, to correct appropriately for regression dilution, replicate measurements of such risk factors may be required at varying intervals after baseline for at least a sample of participants.
SourceAvailable from: Paul Downey[Show abstract] [Hide abstract]
ABSTRACT: Cathie Sudlow and colleagues describe the UK Biobank, a large population-based prospective study, established to allow investigation of the genetic and non-genetic determinants of the diseases of middle and old age.PLoS Medicine 03/2015; 12(3):e1001779. DOI:10.1371/journal.pmed.1001779 · 14.00 Impact Factor
[Show abstract] [Hide abstract]
ABSTRACT: Blood pressure (BP) trajectories derived from measurements repeated over years have low measurement error and may improve cardiovascular disease prediction compared to single, average, and usual BP (single BP adjusted for regression dilution). We characterized 10-year BP trajectories and examined their association with cardiovascular mortality, all-cause mortality, and life years lost. Data from 2 prospective and nearly extinct cohorts of middle-aged men-the Minnesota Business and Professional Men Study (n=261) and the Zutphen Study (n=632)-were used. BP was measured annually during 1947-1957 in Minnesota and 1960-1970 in Zutphen. BP trajectories were identified by latent mixture modeling. Cox proportional hazards and linear regression models examined BP trajectories with cardiovascular mortality, all-cause mortality, and life years lost. Associations were adjusted for age, serum cholesterol, smoking, and diabetes mellitus. Mean initial age was about 50 years in both cohorts. After 10 years of BP measurements, men were followed until death on average 20 years later. All Minnesota men and 98% of Zutphen men died. Four BP trajectories were identified, in which mean systolic BP increased by 5 to 49 mm Hg in Minnesota and 5 to 20 mm Hg in Zutphen between age 50 and 60. The third systolic BP trajectories were associated with 2 to 4 times higher cardiovascular mortality risk, 2 times higher all-cause mortality risk, and 4 to 8 life years lost, compared to the first trajectory. Ten-year BP trajectories were the strongest predictors, among different BP measures, of cardiovascular mortality, all-cause mortality, and life years lost in Minnesota. However, average BP was the strongest predictor in Zutphen. © 2015 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley Blackwell.Journal of the American Heart Association 02/2015; 4(3). DOI:10.1161/JAHA.114.001378 · 2.88 Impact Factor
[Show abstract] [Hide abstract]
ABSTRACT: -While physical activity has generally been associated with reduced risk of vascular disease, there is limited evidence about the effects of the frequency and duration of various activities on incidence of particular types of vascular disease. -In 1998, on average, 1.1 million women without prior vascular disease reported their frequency of physical activity and many other personal factors. Three years later, they were asked about hours spent doing walking, cycling, gardening and housework. Women were followed by record linkage to NHS cause-specific hospital admissions and death records. Cox regression was used to calculate adjusted relative risks for first vascular events in relation to physical activity. During an average of 9 years follow-up, 49,113 women had a first coronary heart disease event (CHD), 17,822 had a first cerebrovascular event, and 14,550 had a first venous thromboembolic event (VTE). Compared to inactive women, those reporting moderate activity had significantly lower risks of all three conditions (p<0.001 for each). However, women reporting strenuous physical activity daily had higher risks of CHD (p=0.002), cerebrovascular disease (p<0.001) and VTE (p<0.001) than those reporting doing such activity 2-3 times per week. Risks did not differ between hemorrhagic and ischemic stroke, or between VTE with or without pulmonary embolism. -Moderate physical activity is associated with a lower risk of CHD, VTE and cerebrovascular disease than inactivity. However, among active women there is little to suggest progressive reductions in risk of vascular diseases with increasing frequency of activity.Circulation 02/2015; DOI:10.1161/CIRCULATIONAHA.114.010296 · 14.95 Impact Factor