RESUMEN
Observational studies often seek to estimate the causal relevance of an exposure to an outcome of interest. However, many possible biases can arise when estimating such relationships, in particular bias because of confounding. To control for confounding properly, careful consideration of the nature of the assumed relationships between the exposure, the outcome, and other characteristics is required. Causal diagrams provide a simple graphic means of displaying such relationships, describing the assumptions made, and allowing for the identification of a set of characteristics that should be taken into account (i.e., adjusted for) in any analysis. Furthermore, causal diagrams can be used to identify other possible sources of bias (such as selection bias), which if understood from the outset, can inform the planning of appropriate analyses. In this article, we review the basic theory of causal diagrams and describe some of the methods available to identify which characteristics need to be taken into account when estimating the total effect of an exposure on an outcome. In doing so, we review the concept of collider bias and show how it is inappropriate to adjust for characteristics that may be influenced, directly or indirectly, by both the exposure and the outcome of interest. A motivating example is taken from the Study of Heart and Renal Protection, in which the relevance of smoking to progression to ESRD is considered.
Asunto(s)
Fallo Renal Crónico/epidemiología , Estudios Observacionales como Asunto/estadística & datos numéricos , Proyectos de Investigación , Fumar/epidemiología , Sesgo , Factores de Confusión Epidemiológicos , Interpretación Estadística de Datos , Humanos , Fallo Renal Crónico/etiología , Fumar/efectos adversosRESUMEN
Among those with moderate-to-advanced chronic kidney disease, the relationship between blood pressure (BP) and cardiovascular disease seems U shaped but is loglinear in apparently healthy adults. The SHARP (Study of Heart and Renal Protection) randomized 9270 patients with chronic kidney disease to ezetimibe/simvastatin versus matching placebo and measured BP at each follow-up visit. Cox regression was used to assess the association between BP and risk of cardiovascular disease among (1) those with a self-reported history of cardiovascular disease and (2) those with no such history and, based on plasma troponin-I concentration, a low probability of subclinical cardiac disease. A total of 8666 participants had a valid baseline BP and troponin-I measurement, and 2188 had at least 1 cardiovascular event during follow-up. After adjustment for relevant confounders, the association between systolic BP and cardiovascular events was U shaped, but among participants without evidence of previous cardiovascular disease, there was a positive loglinear association throughout the range of values studied. Among those with the lowest probability of subclinical cardiac disease, each 10 mm Hg higher systolic BP corresponded to a 27% increased risk of cardiovascular disease (hazard ratio, 1.27; 95% confidence interval, 1.11-1.44). In contrast, the relationship between diastolic BP and cardiovascular risk remained U shaped irrespective of cardiovascular disease history or risk of subclinical disease. In conclusion, the lack of a clear association between systolic BP and cardiovascular risk in this population seems attributable to confounding, suggesting that more intensive systolic BP reduction may be beneficial in such patients. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00125593.