Commentary: Berkson's Bias reviewed

Division of Epidemiology, Institute for Environmental Medicine, Karolinska Institutet, Stockholm, Sweden.
European Journal of Epidemiology (Impact Factor: 5.34). 02/2003; 18(12):1109-12. DOI: 10.1023/B:EJEP.0000006552.89605.c8
Source: PubMed
9 Reads
  • Source
    • "Suspected examples of severe Berksonian bias have been shown to cause extreme downward bias, to a 10-fold decrease in effect estimate [23] [24]. Exploration of the potential impact of selection bias in the electromagnetic fields and leukemia literature has found that this type of bias could result in a twofold increase in effect estimates [25]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Selection bias is a form of systematic error that can be severe in compromised study designs such as case-control studies with inappropriate selection mechanisms or follow-up studies that suffer from extensive attrition. External adjustment for selection bias is commonly undertaken when such bias is suspected, but the methods used can be overly simplistic, if not unrealistic, and fail to allow for simultaneous adjustment of associations of the exposure and covariates with the outcome, when of interest. Internal adjustment for selection bias via inverse probability weighting allows bias parameters to vary with the levels of covariates but has only been formalized for longitudinal studies with covariate data on patients up until loss to follow-up. Methods: We demonstrate the use of inverse probability weighting and externally obtained bias parameters to perform internal adjustment of selection bias in studies lacking covariate data on unobserved participants. Results: The "true" or selection-adjusted odds ratio for the association between exposure and outcome was successfully obtained by analyzing only data on those in the selected stratum (i.e., responders) weighted by the inverse probability of their being selected as function of their observed covariate data. Conclusions: This internal adjustment technique using user-supplied bias parameters and inverse probability weighting for selection bias can be applied to any type of observational study.
    Annals of Epidemiology 08/2014; 24(10). DOI:10.1016/j.annepidem.2014.07.014 · 2.00 Impact Factor
  • Source
    • "The reference lists of the citations identified in these searches, including review articles, were also examined. General population-based samples are the focus of this review because these study designs provide the best methodology for understanding the natural history of frailty and depression; clinic-based populations probably reflect a host of selection factors that make them inappropriate for exploring these epidemiologic questions (Schwartzbaum et al., 2003). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Many of the symptoms, consequences, and risk factors for frailty are shared with late-life depression. However, thus far, few studies have addressed the conceptual and empirical interrelationships between these conditions. This review synthesizes existing studies that examined depression and frailty among older adults and provides suggestions for future research.MethodsA search was conducted using PubMed for publications through 2010. Reviewers assessed the eligibility of each report and abstracted information on study design, sample characteristics, and key findings, including how depression and frailty were conceptualized and treated in the analysis.ResultsOf 133 abstracted articles, 39 full-text publications met inclusion criteria. Overall, both cross-sectional (n = 16) and cohort studies (n = 23) indicate that frailty, its components, and functional impairment are risk factors for depression. Although cross-sectional studies indicate a positive association between depression and frailty, findings from cohort studies are less consistent. The majority of studies included only women and non-Hispanic Whites. None used diagnostic measures of depression or considered antidepressant use in the design or analysis of the studies.ConclusionsA number of empirical studies support for a bidirectional association between depression and frailty in later life. Extant studies have not adequately examined this relationship among men or racial/ethnic minorities, nor has the potential role of antidepressant medications been explored. An interdisciplinary approach to the study of geriatric syndromes such as late-life depression and frailty may promote cross-fertilization of ideas leading to novel conceptualization of intervention strategies to promote health and functioning in later life. Copyright © 2011 John Wiley & Sons, Ltd.
    International Journal of Geriatric Psychiatry 09/2012; 27(9). DOI:10.1002/gps.2807 · 2.87 Impact Factor
  • Source
    • "Example 2.1 Hospitalisation bias, also known as Berkson's bias, has been extensively studied in the epidemiologic literature (Schwartzbaum et al., 2003). This type of bias arises when the exposure is a medical condition and hence also a reason for hospitalisation , and only hospital-based controls are used. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Retrospective case-control studies are more susceptible to selection bias than other epidemiologic studies as by design they require that both cases and controls are representative of the same population. However, as cases and control recruitment processes are often different, it is not always obvious that the necessary exchangeability conditions hold. Selection bias typically arises when the selection criteria are associated with the risk factor under investigation. We develop a method which produces bias-adjusted estimates for the odds ratio. Our method hinges on 2 conditions. The first is that a variable that separates the risk factor from the selection criteria can be identified. This is termed the "bias breaking" variable. The second condition is that data can be found such that a bias-corrected estimate of the distribution of the bias breaking variable can be obtained. We show by means of a set of examples that such bias breaking variables are not uncommon in epidemiologic settings. We demonstrate using simulations that the estimates of the odds ratios produced by our method are consistently closer to the true odds ratio than standard odds ratio estimates using logistic regression. Further, by applying it to a case-control study, we show that our method can help to determine whether selection bias is present and thus confirm the validity of study conclusions when no evidence of selection bias can be found.
    Biostatistics 06/2008; 10(1):17-31. DOI:10.1093/biostatistics/kxn010 · 2.65 Impact Factor
Show more