Browsing by Subject "Data Interpretation, Statistical"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Open Access
    A random effects variance shift model for detecting and accommodating outliers in meta-analysis
    (BioMed Central Ltd, 2011) Gumedze, Freedom; Jackson, Dan
    BACKGROUND:Meta-analysis typically involves combining the estimates from independent studies in order to estimate a parameter of interest across a population of studies. However, outliers often occur even under the random effects model. The presence of such outliers could substantially alter the conclusions in a meta-analysis. This paper proposes a methodology for identifying and, if desired, downweighting studies that do not appear representative of the population they are thought to represent under the random effects model. METHODS: An outlier is taken as an observation (study result) with an inflated random effect variance. We used the likelihood ratio test statistic as an objective measure for determining whether observations have inflated variance and are therefore considered outliers. A parametric bootstrap procedure was used to obtain the sampling distribution of the likelihood ratio test statistics and to account for multiple testing. Our methods were applied to three illustrative and contrasting meta-analytic data sets. RESULTS: For the three meta-analytic data sets our methods gave robust inferences when the identified outliers were downweighted. CONCLUSIONS: The proposed methodology provides a means to identify and, if desired, downweight outliers in meta-analysis. It does not eliminate them from the analysis however and we consider the proposed approach preferable to simply removing any or all apparently outlying results. We do not however propose that our methods in any way replace or diminish the standard random effects methodology that has proved so useful, rather they are helpful when used in conjunction with the random effects model.
  • Thumbnail Image
    Item
    Open Access
    Risks and benefits of hormone therapy: has medical dogma now been overturned?
    (2014) Shapiro, S; de Villiers, T J; Pines, A; Sturdee, D W; Baber, R J; Panay, N; Stevenson, J C; Mueck, A O; Burger, H G
    BACKGROUND In an integrated overview of the benefits and risks of menopausal hormone therapy (HT), the Women's Health Initiative (WHI) investigators have claimed that their 'findings … do not support use of this therapy for chronic disease prevention'. In an accompanying editorial, it was claimed that 'the WHI overturned medical dogma regarding menopausal [HT]'. OBJECTIVES To evaluate those claims. METHODS Epidemiological criteria of causation were applied to the evidence. RESULTS A 'global index' purporting to summarize the overall benefit versus the risk of HT was not valid, and it was biased. For coronary heart disease, an increased risk in users of estrogen plus progestogen (E + P), previously reported by the WHI, was not confirmed. The WHI study did not establish that E+ P increases the risk of breast cancer; the findings suggest that unopposed estrogen therapy (ET) does not increase the risk, and may even reduce it. The findings for stroke and pulmonary embolism were compatible with an increased risk, and among E+ P users there were credible reductions in the risk of colorectal and endometrial cancer. For E+ P and ET users, there were credible reductions in the risk of hip fracture. Under 'worst case' and 'best case' assumptions, the changes in the incidence of the outcomes attributable to HT were minor. CONCLUSIONS Over-interpretation and misrepresentation of the WHI findings have damaged the health and well-being of menopausal women by convincing them and their health professionals that the risks of HT outweigh the benefits.