Facilitating meta-analyses by deriving relative effect and precision estimates for alternative comparisons from a set of estimates presented by exposure level or disease category.
ABSTRACT Epidemiological studies relating a particular exposure to a specified disease may present their results in a variety of ways. Often, results are presented as estimated odds ratios (or relative risks) and confidence intervals (CIs) for a number of categories of exposure, for example, by duration or level of exposure, compared with a single reference category, often the unexposed. For systematic literature review, and particularly meta-analysis, estimates for an alternative comparison of the categories, such as any exposure versus none, may be required. Obtaining these alternative comparisons is not straightforward, as the initial set of estimates is correlated. This paper describes a method for estimating these alternative comparisons based on the ideas originally put forward by Greenland and Longnecker, and provides implementations of the method, developed using Microsoft Excel and SAS. Examples of the method based on studies of smoking and cancer are given. The method also deals with results given by categories of disease (such as histological types of a cancer). The method allows the use of a more consistent comparison when summarizing published evidence, thus potentially improving the reliability of a meta-analysis.
- SourceAvailable from: Toni Alterman[show abstract] [hide abstract]
ABSTRACT: To determine the relative risk (RR) of lung cancer in lifetime never smokers associated with environmental tobacco smoke (ETS) exposure. Multicenter population-based case-control study. Five metropolitan areas in the United States: Atlanta, Ga, Houston, Tex, Los Angeles, Calif, New Orleans, La, and the San Francisco Bay Area, Calif. Female lifetime never smokers: 653 cases with histologically confirmed lung cancer and 1253 controls selected by random digit dialing and random sampling from the Health Care Financing Administration files for women aged 65 years and older. The RR of lung cancer, estimated by adjusted odds ratio (OR) with 95% confidence interval (CI), associated with ETS exposure. Tobacco use by spouse(s) was associated with a 30% excess risk of lung cancer: all types of primary lung carcinoma (adjusted OR = 1.29; P < .05), pulmonary adenocarcinoma (adjusted OR = 1.28; P < .05), and other primary carcinomas of the lung (adjusted OR = 1.37; P = .18). An increasing RR of lung cancer was observed with increasing pack-years of spousal ETS exposure (trend P = .03), such that an 80% excess risk of lung cancer was observed for subjects with 80 or more pack-years of exposure from a spouse (adjusted OR = 1.79; 95% CI = 0.99 to 3.25). The excess risk of lung cancer among women ever exposed to ETS during adult life in the household was 24%; in the workplace, 39%; and in social settings, 50%. When these sources were considered jointly, an increasing risk of lung cancer with increasing duration of exposure was observed (trend P = .001). At the highest level of exposure, there was a 75% increased risk. No significant association was found between exposure during childhood to household ETS exposure from mother, father, or other household members; however, women who were exposed during childhood had higher RRs associated with adult-life ETS exposures than women with no childhood exposure. At the highest level of adult smoke-years of exposure, the ORs for women with and without childhood exposures were 3.25 (95% CI, 2.42 to 7.46) and 1.77 (95% CI, 0.98 to 3.19), respectively. Exposure to ETS during adult life increases risk of lung cancer in lifetime nonsmokers.JAMA The Journal of the American Medical Association 06/1994; 271(22):1752-9. · 29.98 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: We discuss the problem of describing multiple group comparisons in survival analysis using the Cox model, and in matched case-control studies. The standard method of comparing the risk in each group with a baseline group is unsatisfactory because the standard errors and confidence limits relate to correlated parameters, all dependent on precision within the baseline group. We describe the construction of standard errors for the parameters of all groups, without the need to select a baseline group. These standard errors can be regarded as relating to roughly independent parameters, so that groups can be compared efficiently without knowledge of the covariances. The method should assist in graphical presentation of relative risks, and in the combination of results from published studies. Two examples are presented.Statistics in Medicine 08/1991; 10(7):1025-35. · 2.04 Impact Factor
- American Journal of Epidemiology 09/2000; 152(4):393-4. · 4.78 Impact Factor