Guidelines for Application of Metaanalysis in Environmental Epidemiology

Blair, Aaron; Burg, JeAnne R.; Foran, Jeffery A.; Gibb, Herman J.; Greenland, Sander; Morris, Robin; Raabe, Gerhard K.; Savitz, David A.; Teta, J.; Wartenberg, Daniel; Wong, Otto; Zimmerman, Ryan M.

doi:10.1006/rtph.1995.1084

Cited by 101 publications

(75 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…7 In the meta-analysis, no assessment of the methodological quality of the studies was made, because this is highly arbitrary. 21,47,48 The advantage of this meta-analysis is that it clearly identifies sources that could potentially modify the relationship between insulin and CVD. Moreover, it reveals issues that vary greatly between studies (eg, number and type of confounding or intermediate variables) or are in need of improvement (eg, correct presentation of significant as well as nonsignificant data).…”

Section: Discussionmentioning

confidence: 99%

Insulin and Risk of Cardiovascular Disease

et al. 1998

View full text Add to dashboard Cite

Background-Our purposes were to estimate the strength of the longitudinal relationship between hyperinsulinemia and cardiovascular diseases (CVD) from the available literature and to identify study characteristics that modify this relationship. Methods and Results-Articles were identified by means of a MEDLINE and Embase search and citation tracking. Eligible studies were prospective population-based cohort studies and nested case-control studies on the relationship between, on the one hand, fasting or nonfasting insulin levels and, on the other hand, myocardial infarction, death from coronary heart disease, and/or ECG abnormalities. Data were extracted pertaining to insulin measurements, type of outcome studied, adjustment for confounding, sex, mean age of the study population, follow-up period, insulin assay, and ethnic background (white or nonwhite). Associations of insulin and CVD were reexpressed in a uniform manner, an estimate of relative risk (RR) and 95% CI, to be used in meta-regression analyses. Twelve of 17 potentially eligible articles provided sufficient information. Overall, a weak positive association was found. The meta-analysis resulted in an estimated summary RR (95% CI) of 1.18 (1.08 to 1.29) for differences in insulin level, equivalent to the difference between the 75th and the 25th percentiles of the general population in the Netherlands. Ethnic background and type of insulin assay modified the relationship between insulin and CVD with borderline significance. Conclusions-Hyperinsulinemia is a weak risk indicator for the occurrence of CVD. The relationship between hyperinsulinemia and CVD was modified by ethnic background and by the type of insulin assay involved. (Circulation. 1998;97:996-1001.)

show abstract

Section: Discussionmentioning

confidence: 99%

Insulin and Risk of Cardiovascular Disease

et al. 1998

View full text Add to dashboard Cite

show abstract

“…The wire code/distance and calculated-exposure results moved closer to the null, largely owing to the results of Linet et al 4' and Tynes and Haldorsen.44…”

Section: All Studies Combinedmentioning

confidence: 64%

Residential magnetic fields and childhood leukemia: a meta-analysis.

Wartenberg¹

1998

Am J Public Health

View full text Add to dashboard Cite

OBJECTIVES: This article uses meta-analysis methodology to examine the statistical consistency and importance of random variation among results of epidemiologic studies of residential magnetic field exposure and childhood leukemia. METHODS: A variety of meta-analytic statistical methods were applied to all available studies combined and on sub-groups of studies chosen by exposure characteristics. Sample sizes and fail-safe n's were calculated to determine the robustness of results and the potential role of publication bias. RESULTS: Most studies show elevated but not statistically significant odds ratios. Results for exposures assessed by wire codes, distance, and/or historically reconstructed fields are relatively consistent, homogeneous, and positive, while those for direct magnetic field measurements are consistent, homogeneous, and marginally protective. Several unpublished studies, or a single unpublished study with several hundred subjects, would be needed to nullify the observed data. CONCLUSIONS: The observed results identify a consistent risk that cannot be explained by random variation. The data supporting magnetic fields as the principal risk factor are suggestive but inconsistent. Additional studies using innovative designs that focus on highly exposed children offer the most hope of untangling this issue.

show abstract

“…A systematic review provides an optimal opportunity to place studies in a broader context 16. Sometimes the interpretation of a meta‐analysis is straightforward, when all studies give the same picture.…”

Section: Answering the Research Questionmentioning

confidence: 99%

Meta‐analysis: Key features, potentials and misunderstandings

Dekkers

2018

Research and Practice in Thrombosis and Haemostasis

View full text Add to dashboard Cite

A meta‐analysis consists of a systematic approach to combine different studies in one design. Preferably, a protocol is written and published spelling out the research question, eligibility criteria, risk of bias assessment, and statistical approach. Included studies are likely to display some diversity regarding populations, calendar period, or treatment settings. Such diversity should be considered when deciding whether to combine (some) studies in a formal meta‐analysis. Statistically, the fixed effect model assumes that all studies estimate the same underlying true effect. This assumption is relaxed in a random effects model and given the expected study diversity a random effects approach will often be more realistic. In the absence of statistical heterogeneity, fixed and random effects models give identical estimates. Meta‐analyses are especially useful to provide a broader scope of the literature; they should carefully explore sources of between study heterogeneity and may show a treatment effect or an exposure–outcome association where individual studies are not powered. However, its validity largely depends on the validity of included studies.

show abstract

Guidelines for Application of Metaanalysis in Environmental Epidemiology

Cited by 101 publications

References 0 publications

Insulin and Risk of Cardiovascular Disease

Insulin and Risk of Cardiovascular Disease

Residential magnetic fields and childhood leukemia: a meta-analysis.

Meta‐analysis: Key features, potentials and misunderstandings

Contact Info

Product

Resources

About