Body mass index and 20 specific cancers: re-analyses of dose–response meta-analyses of observational studies
- 28 December 2017
- journal article
- research article
- Published by Elsevier BV in Annals of Oncology
- Vol. 29 (3), 749-757
- https://doi.org/10.1093/annonc/mdx819
Abstract
Objectives were to provide an overview and understand the strength of evidence and extent of potential biases and validity of claimed associations between body mass index (BMI) and risk of developing cancer. We carried out an umbrella review and comprehensively re-analyzed the data of dose–response meta-analyses on associations between BMI and risk of 20 specific cancers (bladder, brain, breast, colonic, rectal, endometrial, gallbladder, gastric, leukemia, liver, lung, melanoma, multiple myeloma, non-Hodgkins lymphoma, esophagus, ovarian, pancreatic, prostate, renal, thyroid) by adding big data or missed individual studies. Convincing evidence for an association was defined as a strong statistical significance in fixed-effects and random-effects meta-analyses at P < 0.001, 95% prediction interval (PI) excluded null, there was no large between-study heterogeneity and no small study effects. Suggestive evidence was defined as meeting the significance threshold for the random summary effects of P < 0.05, but 95% PI included the null. Weak evidence was defined as meeting the significance threshold for the random summary effects at a P < 0.05, but 95% PI included the null and there was large between-study heterogeneity or there were small study effects. Convincing evidence for an association with BMI was detectable for six cancers (leukemia, multiple myeloma, pancreatic, endometrial, rectal, and renal cell carcinoma). Suggestive evidence was detectable for malignant melanoma, non-Hodgkins lymphoma, and esophageal adenocarcinoma. Weak evidence was detectable for brain and central nervous system tumors, breast, colon, gall bladder, lung, liver, ovarian, and thyroid cancer. No evidence was detectable for bladder, gastric, and prostate cancer. The association of increased BMI and cancer is heterogeneous across cancer types. Leukemia, multiple myeloma, pancreatic, endometrial, rectal, and renal cell carcinoma are convincingly associated with an increased BMI by dose–response meta-analyses.Keywords
This publication has 23 references indexed in Scilit:
- Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-years for 32 Cancer Groups, 1990 to 2015JAMA Oncology, 2017
- The Global Burden of Cancer 2013JAMA Oncology, 2015
- Obesity and Risk of Bladder Cancer: A Dose-Response Meta-Analysis of 15 Cohort StudiesPLOS ONE, 2015
- Environmental risk factors and multiple sclerosis: an umbrella review of systematic reviews and meta-analysesThe Lancet Neurology, 2015
- Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010The Lancet, 2012
- The False-positive to False-negative Ratio in Epidemiologic StudiesEpidemiology, 2011
- A Re-Evaluation of Random-Effects Meta-AnalysisJournal of the Royal Statistical Society Series A: Statistics in Society, 2008
- Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studiesThe Lancet, 2008
- What is heterogeneity and is it important?BMJ, 2007
- Bias in meta-analysis detected by a simple, graphical testBMJ, 1997