Fruit and vegetable consumption and mortality from all causes, cardiovascular disease, and cancer: systematic review and dose-response meta-analysis of prospective cohort studies
Wang et al. 1 performed a meta-analysis to investigate the association between fruit and vegetable consumption and mortality from all causes, cardiovascular disease and cancer. However, I have three different opinions.
First, they described that “Pooled hazard ratios of all-cause mortality were 0.95 (95% confidence interval 0.92 to 0.98) for an increment of one serving a day of fruit and vegetables (P=0.001), 0.94 (0.90 to 0.98) for fruit (P=0.002), and 0.95 (0.92 to 0.99) for vegetables (P=0.006)”. In my opinion, those descriptions above are incorrect. The authors used generalized least squares trend estimation method developed by Greenland and Longnecker2,3 to estimate the dose-response relationship. Based on their results, there was evidence of curvilinear associations between fruit and vegetable combined or alone and all cause, cardiovascular, and cancer mortality (all P less than 0.05 for non-linearity). Thus, it is wrong to further emphasize the linear dose-response associations and present pooled relative risk for an increment of one serving a day of fruit and vegetable combined, fruit and vegetable alone since those associations did not conform to linear dose-response relationship. And the authors should describe the curvilinear associations in detail.
Second, I noted that the protective effects of fruit and vegetable consumption on risk of total, cardiovascular, and cancer mortality were very small, the decreased risk ranging from 5% to 6%. In addition, the upper limit of confidence interval ranged from 0.98 to 0.99, approximately equal to 1. Thus, the significant association might be due to confounding factors or residual error.
Third, I also query about the use of 77 g for vegetables and 80 g for fruits to define one serving or each time intake. Wang et al. 1 cited the article by He FJ et al.4 to support their definition, but to my knowledge, 80 g for vegetables and 100 g for fruits were more widely accepted to quantify one serving or each time intake 5.
References
1. Wang X, Ouyang Y, Liu J, Zhu M, Zhao G, Bao W, Hu FB. Fruit and vegetable consumption and mortality from all causes, cardiovascular disease, and cancer: systematic review and dose-response meta-analysis of prospective cohort studies. BMJ 2014;349:g4490.
2. Greenland S, Longnecker MP. Methods for trend estimation from summarizeddose-response data, with applications to meta-analysis. Am J Epidemiol1992;135:1301-9.
3. Orsini N, Bellocco R, Greenland S. Generalized least squares for trend estimation ofsummarized dose-response data. Stata J 2006;6:40-57.
4. He FJ, Nowson CA, MacGregor GA. Fruit and vegetable consumption and stroke: meta-analysis of cohort studies. Lancet 2006;367:320-6.
5. Riboli E, Norat T. Epidemiologic evidence of the protective effect of fruit and vegetables on cancer risk. Am J Clin Nutr. 2003;78:559S-69S.
Competing interests:
No competing interests
07 August 2014
Liu Chibo
Doctor
Department of Clinical Laboratory, Taizhou Municipal Hospital
Rapid Response:
Wang et al. 1 performed a meta-analysis to investigate the association between fruit and vegetable consumption and mortality from all causes, cardiovascular disease and cancer. However, I have three different opinions.
First, they described that “Pooled hazard ratios of all-cause mortality were 0.95 (95% confidence interval 0.92 to 0.98) for an increment of one serving a day of fruit and vegetables (P=0.001), 0.94 (0.90 to 0.98) for fruit (P=0.002), and 0.95 (0.92 to 0.99) for vegetables (P=0.006)”. In my opinion, those descriptions above are incorrect. The authors used generalized least squares trend estimation method developed by Greenland and Longnecker2,3 to estimate the dose-response relationship. Based on their results, there was evidence of curvilinear associations between fruit and vegetable combined or alone and all cause, cardiovascular, and cancer mortality (all P less than 0.05 for non-linearity). Thus, it is wrong to further emphasize the linear dose-response associations and present pooled relative risk for an increment of one serving a day of fruit and vegetable combined, fruit and vegetable alone since those associations did not conform to linear dose-response relationship. And the authors should describe the curvilinear associations in detail.
Second, I noted that the protective effects of fruit and vegetable consumption on risk of total, cardiovascular, and cancer mortality were very small, the decreased risk ranging from 5% to 6%. In addition, the upper limit of confidence interval ranged from 0.98 to 0.99, approximately equal to 1. Thus, the significant association might be due to confounding factors or residual error.
Third, I also query about the use of 77 g for vegetables and 80 g for fruits to define one serving or each time intake. Wang et al. 1 cited the article by He FJ et al.4 to support their definition, but to my knowledge, 80 g for vegetables and 100 g for fruits were more widely accepted to quantify one serving or each time intake 5.
References
1. Wang X, Ouyang Y, Liu J, Zhu M, Zhao G, Bao W, Hu FB. Fruit and vegetable consumption and mortality from all causes, cardiovascular disease, and cancer: systematic review and dose-response meta-analysis of prospective cohort studies. BMJ 2014;349:g4490.
2. Greenland S, Longnecker MP. Methods for trend estimation from summarizeddose-response data, with applications to meta-analysis. Am J Epidemiol1992;135:1301-9.
3. Orsini N, Bellocco R, Greenland S. Generalized least squares for trend estimation ofsummarized dose-response data. Stata J 2006;6:40-57.
4. He FJ, Nowson CA, MacGregor GA. Fruit and vegetable consumption and stroke: meta-analysis of cohort studies. Lancet 2006;367:320-6.
5. Riboli E, Norat T. Epidemiologic evidence of the protective effect of fruit and vegetables on cancer risk. Am J Clin Nutr. 2003;78:559S-69S.
Competing interests: No competing interests