Table 1. Dietary Factors, Related Cancer Outcomes, Cancer Relative Risks, and Effect Estimates on Body Mass Indexa
Dietary Factors
|
Cancers associated via direct risk
|
Unit of RR
|
Cancer RR (95% UI) Per Unit of RRb
|
Effect Estimates on BMI
kg/m2 (95% UI)
(Per 1 serving/d)c
|
Fruits
|
Mouth and pharynx, and Larynx cancer
|
1 serving/day (100 g)
|
0.95 (0.91, 1.00)
|
For baseline BMI <25: -0.06 (-0.08, -0.04)
For baseline BMI ≥25: -0.11 (-0.16, -0.06)
|
Vegetables
|
Mouth and pharynx, and Larynx cancer
|
1 serving/day (100 g)
|
0.91 (0.87, 0.96)
|
For baseline BMI <25: -0.03 (-0.04, -0.01)
For baseline BMI ≥25: -0.06 (-0.09, -0.02)
|
Whole grains
|
Colorectal cancer
|
1 serving/day (90 g)
|
0.83 (0.78, 0.89)
|
For baseline BMI <25: -0.05 (-0.07, -0.03)
For baseline BMI ≥25: -0.08 (-0.10, -0.06)
|
Processed meats
|
Colorectal and Stomach cancer
|
1 serving/day (50 g)
|
Colorectal: 1.16 (1.08, 1.26)
Stomach: 1.18
(1.01, 1.38)
|
For baseline BMI <25: 0.13 (0.07, 0.19)
For baseline BMI ≥25: 0.16 (0.11, 0.21)
|
Red meats
|
Colorectal cancer
|
1 serving/day (100 g)
|
1.12 (1.00, 1.25)
|
For baseline BMI <25: 0.13 (0.07, 0.20)
For baseline BMI ≥25: 0.23 (0.14, 0.32)
|
Dairy
|
Colorectal cancer
|
1.6 servings/day (400 g)
|
0.87
(0.83, 0.90)
|
No effect estimates of dairy products on BMI
|
SSBs
|
No cancers associated with direct risk
|
No direct RR of SSB on cancer
|
For baseline BMI <25: 0.09 (0.05, 0.14)
For baseline BMI ≥25: 0.23 (0.14, 0.32)
|
Adapted with permission from Zhang FF, Cudhea F, Shan Z, et al. Preventable Cancer Burden Associated with Poor Diet in the United States. JNCI Cancer Spectrum. 2019;3(2).
RR=Relative Risk; BMI=Body Mass Index; UI=Uncertainty Interval
aThis table is adapted from Zhang et al. (2019)[6]
bRR estimates were based on meta-analyses of prospective cohort studies with limited evidence of bias from confounding, where the associations were multivariable adjusted and independent of obesity.
cObesity is associated with an increased risk of 13 cancers (colorectal, stomach, post-menopausal breast, ovary, corpus uteri, advanced prostate, liver, kidney, pancreas, gallbladder, thyroid, multiple myeloma, esophageal adenocarcinoma). Although there is no direct RR for SSB and cancer, SSB can increase the risk of cancer mediated through obesity. There are no effect estimates of dairy products on obesity.
Table 2. Estimated five-year medical costs (2018 billion $) of new cancer cases diagnosed in 2015, total and attributable to suboptimal diet, by cancer site
Cancer site
|
Number of new cancer cases (95% UI)a
|
5-year medical costb of cancer care attributable to suboptimal diet
(95% UI)c
|
5-year medical cost of cancer care
(95% UI)d
|
|
|
Directe
|
Indirectf
|
Total
|
% of total 5-year medical cost of cancer careg
|
|
Post-menopausal breast
|
1,900
(1,443-2,392)
|
0 (0-0)
|
0.09
(0.08-0.12)
|
0.09
(0.08-0.12)
|
0.86
(0.7-1.04)
|
10.45 (10.29-10.62)
|
Ovary
|
107
(64-157)
|
0 (0-0)
|
0.01
(0.01-0.02)
|
0.01
(0.01-0.02)
|
0.46
(0.31-0.64)
|
2.81 (2.74-2.88)
|
Corpus uteri
|
1,935
(1,561-2,313)
|
0 (0-0)
|
0.11
(0.09-0.13)
|
0.11
(0.09-0.13)
|
3.39
(2.88-3.92)
|
3.17 (3.08-3.27)
|
Advanced Prostate
|
327
(146-536)
|
0 (0-0)
|
0.03
(0.02-0.04)
|
0.03
(0.02-0.04)
|
0.7
(0.44-0.96)
|
3.51 (3.33-3.71)
|
Kidney
|
1,408
(1,191-1,632)
|
0 (0-0)
|
0.11
(0.09-0.13)
|
0.11
(0.09-0.13)
|
2.3
(1.96-2.66)
|
4.64 (4.47-4.82)
|
Liver
|
803
(630-997)
|
0 (0-0)
|
0.07
(0.05-0.09)
|
0.07
(0.05-0.09)
|
2.3
(1.83-2.82)
|
2.94 (2.8-3.08)
|
Pancreas
|
404
(314-501)
|
0 (0-0)
|
0.04
(0.03-0.05)
|
0.04
(0.03-0.05)
|
0.81
(0.65-0.99)
|
4.6 (4.46-4.74)
|
Multiple myeloma
|
205
(143-277)
|
0 (0-0)
|
0.03
(0.02-0.05)
|
0.03
(0.02-0.05)
|
0.74
(0.52-1)
|
4.41 (4.25-4.57)
|
Thyroid
|
264
(195-337)
|
0 (0-0)
|
0.01
(0.01-0.02)
|
0.01
(0.01-0.02)
|
0.51
(0.38-0.64)
|
2.06 (1.87-2.27)
|
Gallbladder
|
74
(57-92)
|
0 (0-0)
|
0.01
(0.01-0.01)
|
0.01
(0.01-0.01)
|
1.81
(1.49-2.11)
|
0.38 (0.35-0.42)
|
Esophageal adenocarcinoma
|
368
(276-469)
|
0 (0-0)
|
0.04
(0.03-0.05)
|
0.04
(0.03-0.05)
|
3.46
(2.93-4.04)
|
1.06 (1.02-1.11)
|
Larynx
|
3,034
(2,177-3,820)
|
0.26
(0.17-0.37)
|
0 (0-0)
|
0.26
(0.17-0.37)
|
25.57
(17.13-34.32)
|
0.98 (0.92-1.04)
|
Mouth and pharynx
|
12,730
(9,642-15,750)
|
1.09
(0.7-1.54)
|
0 (0-0)
|
1.09
(0.7-1.54)
|
25.55
(16.95-35)
|
4.06 (3.92-4.2)
|
Colorectal
|
60,748
(51,832-69,462)
|
5.28
(4.35-6.21)
|
0.07
(0.03-0.12)
|
5.32
(4.43-6.22)
|
36.88
(31.25-42.36)
|
13.8 (13.56-14.05)
|
Stomach
|
151
(98-213)
|
0.2
(0.06-0.41)
|
0.01
(0.01-0.02)
|
0.21
(0.07-0.43)
|
7.94
(2.62-15.52)
|
2.55 (2.46-2.65)
|
All cancer sites
|
84,459
(69,766-98,947)
|
6.83
(5.27-8.52)
|
0.64
(0.47-0.84)
|
7.44
(5.79-9.27)
|
11.58
(9.31-13.97)
|
61.44 (59.52-63.44)
|
This table is original to the manuscript
UI = Uncertainty Interval. 95% UIs were obtained through Monte Carlo simulation using 1000 iterations.
aFor each cancer type, new cancer cases attributable to diet were estimated using the Comparative Risk Assessment model using cancer incidence and Joint Population Attributable fraction. For cancer with more than one associated dietary factor, joint PAFs were computed using the proportional multiplication formula for cumulative effects, such that the joint PAF is less than the sum of the PAFs for the contributing dietary factors. The cancer incidence in 2015 was used to generate the estimates of new cancer cases attributable to diet. 95% UIs were obtained through Monte Carlo simulation using 1000 iterations.
b All costs are reported in 2018 billion $.
cThe 5-year medical costs for cancers attributable to diet were computed using a Markov cohort model that incorporates cancer incidence, Population attributable fraction, monthly phase-specific costs of cancer care and monthly probability of death due to cancer and other causes. 95% UIs were obtained through Monte Carlo simulation using 1000 iterations.
dThe total 5-year medical costs attributable to diet were computed using a Markov cohort model that incorporates cancer incidence, monthly phase-specific costs of cancer care and monthly probability of death due to cancer and other causes. 95% UIs were obtained through Monte Carlo simulation using 1000 iterations.
eThese are the 5-year medical costs for cancers attributable to direct effect of dietary factors.
fThese are the 5-year medical costs for cancers attributable to obesity-mediated effect of dietary factors. Total costs are a sum of medical costs for cancers attributable to both direct and obesity-mediated effects of dietary factors.
gComputed as (Total 5-year costs of cancer care attributable to diet/5-year medical cost of cancer care for cases diagnosed in the same year)*100.
Table 3. Estimated five-year medical costs (2018 billion $) of new cancers diagnosed in 2015 attributable to suboptimal diet, by dietary factors and type of association
Dietary factora
|
5-year medical costb of cancer care attributable to dietary factor
(95% UI)
|
|
Direct associationc
|
Obesity-mediated associationd
|
Total
|
Low Consumption of whole grain
|
2.66
(1.88-3.55)
|
0.09
(0.05-1.4)
|
2.76
(1.89-4.96)
|
Low consumption of fruits
|
0.46
(0.15-0.94)
|
0.16
(0.09-0.27)
|
0.62
(0.23-1.22)
|
Low consumption of vegetables
|
0.99
(0.54-1.48)
|
0.1
(0.05-0.19)
|
1.09
(0.59-1.67)
|
Low consumption of dairy
|
1.83
(1.35-2.34)
|
0
(0-0)
|
1.82
(1.34-2.36)
|
High consumption of sugar-sweetened beverages
|
0 (0-0)
|
0.13
(0.05-0.25)
|
0.13
(0.05-0.25)
|
High consumption of red meat
|
0.47
(0.14-1.03)
|
0.06
(0.03-0.09)
|
0.54
(0.17-1.11)
|
High consumption of processed meat
|
1.4
(0.72-2.29)
|
0.11
(0.05-0.21)
|
1.5
(0.77-2.48)
|
This table is original to the manuscript
UI = Uncertainty Intervals. 95% UIs were obtained through Monte Carlo simulation using 1000 iterations.
aThe 5-year medical costs for cancers attributable to each dietary factor were obtained by summing the costs for all cancer sites attributable to the particular dietary factor. 95% UIs were obtained through Monte Carlo simulation using 1000 iterations.
bAll costs are reported in 2018 billion $
cThese are the 5-year medical costs for cancers attributable to direct effect of each dietary factor.
dThese are the 5-year medical costs for cancers attributable to obesity-mediated effect of dietary factors. Total costs are a sum of medical costs for cancers attributable to both direct and obesity-mediated effects of each dietary factor.
Table 4. Estimated total and per-patient five-year medical costs of new cancer cases diagnosed in 2015 attributable to suboptimal diet, by demographic factors
Demographic groupa
|
5-year medical costb of cancer care attributable to suboptimal diet
|
|
Total cost in 2018 billion $ (95% UI)
|
Cost per cancer patient in 2018 $ (95% UI)c
|
Race
|
White
|
5.96
(4.64-7.43)
|
8,963
(6,968-11,172)
|
|
Black
|
1.08
(0.85-1.34)
|
11,068
(8,699-13,670)
|
|
Other
|
0.38
(0.29-0.48)
|
9,469
(7,338-11,895)
|
Sex
|
Males
|
4.53
(3.48-5.69)
|
14,575
(11,195-18,302)
|
|
Females
|
2.9
(2.3-3.56)
|
5,868
(4,664-7,212)
|
Age
|
20-44 years
|
0.52
(0.41-0.64)
|
11,858
(9,384-14,550)
|
|
45-54 years
|
1.27
(0.99-1.57)
|
11,786
(9,235-14,620)
|
|
55-64 years
|
2.16
(1.67-2.7)
|
9,754
(7,535-12,193)
|
|
65+
|
3.48
(2.71-4.34)
|
8,063
(6,278-10,054)
|
This table is original to the manuscript
UI = Uncertainty Intervals. 95% UIs were obtained through Monte Carlo simulation using 1000 iterations.
aThe 5-year medical costs for cancers attributable to suboptimal diet for each demographic group were obtained by summing the costs across all cancer sites for each demographic group. 95% UIs were obtained through Monte Carlo simulation using 1000 iterations.
bThe cost per cancer patient is obtained by dividing the cost for each demographic group by the cancer incidence for each demographic group.