University of Southern California Dissertations and Theses

Red and processed meat consumption and colorectal cancer risk: meta-analysis of case-control studies

(USC Thesis Other)

Red and processed meat consumption and colorectal cancer risk: meta-analysis of case-control studies

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content i

RED AND PROCESSED MEAT CONSUMPTION AND COLORECTAL CANCER RISK:
META-ANALYSIS OF CASE-CONTROL STUDIES

by
Lu Tan

A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(APPLIED BIOSTATISTICS AND EPIDEMIOLOGY)

December 2017

Copyright 2017 Lu Tan

ii
ACKNOWLEDGEMENTS

I would like to thank my advisor Professor Mariana Stern, for her guidance and support
throughout my research and writing process. I would like to thank Dr. Sabrina Schlesinger for
answering my analysis and methodological questions patiently. I would also like to thank
committee members, Professor Anna Wu and Professor Wendy Mack, for providing valuable
suggestions. Finally, I would like to thank my family and friends for all their support and
encouragement with love.

iii
TABLE OF CONTENTS

1. List of Tables and Figures iv
2. Abstract v
3. Introduction 1
4. Methods 3
a. Data source and search 3
b. Inclusion criteria and data extraction 3
c. Statistical analysis 4
5. Results 7
a. Results of search and study selection 7
b. Red meat 7
c. Processed meat 8
d. Red meat including processed meat 8
e. Individual red meat 9
f. Small study or Publication bias 10
g. Cumulative meta-analyses 10
h. Subgroup analyses 11
6. Discussion 33
7. References 39

iv
LIST OF TABLES AND FIGURES
1. Figure 1: PRISMA flow diagram of systematic literature search on red and processed
meat and the risk of colorectal, colon, and rectal cancer 16

2. Table 1: Characteristics of studies included in the meta-analysis 17

3. Table 2: Summary odds ratios of random effects meta-analyses of red meat, processed meat,
red including processed meat, and colorectal cancer for all studies and
by subgroups. 19

4. Table 3: Summary meta-analyses odds ratios of highest versus lowest intake of
individual red meat (beef and pork) and colorectal cancer. 22

5. Figure 2: Random effects meta-analysis of the highest versus the lowest consumption
of red meat and the risk of Colon, Rectal, and Colorectal cancers. 23

6. Figure 3: Dose-response meta-analysis of red meat consumption and the risk of Colon,
Rectal, and Colorectal cancers per increase of 100 gr per day. 24

7. Figure 4: Random effects meta-analysis of the highest versus the lowest consumption
of processed meat and the risk of Colon, Rectal, and Colorectal cancers. 25

8. Figure 5: Dose-response meta-analysis of processed meat consumption and the risk of
Colon, Rectal, and Colorectal cancers per increase of 100 gr per day. 26

9. Figure 6: Random effects meta-analysis of the highest versus the lowest consumption
of red meat including processed meat and the risk of Colon, Rectal, and
Colorectal cancers. 27

10. Figure 7: Dose-response meta-analysis of consumption of red meat including processed
meat and the risk of Colon, Rectal, and Colorectal cancers per increase of 100
gr per day. 28

11. Figure 8: Random effects meta-analysis of the highest versus the lowest consumption
of beef and the risk of Colon, Rectal, and Colorectal cancers. 29

12. Figure 9: Random effects meta-analysis of the highest versus the lowest consumption
of pork and the risk of Colon, Rectal, and Colorectal cancers. 30

13. Figure 10: Cumulative random-effects meta-analysis of highest versus lowest level red
meat consumption by cancer type, publication year, and variable
definition 31

14. Figure 11: Funnel Plots with P-values from Beggar’s test 38

v
ABSTRACT

Background: Red and processed meat consumption have been linked to several cancers,
including colorectal cancer (CRC). In 2015, the International Agency for Research on Cancer
classified red meat as a probable carcinogen and processed meat as carcinogenic. Although a
meta-analysis of prospective studies has been published, a comparable meta-analysis of case-
control studies is currently lacking, in particular, with consideration of the great variability found
across studies regarding analytical approaches, study design, and overall study quality.

Methods and Findings: Case-control studies on risk of colorectal cancer in relation to red and/or
processed meat consumption were obtained by searching in PubMed until November 2016. For
studies that met our selection criteria (see Methods), study-specific ORs were pooled using
fixed-effects model or random-effects models. The highest versus the lowest intake comparison,
dose-response analyses, and stratified analyses were conducted to estimate the red/processed
meat and CRC risk association, and to assess the role of potential confounders on the observed
associations. The summary meta-analysis random effects ORs of the highest versus lowest red
meat consumption comparison for colorectal and colon cancer were 1.24 (95% CI: 1.04-1.49)
and 1.22 (95% CI: 1.02-1.45). Similar results were obtained for processed meat. In dose-
response analyses, with every 100 g/day increment of processed meat consumption, the odds of
having colorectal cancer increases by 197% (95% CI: 59%-455%). The summary ORs of the
highest versus lowest red including processed meat consumption comparison for CRC, colon,
and rectal cancer were 1.51 (95% CI: 1.22-1.89), 1.30 (95% CI: 1.05-1.60), and 1.44 (95% CI:
1.08-1.93), respectively. In dose-response analyses, red meat including processed meat was
vi
positively associated with increased risk of colorectal cancer (summary OR = 1.62, 95% CI =
1.25-2.11). In general, inclusion or exclusion of confounders (total energy intake, BMI, Physical
activity, Fruits and/or vegetable intake, Alcohol, and smoking) seemed to have little influence on
the associations for processed meat. In some instances, marginally higher for studies that did not
include confounders. Studies of red meat and processed meat that considered total energy intake
adjustment showed a stronger positive association with colorectal cancer (OR
red meat and CRC
= 1.32,
95% CI = 1.05-1.65; OR
processed meat and CRC
= 1.52, 95% CI = 1.24-1.86) than those that did not
(OR
red meat and CRC
= 1.03, 95% CI = 1.01-1.04; OR
processed meat and CRC
= 1.37, 95% CI = 0.79-2.40).
For studies of red meat including processed meat, the positive association was slightly stronger
for those studies that did not adjust (OR
red including processed meat and CRC
= 1.69, 95% CI = 1.44-1.99),
and did not reach statistical significance among those that did (OR
red including processed meat and CRC
=
1.37, 95% CI = 0.98-1.90).

Conclusion: High intake of red meat, as well as processed meat, is associated with increased risk
of colorectal, colon, and rectal cancer. Our findings highlight that consideration of the type of
controls used, FFQ validation, and adjustment for energy intake, BMI, physical activity, fruits
and vegetables, alcohol, and smoking, seem critical to obtain estimates most comparable to those
from prospective cohort studies.

1

INTRODUCTION
Colorectal cancer (CRC), which consists of colon cancer and rectal cancer, is one of the most
common cancers worldwide [1]. Approximately 10% of cancer-related mortality in developed
countries is due to colorectal cancer [2]. Based on data in 2012, the case fatality rate for
colorectal cancer in less developed regions of the world (57.9%) is higher than more developed
regions (45.2%) [3]. In the US, colorectal cancer is the third common cancer in both men and
women, with an estimated 95,520 cases of colon cancer, 39,910 cases of rectal cancer, and
50,260 deaths in 2017 [1].
Beef, veal, pork, and lamb are the most commonly consumed red meats. Processed meat
is defined as meat that has been smoked, salted, preserved, or transformed by other processes [4].
The proportions of red meat versus processed meat intake varies in different countries and
populations are quite distinct given the diversity of diets and eating habits. Based on a review of
over 800 studies, which included large prospective cohort studies, red meat consumption was
classified as probably carcinogenic (Group 2A) and processed meats were classified as
carcinogenic to humans (Group 1) by the International Agency for Research on Cancer (IARC)
[5]. This Group 2A classification of red meat was driven by the evidence for colorectal, prostate
and pancreatic cancer whereas the Group 1 classification of processed meat was based largely on
the evidence on colorectal cancer, and to a lesser extent stomach cancer.
The results of a previous meta-analysis of 30 prospective cohort studies showed that high
consumption of red and/or processed meat is linked to increased risk of colorectal, colon, and
rectal cancer [6]. Many more case-control studies than prospective cohort studies have examined
the role of red meat and processed meat on colorectal cancer risk. Yet, there are no up-to-date

2
meta-analyses of case-control studies. Importantly, prior meta-analysis of case-control studies
has not taken into account important sources of variability across studies that may bias results.
Based on observational epidemiological studies, we present here an up-to-date meta-
analysis of case-control studies. This current meta-analysis carefully considered carefully
possible sources of confounding and misclassification of exposure. Chief among them was
consideration of total energy intake, which relates to diseases risk and also to red meat
consumption, itself an important source of total calories [7]. We also considered the use of
hospital-based controls, which might lead to over- and under- estimates of association depending
on the specific sources of and type of controls were selected in specific studies. Lastly, we
considered the use of non-validated questionnaires, which can be an important source of
exposure misclassification.
In this meta-analysis, we examined the association between consumption of red meat and
processed meat and risk of colorectal cancer based on 61 case-control publications that met our
eligibility criteria, with 16 that reported on unprocessed red meat, 21 that reported on processed
meat, and 23 that reported on red meat including processed meat.

3
METHODS

Data source and search
A systematic search was performed in MEDLINE for publications that reported on red meat
and/or processed meat and CRC in PubMed until November 2016 (keywords included ‘red meat’,
‘processed meat’, ‘meat’, ‘cured meat’, ‘case-control’, ‘colorectal cancer’, etc.). The reporting of
this meta-analysis followed the checklist of the Meta-analysis of Observational Studies in
Epidemiology (MOOSE) [8].

Inclusion criteria and data extraction
We included in our meta-analysis case-control studies that reported estimates of association
between red meat or processed meat or both and colon, rectal or CRC risk, had at least 100 cases,
and provided clear definitions of red meat or processed meat to rule out the inclusion of poultry
or fish.
We classified articles as informative or less informative based on whether they met
certain criteria pertaining to adjustment of confounders, study design, and type of questionnaire
used. Informative publications had to meet the following criteria: 1) unambiguous definition of
red meat or processed meat; 2) use of population-based cases and controls, or hospital-based
cases with population-based controls; 3) use of a validated dietary instrument; 4) important
potential confounders, included body mass index (BMI), smoking, alcohol and physical activity,
in particular total energy intake. Some studies that met the first two criteria but showed
limitations in others were also considered as informative. Those studies that showed important

4
limitations in the last three criteria and/or did not clearly define “red meat” were classified as less
informative.
The article selection, data extraction, and data checking were conducted by several
trained reviewers at IARC and the University of Southern California. The systematic literature
searches on red and/or processed meat and the risk of colorectal, colon, and rectal cancer are
shown in Figure 1. Characteristics of studies included in the meta-analysis are shown in Table 1.

Statistical analysis
We pooled odds ratio estimates using fixed-effects or random-effects models. Publications with
duplicated data were excluded from our analysis. We used the fixed-effects model to pooled
summary odds ratios within a paper when it reported odds ratios of females and males, colon and
rectal cancer, etc., separately. The random-effects model was used between studies to better
capture heterogeneity [9]. We conducted meta-analyses comparing the highest intake level to the
lowest intake (as defined in each study) level of red or processed meat considering the following
three variable definitions: unprocessed red meat only (“red meat”), processed meat only, red
meat including processed meat or red meat unclearly defined as to processed meat inclusion. For
the subset of studies with available data we conducted analyses of beef or pork separately.
We conducted dose-response meta-analysis by pooling odds ratio estimates per 100
grams of intake. For studies that reported results on a continuous scale, we used the continuous
results directly; for studies that reported only on categorical scale, we use the study-specific
mean or median intake of exposure in each category, or the midpoint dose from the range of the
exposure for each n-tile, to compute an odds ratio from the categorical data using generalized
least-squares for trend estimation [10]. Zero consumption was used as a lower boundary when

5
the lowest category was open-ended; and we used the same ranges as the lower nearest category
to estimate the upper value when the highest category was open-ended. We converted
consumption “times” or “servings” into grams using 120 g as a standard portion size for red meat
or red/processed meat combined and 50 g for processed meat only [11]. For studies that used
energy-adjusted estimates (grams or servings/kcal/day) [12-14], we converted to grams/day
using a mean energy intake of 2,000 kcal.
Summary odds ratio estimates with corresponding 95% CIs were derived with the method
of DerSimonian and Laird, using the assumption of a random effects model that incorporates
between-study variability in effect size [9]. For those studies that only reported odds ratios by
gender [15-27], pooled odds ratios were derived using a fixed-effect model and then included in
the meta-analysis of both sexes combined. Similarly, for studies that reported odds ratios
specifically for proximal colon cancer versus distal colon cancer [23, 28], we derived a pooled
“colorectal cancer” OR using the same method. One paper only provided ORs for red meat that
using different cooking methods [29]; therefore, we used a fixed-effects model to obtain the
pooled OR for this study. We also conducted separate analyses for each meat type by tumor site.
In the individual meat meta-analysis, the results of “beef”, “beef steak”, “fatty beef”, and “lean
beef” were combined and reported as “beef”. One paper [23] reported ORs of “beef, pork, or
lamb as main dish” and we applied these ORs in the meta-analysis of each subgroup of beef,
pork, and lamb.
We conducted stratified analyses by study design (population-based versus hospital-
based), informativeness of the study (informative versus less informative studies), gender,
geographic region (North America, South America, Europe, Asia Pacific, and Australia), and
cancer sites (Colon cancer, Proximal Colon cancer, Distal Colon cancer, and Rectal cancer).

6
The I
2
statistic was computed to assess potential heterogeneity in meat and colorectal
cancer associations between studies. A cumulative meta-analysis, giving plots to display
summary meat and colorectal cancer associations as accumulated over calendar time, was
conducted for highest intake of meats versus lowest intake level comparison. Begg’s funnel plots
and the Egger’s test were used to assess possible publications bias [30]. Influence analysis was
used to examine the influence of individual studies by excluding each study in turn. All statistical
analyses were conducted using Stata, version 14.0 (Stata Corp, College Station, Texas). A two-
sided P<0.05 was considered statistically significant.

7
RESULTS
Results of search and study selection
A total of 146 publications were identified that examined the relationship between red and/or
processed meat consumption and risk of CRC, colon, and/or rectal cancer (Figure 1). Of these we
excluded: 9 publications that had less than 100 cases; 10 publications that reported on dietary
patterns, dietary diversity or examined red meat in combination with other foods; 21 publications
with unclear definition of “meat”, or if “meat” definition included poultry; 13 that did not
provide association estimates or sufficient data to obtain them, 17 that only reported on estimated
carcinogens derived from meat but not on meat variables and 15 publications that overlapped
with previous studies (duplicated data). Therefore, a total of 61 case-control studies were
included in the meta-analysis of highest versus the lowest intake; 25 of these studies had
sufficient data to be included in the dose-response meta-analysis (Table 2). We summarize below
our findings for each of the three variables we investigated.

Red meat
Sixteen case-control studies that reported an association between unprocessed red meat
consumption and colorectal cancer risk were included in the meta-analysis of highest versus
lowest intake [12, 14, 17, 19, 21, 26, 28, 31-39]. Of them, 10 were included in the dose-response
analyses [12, 14, 17, 28, 31-32, 35, 37-39].
The meta-analysis ORs of the highest versus lowest red meat consumption comparison
for CRC, colon, and rectal cancer risk were 1.24 (95% CI: 1.04-1.49), 1.22 (95% CI: 1.02-1.45),
and 1.21 (95% CI: 0.93-1.56), respectively (Table 2, Figure 2). In dose-response analyses, we
observed a non-statistically significant positive trend with consumption of red meat and

8
colorectal cancer (OR for 100 gm increment per day = 1.40; 95% CI = 0.90-2.16) and no
evidence of increased risk with colon or rectal cancer (Figure 3). Influence analyses did not show
any individual publication that dominated the summary odds ratio estimates.

Processed meat
Twenty-one case-control studies that reported an association between processed meat
consumption and colorectal cancer risk were included in the highest versus the lowest meta-
analysis [12, 14-15, 19-20, 22-23, 25, 28, 34, 36, 40-49]. 7 of these studies were included in the
dose-response analyses [12, 14, 22, 28, 40, 44, 47].
The meta-analysis ORs of the highest versus lowest processed meat consumption comparison
for CRC, colon, and rectal cancer were 1.48 (95% CI: 1.25-1.75), 1.27 (95% CI: 1.00-1.60), and
1.24 (95% CI: 0.92-1.68), respectively (Table 2, Figure 4. In dose-response analyses, processed
meat was statistically significantly associated with increased risk of CRC (OR per 100 gm
increase per day= 2.97, 95% CI = 1.59-5.55). A positive association of lower magnitude (OR per
100 gm increase per day= 1.43) was observed for colon cancer but it was not statistically
significant. No evidence of dose-response association was observed with rectal cancer (Figure 5).
Influence analyses did not show any dominant individual publication on the summary meta-
analysis odds ratio estimates.

Red meat including processed meat
Twenty-three case-control studies that reported on intake of red meat that included processed
meat, or red meat without a clear inclusion of processed meat and colorectal cancer risk, were
included in the highest versus lowest meta-analysis [13, 18, 20, 22, 25, 29, 32, 40-41, 50-57, 58-

9
61, 62-63]. Of these, 14 publications were included in the dose-response analyses [13, 18, 40, 42,
52, 55-59, 62, 64].
The meta-analysis ORs of the highest versus lowest red and processed meats
consumption comparison for CRC, colon, and rectal cancer were 1.51 (95% CI: 1.22-1.89), 1.30
(95% CI: 1.05-1.60), and 1.44 (95% CI: 1.08-1.93), respectively (Table 2, Figure 6). In dose-
response analyses, consumption of red meat including processed meat was statistically
significantly associated with increased risk of colorectal cancer (OR for 100 gm increment per
day = 1.62, 95% CI = 1.25-2.11). For colon cancer we observed a positive association that was
not statistically significant. There was no evidence of a dose-response association for rectal
cancer (Figure 7). Influence analyses did not show any dominant individual publication on the
summary odds ratio estimates.

Individual red meat
Beef
Eight case-control studies that reported an association between beef consumption and colorectal
cancer risk, were included in the highest versus the lowest meta-analysis [14, 23, 42, 45-46, 64,
65-66]. In highest versus lowest beef consumption comparison, the summary ORs for the
colorectal, colon, and rectal cancer were 1.04, 1.13, and 1.11, respectively (Table 3, Figure 8).
None of these associations were statistically significant.

Pork
Six case-control studies that reported on pork consumption and colorectal cancer risk, were
included in the highest versus lowest intake meta-analysis [14, 23, 45-46, 65-66]. Pork
consumption was not associated with increased risk of CRC (OR = 1.07, 95% CI = 0.95-1.21), or

10
colon cancer (OR = 1.11, 95% CI = 0.96-1.28). A non-statistically significant positive
association was observed with rectal cancer (OR = 1.23, 95% CI = 0.88-1.73) (Table 3, Figure 9).

Small study or Publication bias
In the highest versus the lowest intake meta-analysis, the results of Egger's test for small-study
effects showed that there were significant small-study effects (with P values from 0.001 to 0.048);
Results of Begg’s funnel plots showed evidence of publication bias for studies on all three meat
subtypes (red meat, processed meat, and red meat including processed meat). In the dose-
response analyses, evidences of both small-study effects and publication bias existed in the
studies on red meat (Egger’s test P=0.021, asymmetrical funnel plot) and processed meat
(Egger’s test P=0.001, asymmetrical funnel plot); Results of Begg’s funnel plots showed no
evidence of publication bias for studies on red meat intake but did observe evidence for
publication bias in analyses of processed meat and red meat including processed meat (Figure
11). Results of I
2
statistic for most strata suggested a large amounts of heterogeneity across
studies.

Cumulative meta-analyses
We conducted a cumulative meta-analysis using the ORs of the highest versus lowest level meat
consumption, sorted first by year of publication and by the number of cases (Figure 10). For
analyses of colorectal cancer, no obvious heterogeneity was detected among publications after
the year 2000 for red meat [12, 14, 28, 34-35, 67], processed meat [12, 14, 28, 34, 40-41, 46-48],
and red meat including processed meat [13, 57, 32, 62-63]. The cumulative ORs estimates from
these publications for red meat, processed meat, and red meat including processed meat showed

11
a statistically significant increased risk for colorectal cancer with high consumption of each type
of meat in our analysis. For analyses of colon cancer, publications before 2010 demonstrated a
high degree of heterogeneity on red meat [31, 15, 17, 21, 28, 36-37], processed meat [15, 20, 23,
25, 28, 36], and red meat including processed meat [25, 40, 50-53, 55-57, 68]. A high percentage
of null results of processed meat and colon cancer were seen in articles published before 2000
[15, 20, 25, 36]. For cumulative analyses of estimates of rectal cancer on red meat [12, 28, 31, 37,
39], processed meat [12, 22, 25, 28, 49], and red meat including processed meat [22, 25, 32, 50,
61], publications demonstrated a high degree of heterogeneity over time.

Subgroup analyses
Publications with clear red meat and/or processed meat definition, population-based controls,
validated dietary instrument, and considering potential confounders (total energy intake, BMI,
physical activity, fruits and vegetables, alcohol, and smoking) were defined as “informative”.
Studies were defined as “less informative” if they had limitations in the type of questionnaire
used or important limitations in the appropriate confounding adjustment. To investigate the
potential factors that may play important roles and may affect the quality of a case-control study
of red meat and processed meats on CRC, we conducted subgroup analyses by our classification
of study quality (informative or less informative), study design (hospital-based or population
based controls), use of a validated FFQ, and control of individual potential confounders among
different meat type, geographic regions, and tumor sites. Results of subgroup analyses are shown
in Table 1 for both highest versus lowest meta-analysis and dose-response analysis. For
geographic regions analysis, we did not report results of South America and Australia due to
insufficient data.

12
For stratified analyses of dose-response associations, the number of studies was too low
for meaningful comparisons, and for most strata, results of I
2
statistic suggested a large amount
of heterogeneity across studies; therefore, those were excluded.

Study quality. Comparing high to low intake of red meat among informative studies, there was a
statistically significant positive association with colorectal cancer among informative studies
(OR
for high/low comparison
= 1.25; 95% CI = 1.04-1.51), which was not present among less
informative studies (OR
for high/low comparison
= 1.15; 95% CI = 0.57-2.31) (Table 2). For processed
meat, both informative and less informative studies showed a statistically significant positive
association with high intake and colorectal cancer risk, albeit among less informative studies the
estimate was slightly higher, and number of studies slightly lower. For red meat including
processed meat estimates of association were stronger among less informative publications (OR

for high/low comparison
=1.71, 95% CI=1.33-2.18) relative to informative publications (OR
for high/low
comparison
=1.25, 95% CI=0.88-1.78) (Table 2).

Study design (source of controls). Comparing high to low intake of red meat, there was a
statistically significant positive association with colorectal cancer among studies using
population-based controls (OR
for high/low comparison
= 1.25, 95% CI = 1.04-1.51), which was not
observed among hospital-based studies. For processed meat, both hospital-based and population-
based studies showed a statistically significant positive association with colorectal cancer (OR
for
high/low comparison hospital-based
= 1.52, 95% CI = 1.09-2.12; OR
for high/low comparison population-based
= 1.46, 95%
CI = 1.18-1.82) (Table 2). Similarly, for red meat including processed meats, both population-
based and hospital-based studies showed positive associations with colorectal cancer risk, albeit

13
estimates for hospital-based studies were slightly larger, and fewer studies were available (Table
2).

Adjustment of confounders. Studies of red meat that considered total energy intake adjustment
showed a stronger positive association with colorectal cancer than those that did not. Similarly,
for processed meat, there was a statistically significant positive association with colorectal
cancer among studies that considered total energy adjustment (OR

= 1.52, 95% CI = 1.24-1.86),
which was not seen among studies without total energy adjustment where a positive association
was also observed but of lower magnitude a no statistical significance (Table 2). For studies of
red meat including processed meat, both studies that adjusted for total energy or did not seen an
overall positive association with colorectal cancer risk, although it was slightly stronger for those
studies that did not adjust, and did not reach statistical significance among those that did (Table
2).
Considering individual confounders, in general, inclusion or exclusion of confounders
seemed to have little influence on the associations for processed meat, with estimates being
comparable, and in some instances, marginally higher for studies that did not include
confounders (Table 2). For red meat, consideration of BMI, or physical activity, or alcohol or
smoking, led to statistically significant meta-ORs for colorectal cancer risk that were not
observed among studies that did not consider these confounders (although in many instances
estimates were of similar magnitude). The opposite was observed for studies that reported on red
meat including processed meats, with studies that did not adjust for physical activity, fruits and
vegetables, alcohol and smoking showing statistically significant meta-ORs and those that did

14
not, although in many instances estimates were of similar magnitude, and almost identical meta-
ORs were observed for studies that considered or did not considered BMI adjustment (Table 2).

Questionnaire validation. Slightly greater meta-ORs between red meat, processed meat, and red
meat including processed meat and colorectal cancer were observed among studies that did not
use validated questionnaire relative to studies that used validated questionnaires, although for
processed meats it did not have an impact on the statistical significance of the association (Table
2).

Geographic localization. The results of our meta-analysis by geographic region shows that
people in North America who consume the highest level of red meat, processed meat, and red
including processed meat have 32% (95% CI: 8%-60%), 66% (95% CI: 29%-114%), and 39%
(95% CI: 4%-86%) higher chance of developing CRC, respectively, compared to those who
consume the lowest level of processed meat. In Europe, higher magnitude associations were
observed for red meat and red meat including processed meat consumption (OR
red meat and
CRC
=1.50, 95% CI=0.68-3.30; OR
red meat including processed meat and CRC
=1.73, 95% CI=1.21-2.48). A
lower magnitude association was observed for processed meat consumption (OR
processed meat and
CRC
=1.55, 95% CI=1.17-2.05). A positive association of similar magnitude was observed in Asia
Pacific compare to North America, but they were not statistically significant. In dose-response
meta-analysis, consumption of each type of meat and colorectal cancer had a positive trend with
CRC (OR for 100 gm increment of red meat per day = 1.23; 95% CI = 0.99-1.54; OR for 100 gm
increment of processed meat per day = 5.59; 95% CI = 1.04-30.04; OR for 100 gm increment of
red including processed meat per day = 1.53; 95% CI = 1.10-2.14). Positive trends with higher

15
magnitude was observed in Europe (OR for 100 gm increment of red meat per day = 3.35; 95%
CI = 2.06-5.43; OR for 100 gm increment of processed meat per day = 3.86; 95% CI = 0.22-
67.55; OR for 100 gm increment of red including processed meat per day = 2.68; 95% CI = 0.48-
15.06). No significant trend was observed in Asia Pacific area.

16
Figure 1. PRISMA flow diagram of systematic literature search on red and processed meat and
the risk of colorectal, colon, and rectal cancer

Publications identified in PubMed
until November 2016 and via hand
searching
146 reported on
red/processed meat and
colorectal cancer
incidence
Exclusion:
- irrelevant title and abstract
- reported on exposures other than red meat
and/or processed meat
- studies focused only on adenomas, or
colorectal cancer mortality

Excluded publications:
9 with < 100 cases
10 reported on dietary patterns, dietary
diversity or examined red meat in combination
with other foods only
21 had unclear definition of “meat”, or “meat”
definition included poultry
13 did not provide association estimates or
insufficient data to obtain them
17 only reported on estimated carcinogens
derived from meat but not on meat variables.
15 overlapped with previous reports from the
same study

61 publications included in the meta-
analysis of case-control studies
9 publications reported on red meat only
5 publications reported on processed
meat only
25 publications reported on red meat
including processed meats
12 publications reported on red meat,
also on processed meat separately
10 publications reported on red meat
including processed meats, also on
processed meat separately

17

Table 1. Characteristics of studies included in the meta-analysis
First Author Year Data Source of
Controls
Subjects
Country
Cases
Number
Controls
Number
Informative
Boutron-Ruault 1999 RM
a
, PM
b
Population France 171 309
Butler 2003 RPM
c
Population USA 620 1038
Chiu 2003 RM Population China 931 1552
Cotterchio 2008 RPM Population Canada 1095 1890
Gerhardsson 1991 RM, PM Population Sweden 559 505
Hu J 2007 RM, PM Population Canada 1723 3097
Joshi 2009 RPM Population USA 577 361
Joshi 2015 RM, PM Population USA, Canada 3350 3504
Juarranz Sanz 2004 RM, PM Population Spain 196 196
Kampman 1995 RM Population Netherlands 232 259
Kampman 1999 RPM, PM Population USA 1542 1860
Kimura 2007 RM, PM Population Japan 840 833
Kune 1987 RPM Population Australia 715 727
Le Marchand 1997 RM, PM Population Hawaii 1192 1192
Le Marchand 2001 RM Population Hawaii 727 727
Le Marchand 2002 RPM Population Hawaii 727 727
Miller 2013 RM, PM Population USA 982 1033
Murtaugh 2004 RPM, PM Population USA 952 1205
Murtaugh 2005 RPM Population USA 952 1205
Navarro 2003 RPM, PM Hospital Argentina 287 564
Navarro 2004 RPM Hospital Argentina 296 597
Nowell 2002 RPM Population USA 156 366
Rosato 2013 RPM, PM Hospital Switzerland 329 1361
Satia-Abouta 2004 RPM Population USA 613 996
Shannon 1996 RPM Population USA 424 414
Squires 2010 RPM, PM Population Canada 518 686
Steinmetz 1993 RM, PM Population Australia 220 438
Tabatabaei 2011 RPM Population Australia 567 713
Turner 2004 RM Hospital UK 484 738
Tuyns 1988 RPM, PM Population Europe 818 2851
Williams 2010 RM, PM Population USA 945 959
Less Informative
Abu Mweiss 2014 RPM Hospital Jordan 167 240
Benito 1990 PM Population Spain 286 498
Bidoli 1992 RPM Hospital Italy 248 699
Centonze 1994 RM, PM Population Italy 119 119
Chen K 2006 RM Population China 140 343
De Stefani 1997 RPM, PM Hospital Uruguay 250 500
De Stefani 2012 PM Hospital Uruguay 361 2532
Di Maso 2013 RM Hospital Italy,Switzerland 2390 4943
Evans 2002 RPM Population UK 512 512
Faivre 1997 PM Population France 171 309

18
a. RM: Red meat
b. PM: Processed meat
c. RPM: Red meat including Processed meat
Fernandez 1997 RPM Hospital Italy 112 108
Franceschi 1997 RPM, PM Hospital Italy 1225 4154
Freedman 1996 RPM Hospital USA 163 326
Guo 2015 RPM Hospital China 600 600
Hu X 2013 RPM Hospital China 400 400
Hu X 2014 RPM, PM Hospital China 400 400
Iscovich 1992 RM, PM Population Argentina 110 220
Kotake 1995 RPM Hospital Japan 363 363
Kury 2007 RPM Hospital France 1023 1121
La Vecchia 1996 RPM Hospital Italy 1326 2024
Lee 1989 RPM Hospital Singapore 203 426
Levi 2004 PM Hospital Switzerland 323 611
Lohsoonthorn 1995 RPM, PM Hospital Thailand 279 279
Macquart-Moulin 1986 PM Hospital France 399 399
Manousos 1983 RM Hospital Greece 100 100
Morita 2009 RPM Population Japan 685 833
Muscat 1994 RM Hospital USA 511 500
Saebo 2008 RPM Population Norway 198 222
Seow 2002 RPM Population Singapore 121 222
Tavani 2000 RM Hospital Italy 828 7990

19

Table 2. Summary odds ratios of random effects meta-analyses of red meat, processed meat, red including processed meat, and colorectal cancer for
all studies and by subgroups.

Red meat Processed meat Red meat including
Processed meat

Pooled ORs (95% CI)
P value

n
Heterogeneity
I
2
, P value
Pooled ORs (95% CI)
P value

n
Heterogeneity
I
2
, P value
Pooled ORs (95% CI)
P value

n
Heterogeneity
I
2
, P value
Highest versus lowest meta-analysis
Colorectal cancer 1.24 (1.04-1.49), 0.017 8 77.5%, <0.001 1.48 (1.25-1.75), <0.001 15 85.4%, <0.001 1.51 (1.22-1.89), <0.001 16 80.6%, <0.001
Colon cancer 1.22 (1.02-1.45), 0.029 11 68.8%, <0.001 1.27 (1.00-1.60), 0.050 8 70.2%, 0.001 1.30 (1.05-1.60), 0.016 10 70.9%, <0.001
Proximal colon cancer 1.44 (0.76-2.72), 0.261 1 - 1.45 (1.12-1.89), 0.005 2 0.0%, 0.403 - 0 -
Distal colon cancer 1.23 (0.75-2.01), 0.408 1 - 1.41 (1.12-1.77), 0.003 2 0.0%, 0.752 - 0 -
Rectal cancer 1.21 (0.93-1.56), 0.157 6 73.0%, 0.002 1.24 (0.92-1.68), 0.148 5 60.1%, 0.040 1.44 (1.08-1.93), 0.013 6 68.7%, 0.007
By study quality
informative
Colorectal cancer 1.25 (1.04-1.51), 0.020 7 80.7%, <0.001 1.40 (1.17-1.67), <0.001 9 80.4%, <0.001 1.25 (0.88-1.78), 0.214 5 88.1%, <0.001
Colon cancer 1.17 (0.95-1.44), 0.135 7 58.8%, 0.024 1.31 (1.15-1.48), <0.001 5 0.0%, 0.412 1.18 (0.92-1.53), 0.197 6 64.0%, 0.016
Rectal cancer 0.93 (0.73-1.18), 0.543 3 22.8%, 0.274 1.19 (0.98-1.45), 0.076 3 0.0%, 0.975 1.19 (0.92-1.55), 0.186 3 24.0%, 0.268
Less informative
Colorectal cancer 1.15 (0.57-2.31), 0.695 1 - 1.61 (1.04-2.51), 0.034 6 88.1%, <0.001 1.71 (1.33-2.18), <0.001 10 61.3%, 0.006
Colon cancer 1.27 (0.89-1.82), 0.194 4 79.7%, 0.002 1.18 (0.43-3.22), 0.748 3 89.3%, <0.001 1.47 (1.06-2.06), 0.023 4 69.8%, 0.019
Rectal cancer 1.45 (1.25-1.69), <0.001 3 0.0%, 0.370 1.25 (0.40-3.88), 0.704 2 88.4%, 0.003 1.67 (1.02-2.74), 0.042 3 63.8%, 0.063
By study design
Hospital-based
controls

Colorectal cancer 1.15 (0.57-2.31), 0.695 1 - 1.52 (1.09-2.12), 0.013 6 82.2%, <0.001 1.72 (1.47-2.01), <0.001 4 0.0%, 0.487
Colon cancer 1.90 (1.53-2.35), <0.001 1 - 1.90 (0.92-3.93), 0.083 2 75.4%, 0.044 1.80 (1.13-2.87), 0.014 1 -
Rectal cancer 1.70 (1.31-2.21), <0.001 1 - 1.25 (0.40-3.88), 0.704 2 88.4%, 0.003 1.30 (0.64-2.65), 0.471 1 -
Population-based
controls

Colorectal cancer 1.25 (1.04-1.51), 0.020 7 80.7%, <0.001 1.46 (1.18-1.82), 0.001 9 86.0%, <0.001 1.41 (1.07-1.87), 0.016 11 81.7%, <0.001
Colon cancer 1.16 (1.00-1.34), 0.056 10 45.2%, 0.058 1.16 (0.93-1.45), 0.185 6 60.9%, 0.025 1.26 (1.01-1.57), 0.044 9 72.2%, <0.001
Rectal cancer 1.10 (0.85-1.44), 0.467 5 64.5%, 0.024 1.19 (0.98-1.45), 0.076 3 0.0%, 0.975 1.47 (1.06-2.03), 0.020 5 75.0%, 0.003
By FFQ validation
Yes
Colorectal cancer 1.15 (0.99-1.33), 0.065 6 66.2%, 0.011 1.39 (1.19-1.64), <0.001 12 80.5%, <0.001 1.34 (0.97-1.87), 0.079 6 86.1%, <0.001
Colon cancer 1.13 (0.94-1.37), 0.201 9 50.6%, 0.040 1.31 (1.15-1.48), <0.001 5 0.0%, 0.412 1.18 (0.92-1.53), 0.197 6 604%, 0.016
Rectal cancer 0.97 (0.76-1.24), 0.838 4 24.9%, 0.262 1.19 (0.98-1.45), 0.076 3 0.0%, 0.975 1.19 (0.92-1.55), 0.186 3 24.0%, 0.268
No/Unknown
Colorectal cancer 1.73 (0.89-3.38), 0.107 2 65.3%, 0.089 1.76 (1.08-2.87), 0.023 3 81.3%, 0.005 1.66 (1.24-2.21), 0.001 9 70.2%, 0.001
Colon cancer 1.51 (0.98-2.33), 0.062 2 91.1%, 0.001 1.18 (0.43-3.22), 0.748 3 89.3%, <0.001 1.47 (1.06-2.06), 0.023 4 69.8%, 0.019
Rectal cancer 1.49 (1.19-1.86), <0.001 2 49.4%, 0.160 1.25 (0.40-3.88), 0.704 2 88.4%, 0.003 1.67 (1.02-2.74), 0.042 3 63.8%, 0.063
Cont.
20
By confounding adjustment
Total energy intake
Colorectal cancer 1.32 (1.05-1.65), 0.016 6 69.2%, 0.006 1.52 (1.24-1.86), <0.001 12 79.1%, <0.001 1.37 (0.98-1.90), 0.064 8 87.3%, <0.001
Colon cancer 1.23 (0.96-1.59), 0.105 7 78.7%, <0.001 1.43 (1.18-1.74), <0.001 6 54.6%, 0.051 1.29 (1.02-1.62), 0.030 9 74.0%, <0.001
Rectal cancer 1.14 (0.76-1.70), 0.528 4 81.9%, 0.001 1.24 (0.92-1.68), 0.148 5 60.1%, 0.040 1.52 (1.11-2.09), 0.009 5 73.7%, 0.004
No energy intake
Colorectal cancer 1.03 (1.01-1.04), 0.001 2 0.0%, 0.749 1.37 (0.79-2.40), 0.262 3 91.4%, <0.001 1.69 (1.44-1.99), <0.001 7 0.0%, 0.442
Colon cancer 1.21 (1.06-1.39), 0.006 4 0.0%, 0.414 0.66 (0.31-1.38), 0.270 2 65.6%, 0.088 1.41 (0.87-2.30), 0.168 1 -
Rectal cancer 1.35 (1.13-1.62), 0.001 2 0.0%, 0.921 - 0 - 0.97 (0.49-1.94), 0.931 1 -
BMI adjustment
Colorectal cancer 1.21 (1.02-1.44), 0.030 5 42.5%, 0.138 1.52 (1.21-1.91), <0.001 9 78.6%, <0.001 1.52 (1.06-2.18), 0.024 6 90.2%, <0.001
Colon cancer 1.18 (1.01-1.38), 0.035 8 52.2%, 0.041 1.37 (1.13-1.67), 0.001 7 56.3%, 0.033 1.24 (0.97-1.58), 0.084 8 68.4%, 0.002
Rectal cancer 1.07 (0.80-1.44), 0.641 4 72.5%, 0.012 1.24 (0.92-1.68), 0.148 5 60.1%, 0.040 1.55 (0.97-2.46), 0.065 4 71.8%, 0.014
No BMI adjustment
Colorectal cancer 1.39 (0.78-2.49), 0.266 3 87.8%, <0.001 1.44 (1.09-1.90), 0.011 6 85.7%, <0.001 1.51 (1.15-1.97), 0.003 9 60.3%, 0.010
Colon cancer 1.19 (0.62-2.28), 0.601 3 75.2%, 0.018 0.43 (0.21-0.89), 0.022 1 - 1.57 (1.32-1.88), <0.001 2 0.0%, 0.636
Rectal cancer 1.66 (1.30-2.12), <0.001 2 0.0%, 0.609 - 0 - 1.39 (0.90-2.16), 0.136 2 45.2%, 0.177
Physical activity (+)
Colorectal cancer 1.21 (1.02-1.44), 0.030 5 42.5%, 0.138 1.41 (1.11-1.78), 0.004 7 79.8%, <0.001 1.24 (0.86-1.80), 0.251 6 88.0%, <0.001
Colon cancer 1.11 (0.87-1.43), 0.406 5 69.8%, 0.010 1.34 (1.18-1.53), <0.001 4 0.0%, 0.602 1.14 (0.92-1.42), 0.230 7 58.3%, 0.026
Rectal cancer 0.93 (0.73-1.18), 0.543 3 22.8%, 0.274 1.19 (0.98-1.45), 0.076 3 0.0%, 0.975 1.19 (0.92-1.55), 0.186 3 24.0%, 0.268
Physical activity (-)
Colorectal cancer 1.39 (0.78-2.49), 0.266 3 87.8%, <0.001 1.57 (1.14-2.16), 0.005 8 88.3%, <0.001 1.72 (1.40-2.11), <0.001 9 46.8%, 0.058
Colon cancer 1.34 (1.03-1.74), 0.030 6 66.7%, 0.010 1.12 (0.54-2.31), 0.767 4 86.1%, <0.001 1.67 (1.32-2.11), <0.001 3 22.7%, 0.274
Rectal cancer 1.45 (1.25-1.69), <0.001 3 0.0%, 0.370 1.25 (0.40-3.88), 0.704 2 88.4%, 0.003 1.67 (1.02-2.74), 0.042 3 63.8%, 0.063
Fruits and veg (+)
Colorectal cancer 1.15 (0.99-1.33), 0.065 6 66.2%, 0.011 1.36 (1.13-1.62), 0.001 9 80.4%, <0.001 1.46 (0.96-2.20), 0.075 5 90.9%, <0.001
Colon cancer 1.23 (1.00-1.52), 0.055 7 79.3%, <0.001 1.43 (1.09-1.86), 0.009 5 62.1%, 0.032 1.35 (0.97-1.87), 0.075 6 76.3%, 0.001
Rectal cancer 1.19 (0.89-1.58), 0.236 5 78.3%, 0.001 1.24 (0.92-1.68), 0.148 5 60.1%, 0.040 1.55 (0.97-2.46), 0.065 4 71.8%, 0.014
Fruits and veg (-)
Colorectal cancer 1.73 (0.89-3.38), 0.107 2 65.3%, 0.089 1.72 (1.15-2.57), <0.001 6 86.7%, <0.001 1.55 (1.22-1.96), <0.001 10 59.9%, 0.008
Colon cancer 1.20 (0.89-1.62), 0.236 4 0.0%, 0.404 0.90 (0.47-1.74), 0.762 3 84.2%, 0.002 1.27 (0.96-1.67), 0.091 4 60.9%, 0.053
Rectal cancer 1.40 (0.70-2.81), 0.344 1 - - 0 - 1.39 (0.90-2.16), 0.136 2 45.2%, 0.177
Alcohol (+)
Colorectal cancer 1.24 (1.02-1.51), 0.029 4 51.6%, 0.102 1.61 (1.23-2.12), 0.001 7 82.7%, <0.001 1.30 (0.84-1.99), 0.237 4 91.8%, <0.001
Colon cancer 1.33 (1.12-1.59), 0.001 7 62.3%, 0.014 1.38 (1.11-1.71), 0.003 6 63.4%, 0.018 1.23 (0.79-1.92), 0.353 3 76.4%, 0.015
Rectal cancer 1.19 (0.89-1.58), 0.236 5 78.3%, 0.001 1.45 (0.97-2.18), 0.073 3 60.1%, 0.081 - 0 -
Alcohol (-)
Colorectal cancer 1.28 (0.85-1.93), 0.240 4 81.7%, 0.001 1.37 (1.09-1.72), 0.006 8 81.5%, <0.001 1.62 (1.28-2.05), <0.001 11 62.1%, 0.003
Colon cancer 0.94 (0.57-1.54), 0.804 4 77.2%, 0.004 0.76 (0.26-2.21), 0.613 2 81.3%, 0.021 1.33 (1.03-1.71), 0.026 7 69.0%, 0.004
Rectal cancer 1.40 (0.70-2.81), 0.344 1 - 0.98 (0.57-1.67), 0.939 2 62.0%, 0.105 1.44 (1.08-1.93), 0.013 6 68.7%, 0.007
Smoking (+)
Colorectal cancer 1.21 (1.02-1.44), 0.030 5 42.5%, 0.138 1.54 (1.19-1.99), 0.001 8 80.7%, <0.001 1.28 (0.90-1.82), 0.166 7 86.7%, <0.001
Colon cancer 1.18 (0.95-1.46), 0.142 8 75.6%, <0.001 1.46 (1.07-1.99), 0.017 4 71.5%, 0.015 0.95 (0.81-1.11), 0.542 4 0.0%, 0.989
Cont.
21
Rectal cancer 1.21 (0.93-1.56), 0.157 6 73.0%, 0.002 1.36 (1.03-1.79), 0.029 4 51.8%, 0.101 1.25 (0.81-1.92), 0.307 2 59.5%, 0.116
Smoking (-)
Colorectal cancer 1.39 (0.78-2.49), 0.266 3 87.8%, <0.001 1.42 (1.11-1.81), 7 84.1%, <0.001 1.72 (1.38-2.15), <0.001 8 52.3%, 0.041
Colon cancer 1.40 (1.07-1.83), 0.014 3 16.7%, 0.301 1.00 (0.62-1.61), 4 76.3%, 0.005 1.63 (1.35-1.97), <0.001 6 27.3%, 0.230
Rectal cancer - 0 - 0.68 (0.36-1.29), 1 - 1.59 (1.10-2.31), 0.015 4 49.0%, 0.118
By geographic region
North America
Colorectal cancer 1.32 (1.08-1.60), 0.005 7 54.5%, 0.040 1.66 (1.29-2.14), <0.001 8 78.1%, <0.001 1.39 (1.04-1.86), 0.024 5 63.7%, 0.026
Colon cancer 0.95 (0.68-1.33), 0.766 3 72.1%, 0.028 1.33 (1.18-1.50), <0.001 5 0.0%, 0.553 1.10 (0.92-1.31), 0.296 10 49.9%, 0.036
Rectal cancer 0.94 (0.63-1.39), 0.747 2 56.7%, 0.129 1.20 (1.02-1.41), 0.025 4 0.0%, 0.999 1.11 (0.97-1.27), 0.142 5 0.0%, 0.596
Europe
Colorectal cancer 1.50 (0.68-3.30), 0.315 2 93.9%, <0.001 1.55 (1.17-2.05), 0.002 7 86.7%, <0.001 1.73 (1.21-2.48), 0.003 5 56.4%, 0.057
Colon cancer 1.44 (1.06-1.95), 0.018 5 71.6%, 0.007 - 0 - 1.60 (1.29-1.98), <0.001 1 -
Rectal cancer 1.49 (1.19-1.86), <0.001 2 49.4%, 0.160 - 0 - 1.38 (1.01-1.89), <0.001 1 -
Asia Pacific
Colorectal cancer 1.13 (0.80-1.60), 0.493 1 - 1.65 (0.82-3.32), 0.163 2 89.4%, 0.002 1.41 (0.99-2.01), 0.061 5 69.5%, 0.011
Colon cancer 1.42 (1.20-1.68), <0.001 5 0.0%, 0.685 1.26 (0.89-1.79), 0.195 1 - 1.12 (0.78-1.60), 0.532 2 34.2%, 0.218
Rectal cancer 1.11 (0.76-1.63), 0.578 2 0.0%, 0.443 1.14 (0.73-1.78), 0.562 1 - 0.97 (0.49-1.94), 0.931 1 -

Dose-response meta-analysis (per 100 grams per day for red meat, processed meat, and red and processed meats)
All studies
Colorectal cancer 1.4 (0.90-2.16), 0.133 5 75.0%, 0.003 2.97 (1.59-5.55), 0.001 7 86.8%, <0.001 1.62 (1.25-2.11), 0.003 9 53.1%, 0.029
Colon cancer 1.13 (0.92-1.39), 0.241 9 46.0%, 0.063 1.43 (0.57-3.58), 0.450 2 49.6%, 0.159 1.31 (0.93-1.85), 0.121 7 78.4%, <0.001
Proximal colon cancer - 0 - 5.25 (0.21-132.89), 0.314 1 - - 0 -
Distal colon cancer - 0 - 4.45 (0.25-78.85), 0.309 1 - - 0 -
Rectal cancer 1.18 (0.86-1.61), 0.309 5 62.6%, 0.030 0.83 (0.62-1.13), 0.240 5 0.0%, 0.433 1.09 (0.85-1.42),0.492 3 0.0%, 0.850
By geographic region
North America
Colorectal cancer 1.23 (0.99-1.54), 0.061 3 0.0%, 0.812 5.59 (1.04-30.04), 0.045 4 88.5%, <0.001 1.53 (1.10-2.14), 0.011 5 56.8%, 0.055
Colon cancer 0.92 (0.60-1.39), 0.677 4 51.6%, 0.102 3.16 (0.72-13.82), 0.127 1 - 1.68 (0.92-3.07), 0.091 4 83.6%, <0.001
Rectal cancer 0.77 (0.53-1.11), 0.161 2 0.0%, 0.522 2.88 (0.69-11.98), 0.146 3 0.0%, 0.795 1.09 (0.85-1.42), 0.492 3 0.0%, 0.850
Europe
Colorectal cancer 3.35 (2.06-5.43), <0.001 1 - 3.86 (0.22-67.55), 0.354 2 94.3%, <0.001 2.68 (0.48-15.06), 0.263 1 -
Colon cancer 1.28 (0.98-1.69), 0.071 4 53.3%, 0.093 1.08 (0.86-1.35), 0.499 1 - 0.98 (0.93-1.03), 0.438 1 -
Rectal cancer 1.38 (1.01-1.89), 0.045 3 69.7%, 0.069 0.78 (0.57-1.06), 0.116 1 - - 0 -
Asia Pacific
Colorectal cancer 0.94 (0.53-1.67), 0.844 1 - 2.40 (0.32-18.30), 0.397 1 - 1.66 (0.37-7.48), 0.510 3 71.6%, 0.030
Colon cancer 0.31 (0.00-70.87), 0.672 1 - 4.79 (0.56-41.01), 0.153 1 - 0.96 (0.25-3.74), 0.958 2 63.3%, 0.099
Rectal cancer 11.38 (0.07-1750.27), 0.344 1 - 1.69 (0.11-25.49), 0.705 1 - - 0 -

22
Table3. Summary meta-analyses odds ratios of highest versus lowest intake of individual red meat (beef and pork) and colorectal cancer.

Beef Pork
Pooled ORs (95% CI)
P value

n
Heterogeneity
I
2
, P value
Pooled ORs (95% CI)
P value

n
Heterogeneity
I
2
, P value
Colorectal cancer 1.04 (0.83-1.31), 0.70 6 50.3%, 0.07 1.07 (0.95-1.21), 0.24 4 0.0%, 0.85
Colon cancer 1.13 (0.93-1.38), 0.21 3 35.4%, 0.21 1.11 (0.96-1.28), 0.16 3 12.2%, 0.32
Proximal colon cancer 1.30 (0.94-1.79), 0.11 1 - 1.30 (0.94-1.79), 0.11 1 -
Distal colon cancer 1.14 (0.87-1.49), 0.34 1 - 1.14 (0.87-1.49), 0.34 1 -
Rectal cancer 1.11 (0.82-1.52), 0.50 2 10.8%, 0.29 1.23 (0.88-1.73), 0.23 2 36.3%, 0.21
Overall 1.10 (0.99-1.22), 0.08 13 78.1%, 0.00 1.23 (1.11-1.36), 0.01 11 70.4%, 0.04

23
Figure 2. Random effects meta-analysis of the highest versus the lowest consumption of red
meat and the risk of Colon, Rectal, and Colorectal cancers.

24
Figure 3. Dose-response meta-analysis of red meat consumption and the risk of Colon, Rectal,
and Colorectal cancers per increase of 100 gr per day.

25
Figure 4. Random effects meta-analysis of the highest versus the lowest consumption of
processed meat and the risk of Colon, Rectal, and Colorectal cancers.

26
Figure 5. Dose-response meta-analysis of processed meat consumption and the risk of Colon,
Rectal, and Colorectal cancers per increase of 100 gr per day.

27

Figure 6. Random effects meta-analysis of the highest versus the lowest consumption of red
meat including processed meat and the risk of Colon, Rectal, and Colorectal cancers.

28
Figure 7. Dose-response meta-analysis of consumption of red meat including processed meat
and the risk of Colon, Rectal, and Colorectal cancers per increase of 100 gr per day.

29
Figure 8. Random effects meta-analysis of the highest versus the lowest consumption of beef
and the risk of Colon, Rectal, and Colorectal cancers.

30
Figure 9. Random effects meta-analysis of the highest versus the lowest consumption of pork
and the risk of Colon, Rectal, and Colorectal cancers.

31

Figure 10. Cumulative random-effects meta-analysis of highest versus lowest level red meat consumption by cancer type, publication year, and variable
definition

A. Red meat, processed meat, and red including processed meat consumption and Colorectal cancer

B. Red meat, processed meat, and red including processed meat consumption and Colon cancer

32
C. Red meat, processed meat, and red including processed meat consumption and Rectal cancer

33
DISCUSSION

General findings
According to our findings, people who consume the highest level of red meat have 24% (95% CI:
4%-49%) and 22% (95% CI: 2%-45%) higher chance of developing colorectal and colon cancer
when compared to those who consume the lowest level, respectively. Dose-response analyses
showed an increase of 40% per 100 grams of red meat consumed, however, this association was
not statistically significant (P value = 0.133). Our findings also suggest that people who consume
the highest level of processed meat have 48% (95% CI: 25%-75%) and 27% (95% CI: 0%-60%)
higher chance of developing colorectal and colon cancer, respectively, compared to those who
consume the lowest level of processed meat. In dose-response analysis, the risk of having
colorectal cancer increased by 197% (95% CI: 59%-455%) for every 100 g/d increase of
processed meat, whereas the association seemed to be mostly restricted to colon cancer, there
were not enough studies to conclude this with confidence.
When considering studies that defined red meat including processed meat, or did not
define red meat clearly, we also observed that people with the highest level of consumption have
51% (95% CI: 22%-89%), 30% (95% CI: 5%-60%), and 44 % (95% CI: 8%-93%) higher risk of
developing CRC, colon or rectal cancer, respectively, compared to those with the lowest
consumption level. For every 100 g/d increase of red meat (including processed meat), the risk of
colorectal cancer increased by 62% (95% CI: 25%-111%).
Our estimates from the highest versus the lowest intake meta-analysis and dose-response
meta-analyses for red meat only, processed meat only, and red meat defined to include processed
meat are all higher than those reported in a previous meta-analysis of prospective studies [6]. The
previous study reported summary relative risk (RR) of colorectal cancer for the highest versus

34
the lowest intake of 1.10 (95% CI: 1.00-1.21) for red meat, 1.17 (95% CI: 1.09-1.25) for
processed meat, and 1.22 (95% CI: 1.11-1.34) for red meat including processed meat. The
prospective studies meta-analysis also reported RRs of colorectal cancer 1.17 (95% CI: 1.05-
1.31), and 1.18 (95% CI: 1.10-1.28), and 1.14 (95% CI: 1.04-1.24), for every 100 g/day increase
consume of red meat, processed meat, and red meat including processed meat, respectively [6].
The result is similar for colon cancer. No significant association of red meat and rectal cancer
was observed [6]. We observed that when restricting studies to those we deemed to be
“informative” based on study quality, the pooled OR estimates we obtained were closer to those
in the previous meta-analysis of cohort studies. This finding emphasizes the importance of
considering well-conducted case-control studies when synthesizing evidence of dietary factors
and cancer risk using case-control approaches.

Stratified analyses
In general, for analysis of unprocessed red meat we observed that informative studies showed
stronger estimates of association with colorectal cancer risk compared to less informative studies,
although we only had one less informative study to compare. This was consistent with the results
of analyses stratified by considering individual factors that contribute to the overall quality of the
study, such as use of population-based controls, adjustment for energy intake, BMI, alcohol and
smoking. For analyses of processed meat, estimates seemed to be less influenced by study
quality and we observed similar estimates across different strata defined by study quality.
Overall, for each of the strata we observed that studies that did not consider these confounders,
or had not used a validated questionnaire had higher pooled ORs; In most instances estimates
were more precise for studies that did included these confounders. The only exception seems to

35
be studies that considered red meat including processed meat, where we observed that studies
that lack consideration of key confounders showed greater meta-ORs and more precise estimates
of association with colorectal cancer. Given that studies included in this variable group include
studies with unclear red meat definition, it is difficult to interpret these findings, as the exposure
variable is very heterogeneous.
The results of our meta-analysis by geographic region shows that significant positive
associations and trends between each type of meat and CRC were observed among people in
North America. Higher magnitude associations and trends were observed for red meat and red
meat including processed meat consumption and a lower magnitude association was observed for
processed meat consumption in Europe compared to North America. For those who living in the
Asia Pacific area, no necessarily relationship between meat consumption and colorectal cancer.

Selective reporting and publication bias
Through results of Egger’s test and visual inspection of funnel plots we noted that larger studies
with smaller standard errors gathered at the top of the graph, suggesting evidence of possible
publication bias for all three exposure variables considered in the highest versus the lowest meta-
analysis. In the dose-response analysis, we observed no evidence of publication bias for red meat
intake but did observe evidence for processed meat and red meat including processed meat.
As time goes by, more studies and studies with high quality were published. In our
cumulative meta-analyses, the ORs estimates of each type of meat consumption and colorectal
cancer tended to be stable over time and show positive associations when those values are stable.
Hence, our overall results of the cumulative meta-analysis are more likely to be real and less
likely to be driven by publication bias, influential older publications, or study quality.

36

Strengths and limitations of the study
Our study has several strengths including, an up-to-date inclusion of all available studies,
consideration of study quality as determined by study design, exposure definition, exposure
assessment and consideration of confounders. Moreover, we used two approaches to capture
meat exposure by considering highest and lowest intake levels and also by conducting a dose-
response analysis. However, we highlight a possible weakness with the first approach, as meta-
analyses of high versus low consumption levels can have substantial heterogeneity as different
studies used different rules and criteria to divide meat consumption levels. Moreover, several
studies were excluded from dose-response meta-analysis due to insufficient data/information,
which included 1) those were the exposure had not been quantified [16, 25, 29, 32, 34, 36, 43,
49-50, 53, 61, 67, 69-70, 71-74]; 2) those that only reported individual meat types [60, 65-66,
69-71, 74-76]; and 3) those where the exposed number in each category level was not reported
[69, 20, 25-26, 29, 32, 34, 36, 43, 49-50, 53-54, 60-61, 66-67, 70, 74, 76]. The results of dose-
response analysis that excluded these publications were not entirely consistent with those from
highest versus lowest analysis, which suggested that the exclusion of these studies might have
biased the results. In addition, due to different units and number of categories of meat
consumption that were used in different papers, it is hard to draw a conclusion regarding risk
associated with a specific level of consumption.

Final Conclusions
This meta-analyses shows that both high intake of unprocessed red meat and processed meat are
associated with increased risk of CRC. Individuals in the highest category of red meat

37
consumption are 24% more likely to develop colorectal cancer than those in the lowest category.
Similarly, individuals in the highest category of processed meat consumption are 48% more
likely to develop colorectal cancer than those in the lowest consumption category, with, a 197%
increase for every 100 g/day increment of processed meat consumption. Our findings highlight
that use of population-based controls and a validated questionnaire, as well as consideration of
appropriate confounders are important to obtain association estimates of red meat and processed
meat consumption that are more comparable to those from prospective studies.

38
Figure 11. Funnel Plots with P-values from Beggar’s test

Funnel plots of red meat (a. highest vs. lowest meta-analysis, P=0.005; b. dose-response meta-
analysis, P=0.913), processed meat (c. highest vs. lowest meta-analysis, P=0.001; d. dose-
response meta-analysis, P=0.001), and red meat including processed meat (e. highest vs. lowest
meta-analysis, P=0.048; f. dose-response meta-analysis, P=0.021) consumption and the risk of
colon, rectal, and colorectal cancers.

39
REFERENCES

[1] Cancer Facts & Figures 2017. (2017). Cancer.org. Retrieved 8 May 2017, from
https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-
figures-2017.html

[2] Shibboleth Authentication Request. (2017). Www-nature-com.libproxy2.usc.edu. Retrieved 8
May 2017, from https://www-nature-com.libproxy2.usc.edu/articles/nrdp201566

[3] Fact Sheets by Cancer. (2017). Globocan.iarc.fr. Retrieved 8 May 2017, from
http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx?cancer=colorectal

[4] (Bouvard, V., Loomis, D., Guyton, K., Grosse, Y., El Ghissassi, F., Benbrahim-Tallaa, L., . . .
International Agency for Research on Cancer Monograph Working Group. (2015).
Carcinogenicity of consumption of red and processed meat. Lancet Oncology, 16(16), 1599-
1600. doi:10.1016/S1470-2045(15)00444-1)

[5] IARC Monographs evaluate consumption of red meat and processed meat. (2015) (1st ed., p.
1). Retrieved from https://www.iarc.fr/en/media-centre/pr/2015/pdfs/pr240_E.pdf

[6] Chan, D. S., Lau, R., Aune, D., Vieira, R., Greenwood, D. C., Kampman, E., & Norat, T.
(2011). Red and processed meat and colorectal cancer incidence: Meta-analysis of
prospective studies. Plos One, 6(6), e20456-e20456. doi:10.1371/journal.pone.0020456

[7] Willett, W. (1998). Nutritional epidemiology: Walter willett (2nd;CSZA; ed.). New York:
Oxford University Press.

[8] Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D, Becker BJ,
Sipe TA, Thacker SB, for the Meta-analysis Of Observational Studies in Epidemiology
(MOOSE) Group. Meta-analysis of Observational Studies in Epidemiology A Proposal for
Reporting. JAMA. 2000;283(15):2008–2012. doi:10.1001/jama.283.15.2008

[9] DerSimonian, R., & Laird, N. (2015). Meta-analysis in clinical trials revisited. Contemporary
Clinical Trials, 45(Pt A), 139-145. doi:10.1016/j.cct.2015.09.002

[10] Orsini, N., Bellocco, R., & Greenland, S. (2006). Generalized least squares for trend
estimation of summarized dose-response data. Stata Journal, 6(1), 40.

[11] GLADE, M. (2008). World cancer research Fund/American institute for cancer research,
food, nutrition, physical activity and the prevention of cancer: A global perspective ,
american institute for cancer research, washington, D.C (2007) ISBN: 978-0-9722522-2-5.
Nutrition, 24(4), 393-398. doi:10.1016/j.nut.2007.12.019

[12] Miller, P. E., Lazarus, P., Lesko, S. M., Cross, A. J., Sinha, R., Laio, J., . . . Hartman, T. J.
(2013). Meat-related compounds and colorectal cancer risk by anatomical subsite. Nutrition
and Cancer, 65(2), 202-226. doi:10.1080/01635581.2013.756534

40

[13] Morita, M., Marchand, L. L., Kono, S., Yin, G., Toyomura, K., Nagano, J., . . . Imaizumi, N.
(2009). Genetic polymorphisms of CYP2E1 and risk of colorectal cancer: The fukuoka
colorectal cancer study. Cancer Epidemiology Biomarkers & Prevention, 18(1), 235-241.
doi:10.1158/1055-9965.EPI-08-0698

[14] Joshi, A. D., Kim, A., Lewinger, J. P., Ulrich, C. M., Potter, J. D., Cotterchio, M., . . . Stern,
M. C. (2015). Meat intake, cooking methods, dietary carcinogens, and colorectal cancer risk:
Findings from the colorectal cancer family registry. Cancer Medicine, 4(6), 936-952.
doi:10.1002/cam4.461 Cancel

[15] Steinmetz, K. A., & Potter, J. D. (1993). Food-group consumption and colon cancer in the
adelaide case-control study. II. meat, poultry, seafood, dairy foods and eggs. International
Journal of Cancer, 53(5), 720-727. doi:10.1002/ijc.2910530503

[16] Muscat, J. E., & Wynder, E. L. (1994). The consumption of well-done red meat and the risk
of colorectal cancer. American Journal of Public Health, 84(5), 856-858.
doi:10.2105/AJPH.84.5.856

[17] Kampman, E., Verhoeven, D., Sloots, L., & Pieter van 't Veer. (1995). Vegetable and
animal products as determinants of colon cancer risk in dutch men and women. Cancer
Causes & Control, 6(3), 225-234. doi:10.1007/BF00051794

[18] Shannon, J., White, E., Shattuck, A. L., & Potter, J. D. (1996). Relationship of food groups
and water intake to colon cancer risk. Cancer Epidemiology Biomarkers & Prevention, 5(7),
495.

[19] Marchand, L. L., Wilkens, L. R., Hankin, J. H., Kolonel, L. N., & Lyu, L. (1997). A case-
control study of diet and colorectal cancer in a multiethnic population in hawaii (united
states): Lipids and foods of animal origin. Cancer Causes & Control, 8(4), 637-648.
doi:10.1023/A:1018406716115

[20] Kampman, E., Slattery, M. L., Bigler, J., Leppert, M., Samowitz, W., Caan, B. J., & Potter,
J. D. (1999). Meat consumption, genetic susceptibility, and colon cancer risk: A united states
multicenter case-control study. Cancer Epidemiology Biomarkers & Prevention, 8(1), 15-24.

[21] Chiu, B. C., Ji, B., Dai, Q., Gridley, G., McLaughlin, J. K., Gao, Y., . . . Chow, W. (2003).
Dietary factors and risk of colon cancer in shanghai, china. Cancer Epidemiology
Biomarkers & Prevention, 12(3), 201.

[22] Murtaugh, M. A., Ma, K., Sweeney, C., Caan, B. J., & Slattery, M. L. (2004). Meat
consumption patterns and preparation, genetic variants of metabolic enzymes, and their
association with rectal cancer in men and women. The Journal of Nutrition, 134(4), 776-784.

[23] Hu, J., Morrison, H., Mery, L., DesMeules, M., Macleod, M., & Canadian Cancer Registries
Epidemiology Research Group. (2007). Diet and vitamin or mineral supplementation and

41
risk of colon cancer by subsite in canada. European Journal of Cancer Prevention : The
Official Journal of the European Cancer Prevention Organisation (ECP), 16(4), 275-291.
doi:10.1097/01.cej.0000228411.21719.25

[24] Squires, J., Roebothan, B., Buehler, S., Sun, Z., Cotterchio, M., Younghusband, B., . . .
Wang, P. P. (2010). Pickled meat consumption and colorectal cancer (CRC): A case-control
study in newfoundland and labrador, canada. Cancer Causes & Control, 21(9), 1513-1521.
doi:10.1007/s10552-010-9580-z

[25] Destefani, E., Deneopellegrini, H., Mendilaharsu, M., & Ronco, A. (1997). Meat intake,
heterocyclic amines and risk of colorectal cancer. International Journal of Oncology, 10(3),
573.

[26] Steinmetz, K. A., & Potter, J. D. (1993). Food-group consumption and colon cancer in the
adelaide case-control study. II. meat, poultry, seafood, dairy foods and eggs. International
Journal of Cancer, 53(5), 720-727. doi:10.1002/ijc.2910530503

[27] De Stefani, E., Ronco, A. L., Boffetta, P., Deneo-Pellegrini, H., Correa, P., Acosta, G., &
Mendilaharsu, M. (2012). Nutrient-derived dietary patterns and risk of colorectal cancer: A
factor analysis in uruguay. Asian Pacific Journal of Cancer Prevention : APJCP, 13(1), 231-
235. doi:10.7314/APJCP.2012.13.1.231

[28] Kimura, Y., Kono, S., Toyomura, K., Nagano, J., Mizoue, T., Moore, M. A., . . . Imaizumi,
N. (2007). Meat, fish and fat intake in relation to subsite-specific risk of colorectal cancer:
The fukuoka colorectal cancer study. Cancer Science, 98(4), 590-597. doi:10.1111/j.1349-
7006.2007.00425.x

[29] Tabatabaei, S. M., Fritschi, L., Knuiman, M. W., Boyle, T., Iacopetta, B. J., Platell, C., &
Heyworth, J. S. (2011). Meat consumption and cooking practices and the risk of colorectal
cancer. European Journal of Clinical Nutrition, 65(6), 668-675. doi:10.1038/ejcn.2011.17

[30] Jonathan A C Sterne, Sutton, A. J., John P A Ioannidis, Terrin, N., Jones, D. R., Lau, J., . . .
Julian P T Higgins. (2011). Recommendations for examining and interpreting funnel plot
asymmetry in meta-analyses of randomised controlled trials. BMJ: British Medical Journal,
343(7818), 302-307. doi:10.1136/bmj.d4002

[31] Tavani, A., La Vecchia, C., Gallus, S., Lagiou, P., Trichopoulos, D., Levi, F., & Negri, E.
(2000). Red meat intake and cancer risk: A study in italy. International Journal of Cancer,
86(3), 425-428. doi:10.1002/(SICI)1097-0215(20000501)86:3<425::AID-IJC19>3.0.CO;2-S

[32] Le Marchand, L., Hankin, J. H., Pierce, L. M., Sinha, R., Nerurkar, P. V., Franke, A. A., . . .
Chang, W. (2002). Well-done red meat, metabolic phenotypes and colorectal cancer in
hawaii. Mutation Research - Fundamental and Molecular Mechanisms of Mutagenesis, 506,
205-214. doi:10.1016/S0027-5107(02)00167-7

42
[33] Muscat, J. E., & Wynder, E. L. (1994). The consumption of well-done red meat and the risk
of colorectal cancer. American Journal of Public Health, 84(5), 856-858.
doi:10.2105/AJPH.84.5.856

[34] Juarranz Sanz, M., Soriano Llora, T., Calle Purón, M. E., Martínez Hernández, D., González
Navarro, A., & Domínguez Rojas, V. (2004). Influence of the diet on the development of
colorectal cancer in a population of madrid. Revista Clínica Española, 204(7), 355.

[35] Turner, F., Smith, G., Sachse, C., Lightfoot, T., Garner, R. C., Wolf, C. R., . . . on behalf of
the Colorectal Cancer Study Group. (2004). Vegetable, fruit and meat consumption and
potential risk modifying genes in relation to colorectal cancer. International Journal of
Cancer, 112(2), 259-264. doi:10.1002/ijc.20404

[36] Iscovich, J. M., L'Abbé, K. A., Castelleto, R., Calzona, A., Bernedo, A., Chopita, N. A., . . .
Kaldor, J. (1992). Colon cancer in argentina. I: Risk from intake of dietary items.
International Journal of Cancer, 51(6), 851-857. doi:10.1002/ijc.2910510603

[37] Chen, K., Fan, C., Jin, M., Song, L., Xu, H., He, H., & Tong, F. (2006). A case-control
study on the association between the genetic polymorphism of sulfotransferase 1A1, diet and
susceptibility of colorectal cancer. Zhonghua Zhong Liu Za Zhi [Chinese Journal of
Oncology], 28(9), 670-673.

[38] Williams, C. D., Satia, J. A., Adair, L. S., Stevens, J., Galanko, J., Keku, T. O., & Sandler,
R. S. (2010). Associations of red meat, fat, and protein intake with distal colorectal cancer
risk. Nutrition and Cancer, 62(6), 701-709. doi:10.1080/01635581003605938

[39] Di Maso, M., Talamini, R., Bosetti, C., Montella, M., Zucchetto, A., Libra, M., . . . Polesel,
J. (2013). Red meat and cancer risk in a network of case-control studies focusing on cooking
practices. Annals of Oncology : Official Journal of the European Society for Medical
Oncology / ESMO, 24(12), 3107-3112. doi:10.1093/annonc/mdt392

[40] Squires, J., Roebothan, B., Buehler, S., Sun, Z., Cotterchio, M., Younghusband, B., . . .
Wang, P. P. (2010). Pickled meat consumption and colorectal cancer (CRC): A case-control
study in newfoundland and labrador, canada. Cancer Causes & Control, 21(9), 1513-1521.
doi:10.1007/s10552-010-9580-z

[41] Rosato, V., Bosetti, C., Levi, F., Polesel, J., Zucchetto, A., Negri, E., & Vecchia, C. L.
(2013). Risk factors for young-onset colorectal cancer. Cancer Causes & Control, 24(2),
335-341. doi:10.1007/s10552-012-0119-3

[42] Centonze, S., Boeing, H., Leoci, C., Guerra, V., & Misciagna, G. (1994). Dietary habits and
colorectal cancer in a low-risk area. results from a population-based case-control study in
southern italy. Nutrition and Cancer, 21(3), 233-246. doi:10.1080/01635589409514322

[43] Faivre, J., Boutron, M. C., Senesse, P., Couillault, C., Belighiti, C., & Meny, B. (1997).
Environmental and familial risk factors in relation to the colorectal adenoma--carcinoma

43
sequence: Results of a case-control study in burgundy (france). European Journal of Cancer
Prevention : The Official Journal of the European Cancer Prevention Organisation (ECP),
6(2), 127-131.

[44] Franceschi, S., Favero, A., La Vecchia, C., Negri, E., Conti, E., Montella, M., . . . Decarli, A.
(1997). Food groups and risk of colorectal cancer in italy. International Journal of Cancer,
72(1), 56-61. doi:10.1002/(SICI)1097-0215(19970703)72:1<56::AID-IJC8>3.0.CO;2-3

[45] Boutron-Ruault, M. C., Senesse, P., Faivre, J., Chatelain, N., Belghiti, C., & Méance, S.
(1999). Foods as risk factors for colorectal cancer: A case-control study in burgundy (france).
European Journal of Cancer Prevention : The Official Journal of the European Cancer
Prevention Organisation (ECP), 8(3), 229-235. doi:10.1097/00008469-199906000-00011

[46] Navarro, A., Díaz, M. P., Muñoz, S. E., Lantieri, M. J., & Eynard, A. R. (2003).
Characterization of meat consumption and risk of colorectal cancer in cordoba, argentina.
Nutrition, 19(1), 7-10. doi:10.1016/S0899-9007(02)00832-8

[47] Levi, F., Pasche, C., Lucchini, F., Bosetti, C., & La Vecchia, C. (2004). Processed meat and
the risk of selected digestive tract and laryngeal neoplasms in switzerland. Annals of
Oncology : Official Journal of the European Society for Medical Oncology / ESMO, 15(2),
346-349. doi:10.1093/annonc/mdh060

[48] Hu, X., Yuan, P., Luan, R., Li, X., Liu, W., Feng, F., . . . Yang, Y. (2013). Calpain-10
SNP43 and SNP19 polymorphisms and colorectal cancer: A matched case-control study.
Asian Pacific Journal of Cancer Prevention, 14(11), 6673-6680.
doi:10.7314/APJCP.2013.14.11.6673

[49] De Stefani, E., Ronco, A. L., Boffetta, P., Deneo-Pellegrini, H., Correa, P., Acosta, G., &
Mendilaharsu, M. (2012). Nutrient-derived dietary patterns and risk of colorectal cancer: A
factor analysis in uruguay. Asian Pacific Journal of Cancer Prevention : APJCP, 13(1), 231-
235. doi:10.7314/APJCP.2012.13.1.231

[50] Lee, H. P., Gourley, L., Duffy, S. W., Estève, J., Lee, J., & Day, N. E. (1989). Colorectal
cancer and diet in an asian population--a case-control study among singapore chinese.
International Journal of Cancer, 43(6), 1007-1016. doi:10.1002/ijc.2910430609

[51] Fernandez, E., D'Avanzo, B., Negri, E., Franceschi, S., & Vecchia, C. L. (1996). Diet
diversity and the risk of colorectal cancer in northern italy. Cancer Epidemiology
Biomarkers & Prevention, 5(6), 433.

[52] Seow, A., Yuan, J., Sun, C., David Van Den Berg, Lee, H., & Yu, M. C. (2002). Dietary
isothiocyanates, glutathione S-transferase polymorphisms and colorectal cancer risk in the
singapore chinese health study. Carcinogenesis, 23(12), 2055-2061.
doi:10.1093/carcin/23.12.2055

44
[53] Evans, R. C., Fear, S., Ashby, D., Hackett, A., Williams, E., van der Vliet, M., . . . Rhodes, J.
M. (2002). Diet and colorectal cancer: An investigation of the lectin/galactose hypothesis.
Gastroenterology, 122(7), 1784-1792. doi:10.1053/gast.2002.33659

[54] Küry, S., Buecher, B., Robiou-du-Pont, S., Scoul, C., Sébille, V., Colman, H., . . . Bézieau,
S. (2007). Combinations of cytochrome P450 gene polymorphisms enhancing the risk for
sporadic colorectal cancer related to red meat consumption. Cancer Epidemiology
Biomarkers & Prevention, 16(7), 1460-1467. doi:10.1158/1055-9965.EPI-07-0236

[55] SÃEBÃ, M., SKJELBRED, C. F., LI, K. B., INGER MARIE BOWITZ LOTHE, HAGEN,
P. C., JOHNSEN, E., . . . KURE, E. H. (2008). CYP1A2 164 A àC polymorphism,
cigarette smoking, consumption of well-done red meat and risk of developing colorectal
adenomas and carcinomas. Anticancer Research, 28(4C), 2289-2295.

[56] Cotterchio, M., Boucher, B. A., Manno, M., Gallinger, S., Okey, A. B., & Harper, P. A.
(2008). Red meat intake, doneness, polymorphisms in genes that encode carcinogen-
metabolizing enzymes, and colorectal cancer risk. Cancer Epidemiology Biomarkers &
Prevention, 17(11), 3098-3107. doi:10.1158/1055-9965.EPI-08-0341

[57] Joshi, A. D., Corral, R., Siegmund, K. D., Haile, R. W., Le Marchand, L., Martínez, M.
E., . . . Stern, M. C. (2009). Red meat and poultry intake, polymorphisms in the nucleotide
excision repair and mismatch repair pathways and colorectal cancer risk. Carcinogenesis,
30(3), 472-479. doi:10.1093/carcin/bgn260

[58] Hu, X., Yuan, P., Yan, J., Feng, F., Li, X., Liu, W., & Yang, Y. (2013). Gene
polymorphisms of ADIPOQ +45T>G, UCP2 -866G>A, and FABP2 Ala54Thr on the risk of
colorectal cancer: A matched case-control study. PloS One, 8(6), e67275.
doi:10.1371/journal.pone.0067275

[59] Abu Mweis, S. S., Tayyem, R. F., Shehadah, I., Bawadi, H. A., Agraib, L. M., Bani-Hani, K.
E., . . . Al-Nusairr, M. (2015). Food groups and the risk of colorectal cancer: Results from a
jordanian case-control study. European Journal of Cancer Prevention : The Official Journal
of the European Cancer Prevention Organisation (ECP), 24(4), 313-320.
doi:10.1097/CEJ.0000000000000089

[60] Guo, X., Liu, J., You, L., Li, G., Huang, Y., & Li, Y. (2015). Association between
adiponectin polymorphisms and the risk of colorectal cancer. Genetic Testing and Molecular
Biomarkers, 19(1), 9-13. doi:10.1089/gtmb.2014.0238

[61] La Vecchia, C., Ferraroni, M., Mezzetti, M., Enard, L., Negri, E., Franceschi, S., & Decarli,
A. (1996). Attributable risks for colorectal cancer in northern italy. International Journal of
Cancer, 66(1), 60-64. doi:10.1002/(SICI)1097-0215(19960328)66:1<60::AID-
IJC11>3.0.CO;2-F

[62] Butler, L. M., Sinha, R., Millikan, R. C., Martin, C. F., Newman, B., Gammon, M. D., . . .
Sandler, R. S. (2003). Heterocyclic amines, meat intake, and association with colon cancer

45
in a population-based study. American Journal of Epidemiology, 157(5), 434-445.
doi:10.1093/aje/kwf221

[63] Satia Abouta, J., Galanko, J. A., Martin, C. F., Ammerman, A., & Sandler, R. S. (2004).
Food groups and colon cancer risk in African Americans and caucasians. International
Journal of Cancer, 109(5), 728-736. doi:10.1002/ijc.20044

[64] Freedman, A. N., Michalek, A. M., Marshall, J. R., Mettlin, C. J., Petrelli, N. J., Black, J.
D., . . . Asirwatham, J. E. (1996). Familial and nutritional risk factors for p53 overexpression
in colorectal cancer. Cancer Epidemiology Biomarkers & Prevention, 5(4), 285.

[65] Lohsoonthorn, P., & Danvivat, D. (1995). Colorectal cancer risk factors: A case-control
study in bangkok. Asia-Pacific Journal of Public Health, 8(2), 118-122.
doi:10.1177/101053959500800211

[66] Kotake, K., Koyama, Y., Nasu, J., Fukutomi, T., & Yamaguchi, N. (1995). Relation of
family history of cancer and environmental factors to the risk of colorectal cancer: A case-
control study. Japanese Journal of Clinical Oncology, 25(5), 195.

[67] Marchand, L. L., Hankin, J. H., Wilkens, L. R., Pierce, L. M., Franke, A., Kolonel, L. N., . . .
Donlon, T. (2001). Combined effects of well-done red meat, smoking, and rapid N-
acetyltransferase 2 and CYP1A2 phenotypes in increasing colorectal cancer risk. Cancer
Epidemiology Biomarkers & Prevention, 10(12), 1259-1266.

[68] Küry, S., Buecher, B., Robiou-du-Pont, S., Scoul, C., Sébille, V., Colman, H., . . . Bézieau,
S. (2007). Combinations of cytochrome P450 gene polymorphisms enhancing the risk for
sporadic colorectal cancer related to red meat consumption. Cancer Epidemiology
Biomarkers & Prevention, 16(7), 1460-1467. doi:10.1158/1055-9965.EPI-07-0236

[69] Navarro, A., Muñoz, S. E., Lantieri, M. J., del Pilar Diaz, M., Cristaldo, P. E., de Fabro, S.
P., & Eynard, A. R. (2004). Meat cooking habits and risk of colorectal cancer in córdoba,
argentina. Nutrition, 20(10), 873-877. doi:10.1016/j.nut.2004.06.008

[70] Manousos, O., Day, N. E., Trichopoulos, D., Gerovassilis, F., Tzonou, A., &
Polychronopoulou, A. (1983). Diet and colorectal cancer: A case-control study in greece.
International Journal of Cancer, 32(1), 1-5. doi:10.1002/ijc.2910320102

[71] Gerhardsson de Verdier, M., Hagman, U., Peters, R. K., Steineck, G., & Overvik, E. (1991).
Meat, cooking methods and colorectal cancer: A case-referent study in stockholm.
International Journal of Cancer, 49(4), 520-525. doi:10.1002/ijc.2910490408

[72] Bidoli, E., Franceschi, S., Talamini, R., Barra, S., & La Vecchia, C. (1992). Food
consumption and cancer of the colon and rectum in north-eastern italy. International Journal
of Cancer, 50(2), 223-229. doi:10.1002/ijc.2910500211

46
[73] Nowell, S., Coles, B., Sinha, R., MacLeod, S., Luke Ratnasinghe, D., Stotts, C., . . . Lang, N.
P. (2002). Analysis of total meat intake and exposure to individual heterocyclic amines in a
case-control study of colorectal cancer: Contribution of metabolic variation to risk. Mutation
Research - Fundamental and Molecular Mechanisms of Mutagenesis, 506, 175-185.
doi:10.1016/S0027-5107(02)00164-1

[74] Fernandez, E., La Vecchia, C., D'Avanzo, B., Negri, E., & Franceschi, S. (1997). Risk
factors for colorectal cancer in subjects with family history of the disease. British Journal of
Cancer, 75(9), 1381-1384. doi:10.1038/bjc.1997.234

[75] Kune, S., Kune, G. A., & Watson, L. F. (1987). Case-control study of dietary etiological
factors: The melbourne colorectal cancer study. Nutrition and Cancer, 9(1), 21-42.
doi:10.1080/01635588709513908

[76] Tuyns, A. J., Kaaks, R., & Haelterman, M. (1988). Colorectal cancer and the consumption
of foods: A case-control study in belgium. Nutrition and Cancer, 11(3), 189-204.
doi:10.1080/01635588809513986

[77] Di Maso, M., Talamini, R., Bosetti, C., Montella, M., Zucchetto, A., Libra, M., . . . Polesel,
J. (2013). Red meat and cancer risk in a network of case-control studies focusing on cooking
practices. Annals of Oncology, 24(12), 3107-3112. doi:10.1093/annonc/mdt392

Abstract (if available)

Abstract Background: Red and processed meat consumption have been linked to several cancers, including colorectal cancer (CRC). In 2015, the International Agency for Research on Cancer classified red meat as a probable carcinogen and processed meat as carcinogenic. Although a meta-analysis of prospective studies has been published, a comparable meta-analysis of case-control studies is currently lacking, in particular, with consideration of the great variability found across studies regarding analytical approaches, study design, and overall study quality. ❧ Methods and Findings: Case-control studies on the risk of colorectal cancer in relation to red and/or processed meat consumption were obtained by searching in PubMed until November 2016. For studies that met our selection criteria (see Methods), study-specific ORs were pooled using the fixed-effects model or random-effects models. The highest versus the lowest intake comparison, dose-response analyses, and stratified analyses were conducted to estimate the red/processed meat and CRC risk association, and to assess the role of potential confounders on the observed associations. The summary meta-analysis random effects ORs of the highest versus lowest red meat consumption comparison for colorectal and colon cancer were 1.24 (95% CI: 1.04-1.49) and 1.22 (95% CI: 1.02-1.45). Similar results were obtained for processed meat. In dose-response analyses, with every 100 g/day increment of processed meat consumption, the odds of having colorectal cancer increases by 197% (95% CI: 59%-455%). The summary ORs of the highest versus lowest red including processed meat consumption comparison for CRC, colon, and rectal cancer were 1.51 (95% CI: 1.22-1.89), 1.30 (95% CI: 1.05-1.60), and 1.44 (95% CI: 1.08-1.93), respectively. In dose-response analyses, red meat including processed meat was positively associated with increased risk of colorectal cancer (summary OR = 1.62, 95% CI = 1.25-2.11). In general, inclusion or exclusion of confounders (total energy intake, BMI, Physical activity, Fruits and/or vegetable intake, Alcohol, and smoking) seemed to have little influence on the associations for processed meat. In some instances, marginally higher for studies that did not include confounders. Studies of red meat and processed meat that considered total energy intake adjustment showed a stronger positive association with colorectal cancer (OR red meat and CRC= 1.32, 95% CI = 1.05-1.65

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Meat intake, polymorphisms in the NER and MMR pathways and colorectal cancer risk

PDF

Fish consumption and risk of colorectal cancer

PDF

Carcinogen metabolism genes, meat intake, and colorectal cancer risk

PDF

Genetic variation in the base excision repair pathway, environmental risk factors and colorectal adenoma risk

PDF

Disparities in colorectal cancer survival among Latinos in California

PDF

Diet quality and pancreatic cancer incidence in the multiethnic cohort

PDF

Pathogenic variants in cancer predisposition genes and risk of non-breast multiple primary cancers in breast cancer patients

PDF

Dietary and supplementary folate intake and prostate cancer risk

PDF

Origins of the gender disparity in bladder cancer risk: a SEER analysis

PDF

Risk factors of pelvic floor disorders in the multiethnic cohort study

PDF

The impact of the COVID-19 pandemic on cancer care delivery

PDF

Body size and the risk of prostate cancer in the multiethnic cohort

PDF

The evaluation of the long-term effectiveness of zero/low fluoroscopy workflow in ablation procedures for the treatment of paroxysmal and persistent atrial fibrillation

PDF

The risk estimates of pneumoconiosis and its relevant complications: a systematic review and meta-analysis

PDF

Dietary carcinogens and genetic variation in their metabolism: epidemiological studies on the risk of selected cancers

PDF

Associations between isoflavone soy protein (ISP) supplementation and breast cancer in postmenopausal women in the Women’s Isoflavone Soy Health (WISH) clinical trial

PDF

An analysis of disease-free survival and overall survival in inflammatory breast cancer

PDF

Prostate cancer disparities among Californian Latinos by country of origin: clinical characteristics, incidence, treatment received and survival

PDF

Preeclampsia and occurrence of neurological outcomes in the child: a meta-analysis

PDF

The effects of tobacco exposure on hormone levels and breast cancer risk among young women

Asset Metadata

Creator Tan, Lu (author)

Core Title Red and processed meat consumption and colorectal cancer risk: meta-analysis of case-control studies

School Keck School of Medicine

Degree Master of Science

Degree Program Applied Biostatistics and Epidemiology

Publication Date 10/30/2017

Defense Date 10/27/2017

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag colorectal cancer,Meat,meta-analysis,OAI-PMH Harvest,processed meat,red meat

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Stern, Mariana (committee chair), Mack, Wendy Jean (committee member), Wu, Anna (committee member)

Creator Email lukis-77@hotmail.com,lutan@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-447723

Unique identifier UC11265586

Identifier etd-TanLu-5861.pdf (filename),usctheses-c40-447723 (legacy record id)

Legacy Identifier etd-TanLu-5861.pdf

Dmrecord 447723

Document Type Thesis

Rights Tan, Lu

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA