Causes of cancer in the world: Comparative risk assessment of nine behavioural and environmental risk factors

Article (PDF Available)inThe Lancet 366(9499):1784-93 · December 2005with178 Reads
DOI: 10.1016/S0140-6736(05)67725-2 · Source: PubMed
Abstract
With respect to reducing mortality, advances in cancer treatment have not been as effective as those for other chronic diseases; effective screening methods are available for only a few cancers. Primary prevention through lifestyle and environmental interventions remains the main way to reduce the burden of cancers. In this report, we estimate mortality from 12 types of cancer attributable to nine risk factors in seven World Bank regions for 2001. We analysed data from the Comparative Risk Assessment project and from new sources to assess exposure to risk factors and relative risk by age, sex, and region. We applied population attributable fractions for individual and multiple risk factors to site-specific cancer mortality from WHO. Of the 7 million deaths from cancer worldwide in 2001, an estimated 2.43 million (35%) were attributable to nine potentially modifiable risk factors. Of these, 0.76 million deaths were in high-income countries and 1.67 million in low-and-middle-income nations. Among low-and-middle-income regions, Europe and Central Asia had the highest proportion (39%) of deaths from cancer attributable to the risk factors studied. 1.6 million of the deaths attributable to these risk factors were in men and 0.83 million in women. Smoking, alcohol use, and low fruit and vegetable intake were the leading risk factors for death from cancer worldwide and in low-and-middle-income countries. In high-income countries, smoking, alcohol use, and overweight and obesity were the most important causes of cancer. Sexual transmission of human papilloma virus is a leading risk factor for cervical cancer in women in low-and-middle-income countries. Reduction of exposure to key behavioural and environmental risk factors would prevent a substantial proportion of deaths from cancer.
Articles
1784 www.thelancet.com Vol 366 November 19, 2005
Introduction
Worldwide, between 1990 and 2001, mortality rates from
all cancers fell by 17% in those aged 30–69 years and
rose by 0·4% in those aged older than 70 years.
1,2
This
deline was lower than the fall in mortality rates from
cardiovascular diseases, which declined by 9% and 14%
in men aged 30–69 years and 70 years or older,
respectively, and of 15% and 11% for women in the
same age groups.
1,2
Age-specific mortality rates for chronic diseases are
driven by changes in exposure to risk factors, and by
availability of screening systems and treatment.
Advances in primary and secondary prevention and in
treatment have been effective in reducing mortality from
cardiovascular diseases over the past few decades, at least
in developed countries.
3
Therapeutic interventions have
had less success in reducing deaths from most cancers.
For example, in the USA, age-adjusted death rates for all
cancers combined fell slightly in the 1990s (on average
1·5% per year in men and 0·6% per year in women
between 1992 and 1999).
4
The fall in overall cancer
mortality in men was mainly a result of reductions in
mortality from lung, prostate, and colorectal cancers.
4
In
women, mortality due to lung cancer increased in the
1990s, but there was a decrease in death rates from breast
and colorectal cancers.
4
The fall in lung-cancer mortality
in men was mostly due to lower incidence,
4
itself a result
of a reduction in smoking. The causes of decreased
mortality from prostate cancer remain uncertain, since
incidence figures over time might not be comparable
because of changes in diagnostic techniques and death
certification.
5
Rates of death from colorectal cancer might
have fallen because of early detection and removal of
precancerous polyps, early detection of tumours, and
improved treatment.
6
For breast cancer in women,
increased coverage of mammography screening
7
and
successful treatment with multi-agent chemotherapy and
tamoxifen have been effective in reducing mortality.
8
Treatment has also been effective for some cancers in
children and young adults—eg, leukaemia and testicular
cancer.
9,10
5-year survival rates remain relatively low,
however, for lung, oesophageal, liver, stomach, and
pancreatic cancers (all 25%).
9
Although a combination of screening and treatment is
increasingly effective in reducing mortality from some
cancers, limitations in availability of clinical inter-
ventions for other cancers, and in access to and use of
existing technologies, clearly constrain the effects of
treatment on population trends in cancer mortality, even
in developed countries. As such, primary prevention
through lifestyle and environmental interventions might
offer the best option for reducing the large and
increasing burden of cancers worldwide. Policies and
programmes to implement such interventions depend
Lancet 2005; 366: 1784–93
*Members listed at end of paper
Harvard School of Public
Health, Boston, MA, USA, and
Initiative for Global Health,
Harvard University, Cambridge,
MA, USA (G Danaei MD,
M Ezzati PhD,
Prof CJLMurray MD); Clinical
Trials Research Unit (CTRU),
University of Auckland,
Auckland, New Zealand
(S Vander Hoorn MSc); and
School of Population Health,
University of Queensland,
Brisbane, Australia
(Prof A D Lopez PhD)
Correspondence to:
Dr Majid Ezzati, Department of
Population and International
Health, Harvard School of Public
Health, 665 Huntington Avenue,
Boston, MA 02115, USA
mezzati@hsph.harvard.edu
Causes of cancer in the world: comparative risk assessment
of nine behavioural and environmental risk factors
Goodarz Danaei, Stephen Vander Hoorn, Alan D Lopez, Christopher J L Murray, Majid Ezzati, and the Comparative Risk Assessment collaborating
group (Cancers)*
Summary
Introduction With respect to reducing mortality, advances in cancer treatment have not been as effective as those for
other chronic diseases; effective screening methods are available for only a few cancers. Primary prevention through
lifestyle and environmental interventions remains the main way to reduce the burden of cancers. In this report, we
estimate mortality from 12 types of cancer attributable to nine risk factors in seven World Bank regions for 2001.
Methods We analysed data from the Comparative Risk Assessment project and from new sources to assess exposure
to risk factors and relative risk by age, sex, and region. We applied population attributable fractions for individual and
multiple risk factors to site-specific cancer mortality from WHO.
Findings Of the 7 million deaths from cancer worldwide in 2001, an estimated 2·43 million (35%) were attributable
to nine potentially modifiable risk factors. Of these, 0·76 million deaths were in high-income countries and
1·67 million in low-and-middle-income nations. Among low-and-middle-income regions, Europe and Central Asia
had the highest proportion (39%) of deaths from cancer attributable to the risk factors studied. 1·6 million of the
deaths attributable to these risk factors were in men and 0·83 million in women. Smoking, alcohol use, and low fruit
and vegetable intake were the leading risk factors for death from cancer worldwide and in low-and-middle-income
countries. In high-income countries, smoking, alcohol use, and overweight and obesity were the most important
causes of cancer. Sexual transmission of human papilloma virus is a leading risk factor for cervical cancer in women
in low-and-middle-income countries.
Interpretation Reduction of exposure to key behavioural and environmental risk factors would prevent a substantial
proportion of deaths from cancer.
Articles
www.thelancet.com Vol 366 November 19, 2005 1785
on reliable and comparable analyses of the effect of risk
factors for cancer at the population level.
Several reports have quantified the effects of risk
factors on cancer incidence and mortality, as
summarised in the World Cancer Report.
11
Most of these
studies, however, are restricted to one risk factor, one
site of cancer, or one population.
12–18
Pisani and
colleagues
19
estimated the worldwide proportion of
cancer incidence attributable to selected infectious
agents to be about 16%. Parkin and colleagues
20
presented attributable fractions for several risk factor-
cancer site combinations based on a review of published
studies. Our aim was to estimate the worldwide and
regional mortality from site-specific cancers attributable
to specific risk factors, individually and jointly.
Methods
Risk factors assessed
We used systematic reviews and meta-analyses of the
evidence on risk factor exposure and relative risk from
the Comparative Risk Assessment project
21,22
and data
from new sources to quantify the effect of a group of risk
factors on cancer mortality (table 1).
23–31
We selected the
risk factors on the basis of the following criteria: likely to
be a leading cause of worldwide or regional disease
burden; not too specific (eg, specific types of fruits and
vegetables) or too broad (eg, diet as a whole) for
comparable definition and quantification of exposure in
different populations; high likelihood of causality;
reasonably complete data on population exposure and
risk levels, or appropriate methods for extrapolation
when necessary; and potentially modifiable.
Procedures
For every risk factor, an expert working group undertook
comprehensive and systematic reviews of published
studies and other sources (eg, government reports,
international databases) to obtain data on the risk-factor
exposure and relative risk (RR). The groups also obtained
primary data, and undertook reanalyses of original data
Exposure variable Theoretical-minimum-risk Cancer sites affected (age groups assessed)†
exposure distribution
Diet and physical inactivity
Overweight and BMI (kg/m
2
) 21 SD 1 kg/m
2
Corpus uteri cancer, colorectal cancers (30 years), post-
obesity (high BMI)
23
menopausal breast cancer (45 years), gallbladder cancer,
kidney cancer
Low fruit and vegetable Fruit and vegetable intake per day 600 SD 50 g intake per Colorectal cancer, stomach cancer, lung cancer,
intake
24
day for adults oesophageal cancer (15 years)
Physical inactivity
25
Three categories: inactive, insufficiently active (2·5 h per week of 2·5 h per week of Breast cancer, colorectal cancer (15 years), prostate cancer
moderate-intensity activity, or 4000 KJ per week), and sufficiently moderate-intensity
active. Activity in spare time, work, and transport considered activity or equivalent
(400 KJ per week)
Addictive substances
Smoking
26
Current levels of smoking impact ratio (indirect indicator of No smoking Lung cancer, mouth and oropharynx cancer, oesophageal
accumulated smoking risk based on excess lung-cancer mortality) cancer, stomach cancer, liver cancer, pancreatic cancer,
cervix uteri cancer, bladder cancer, leukaemia (30 years)
Alcohol use
27
Current alcohol consumption volumes and patterns No alcohol use‡ Liver cancer, mouth and oropharynx cancer, breast cancer,
oesophageal cancer, selected other cancers (15 years)
Sexual and reproductive health
Unsafe sex
28
Sex with an infected partner without any measures to prevent No unsafe sex Cervix uteri cancer (all ages)§
infection
Environmental risks
Urban air pollution
29
Estimated yearly average particulate matter concentration for 7·5 g/m
3
for PM
2·5
Lung cancer (30 years)
particles with aerodynamic diameters 2·5 microns or 10 microns 15 g/m
3
for PM
10
(PM
2·5
or PM
10
)
Indoor smoke from household Household use of solid fuels No household solid fuel use Lung cancer (coal) (30 years)
use of solid fuels
30
with limited ventilation
Other selected risks
Contaminated injections Exposure to 1 contaminated injection (contamination refers to No contaminated injections Liver cancer (all ages)
in health-care settings
31
potential transmission of hepatitis B virus and hepatitis C virus)
*New exposure data and epidemiological evidence on disease outcomes and relative risks used when new epidemiological analyses allowed improvements compared with original analyses of Comparative Risk Assessment
project—eg, relative risks for site-specific cancers as a result of smoking with better adjustment for potential confounders and new exposure data sources for overweight and obesity. †Italics=outcomes likely to be causal but not
quantified because of insufficient evidence on prevalence or hazard size. Corresponding ICD 9 3-digit codes: bladder cancer 188; breast cancer 174; cervix uteri cancer 180; colorectal cancers 153–154; corpus uteri cancer
179, 182; leukaemia 204–208; liver cancer 155; mouth and oropharynx cancer 140–149; oesophageal cancer 150; pancreatic cancer 157; stomach cancer 151; trachea, bronchus, and lung cancers 162; selected other cancers
210–239. Total deaths from cancer are from WHO. Methods used by WHO
32,33
are based on a combination of vital statistics for countries with complete vital registration and medical certification of deaths, and on a combination
of demographic techniques and information from cancer registries for countries with incomplete vital statistics, with sample registration or surveillance systems, or without vital registration. ‡Alcohol has benefits as well as
harms for different diseases, also depending on patterns of alcohol consumption. A theoretical minimum of zero was chosen for alcohol use because, despite benefits for specific diseases (cardiovascular) in some populations
global and regional burden of disease due to alcohol use was dominated by its effects on neuropsychological diseases and injuries, which are considerably larger than benefits to vascular diseases. Furthermore, no benefits for
neoplastic disease have been noted from alcohol. §A proportion of HPV infections that lead to cervix uteri cancer are transmitted through routes other than sexual contact. PAF for unsafe sex, as defined in the comparative risk
assessment project,
28
measures current population-level cervix cancer mortality that would be reduced, had there never been any sexual transmission of infection—ie, the consequences of past and current exposure, as we do for
accumulated hazards of smoking. By considering health consequences of past and current exposure, nearly all of sexually transmitted diseases are attributable to unsafe sex because, in the absence of sexual transmission in the
past, current infections transmitted through other forms of contact would not occur if infected hosts acquired their infection sexually (and so on in the sequence of past infected hosts).
Table 1: Cancer risk factors, exposure variables, theoretical-minimum-risk exposure distributions, disease outcomes*
Articles
1786 www.thelancet.com Vol 366 November 19, 2005
sources and meta-analyses of epidemiological studies.
The details of data sources and analyses for each risk
factor are provided in references cited in table 1. All
sources cited in table 1 have been peer-reviewed by
external reviewers. The expert working groups presented
exposure and RRs separately for men and women, for
eight age groups (0–4, 5–14, 15–29, 30–44, 45–59, 60–69,
70–79, and 80 years), and for the 14 epidemiological
subregions used in the Global Burden of Disease study;
22
converted to World Bank regions here because they
correspond better to geographically well known regions
of the world (webtables 1 and 2).
Statistical analysis
To estimate the population attributable fractions (PAF)
for individual risk factors, we used the equation shown
in the panel, or its discreet version for risk factors with
categorical exposure data. For every risk factor and
cancer site, we used PAF to estimate the proportional
reduction in site-specific cancer death that would arise if
exposure to the risk factor were reduced to the
counterfactual distribution. The alternative (counter-
factual) scenario used in this study is defined as the
exposure distribution that would result in the lowest
population risk, referred to as the theoretical-minimum-
risk exposure distribution (table 1). This method
provides a quantitative assessment of the potential
reduction in cancer burden in a consistent and
comparable way across risk factors. The theoretical-
minimum-risk exposure distribution is zero for risk
factors for which zero exposure could be defined, and
would indicate minimum risk (eg, the whole population
being lifelong non-smokers). For some risk factors, zero
exposure is an inappropriate choice, either because it is
physiologically impossible (eg, body-mass index [BMI])
or because there are physical lower limits to exposure
reduction (eg, concentration of ambient particulate
matter). For these risk factors, we used the lowest levels
noted in specific populations from epidemiological
studies. For risk factors with protective effects (ie, fruit
and vegetable intake, physical activity) we chose a
counterfactual exposure distribution based on a
combination of levels observed in high-intake
populations and the level up to which health benefits
might accrue.
Because most cancers are caused by multiple risk
factors, PAFs for individual risk factors for the same
cancer site overlap and can add to more than 100%. For
example, more than 70% of Chinese households rely on
solid fuels (coal and biomass) for cooking and heating,
30
and more than 60% of Chinese men smoke. Since
smoking and coal smoke magnify one another’s hazards
for lung cancer,
13
some deaths from lung cancer can be
prevented by removal of smoking or of exposure to
indoor smoke from coal. Such cases would be attributed
to both risk factors. Multicausality also means that a
range of interventions can be used for disease
prevention, with the specific mix being affected by factors
such as cost, technology availability, infrastructure, and
preferences. We therefore also estimated the PAFs for
multiple risk factors, as described in detail elsewhere.
34
We calculated all individual and joint PAFs by sex and
by eight age groups. For every World Bank region, age,
sex, and cancer site, we multiplied the PAF by total
regional site-specific cancer mortality for the year 2001
(from WHO databases) to calculate deaths from site-
specific cancer attributable to the risk factor or group of
risk factors. Attributable deaths were aggregated into
three age groups for presentation (age-specific results
available from the corresponding author).
Role of the funding source
The sponsor of the study had no role in study design,
data collection, data analysis, data interpretation, or
writing of the report. The corresponding author had full
access to all the data in the study and had final
responsibility for the decision to submit for publication.
Results
Table 2 shows the estimated individual and joint
contributions of the selected risk factors to mortality for
each cancer site, for the world, and separately for low-
and-middle-income, and high-income countries. The
joint effects are further divided by region for low-and-
middle-income countries in figure 1, by sex and by
income in figure 2, and by age in table 3. Of the 7 million
global deaths from cancer in 2001, an estimated
2·43 million (35%) were attributable to the joint effect of
the nine risk factors listed in table 1.
Cancers with the largest proportions (60%) attrib-
utable to these risks were cervix uteri cancer, lung
cancer, and oesphagus cancer. The main risk factors for
these cancers included sexual transmission of HPV
leading to persistent infection with oncogenic viruses,
smoking, alcohol use, and low fruit and vegetable intake.
Cancers with the smallest joint PAFs (colorectal cancers
[13%] and leukaemia [9%]) were those with a large
number of risks with heterogeneous and unmeasured
exposure patterns across populations, possibly including
major genetic susceptibilities.
More than a third (37%) of all risk-factor attributable
deaths (908 000 deaths) were from lung cancer,
Panel: Continuous PAF formula
=risk factor exposure level
P()=population distribution of exposure
P()=counterfactual distribution of exposure
RR()=relative risk of mortality from site-specific cancer at exposure level
m=maximum exposure level
RR(
)P(
)d
RR(
)P(
)d
PAF=
=0 =0
m
RR(
)P(
)d
=0
m
m
See Lancet Online
for webtables 1 and 2
For details of data sources and
analyses for each risk factor,
see http://www.who.int/
publications/cra/en/index.html
Articles
www.thelancet.com Vol 366 November 19, 2005 1787
Total deaths PAF (%) and number of attributable cancer PAF due to
deaths (thousands) for individual risk factors joint hazards
of risk factors
Worldwide
Mouth and oropharynx cancers 311 633 Alcohol use (16%, 51), smoking (42%, 131) 52%
Oesophageal cancer 437 511 Alcohol use (26%, 116), smoking (42%, 184), low fruit and vegetable intake (18%, 80) 62%
Stomach cancer 841 693 Smoking (13%, 111), low fruit and vegetable intake (18%, 147) 28%
Colon and rectum cancers 613 740 Overweight and obesity (11%, 69), physical inactivity (15%, 90), low fruit and vegetable intake (2%, 12) 13%
Liver cancer 606 441 Smoking (14%, 85), alcohol use (25%, 150), contaminated injections in health-care settings (18%, 111) 47%
Pancreatic cancer 226 981 Smoking (22%, 50) 22%
Trachea, bronchus, and lung cancers 1 226 574 Smoking (70%, 856), low fruit and vegetable intake (11%, 135), indoor smoke from household use of solid fuels (1%, 16), 74%
urban air pollution (5%, 64)
Breast cancer 472 424 Alcohol use (5%, 26), overweight and obesity (9%, 43), physical inactivity (10%, 45) 21%
Cervix uteri cancer 234 728 Smoking (2%, 6), unsafe sex (100%, 235) 100%
Corpus uteri cancer 70 881 Overweight and obesity (40%, 28) 40%
Bladder cancer 175 318 Smoking (28%, 48) 28%
Leukaemia 263 169 Smoking (9%, 23) 9%
Selected other cancers 145 802 Alcohol use (6%, 8) 6%
All other cancers 1 391 507 None of selected risk factors 0%
All cancers 7 018402 Alcohol use (5%, 351), smoking (21%, 1493), low fruit and vegetable intake (5%, 374), indoor smoke from household 35%
use of solid fuels (0·5%, 16), urban air pollution (1%, 64), overweight and obesity (2%, 139), physical inactivity (2%, 135),
contaminated injections in health-care settings (2%, 111), unsafe sex (3%, 235)
Low-and-middle-income countries
Mouth and oropharynx cancers 271 074 Alcohol use (14%, 38), smoking (37%, 100) 48%
Oesophageal cancer 379 760 Alcohol use (24%, 92) smoking (37%, 141), low fruit and vegetable intake (19%, 73) 58%
Stomach cancer 695 426 Smoking (11%, 74), low fruit and vegetable intake (19%, 130) 27%
Colon and rectum cancers 356 949 Overweight and obesity (9%, 32), physical inactivity (15%, 54), low fruit and vegetable intake (2%, 9) 11%
Liver cancer 504 407 Smoking (11%, 56), alcohol use (23%, 117), contaminated injections in health-care settings (21%, 108) 45%
Pancreatic cancer 116 827 Smoking (15%, 18) 15%
Trachea, bronchus, and lung cancers 770 938 Smoking (60%, 466), low fruit and vegetable intake (13%, 98), indoor smoke from household use of solid fuels (2%, 16), 66%
urban air pollution (7%, 52)
Breast cancer 317 195 Alcohol use (4%, 12), overweight and obesity (7%, 23), physical inactivity (10%, 30) 18%
Cervix uteri cancer 218 064 Smoking (2%, 4), unsafe sex (100%, 218) 100%
Corpus uteri cancer 43 926 Overweight and obesity (37%, 16) 37%
Bladder cancer 116 682 Smoking (21%, 24) 21%
Leukaemia 190 059 Smoking (6%, 11) 6%
Selected other cancers 88 706 Alcohol use (4%, 3) 4%
All other cancers 882 001 None of selected risk factors 0%
All cancers 4 952014 Alcohol use (5%, 262), smoking (18%, 896), low fruit and vegetable intake (6%, 311), indoor smoke from household 34%
use of solid fuels (0·5%, 16), urban air pollution (1%, 52), overweight and obesity (1%, 71), physical inactivity (2%, 84),
contaminated injections in health-care settings (2%, 108), unsafe sex (4%, 218)
High-income countries
Mouth and oropharynx cancers 40 559 Alcohol use (33%, 14), smoking (71%, 29) 80%
Oesophageal cancer 57 752 Alcohol use (41%, 24), smoking (71%, 41), low fruit and vegetable intake (12%, 7) 85%
Stomach cancer 146 267 Smoking (25%, 36), low fruit and vegetable intake (12%, 17) 34%
Colon and rectum cancers 256 791 Overweight and obesity (14%, 37), physical inactivity (14%, 36), low fruit and vegetable intake (1%, 3) 15%
Liver cancer 102 033 Smoking (29%, 29), alcohol use (32%, 33), contaminated injections in health-care settings (3%, 3) 52%
Pancreatic cancer 110 154 Smoking (30%, 33) 30%
Trachea, bronchus, and lung cancers 455 636 Smoking (86%, 391), low fruit and vegetable intake (8%, 36), indoor smoke from household use of solid fuels (0%), 87%
urban air pollution (3%, 12)
Breast cancer 155 230 Alcohol use (9%, 14), overweight and obesity (13%, 20), physical inactivity (9%, 15) 27%
Cervix uteri cancer 16 663 Smoking (11%, 2), unsafe sex (100%, 17) 100%
Corpus uteri cancer 26 955 Overweight and obesity (43%, 12) 43%
Bladder cancer 58 636 Smoking (41%, 24) 41%
Leukaemia 73 110 Smoking (17%, 12) 17%
Selected other cancers 57 095 Alcohol use (8%, 5) 8%
All other cancers 509 507 None of selected risk factors 0%
All cancers 2 066388 Alcohol use (4%, 88), smoking (29%, 596), low fruit and vegetable intake (3%, 64), indoor smoke from household use of 37%
solid fuels (0%, 0), urban air pollution (1%, 12), overweight and obesity (3%, 69), physical inactivity (2%, 51), contaminated
injections in health-care settings (0·5%, 3), unsafe sex (1%, 17)
Results for individual regions available from corresponding author.
Table 2: Individual and joint contributions of risk factors in table 1 to mortality from site-specific cancers
Articles
1788 www.thelancet.com Vol 366 November 19, 2005
12% from liver cancer (283 000 deaths), and 11% from
oesophageal cancer (271 000 deaths), reflecting the
combinations of relatively large joint PAFs applied to
large numbers of deaths from these cancers (table 2). All
three cancers have 5-year survival rates of less than 25%.
Leukaemia (23 000) and corpus uteri cancer (28 000) had
the smallest number of attributable deaths (3% of deaths
from leukaemia were attributed to occupational
exposures,
35
not re-analysed here because of difficulties
in converting exposure to new regions).
Despite having only 15% of the world’s population
(21% of the population aged 30 years), high-income
countries accounted for 29% of the 7 million deaths
from cancer worldwide and 31% of the 2·43 million that
were attributable to the selected risks in table 1; the
remaining 1·67 million attributable deaths occurred in
low-and-middle-income countries. Except for cervix uteri
cancer, joint PAFs were greater in high-income
countries than in low-and-middle-income countries for
all cancer sites. This finding is mostly a result of higher
and longer population exposure to smoking and alcohol
use, and is particularly evident for lung, mouth and
oropharynx, and oesophageal cancers, especially in men.
The joint PAFs for all cancers combined, however, were
similar for both groups of countries (37% in high-
income and 34% in low-and-middle-income regions).
The smaller difference in all-site PAF than in PAFs for
site-specific cancers arises because the contribution of
site-specific cancers to total cancer mortality varies (eg,
stomach and liver cancers caused 695 000 [14% of all
deaths from cancer] and 504 000 [10% of all deaths from
cancer] deaths, respectively, in low-and-middle-income
countries, and 146 000 [7% of all deaths from cancer]
and 102 000 [5% of all deaths from cancer] deaths in
high-income countries). Furthermore, those cancer sites
that were not affected by risk factors in table 1 accounted
for 18% of all deaths from cancer in low-and-middle-
income countries, but 25% in high-income nations.
Lung, liver, and oesophageal cancers had the largest
number of attributable deaths in low-and-middle-
income countries (512 000, 229 000, and 222 000,
respectively). In high-income countries, lung cancer
alone constituted 52% of all risk-factor attributable
deaths from cancer (396 000 deaths); other cancers
accounted for 7% or less, indicating the fairly successful
preventive interventions for some of these cancers (eg,
reduction in liver cancer as a result of reduced exposure
to infectious agents) and the disproportionate rise in
lung cancer mortality in many high-income countries
where smoking prevalence remains high.
In low-and-middle-income countries, the joint PAF for
all cancer sites combined was largest in Europe and
Central Asia (ECA; 39%) and smallest in the Sub-
Saharan Africa (SSA; 24%) and Middle East and North
Africa (MENA; 24%) regions. For men, ECA had the
largest joint PAF (50%) and SSA the smallest (19%),
being strongly affected by the role of smoking, alcohol
Trachea, bronchus, and lung cancers
Stomach cancer
Liver cancer
Oesophageal cancer
Colon and rectum cancers
Breast cancer
Mouth and oropharynx cancers
Cervix uteri cancer
Leukaemia
Pancreatic cancer
Bladder cancer
Corpus uteri cancer
East Asia and Pacific
Europe and Central Asia
Latin America and Caribbean
Middle East and north Africa
South Asia
Sub-Saharan Africa
600 000
800 000
400 000
200 000
0
Number of deaths
Figure 1: Deaths from site-specific cancers attributable to selected risk factors in low-and-middle-income
countries, by region
For every cancer site, solid blocks of colour represent deaths not attributable to risks assessed and broken blocks of
colour represent deaths attributable to selected risk factors in table 1.
Articles
www.thelancet.com Vol 366 November 19, 2005 1789
use, and overweight and obesity; for women, SSA had the
highest PAF (29%) and MENA the smallest (19%). The
main driver of the high PAF for women in SSA was the
large number of deaths from cervix uteri cancer, which is
in turn affected by limited access to cervical screening
above and beyond sexual transmission of HPV.
The largest number of deaths attributable to the
selected risks in table 1 in the low-and-middle-income
countries was in EAP (746 000; 45% of all attributable
deaths from cancer in low-and-middle-income
countries), ECA (324 000; 19%) and South Asia region
(SAR: 322 000; 19%). In all three regions, the total
number of deaths from cancer is large, mostly because
of the large population size in EAP and SAR. The
population of ECA is smaller, but the total number of
deaths from cancer is relatively high and larger
proportions are attributable to the individual and joint
hazards of these risk factors, especially smoking. MENA
and SSA had the smallest number of deaths from cancer
attributable to the risk factors considered here (40 000
and 97 000, respectively), because of both the smaller
total number of deaths from cancer and the lower PAFs
attributable to these risks.
Smoking alone is estimated to have caused 21% of
deaths from cancer worldwide. Alcohol use and low fruit
and vegetable intake caused another 5% each. Smoking
was responsible for a higher fraction of deaths from
cancer in high-income countries (29%) than in low-and-
middle-income regions (18%), because of the shorter
history of smoking and lower prevalence among women.
However, the number of smoking-attributable deaths
from cancer was larger in the low-and-middle-income
countries (896 000 vs 596 000) because of the larger total
number of deaths from cancer. All-cancer-site PAF for
alcohol use was almost equal in the two groups of
countries because of large numbers of alcohol-
attributable deaths in ECA and EAP. In EAP, liver and
oesophagus were the cancer sites with the largest
number of deaths attributable to alcohol use. In addition
to smoking and alcohol use, some risks with well
established interventions also caused considerable
excess cancer mortality in low-and-middle-income
regions: contaminated injections in health-care settings
caused 108 000 deaths in the six regions, 91 000 of them
in EAP; the cancer hazards of indoor smoke from solid
fuels were restricted to EAP and SAR, where coal is
used, causing 16 000 deaths from lung cancer (other
regions use biomass fuels for which lung cancer hazards
have not yet been established
30
). The proportion of all-
site-cancer deaths attributable to overweight or obesity
and physical inactivity was highest in ECA, indicating
the relatively high population exposure to these lifestyle
related risk factors, which coexist with the risk from
smoking and alcohol use.
The risk factors studied caused about twice as many
deaths from cancer in men as in women (1·6 vs
0·83 million); the fraction of deaths from cancer
attributable to these risks was 41% for men versus 27%
for women. Lung cancer contributed the most to the
risk-factor attributable deaths in men (45% of all
attributable deaths) and cervix uteri cancer in women
(28% of all attributable deaths). Smoking and alcohol
use played an important part in male–female differences
in PAFs and attributable deaths. Excluding cervix,
corpus uteri, and breast cancers, and their specific risk
factors, which almost exclusively affect women, the total
number of deaths from major cancers (lung, stomach,
and liver) as well as the number attributable to the major
risks, was higher in men than women (figure 2). This
pattern did not follow for colorectal cancer, where
smoking and alcohol use have no established role. The
largest male–female difference in PAFs was for mouth
or oropharynx cancer, which is strongly affected by
alcohol use and smoking (66% for men vs 23% for
women). Mouth or oropharynx cancer also had the
Trachea, bronchus, and lung cancers
Stomach cancer
Liver cancer
Oesophageal cancer
Colon and rectum cancers
Breast cancer
Mouth and oropharynx cancers
Cervix uteri cancer
Leukaemia
Pancreatic cancer
Bladder cancer
Corpus uteri cancer
High-income countries
Female
Male
Low-and-middle-income countries
600
000
600
000
800
000
400
000
400
000
200
000
200
000
0
Number of deaths
Figure 2: Worldwide deaths from site-specific cancers attributable to selected risk factors by sex
For every cancer site, solid blocks of colour represent deaths not attributable to risks assessed and broken blocks of
colour represent deaths attributable to selected risk factors in table 1.
Articles
1790 www.thelancet.com Vol 366 November 19, 2005
largest sex difference in low-and-middle-income
countries (63% for men vs 17% for women); in high-
income countries, liver cancer had the largest sex
difference in PAFs (59% for men vs 37% for women).
Except for lung cancer in high-income countries, joint
PAFs were largest for those aged 30–69 years. This
finding is in part the result of the cohort effects of
exposure to smoking and alcohol use. More than half of
all deaths from cancer were between ages 30 years and 69
years (table 3). The largest number of attributable deaths
for all cancer sites, except bladder, was also noted in this
age group, pointing to the potential large gains in life
expectancy that could be achieved by reducing exposure
to these risk factors. Leukaemia claimed the largest total
number of deaths from cancer in the youngest age group
(30 years). None of the leukaemia deaths in these ages
were attributed to the risk factors we assessed because
most epidemiological studies only estimate hazards after
the age of 30 years. Although the total number of deaths
from cancer in those younger than age 30 years was
generally low, the low-and-middle-income countries had
noticeably larger numbers of deaths and higher PAFs in
this age group (largely due to the large number of deaths
from liver cancer attributable to contaminated
injections), indicating the success of the cancer control
programmes for young people in high-income nations.
Discussion
More than one in every three of the 7 million deaths
from cancer worldwide is caused by nine potentially
Age Total attributable
deaths
0–29 years 30–69 years 70 years
World
Mouth and oropharynx cancers 6 (15%, 1) 204 (58%, 118) 101 (44%, 44) 163
Oesophageal cancer 2 (38%, 0·8) 264 (66%, 175) 172 (55%, 95) 271
Stomach cancer 9 (21%, 2) 455 (32%, 145) 378 (24%, 92) 240
Colon and rectum cancers 6 (3%, 0·2) 282 (14%, 40) 326 (12%, 40) 79
Liver cancer 16 (36%, 6) 394 (49%, 195) 196 (42%, 82) 283
Pancreatic cancer 0·7 (0%, 0) 111 (23%, 25) 115 (22%, 25) 50
Trachea, bronchus, and lung cancers 5 (11%, 0·5) 684 (74%, 503) 537 (75%, 405) 908
Breast cancer 2 (0%, 0) 305 (20%, 61) 165 (22%, 37) 98
Cervix uteri cancer 11 (100%, 11) 156 (100%, 156) 67 (100%, 67) 235
Corpus uteri cancer 0·4 (0%, 0) 35 (43%, 15) 36 (37%, 13) 28
Bladder cancer 1 (0%, 0) 67 (29%, 19) 106 (27%, 29) 48
Leukaemia 75 (0%, 0) 107 (12%, 13) 81 (12%, 10) 23
All cancers 222 (10%, 21) 3783 (39%, 1467) 3013 (31%, 939) 2427
Low-and-middle-income countries
Mouth and oropharynx cancers 6 (14%, 0·9) 180 (55%, 98) 85 (37%, 32) 131
Oesophageal cancer 2 (38%, 0·8) 235 (64%, 150) 143 (50%, 71) 222
Stomach cancer 8 (21%, 2) 399 (31%, 124) 288 (22%, 64) 191
Colon and rectum cancers 6 (3%, 0·2) 193 (12%, 24) 158 (10%, 16) 40
Liver cancer 16 (36%, 6) 346 (48%, 167) 142 (40%, 56) 229
Pancreatic cancer 0·6 (0%, 0) 68 (18%, 12) 48 (11%, 5) 18
Trachea, bronchus, and lung cancers 4 (11%, 0·5) 482 (68%, 329) 284 (64%, 183) 512
Breast cancer 2 (0%, 0) 227 (18%, 40) 89 (18%, 16) 56
Cervix uteri cancer 11 (100%, 11) 146 (100%, 146) 61 (100%, 61) 218
Corpus uteri cancer 0·4 (0%, 0) 25 (41%, 10) 19 (34%, 6) 16
Bladder cancer 1 (0%, 0) 53 (25%, 13) 62 (18%, 11) 24
Leukaemia 70 (0%, 0) 81 (10%, 8) 39 (7%, 3) 11
All cancers 206 (10%, 20) 2946 (38%, 1123) 1800 (29%, 524) 1668
High-income countries
Mouth and oropharynx cancers 0·2 (34%, 0) 24 (82%, 20) 16 (78%, 13) 32
Oesophageal cancer 0·1 (55%, 0) 29 (87%, 25) 29 (83%, 24) 49
Stomach cancer 0·5 (19%, 0) 56 (37%, 21) 90 (31%, 28) 49
Colon and rectum cancers 0·5 (2%, 0) 88 (18%, 16) 168 (14%, 24) 40
Liver cancer 0·4 (33%, 0·1) 48 (57%, 28) 54 (48%, 26) 53
Pancreatic cancer 0·1 (0%, 0) 43 (30%, 13) 67 (30%, 20) 33
Trachea, bronchus, and lung cancers 0·3 (11%, 0) 202 (86%, 174) 253 (88%, 222) 396
Breast cancer 0·3 (0%, 0) 79 (27%, 22) 76 (27%, 21) 42
Cervix uteri cancer 0·2 (100%, 0·2) 10 (100%, 10) 6 (100%, 6) 17
Corpus uteri cancer 0 (0%, 0) 10 (48%, 5) 17 (41%, 7) 12
Bladder cancer 0 (0%, 0) 14 (43%, 6) 44 (40%, 18) 24
Leukaemia 4 (0%, 0) 26 (19%, 5) 43 (17%, 7) 12
All cancers 16 (4%, 0·6) 837 (41%, 343) 1212 (34%, 415) 759
Numbers in brackets show PAF (%) and number of attributable deaths (thousands) for the joint effects of risk factors in table 1. Only cancer sites affected by risk factors in table 1 are
shown.
Table 3: Total cancer deaths (thousands) by age and cancer site
Articles
www.thelancet.com Vol 366 November 19, 2005 1791
modifiable risk factors, with smoking and alcohol use
having particularly important roles in both high-income
and low-and-middle-income countries. Sexual trans-
mission of HPV, leading to persistent infection with
oncogenic types of virus, as a risk factor for cervix uteri
cancer is a major risk factor for women, especially in the
poorest regions (SSA and SAR) where access to
screening for cervical cancer is limited. Other potentially
modifiable risk factors, which have not been assessed
here, might increase this proportion substantially for
some cancer sites. Our estimate of the proportion of
deaths attributable worldwide to the nine risk factors we
studied is about half of what Doll and Peto estimated
18
by
comparing age-standardised incidence rates from the
USA from 1978 with the lowest reliably observed
incidence rates in other populations. Because Doll and
Peto
18
used comparison of incidence rates, their
estimates include differences in exposure to all known
and unknown risk factors. Furthermore, their estimates
applied to the USA only. Therefore the estimates of Doll
and Peto are not directly comparable to ours.
Some important cancers (eg, prostate, kidney,
melanoma, and lymphomas) were not attributable to any
of the risks we assessed. These cancers are those with
multiple confirmed or suspected environmental and
behavioural risks, with heterogeneous exposure patterns
that make exposure and hazard quantification difficult.
Furthermore, we did not assess some fairly well known
risk factors, such as occupational exposures, which are
responsible for 102 000 deaths from cancer worldwide;
35
Helicobacter pylori exposure in food; and exposure to
ultraviolet light and environmental tobacco smoke. We
excluded these factors for the main part because of the
limitations of deriving detailed exposure estimates from
existing data (eg, extent of exposure to environmental
tobacco smoke depends on public smoking regulation,
exposure to ultraviolet light depends on the time spent
outdoors and use of protection, and exposure to H pylori
depends on methods of food preservation).
There are several sources of uncertainty for exposure
and relative risks in our estimates, especially those that
involve extrapolation of exposure and hazard from one
population to another. These are described in more
detail in the descriptions of the data sources for each risk
factor.
21, 23–31
The sources of uncertainty in methods and
assumptions for estimating joint PAFs, and the
sensitivity of results to these assumptions, have also
been described elsewhere.
34
In addition to uncertainty in
individual and joint PAFs, there is uncertainty in total
site-specific cancer mortality to which the PAFs are
applied.
32,33
75 countries have reasonably complete vital
registration and medical certification of deaths.
32,33
Another 51 countries have incomplete vital statistics, or
use sample registration or surveillance systems. The
remaining 65 countries, mostly in sub-Saharan Africa,
have no reliable data on adult mortality. WHO uses
standard demographic techniques
36
to estimate all-cause
death rates by age for these populations. Cause-of-death
models are then used to estimate the total number of
deaths from cancer for countries with poor data. The
distribution of deaths from cancer by site is based on
regional incidence or mortality patterns from cancer
registries, which report to the International Agency for
Research on Cancer (IARC),
37
or on cancer survival
models when such data are unavailable. IARC also
provides estimates of cancer mortality by site for most
WHO member states. The implications of these
different approaches have been discussed previously,
including an assessment of their comparability.
32,33
The
net result is that the WHO estimates of global cancer
mortality are 11% higher than those of IARC, with larger
differences for SSA, SAR, and MENA.
Decades of biodmedical research in developed nations
have resulted in many effective interventions that affect
cancer incidence and mortality. Examples include
hepatitis B vaccine for liver cancer, screening methods
for cervical cancer,
38,39
faecal occult blood test for colo-
rectal cancer,
40–42
mammography for breast cancer,
43,44
and surgical prevention for those at high risk of
colorectal cancers.
45
Undoubtedly, increased coverage of
the above technologies, especially those that involve
early detection, would help reduce further the burden of
cancers. There has, however, been less success with
respect to other cancers: sputum cytology and chest
radiographs for lung cancer have not been promising,
and multiple chest radiographs might even be harmful;
46
and vaccines for H pylori and HPV are still under
investigation.
47,48
The efficacy of chemotherapy and
radiotherapy varies from cancer to cancer and depends
on multiple technical and biological factors, such as
stage of cancer.
Summarising the changes made in cancer therapy
over the past three decades, Sporn
49
states that the
obsession with curing advanced disease has prevented
progress in the war on cancer. Furthermore, preventive,
screening, and treatment interventions will only affect
population statistics if they are accessible and used,
factors that are highly dependent on cost and health-
system characteristics. These factors limit large-scale
application of these interventions in resource-poor
settings. These limitations further reinforce the
importance of our results for policies and programmes
that modify behavioural and environmental factors to
reduce the burden of cancers.
Contributors
M Ezzati, S Vander Hoorn, A D Lopez, and C J L Murray designed the
study. M Ezzati and S Vander Hoorn developed a framework and
methods for joint-effect analyses. S Vander Hoorn and G Danaei did the
statistical analysis. G Danaei, M Ezzati, and A D Lopez wrote the report.
The Comparative Risk Assessment collaborating group reviewed
scientific evidence and data sources for every risk factor and selected
and summarised data on exposure, outcomes, and hazard. M Ezzati
oversaw the research and the writing of this report.
Conflict of interest statement
We declare that we have no conflict of interest.
Articles
1792 www.thelancet.com Vol 366 November 19, 2005
Comparative Risk Assessment collaborating group (Cancers)
Core, methodology, statistical analysis, editorial and peer review, and
writing—M Ezzati, A Rodgers, A D Lopez, S Vander Hoorn,
C J L Murray.
Overweight and obesity (high BMI)—W P T James, R Jackson-Leach,
C Ni Mhurchu, E Kalamara, M Shayeghi, N J Rigby, C Nishida,
A Rodgers.
Low fruit and vegetable intake—K Lock, J Pomerleau, L Causer,
M McKee.
Physical inactivity—F C Bull, T Armstrong, T Dixon, S Ham, A Neiman,
M Pratt.
Smoking—M Ezzati, A D Lopez.
Alcohol use—J Rehm, R Room, M Monteiro, G Gmel, K Graham,
N Rehn, C T Sempos, U Frick, D Jernigan.
Unsafe sex—E Slaymaker, N Walker, B Zaba, M Collumbien.
Urban air pollution—A J Cohen, H R Anderson, B Ostro, K Dev Pandey,
M Krzyzanowski, N Künzli, K Gutschmidt, C A Pope, I Romieu,
J M Samet, K R Smith.
Indoor smoke from household use of solid fuels—K R Smith, S Mehta,
M Maeusezahl-Feuz.
Contaminated injections in health-care settings—A M Hauri,
G L Armstrong, Y J F Hutin.
Acknowledgments
This work was sponsored by the National Institute of Aging (grant PO1-
AG17625) and by the Disease Control Priorities Project. We thank
Ahmedin Jemal for valuable comments and references on US cancer
mortality trends.
References
1 Murray CJL, Lopez AD. The global burden of disease: a
comprehensive assessment of mortality and disability from
diseases, injuries, and risk factors in 1990 and projected to 2020.
Boston: Harvard School of Public Health, on behalf of the World
Health Organization and the World Bank, 1996.
2 WHO. World health report 2002: reducing risks, promoting healthy
life. Geneva: World Health Organization, 2002.
3 Hunink MG, Goldman L, Tosteson AN, et al. The recent decline in
mortality from coronary heart disease, 1980–1990: the effect of
secular trends in risk factors and treatment. JAMA 1997; 277: 535–42.
4 Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ.
Cancer statistics, 2003. C A Cancer J Clin 2003; 53: 5–26.
5 Chu KC, Tarone RE, Freeman HP. Trends in prostate cancer
mortality among black men and white men in the United States.
Cancer 2003; 97: 1507–16.
6 Weir HK, Thun MJ, Hankey BF, et al. Annual report to the nation
on the status of cancer, 1975–2000, featuring the uses of
surveillance data for cancer prevention and control.
J Natl Cancer Inst 2003; 95: 1276–99.
7 Shapiro S. Periodic screening for breast cancer: the HIP
randomized controlled trial. J Natl Cancer Inst Monogr 1997; 22:
27–30.
8 Early Breast Cancer Trialists’ Collaborative Group. Tamoxifen for
early breast cancer. Cochrane Database Syst Rev 2001; 1:
CD000486.
9 Ries LAG, Eisner MP, Kosary CL, et al, eds. SEER cancer statistics
review, 1975–2002. Bethesda: National Cancer Institute.
http://seer.cancer.gov/csr/1975_2002/ (accessed October, 2005).
10 Doll R. Are we winning the fight against cancer? An
epidemiological assessment. Eur J Cancer 1990; 26: 500–08.
11 Stewart BW, Kleihues P. World cancer report. Lyon: IARC press,
2003.
12 Dikshit RP, Kanhere S. Tobacco habits and risk of lung,
oropharyngeal and oral cavity cancer: a population-based case-
control study in Bhopal, India. Int J Epidemiol 2000; 29: 609–14.
13 Liu BQ, Peto R, Chen ZM, et al. Emerging tobacco hazards in
China: 1, retrospective proportional mortality study of one million
deaths. BMJ 1998; 317: 1411–22.
14 Hirayama T. Life-style and mortality: a large scale census-based
cohort study in Japan. Tokyo: Karger, 1990.
15 Madigan MP, Ziegler RG, Benichou J, Byrne C, Hoover RN.
Proportion of breast cancer cases in the United States explained
by well-established risk factors. J Natl Cancer Inst 1995; 87:
1681–85.
16 Mezzetti M, La Vecchia C, Decarli A, Boyle P, Talamini R,
Franceschi S. Population attributable risk for breast cancer: diet,
nutrition, and physical exercise. J Natl Cancer Inst 1998; 90: 389–94.
17 Engel LS, Chow WH, Vaughan TL, et al. Population attributable
risks of esophageal and gastric cancers. J Natl Cancer Inst 2003; 95:
1404–13.
18 Doll R, Peto R. The causes of cancer: quantitative estimates of
avoidable risks of cancer in the United States today.
J Natl Cancer Inst 1981; 66: 1191–308.
19 Pisani P, Parkin DM, Munoz N, Ferlay J. Cancer and infection:
estimates of the attributable fraction in 1990.
Cancer Epidemiol Biomarkers Prev 1997; 6: 387–400.
20 Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000:
the global picture. Eur J Cancer 2001; 37 (suppl 8): S4–66.
21 Ezzati M, Lopez AD, Rodgers A, Murray CJL. Comparative
quantification of health risks: global and regional burden of disease
attributable to selected major risk factors. Geneva: World Health
Organization, 2004.
22 Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJL.
Selected major risk factors and global and regional burden of
disease. Lancet 2002; 360: 1347–60.
23 James WPT, Jackson-Leach R, Mhurchu CN, et al. Overweight and
obesity (high body mass index). In: Ezzati M, Lopez AD,
Rodgers A, Murray CJ, eds. Comparative quantification of health
risks: global and regional burden of disease attributable to selected
major risk factors. Geneva: World Health Organization, 2004:
497–596.
24 Lock K, Pomerleau J, Causer L, McKee M. Low fruit and vegetable
consumption. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds.
Comparative quantification of health risks: global and regional
burden of disease attributable to selected major risk factors.
Geneva: World Health Organization, 2004: 597–728.
25 Bull FC, Armstrong TP, Dixon T, Ham S, Neiman A, Pratt M.
Physical inactivity. In: Ezzati M, Lopez AD, Rodgers A, Murray CJL,
eds. Comparative quantification of health risks: global and regional
burden of disease attributable to selected major risk factors.
Geneva: World Health Organization, 2004: 729–882.
26 Ezzati M, Lopez AD. Smoking and oral tobacco use. In: Ezzati M,
Lopez AD, Rodgers A, Murray CJL, eds. Comparative quantification
of health risks: global and regional burden of disease attributable to
selected major risk factors. Geneva: World Health Organization,
2004: 883–957.
27 Rehm J, Room R, Monteiro M, et al. Alcohol use. In: Ezzati M,
Lopez AD, Rodgers A, Murray CJL, eds. Comparative quantification
of health risks: global and regional burden of disease attributable to
selected major risk factors. Geneva: World Health Organization,
2004: 959–1108.
28 Slaymaker E, Walker N, Zaba B, Collumbien M. Unsafe sex. In:
Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative
quantification of health risks: global and regional burden of disease
attributable to selected major risk factors. Geneva: World Health
Organization, 2004: 1177–254.
29 Cohen AJ, Anderson HR, Ostro B, et al. Urban air pollution. In:
Ezzati M, Lopez AD, Rodgers A, Murray CJL, eds. Comparative
quantification of health risks: global and regional burden of disease
attributable to selected major risk factors. Geneva: World Health
Organization, 2004: 1353–433.
30 Smith KR, Mehta S, Maeusezahl-Feuz M. Indoor air pollution from
household use of solid fuels. In: Ezzati M, Lopez AD, Rodgers A,
Murray CJL, eds. Comparative quantification of health risks: global
and regional burden of disease attributable to selected major risk
factors. Geneva: World Health Organization, 2004: 1435–94.
31 Hauri AM, Armstrong GL, Hutin YJF. Contaminated injections in
health care settings. In: Ezzati M, Lopez AD, Rodgers A,
Murray CJL, eds. Comparative quantification of health risks:
global and regional burden of disease attributable to selected major
risk factors. Geneva: World Health Organization, 2004: 1803–49.
32 Mathers CD, Shibuya K, Boschi-Pinto C, Lopez AD, Murray CJL.
Global and regional estimates of cancer mortality and incidence by
site: I, application of regional cancer survival model to estimate
cancer mortality distribution by site. BMC Cancer 2002; 2: 36.
33 Shibuya K, Mathers CD, Boschi-Pinto C, Lopez AD, Murray CJL.
Global and regional estimates of cancer mortality and incidence by
site: II, results for the global burden of disease 2000.
BMC Cancer 2002; 2: 37.
Articles
www.thelancet.com Vol 366 November 19, 2005 1793
34 Ezzati M, Vander Hoorn S, Rodgers A, Lopez AD, Mathers CD,
Murray CJL. Estimates of global and regional potential health gains
from reducing multiple major risk factors. Lancet 2003; 362:
271–80.
35 Concha-Barrientos M, Nelson DI, Driscoll T, et al. Selected
occupational risk factors. In: Ezzati M, Lopez AD, Rodgers A,
Murray CJL, eds. Comparative quantification of health risks: global
and regional burden of disease attributable to selected major risk
factors. Geneva: World Health Organization, 2004:
1651–801.
36 Preston SH, Heuveline P, Guillot M. Demography: measuring and
modelling population processes. Oxford: Blackwell, 2005.
37 Ferlay J, Bray FI, Pisani P. Globocan 2000: cancer incidence,
mortality and prevalence worldwide, version 1.0. IARC Cancer
Base 5. Lyon: IARC, 2001.
38 Laara E, Day NE, Hakama M. Trends in mortality from cervical
cancer in the Nordic countries: association with organised
screening programmes. Lancet 1987; 1: 1247–49.
39 Gustafsson L, Ponten J, Zack M, Adami HO. International
incidence rates of invasive cervical cancer after introduction of
cytological screening. Cancer Causes Control 1997; 8: 755–63.
40 Mandel JS, Bond JH, Church TR, et al. Reducing mortality from
colorectal cancer by screening for fecal occult blood. N Engl J Med
1993; 328: 1365–71.
41 Hardcastle JD, Chamberlain JO, Robinson MH, et al. Randomised
controlled trial of faecal-occult-blood screening for colorectal
cancer. Lancet 1996; 348: 1472–77.
42 Kronborg O, Fenger C, Olsen J, Jorgensen OD, Sondergaard O.
Randomised study of screening for colorectal cancer with faecal-
occult-blood test. Lancet 1996; 348: 1467–71.
43 Hackshaw A. EUSOMA review of mammography screening.
Ann Oncol 2003; 14: 1193–95.
44 Freedman DA, Petitti DB, Robins JM. On the efficacy of screening
for breast cancer. Int J Epidemiol 2004; 33: 43–55.
45 Annie Yu HJ, Lin KM, Ota DM, Lynch HT. Hereditary
nonpolyposis colorectal cancer: preventive management.
Cancer Treat Rev 2003; 29: 461–70.
46 Manser RL, Irving LB, Stone C, Byrnes G, Abramson M,
Campbell D. Screening for lung cancer. Cochrane Database Syst Rev
2004; 1: CD001991.
47 Harper DM, Franco EL, Wheeler C, et al. Efficacy of a bivalent L1
virus-like particle vaccine in prevention of infection with human
papillomavirus types 16 and 18 in young women: a randomised
controlled trial. Lancet 2004; 364: 1757–65.
48 Villa LL, Costa RL, Petta CA, et al. Prophylactic quadrivalent
human papillomavirus (types 6, 11, 16, and 18) L1 virus-like
particle vaccine in young women: a randomised double-blind
placebo-controlled multicentre phase II efficacy trial. Lancet Oncol
2005; 6: 271–78.
49 Sporn MB. The war on cancer. Lancet 1996; 347: 1377–81.
    • "Cancer, a chronic disease associated with genetic mutations during the normal processes of cell division controlled by DNA [1]. Viruses, chemical carcinogens, chromosomal rearrangement, tumor suppressor genes, or spontaneous transformation have been implicated in the cause of cancer [2]. "
    [Show abstract] [Hide abstract] ABSTRACT: Isoquinoline alkaloidal content of Berberis vulgaris L. (Berberidaceae) is regarded as the main source of pharmacological activity. Studies showed that berberine (BB) possesses cytotoxic and hepatoprotective properties. The aim of the study was to isolate BB from B. vulgaris and semisynthesis of berberrubine (M1) and diberberine sulphate (Di-BB), investigate their efficiency as cytotoxic, antihepatitis C virus (HCV) agents and acetylcholinesterase enzyme (AChE) inhibitors. M1 and Di-BB were semisynthesized from BB previously isolated from Berberis vulgaris methanolic extract. Structures of the three compounds were elucidated on basis of 1H- and 13C-NMR and LC/MS. Cytotoxicity activity was studied using both normal peripheral blood mononuclear cells as control and three different proliferating cancerous cell lines; HepG-2, Caco-2 and MCF-7. The capability to inhibit viral replication was evidenced by the disappearance of viral RNA-amplified products detected by RT-PCR after incubation for 96 h. In vitro study of cytotoxic activities showed that, B. vulgaris extract was the most efficient on Caco-2 (IC50 3.836 μg/mL) and HepG-2 (IC50 5.55 μg/mL) cells lines, while, BB was the most potent on MCF-7 (IC50 4.433μg/mL) at 48 h. Replication of HCV RNA was inhibited at 100 μg/mL of each of B. vulgaris extract, BB and Di-BB. Finally, Di-BB significantly inhibited AChE activity (IC50: 100 μg/mL). In conclusion, the alkaloids studied possessed potent hepato-protective properties. Furthermore, the AChE inhibiting activity of Di-BB makes it a point of interest for further study on treatment of Alzheimer's disease.
    Full-text · Article · Sep 2016
    • "probability of 1 in 33 chances for a black South African male to develop esophageal cancer (Dandara et al., 2016). Major risk factors that are associated with cancers, including carcinoma of the esophagus, include tobacco smoking, alcohol consumption, nutritional deviancies, and environmental exposures (Danaei et al., 2005). Cancer control plans are needed to improve prevention and care, especially in developing countries, which are presenting with a growing burden. "
    [Show abstract] [Hide abstract] ABSTRACT: Medicinal plants are part of the healthcare systems worldwide, especially in low- and middle-income countries. African lettuce (Launaea taraxacifolia) is cultivated extensively in Africa, from Senegal in the west to Ethiopia and Tanzania in the east, and in Southern Africa. Potential anticancer effects of L. taraxacifolia have been suggested, but little is known about putative molecular mechanisms or potential for herb–drug interactions through inhibition or induction of drug-metabolizing enzymes. We investigated the effects of crude aqueous extracts of L. taraxacifolia on growth kinetics and cell cycle progression of the WHC01 esophageal cancer cells. Antiproliferative and apoptotic effects were evaluated using the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay and flow cytometry, while examining, in parallel, the genes regulating apoptosis and cell cycle in this cell culture model. In addition, we tested the inhibitory and enzyme kinetic effects of the aqueous L. taraxacifolia using recombinant human CYP450 isozyme model systems (CYP1A2, CYP2C9, and CYP2C19). L. taraxacifolia exhibited a significant growth inhibitory effect on the WHC01 cancer cells. Most cell cycle genes were downregulated. Cell cycle analysis showed a G0-G1 cell cycle arrest in WHC01 cells in the presence of L. taraxacifolia extract, accompanied by morphological changes. L. taraxacifolia extract treatment resulted in downregulation of expression levels of CYP1A2 (p < 0.0005) and CYP2C19 (p < 0.003) by 50–70%. L. taraxacifolia extract caused reversible and time-dependent inhibition of the recombinant CYP1A2, CYP2C9, and CYP2C19. This study provides new insights on possible anticancer effects of L. taraxacifolia, a widely used medicinal plant in parts of Africa and across the world especially by patients with cancer. Further mechanistic studies expanding on these observations would be timely and contribute to the field of global precision medicine that requires solid understanding of drug and herb molecular mechanisms of action and drug–herb interaction potentials, given the worldwide use of medicinal plants.
    Full-text · Article · Sep 2016
    • "Deaths from neoplasms currently account for around one-third of total age-standardized mortality rates. The causes of cancer are not yet fully understood, but it has been estimated that in high-income countries, smoking, alcohol use, and overweight and obesity were the most important causes at the turn of the century (Danaei et al. 2005). In terms of the Epidemiologic Transition, the Australian experience of neoplasms since 1990 is characteristic of the Age of Delayed Degenerative Diseases. "
    [Show abstract] [Hide abstract] ABSTRACT: Mortality change in Australia since 1907 is analysed in the light of Epidemiologic Transition theory. Australia began the twentieth century in the second age of the Epidemiologic Transition, the Age of Receding Pandemics. Australia probably moved to the third, the Age of Degenerative and Man-Made Diseases before 1946, which is slightly in advance of most Western countries. Transition to the fourth, the Age of Delayed Degenerative Diseases, is clearly marked by a downturn, in about 1970, in circulatory disease mortality, concurrent with other Western countries. Résumé La théorie de la transition épidémiologique sert de base pour une analyse des changements de mortalité en Australie depuis 1907. Au début du XXe siècle, l'Australie était dans la deuxième phase de la transition épidémiologique, celle du recul des pandémies. Néanmoins, l'Australie entrait probablement avant 1946 dans la troisième phase, celle des maladies dégénératives, ce qui est légèrement en avance sur la plupart des pays occidentaux. La transition vers la quatrième phase, celle des maladies dégénératives retardées, est clairement marqué par un ralentissement depuis environ 1970 dans la mortalité par maladies circulatoires, en même temps que chez d'autres pays occidentaux.
    Article · Jun 2016
Show more