Over one‐third of cancer cases and two‐fifths of cancer deaths in southern China are preventable: Insights from the latest representative population‐based cancer registry data and risk factor survey

Abstract

Assessing the impact of modifiable risk factors on cancer is crucial for prioritizing effective prevention interventions. This study aimed to quantify the cancer burden attributable to modifiable risk factors in Guangdong in 2019, thereby informing targeted prevention strategies tailored to the region. The population attributable fraction (PAF) was estimated by Levin's method, based on the 2010 prevalence of 15 risk factors and the latest representative relative risks. PAFs for specific age groups, sexes, risk factors, and cancer types were then combined to obtain overall PAF. Attributable cancer cases were calculated using cancer burden data in 2019 and PAF estimate. 34.4% of incident cancer and 44.5% of cancer deaths in Guangdong were attributable to 15 potentially modifiable risk factors, with higher PAF among males (44.3% for incidence and 52.0% for mortality) than females (22.9% for incidence cancer and 29.9% for mortality). Ten of the twenty‐two cancers exhibited an attributable fraction exceeding 50%, with five cancers exceeding 70%, including cancers of cervix uteri, nasopharynx, oral cavity and pharynx, liver and larynx. Behavioral factors contributed the most to incident cancer burden, followed by infectious, dietary, and metabolic factors. Nearly one‐third of incident cancer and two‐fifths of cancer deaths in Guangdong were preventable through addressing modifiable risk factors. The findings emphasize the critical need for targeted prevention strategies, particularly focusing on smoking, alcohol consumption, and infection control, to significantly reduce cancer burden. This study underscores the importance of integrating risk factor mitigation into public health policies to advance cancer prevention efforts in the region.

What's new?

This study quantified the cancer burden attributable to modifiable risk factors in Guangdong, the most populous and economically advanced province in southern China. By evaluating 15 modifiable risk factors, the study reveals that over one‐third of cancer cases and two‐fifths of cancer deaths in the region could potentially be prevented. These findings provide important evidence to guide local cancer control policies and highlight the urgent need for targeted prevention strategies, particularly focused on tobacco use, alcohol consumption, and infection‐related cancers. The results also serve as a reference for other rapidly developing regions with similar demographic and socioeconomic profiles.

Abbreviations

CI

confidence interval

CNKI

China National Knowledge Infrastructure

EBV

Epstein–Barr virus

GBD

Global Burden of Disease

GDCDC

Guangdong Provincial Centre for Disease Control and Prevention

HBV

Hepatitis B virus

HCV

Hepatitis C virus

HEFS

high in energy, fat, sugar, and salt

Hp

Helicobacter pylori

HPV

human papillomavirus

IARC

International Agency for Research on Cancer

NIP

National Immunization Programme

PAF

population attributable fraction

RR

relative risk

WCRF/AICR

World Cancer Research Fund/American Institute for Cancer Research

WHO

World Health Organization

1. INTRODUCTION

In 2022, there were approximately 20.0 million incident cancer cases and 9.7 million cancer‐related deaths worldwide. ¹ Cancer remains a leading cause of disease burden and a significant public health challenge globally. Notably, the cancer burden in China is particularly pronounced. China had the highest cancer incidence and mortality globally, with cancer‐related deaths accounting for almost a quarter of all deaths. ¹ , ² Primary prevention is recognized as the most effective strategy for reducing cancer burden. The World Health Organization (WHO) estimated that 30%–50% of cancers could be prevented by avoiding known risk factors. ³ However, there is a substantial number of risk factors associated with cancer, with 1045 carcinogens having been published by the International Agency for Research on Cancer (IARC) to date. ⁴ Consequently, it is imperative to comprehend the contribution of each individual risk factor to the overall cancer burden and to target key risk factors for precise prevention and control. The population attributable fraction (PAF), defined as the proportion of cases that would be prevented if the prevalence of a specific risk factor within the total population were reduced to an ideal reference level, is a valuable metric for quantifying the cancer burden attributable to risk factors. ⁵

Most published studies have focused on specific risk factors or specific cancer types, with fewer addressing the attributable burden for all cancer types. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ Moreover, studies on the Chinese population have predominantly demonstrated the overall national burden. ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ Two studies reported that 45.2% of cancer deaths were attributable to modifiable risk factors across China as a whole. ¹³ , ¹⁵ However, China is a vast country with considerable regional disparities in environmental characteristics and lifestyles. ¹⁶ Guangdong, in particular, stands out as the most economically developed province in China due to its proximity to Hong Kong and Macau, as well as its status as the first economic zone. Meanwhile, Guangdong has the largest and most diverse population in the country, including substantial internal migration and urban–rural population flows, resulting in a complex and dynamic demographic structure. Moreover, Guangdong is characterized by rapid urbanization, a high level of industrialization, and wide gaps in socioeconomic development between urban and rural areas. Due to its unique economic and demographic characteristics, risk factors associated with limited healthcare access as well as rapid social and economic change coexist in Guangdong, resulting in a cancer prevalence spectrum and risk factors that differ from the country as a whole. With the expansion of cancer registry coverage and the availability of representative risk factor survey data in Guangdong, there are now sufficient high‐quality data to estimate the cancer attributable risk. Furthermore, as the fastest‐growing and earliest economic demonstration area in China, the study of PAFs in Guangdong provides important insights for understanding the future cancer burden in the broader Chinese population, making such research essential. Nonetheless, there has been no relevant study focusing on Guangdong so far. Therefore, this study aims to estimate the cancer burden attributable to various risk factors in Guangdong in 2019, so as to prioritize the cancer interventions and provide a reference for the development of region‐specific, cancer evidence‐based prevention strategies.

2. MATERIALS AND METHODS

Based on the IARC Monograph series, the World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) reports, and systematic literature reviews, we identified 15 potentially modifiable risk factors for which there is sufficient or strong (convincing or probable) evidence of an association with cancer. ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ To enhance clarity and enable comparisons with other studies, we categorized the 15 risk factors into four categories based on the classification systems in similar research, ²³ including behavioral factors (smoking, second‐hand smoking, alcohol consumption, and physical inactivity), dietary factors (low fruit intake, low vegetable intake, and red meat consumption), metabolic factors (excess bodyweight and diabetes), and infectious agents (Hepatitis B virus [HBV], Hepatitis C virus [HCV], human papillomavirus [HPV], Helicobacter pylori [Hp], Epstein–Barr virus [EBV], and Clonorchis sinensis). Cancer types that are clearly associated with the above risk factors include cancers of the oral cavity, pharynx, nasopharynx (C00‐C14), esophagus (C15), stomach (C16), colon (C18), rectum (C19‐C20), liver (C22), gallbladder (C23‐C24), pancreas (C25), larynx (C32), lung, bronchus, and trachea (C33‐C34), breast (C50), vulva, vagina (C51‐C52), cervix uteri (C53), corpus uteri (C54), ovary (C56), penis (C60), prostate (C61), kidney (C64), urinary bladder (C67), thyroid (C73), Hodgkin lymphoma (C81), and myeloid leukemia (C92‐C94). A full list of risk factors and related cancer sites is shown in Table S1, Supporting Information. When estimating the cancer attributable risk, we assumed a lag time of approximately 10 years between exposure and cancer development. ¹¹ , ²¹ , ²²

Regarding the relative risk (RR), we obtained data for the Chinese population from published high‐quality meta‐analyses or national cohort studies. When such data were unavailable, we used estimates from Asian or global studies. Sex‐specific RRs were used where available. The RRs applied in the current analysis are presented in Table S2.

2.1. Exposure prevalence data

We obtained the age‐, sex‐, and region‐specific prevalence data for risk factors (excluding infectious agents) from the 2010 Chronic Disease and Risk Factor Surveillance among Adults in Guangdong (detailed in methods section of Data S1). This surveillance employed multistage stratified random cluster sampling, with sample weighting and post‐stratified adjustments to ensure the representativeness of the sample population at the provincial level.

We prioritized using the above representative population‐based exposure rates; however, as the prevalence of infectious agents was not included in the surveillance, these data were derived from meta‐analyses. When prevalence data for Guangdong around 2010 (within ±5 years) were available in the published literature, we preferred to use these sources (including the prevalence of HBV, HCV, Hp, and HPV). ²³ , ²⁴ , ²⁵ Otherwise, we conducted systematic reviews and meta‐analyses of articles published in PubMed, Embase, China National Knowledge Infrastructure (CNKI), and Wanfang. Because the prevalence data of these infectious agents in the general population was not available, we obtained infection rates of EBV in patients with nasopharyngeal cancer (Figure S1) and Hodgkin's lymphoma (Figure S2), as well as HPV in patients with cancers of the oral cavity, pharynx (Figure S3), larynx (Figure S4), vulvar, vaginal (Figure S5), and penis (Figure S6).

2.2. Burden of disease data

The province‐level estimates of cancer burden data in 2019 were obtained from the Guangdong Provincial Centre for Disease Control and Prevention (GDCDC), based on 40 cancer registries covering 43.1% of the province's population (including 19 of 21 prefecture‐level cities, with a total population of 42,757,500). Specifically, the registries‐based incidence and mortality of cancer stratified by sex (male, female), region (urban, rural), cancer site (classified according to the ICD‐10), and age group (20–29, 30–39, 40–49, 50–59, 60–69, and 70 years and above) were extrapolated using sex‐specific, region‐specific, and age‐specific population data for the entire province to estimate the total number of incident cancer cases and cancer deaths in each category. Next, the total number of incident cancer cases and cancer deaths in males and females was calculated by summing cases across all age groups and cancer sites. When analyzing the PAF and attributable cases of a specific cancer type, we only included the cancer types for which there was sufficient or strong evidence of association with the selected risk factors. However, when estimating the PAF and attributable cases for a specific risk factor, all cancers were considered.

2.3. Attributable burden estimation

For categorical variables including behavioral factors, metabolic factors and several infectious agents with available exposure data in general population (HBV, HCV, Hp, HPV [cervical cancer]), PAFs were calculated by Levin's method from the following formula ²⁶ : $\mathrm{PAF}=\frac{{\sum }_{i=1}^n{P}_i\left({\mathrm{RR}}_i-1\right)}{{\sum }_{i=1}^n{P}_i\left({\mathrm{RR}}_i-1\right)+1}$. For continuous variables including dietary factors, PAFs were estimated by the average risk method (modified Levin method) as follows ²⁷ , ²⁸ : $\mathrm{PAF}=\frac{{\exp }^{\left[\mathrm{Ln}\left({\text{Risk}}_{\mathrm{per}\text{unit}}\right)*\text{average level of exposure}\right]}-1}{{\exp }^{\left[\mathrm{Ln}\left({\text{Risk}}_{\mathrm{per}\text{unit}}\right)*\text{average level of exposure}\right]}}$. For infectious agents where representative population prevalence could not be obtained due to insufficient data (HPV [except cervix uteri cancer] and EBV), we used the infection rate in cancer cases ($P_c$) to calculate the cancer burden of the infectious agent retrospectively ²⁹ : $\mathrm{PAF}=P_c\frac{\mathrm{RR}-1}{\mathrm{RR}}$. The 95% confidence intervals (95% CI) for PAFs were estimated based on the delta method from the following formula: $\mathrm{Var}\left(\mathrm{PAF}\right)=\frac{{\left[\exp \left(\ln \mathrm{RR}\right)-1\right]}^2*\mathrm{Var}\left(P\right)+{\left[P*\exp \left(\ln \mathrm{RR}\right)\right]}^2*\mathrm{Var}\left(\ln \mathrm{RR}\right)}{{\left(P*\left[\exp \left(\ln \mathrm{RR}\right)-1\right]+1\right)}^4}$ (detailed in methods section of Data S1).

Subsequently, the PAFs were multiplied by the number of incident cancer cases (or deaths) in each category across the whole province to obtain the attributable incident cases (or deaths) for specific sexes, regions, and age groups. The attributable cases in each age group were summed to obtain the attributable cases for the entire age group, which were then divided by the total cases to obtain the PAF for the whole age group. The PAFs for all sexes and regions were calculated using the same method. We assumed that risk factors are independent except for diabetes and excess bodyweight. To calculate the combined PAF for diabetes and excess bodyweight, we adopted the conservative method proposed by Pearson‐Stuttard and colleagues. ³⁰ A combined PAF for all risk factors by cancer site was calculated according to the following formula: ${\mathrm{PAF}}_{\text{overall}}=1-{\prod }_{r-1}^{\mathrm{R}}\left(1-{\mathrm{PAF}}_r\right)$. The 95% CI of ${\mathrm{PAF}}_{\text{overall}}$ was calculated by a bootstrap simulation method with 5000 times.

All statistical analyses were conducted using R software (version 4.4.2).

3. RESULTS

3.1. Overall cancer attributable risk

In 2019, it was estimated that 34.4% of overall incident cancer cases (90,778 of 263,821) were attributable to potentially modifiable risk factors among adults aged 20 years and older in Guangdong. The corresponding proportion was 44.3% in males and 22.9% in females, as well as 34.1% in urban and 34.7% in rural areas. For each individual cancer type, the PAF ranged from 96.9% for cervical cancer to 3.5% for thyroid cancer. Among the 22 cancers, 10 exhibited an attributable fraction that exceeded 50%, with 5 cancers exceeding 70%, including the cancers of cervix uteri, nasopharynx (87.7%), oral cavity and pharynx (73.0%), liver (70.5%), and larynx (70.4%). The cancer type with the largest attributable cases was lung cancer (24,708), followed by liver cancer (19,352), nasopharyngeal cancer (9967), colorectal cancer (7907), and cervical cancer (6624). Among males, the cancer type with the largest attributable cases was lung cancer (19,324), while among females it was cervical cancer (6624) (Figures 1 and S8–S10 and Tables S3 and S4).

FIGURE 1.

Number and proportion of each cancer site (individually and all combined) attributable to studied risk factors in Guangdong, 2019. (A) incident cancer; (B) cancer death. PAF, population attributable fraction.

As for cancer deaths, it was estimated that 44.5% (57,178 of 128,522) were attributable to potentially modifiable risk factors. By cancer type, the cancers with the largest attributable cases were lung cancer (18,145), liver cancer (16,958), nasopharyngeal cancer (5193), stomach cancer (4031), colorectal cancer (3242), and esophageal cancer (3062) (Figure 1 and Tables S3 and S4).

3.2. Cancer attributable to specific risk factors

Among the four categories of potentially modifiable risk factors, behavioral factors caused the highest proportion of cancer burden (16.9% of incident cases and 23.8% of deaths), followed by infectious agents (12.6% of incident cases and 16.0% of deaths), dietary factors (7.6% of incident cases and 9.9% of deaths), and metabolic factors (7.0% of incident cases and 7.8% of deaths). A similar profile was observed for males; however, infectious agents constituted the highest category for cancer in females. No significant differences were found between urban and rural areas (Figures 2 and S11).

FIGURE 2.

Proportion of incident cancer and cancer death attributable to studied risk factors (divided by four categories) in Guangdong, 2019. (A) incident cancer; (B) cancer death.

By individual risk factor, the greatest attributable proportions of incident cancer were for smoking (18.8%), alcohol (9.1%), HBV (7.0%) in males and HPV (6.0%), diabetes (4.1%), low fruit intake (3.3%) in females. These factors also had the greatest PAFs for cancer death in males; however, low fruit intake was the greatest risk factor in females. As with the overall situation, the top five risk factors for cancer incidence and deaths in both urban and rural areas were smoking, alcohol, diabetes, HBV, and low fruit intake (Tables 1, 2, S5, and S6 and Figures S12–S14).

TABLE 1.

Population attributable fractions (PAF, %) of incident cancer and cancer death by risk factor in Guangdong in 2019.

Site	Total		Male		Female
	Incidence cancer	Cancer death	Incidence cancer	Cancer death	Incidence cancer	Cancer death
Smoking
All	10.6 (7.8, 13.5)	15.5 (11.5, 19.6)	18.8 (13.8, 23.7)	22.5 (16.7, 28.2)	1.1 (0.8, 1.6)	2.1 (1.4, 3.0)
Oral cavity, pharynx	23.9 (15.2, 32.6)	24.4 (15.4, 33.5)	32.2 (21.1, 43.3)	31.2 (20.3, 42.1)	2.4 (0.0, 5.2)	3.0 (0.0, 6.3)
Nasopharynx	9.6 (2.2, 17.0)	9.4 (2.1, 16.6)	12.8 (3.0, 22.6)	12.2 (2.8, 21.7)	0.5 (0.1, 0.9)	0.7 (0.1, 1.3)
Larynx	29.2 (18.9, 39.6)	27.7 (17.7, 37.7)	31.0 (20.1, 41.9)	29.8 (19.2, 40.4)	2.9 (0.0, 6.1)	3.5 (0.0, 7.4)
Lung, bronchus and trachea	34.2 (30.3, 38.2)	37.0 (32.7, 41.3)	49.9 (44.4, 55.4)	48.9 (43.4, 54.4)	5.8 (4.8, 6.9)	7.0 (5.7, 8.2)
Esophagus	14.9 (8.6, 21.7)	15.2 (8.7, 22.0)	18.6 (10.8, 26.5)	18.5 (10.7, 26.2)	0.9 (0.0, 3.4)	1.0 (0.0, 3.9)
Stomach	9.4 (4.7, 14.3)	8.9 (4.4, 13.7)	13.8 (7.1, 20.6)	13.3 (6.8, 19.9)	0.6 (0.0, 1.9)	0.6 (0.0, 2.2)
Colorectum	4.8 (3.2, 6.4)	4.5 (3.0, 6.1)	7.7 (5.2, 10.3)	7.3 (4.9, 9.7)	0.5 (0.3, 0.7)	0.7 (0.4, 0.9)
Pancreas	11.2 (2.0, 20.3)	11.0 (2.0, 20.1)	18.3 (3.5, 33.1)	18.0 (3.4, 32.6)	1.5 (0.0, 3.0)	1.7 (0.0, 3.4)
Liver	11.9 (5.1, 18.7)	11.6 (5.0, 18.3)	14.2 (6.3, 22.1)	14.0 (6.2, 21.8)	1.4 (0.0, 3.1)	1.6 (0.0, 3.4)
Kidney	8.0 (2.7, 13.2)	7.9 (2.6, 13.1)	11.9 (4.1, 19.6)	10.7 (3.6, 17.7)	0.6 (0.2, 1.0)	0.8 (0.2, 1.4)
Urinary bladder	15.4 (9.8, 20.9)	14.0 (8.8, 19.2)	18.3 (11.7, 24.9)	16.9 (10.7, 23.1)	1.7 (0.9, 2.4)	2.4 (1.3, 3.5)
Myeloid leukemia	7.2 (4.8, 9.6)	7.2 (4.8, 9.7)	12.0 (8.1, 16.0)	11.5 (7.7, 15.3)	0.8 (0.5, 1.1)	0.9 (0.5, 1.2)
Cervix uteri	1.2 (0.4, 1.9)	1.5 (0.5, 2.4)	/	/	1.2 (0.4, 1.9)	1.5 (0.5, 2.4)
Ovary	0.5 (0.0, 1.1)	0.7 (0.0, 1.4)	/	/	0.5 (0.0, 1.1)	0.7 (0.0, 1.4)
Second‐hand smoking
All	1.2 (0.7, 1.7)	1.5 (0.8, 2.2)	0.8 (0.2, 1.4)	1.0 (0.2, 1.8)	1.7 (1.4, 2.1)	2.4 (1.9, 2.9)
Lung, bronchus and trachea	6.5 (3.9, 9.2)	5.5 (2.9, 8.1)	3.4 (0.8, 6.1)	3.5 (0.8, 6.1)	12.2 (9.5, 14.8)	10.6 (8.3, 13.0)
Alcohol
All	5.7 (2.7, 9.0)	7.9 (4.0, 11.9)	9.1 (4.6, 13.6)	10.7 (5.6, 15.8)	1.8 (0.5, 3.6)	2.5 (0.9, 4.4)
Oral cavity, pharynx	30.0 (22.4, 37.6)	29.4 (21.9, 37.0)	35.4 (26.8, 44.1)	34.1 (25.6, 42.6)	16.0 (11.0, 21.0)	14.8 (10.1, 19.5)
Nasopharynx	11.7 (3.6, 19.7)	10.7 (3.2, 18.2)	13.9 (4.4, 23.4)	12.7 (3.9, 21.5)	5.4 (1.4, 9.4)	4.7 (1.2, 8.2)
Larynx	24.4 (16.3, 32.5)	23.2 (15.3, 31.0)	25.3 (16.9, 33.6)	24.4 (16.2, 32.5)	12.0 (7.4, 16.6)	9.4 (5.7, 13.1)
Esophagus	24.5 (18.0, 31.0)	24.5 (18.0, 31.0)	28.3 (20.9, 35.7)	28.0 (20.7, 35.4)	10.1 (6.8, 13.3)	9.4 (6.3, 12.5)
Stomach	14.0 (2.8, 25.1)	13.3 (2.6, 24.0)	17.5 (3.8, 31.2)	16.9 (3.6, 30.1)	7.0 (0.9, 13.1)	6.6 (0.8, 12.4)
Colorectum	5.8 (0.7, 10.9)	5.4 (0.6, 10.1)	7.8 (1.0, 14.6)	7.3 (0.9, 13.7)	3.0 (0.3, 5.7)	2.6 (0.2, 4.9)
Liver	23.7 (13.2, 34.2)	23.0 (12.8, 33.3)	26.7 (15.0, 38.3)	26.2 (14.7, 37.7)	10.4 (4.9, 15.9)	9.6 (4.5, 14.8)
Breast	2.8 (0.0, 8.1)	2.6 (0.0, 7.6)	/	/	2.8 (0.0, 8.1)	2.6 (0.0, 7.6)
Physical inactivity
All	0.5 (0.2, 0.7)	0.3 (0.1, 0.5)	0.3 (0.2, 0.5)	0.2 (0.1, 0.4)	0.6 (0.2, 1.0)	0.5 (0.1, 0.9)
Colorectum	2.5 (1.0, 4.0)	2.6 (1.1, 4.2)	2.8 (1.5, 4.2)	2.9 (1.5, 4.2)	2.0 (0.4, 3.7)	2.3 (0.4, 4.2)
Corpus uteri	2.7 (0.6, 4.7)	3.5 (0.8, 6.2)	/	/	2.7 (0.6, 4.7)	3.5 (0.8, 6.2)
Breast	1.5 (0.6, 2.4)	1.8 (0.8, 2.9)	/	/	1.5 (0.6, 2.4)	1.8 (0.8, 2.9)
Low vegetable intake
All	0.7 (0.0, 2.3)	0.7 (0.0, 2.4)	1.0 (0.0, 3.3)	0.9 (0.0, 2.9)	0.4 (0.0, 1.2)	0.5 (0.0, 1.6)
Oral cavity, pharynx	10.7 (0.0, 35.3)	11.4 (0.0, 37.0)	10.5 (0.0, 35.1)	11.0 (0.0, 36.5)	11.2 (0.0, 35.9)	12.6 (0.0, 38.6)
Nasopharynx	10.1 (0.0, 33.9)	10.0 (0.0, 33.3)	10.4 (0.0, 35.1)	10.0 (0.0, 33.8)	9.2 (0.0, 30.5)	9.8 (0.0, 31.5)
Larynx	11.4 (0.0, 37.3)	11.8 (0.0, 38.4)	10.9 (0.0, 36.2)	11.3 (0.0, 37.0)	17.5 (0.0, 52.4)	17.6 (0.0, 53.8)
Low fruit intake
All	4.7 (2.1, 8.3)	6.5 (3.2, 10.8)	5.9 (2.6, 10.7)	7.1 (3.5, 12.1)	3.3 (1.6, 5.5)	5.1 (2.6, 8.3)
Oral cavity, pharynx	17.0 (0.0, 49.9)	17.2 (0.0, 50.0)	17.1 (0.0, 50.0)	17.3 (0.0, 50.1)	16.7 (0.0, 49.5)	17.1 (0.0, 49.8)
Nasopharynx	16.6 (0.0, 49.6)	17.1 (0.0, 49.9)	16.8 (0.0, 49.8)	17.2 (0.0, 50.1)	16.0 (0.0, 48.8)	16.9 (0.0, 49.6)
Larynx	17.2 (0.0, 50.1)	17.4 (0.0, 50.2)	17.3 (0.0, 50.1)	17.4 (0.0, 50.2)	16.2 (0.0, 49.1)	17.3 (0.0, 49.9)
Lung, bronchus and trachea	19.1 (11.5, 26.7)	19.3 (11.7, 26.9)	19.3 (11.7, 26.9)	19.4 (11.8, 26.9)	18.7 (11.1, 26.4)	19.1 (11.5, 26.7)
Red meat
All	2.3 (0.0, 5.9)	2.8 (0.0, 7.1)	3.1 (0.0, 7.7)	3.2 (0.0, 7.8)	1.5 (0.0, 3.8)	2.2 (0.0, 5.6)
Esophagus	23.4 (0.0, 61.5)	23.2 (0.0, 61.3)	24.5 (0.0, 63.8)	24.4 (0.0, 63.6)	19.0 (0.0, 53.0)	18.3 (0.0, 51.5)
Stomach	17.7 (0.6, 35.1)	17.2 (0.5, 34.4)	18.8 (0.8, 36.9)	18.5 (0.7, 36.4)	15.4 (0.1, 31.3)	15.0 (0.1, 30.5)
Colorectum	9.0 (0.0, 25.9)	8.6 (0.0, 25.0)	9.7 (0.0, 27.8)	9.5 (0.0, 27.2)	7.9 (0.0, 23.1)	7.4 (0.0, 21.7)
Excess bodyweight
All	2.5 (1.2, 4.2)	2.3 (1.1, 3.7)	2.3 (1.1, 3.8)	2.2 (1.0, 3.6)	2.8 (1.4, 4.6)	2.5 (1.2, 3.9)
Esophagus	9.2 (4.9, 13.6)	9.2 (4.8, 13.5)	9.5 (5.0, 14.0)	9.4 (4.9, 13.8)	8.4 (4.4, 12.4)	8.3 (4.3, 12.4)
Stomach	6.9 (3.9, 10.0)	6.7 (3.7, 9.7)	7.0 (3.9, 10.2)	6.7 (3.8, 9.7)	6.8 (3.8, 9.8)	6.7 (3.7, 9.7)
Colorectum	5.7 (4.6, 6.9)	5.4 (4.4, 6.5)	5.9 (4.7, 7.0)	5.5 (4.4, 6.6)	5.6 (4.5, 6.7)	5.3 (4.3, 6.4)
Pancreas	3.1 (0.8, 5.4)	3.2 (0.8, 5.6)	3.2 (0.8, 5.5)	3.3 (0.9, 5.8)	3.0 (0.8, 5.3)	3.0 (0.8, 5.3)
Liver	2.8 (0.1, 5.6)	2.8 (0.1, 5.4)	2.9 (0.1, 5.7)	2.8 (0.1, 5.6)	2.6 (0.1, 5.1)	2.5 (0.1, 4.9)
Gallbladder	3.3 (0.9, 5.7)	3.3 (0.9, 5.6)	3.4 (1.0, 5.9)	3.4 (1.0, 5.8)	3.2 (0.9, 5.4)	3.2 (0.9, 5.5)
Kidney	9.8 (8.8, 10.9)	9.2 (8.2, 10.2)	10.2 (9.1, 11.3)	9.2 (8.2, 10.2)	9.1 (8.1, 10.1)	9.1 (8.1, 10.1)
Thyroid	2.4 (0.0, 7.4)	2.7 (0.0, 8.2)	2.8 (0.0, 8.5)	2.8 (0.0, 8.7)	2.3 (0.0, 7.0)	2.6 (0.0, 7.9)
Corpus uteri	17.0 (10.8, 23.2)	16.5 (10.5, 22.6)	/	/	17.0 (10.8, 23.2)	16.5 (10.5, 22.6)
Ovary	10.0 (0.0, 21.1)	10.3 (0.0, 21.7)	/	/	10.0 (0.0, 21.1)	10.3 (0.0, 21.7)
Breast	2.0 (0.9, 3.2)	2.0 (0.9, 3.2)	/	/	2.0 (0.9, 3.2)	2.0 (0.9, 3.2)
Prostate	1.1 (0.0, 8.8)	1.1 (0.0, 8.3)	1.1 (0.0, 8.8)	1.1 (0.0, 8.3)	/	/
Diabetes
All	4.7 (2.6, 6.9)	5.7 (3.6, 7.7)	5.3 (3.3, 7.4)	6.1 (4.2, 8.1)	4.1 (1.9, 6.3)	4.8 (2.6, 7.0)
Esophagus	4.0 (2.7, 5.2)	4.0 (2.8, 5.2)	4.0 (2.8, 5.2)	4.0 (2.8, 5.2)	3.9 (2.7, 5.0)	3.9 (2.7, 5.1)
Stomach	3.1 (0.5, 5.8)	3.2 (0.5, 5.9)	2.2 (0.2, 4.2)	2.2 (0.2, 4.2)	5.0 (1.0, 8.9)	5.1 (1.0, 9.2)
Colorectum	7.5 (3.9, 11.2)	7.6 (4.0, 11.3)	7.6 (4.1, 11.2)	7.7 (4.1, 11.2)	7.4 (3.6, 11.1)	7.6 (3.8, 11.5)
Pancreas	9.5 (4.6, 14.3)	9.8 (4.8, 14.8)	12.3 (7.7, 17.0)	12.8 (8.0, 17.7)	5.6 (0.5, 10.6)	5.8 (0.5, 11.0)
Liver	19.0 (14.8, 23.1)	19.0 (14.8, 23.2)	19.0 (14.8, 23.2)	19.1 (14.8, 23.3)	18.7 (14.6, 22.9)	18.8 (14.6, 23.0)
Gallbladder	7.7 (1.1, 14.3)	7.9 (1.2, 14.7)	8.2 (1.2, 15.2)	8.4 (1.2, 15.5)	7.2 (1.0, 13.4)	7.5 (1.1, 13.8)
Kidney	8.1 (4.6, 11.7)	8.8 (5.0, 12.6)	8.2 (4.5, 11.9)	8.6 (4.7, 12.5)	8.0 (4.8, 11.2)	9.3 (5.6, 13.0)
Urinary bladder	4.3 (2.3, 6.3)	4.3 (2.2, 6.3)	4.4 (2.3, 6.4)	4.3 (2.3, 6.4)	4.0 (2.1, 5.9)	4.1 (2.2, 6.1)
Thyroid	3.3 (1.9, 4.8)	5.8 (3.3, 8.3)	3.3 (1.5, 5.1)	5.0 (2.4, 7.7)	3.3 (2.0, 4.7)	6.3 (3.9, 8.8)
Corpus uteri	18.6 (5.3, 31.8)	21.6 (6.6, 36.5)	/	/	18.6 (5.3, 31.8)	21.6 (6.6, 36.5)
Ovary	5.7 (2.6, 8.9)	6.7 (3.1, 10.4)	/	/	5.7 (2.6, 8.9)	6.7 (3.1, 10.4)
Breast	4.2 (1.2, 7.2)	4.9 (1.4, 8.4)	/	/	4.2 (1.2, 7.2)	4.9 (1.4, 8.4)
Prostate	10.3 (0.0, 21.3)	10.2 (0.0, 21.1)	10.3 (0.0, 21.3)	10.2 (0.0, 21.1)	/	/
Epstein–Barr virus
All	3.5 (3.3, 3.6)	3.7 (3.5, 3.8)	4.7 (4.5, 4.9)	4.2 (4.0, 4.4)	2.0 (1.9, 2.1)	2.7 (2.6, 2.8)
Nasopharynx	79.4 (75.8, 82.6)	79.4 (75.8, 82.5)	79.4 (75.8, 82.6)	79.4 (75.8, 82.6)	79.4 (75.8, 82.5)	79.4 (75.8, 82.5)
Hodgkin lymphoma	51.5 (44.3, 58.7)	51.5 (44.3, 58.7)	51.5 (44.3, 58.7)	51.5 (44.3, 58.7)	51.5 (44.3, 58.7)	51.5 (44.3, 58.7)
Human papillomavirus
All	3.4 (3.1, 3.6)	2.2 (2.0, 2.5)	1.1 (0.8, 1.5)	0.9 (0.6, 1.2)	6.0 (5.8, 6.1)	4.8 (4.7, 5.0)
Oral cavity, pharynx	31.7 (24.7, 39.6)	31.7 (24.7, 39.6)	31.7 (24.7, 39.6)	31.7 (24.7, 39.6)	31.7 (24.7, 39.6)	31.7 (24.7, 39.6)
Larynx	24.7 (15.0, 37.8)	24.7 (15.0, 37.8)	24.7 (15.0, 37.8)	24.7 (15.0, 37.8)	24.7 (15.0, 37.8)	24.7 (15.0, 37.8)
Cervix uteri	96.8 (96.3, 97.4)	96.8 (96.3, 97.4)	/	/	96.8 (96.3, 97.4)	96.8 (96.3, 97.4)
Vulva, vagina	58.7 (51.1, 65.9)	58.7 (51.1, 66.0)	/	/	58.7 (51.1, 65.9)	58.7 (51.1, 66.0)
Penis	44.2 (35.3, 53.4)	44.2 (35.4, 53.5)	44.2 (35.3, 53.4)	44.2 (35.4, 53.5)	/	/
Helicobacter pylori
All	1.0 (0.5, 1.6)	1.5 (0.7, 2.2)	1.3 (0.6, 1.9)	1.5 (0.7, 2.2)	0.8 (0.4, 1.1)	1.5 (0.7, 2.3)
Stomach	25.9 (12.8, 39.0)	25.9 (12.8, 39.0)	25.9 (12.8, 39.0)	25.9 (12.8, 39.0)	25.9 (12.8, 39.0)	25.9 (12.8, 39.0)
Hepatitis B virus
All	4.6 (4.3, 4.9)	8.3 (7.8, 8.9)	7.0 (6.6, 7.5)	10.2 (9.6, 10.9)	1.8 (1.7, 1.9)	4.7 (4.4, 5.0)
Liver	44.4 (41.5, 47.2)	44.4 (41.5, 47.2)	44.4 (41.5, 47.2)	44.4 (41.5, 47.2)	44.4 (41.5, 47.2)	44.4 (41.5, 47.2)
Hepatitis C virus
All	0.1 (0.1, 0.2)	0.3 (0.2, 0.3)	0.2 (0.1, 0.3)	0.3 (0.2, 0.4)	0.1 (0.0, 0.1)	0.1 (0.1, 0.2)
Liver	1.4 (0.9, 1.9)	1.4 (0.9, 1.9)	1.4 (0.9, 1.9)	1.4 (0.9, 1.9)	1.4 (0.9, 1.9)	1.4 (0.9, 1.9)
Clonorchis sinensis
All	0.1 (0.0, 0.3)	0.2 (0.0, 0.5)	0.2 (0.0, 0.4)	0.3 (0.0, 0.6)	0.0 (0.0, 0.1)	0.1 (0.0, 0.3)
Liver	1.2 (0.0, 2.7)	1.2 (0.0, 2.7)	1.2 (0.0, 2.7)	1.2 (0.0, 2.7)	1.2 (0.0, 2.7)	1.2 (0.0, 2.7)

Note: PAF refers to population attributable fraction; “All” refers to PAF of the specific risk factor in total cancer.

TABLE 2.

Attributable cases of incident cancer and cancer death by risk factor in Guangdong in 2019.

Site	Total		Male		Female
	Incidence cancer	Cancer death	Incidence cancer	Cancer death	Incidence cancer	Cancer death
Smoking
All	28,001 (20,476, 35,595)	19,960 (14,770, 25,208)	26,610 (19,549, 33,671)	19,029 (14,157, 23,901)	1391 (926, 1924)	931 (612, 1307)
Oral cavity, pharynx	1110 (706, 1517)	459 (289, 629)	1078 (706, 1450)	445 (289, 601)	31 (0, 67)	13 (0, 28)
Nasopharynx	1089 (250, 1927)	555 (125, 985)	1074 (248, 1900)	545 (124, 965)	14 (1, 26)	10 (1, 19)
Larynx	518 (334, 702)	335 (214, 457)	514 (334, 695)	332 (214, 450)	3 (0, 6)	3 (0, 7)
Lung, bronchus and trachea	16,824 (14,896, 18,752)	12,921 (11,422, 14,420)	15,809 (14,062, 17,556)	12,230 (10,853, 13,606)	1015 (834, 1196)	691 (569, 813)
Esophagus	1017 (584, 1474)	848 (486, 1229)	1004 (584, 1425)	837 (486, 1188)	12 (0,48)	10 (0,40)
Stomach	991 (498, 1512)	657 (327, 1011)	971 (498, 1445)	641 (327, 955)	19 (0,67)	16 (0,55)
Colorectum	1420 (950, 1891)	568 (379, 757)	1356 (908, 1803)	533 (356, 710)	64 (41, 87)	34 (22, 47)
Pancreas	416 (75, 756)	349 (62, 637)	391 (75, 708)	326 (62, 591)	24 (0, 47)	23 (0, 46)
Liver	3258 (1404, 5123)	2804 (1200, 4420)	3186 (1404, 4969)	2730 (1200, 4259)	71 (0, 154)	74 (0, 160)
Kidney	222 (76, 367)	65 (22, 109)	216 (74, 357)	63 (21, 105)	5 (1, 9)	1 (0, 3)
Urinary bladder	645 (410, 880)	231 (145, 317)	632 (403, 862)	223 (140, 305)	12 (6, 17)	8 (4, 11)
Myeloid leukemia	390 (260, 520)	125 (83, 168)	372 (249, 496)	119 (79, 159)	17 (11, 24)	6 (3, 8)
Cervix uteri	79 (28, 130)	28 (10, 46)	/	/	79 (28, 130)	28 (10, 46)
Ovary	18 (0, 37)	8 (0, 18)	/	/	18 (0, 37)	8 (0, 18)
Second‐hand smoking
All	3212 (1924, 4501)	1920 (1026, 2815)	1089 (260, 1919)	864 (202, 1526)	2122 (1663, 2582)	1056 (824, 1288)
Lung, bronchus and trachea	3212 (1924, 4501)	1920 (1026, 2815)	1089 (260, 1919)	864 (202, 1526)	2122 (1663, 2582)	1056 (824, 1288)
Alcohol
All	15,114 (7096, 23,640)	10,167 (5152, 15,273)	12,878 (6497, 19,258)	9067 (4775, 13,359)	2235 (598, 4382)	1099 (376, 1914)
Oral cavity, pharynx	1395 (1040, 1749)	554 (411, 696)	1186 (896, 1475)	486 (365, 608)	208 (143, 274)	67 (45, 88)
Nasopharynx	1324 (407, 2241)	636 (191, 1080)	1164 (367, 1962)	567 (174, 960)	159 (40, 279)	68 (16, 120)
Larynx	432 (288, 577)	281 (186, 375)	419 (280, 558)	271 (180, 363)	13 (8, 18)	9 (5, 12)
Esophagus	1666 (1222, 2110)	1370 (1005, 1735)	1523 (1126, 1921)	1270 (938, 1603)	142 (96, 189)	99 (66, 132)
Stomach	1479 (301, 2656)	979 (193, 1766)	1230 (270, 2191)	810 (172, 1448)	248 (31, 465)	169 (20, 318)
Colorectum	1726 (213, 3239)	671 (80, 1263)	1366 (181, 2551)	539 (69, 1009)	360 (32, 688)	132 (11, 253)
Liver	6508 (3621, 9395)	5569 (3083, 8054)	5987 (3375, 8599)	5120 (2874, 7366)	521 (246, 796)	448 (209, 687)
Breast	580 (0, 1669)	104 (0, 300)	/	/	580 (0, 1669)	104 (0, 300)
Physical inactivity
All	1204 (470, 1939)	433 (171, 696)	494 (260, 728)	209 (111, 308)	709 (209, 1210)	223 (60, 387)
Colorectum	737 (304, 1170)	328 (133, 524)	494 (260, 728)	209 (111, 308)	242 (43, 441)	118 (22, 215)
Corpus uteri	152 (32, 272)	32 (7, 57)	/	/	152 (32, 272)	32 (7, 57)
Breast	314 (132, 495)	72 (30, 114)	/	/	314 (132, 495)	72 (30, 114)
Low vegetable intake
All	1843 (0, 6154)	948 (0, 3134)	1404 (0, 4720)	730 (0, 2441)	438 (0, 1434)	218 (0, 692)
Oral cavity, pharynx	498 (0, 1642)	214 (0, 696)	352 (0, 1174)	157 (0, 520)	146 (0, 468)	57 (0, 175)
Nasopharynx	1142 (0, 3851)	590 (0, 1972)	870 (0, 2944)	446 (0, 1507)	272 (0, 907)	144 (0, 465)
Larynx	201 (0, 660)	143 (0, 465)	181 (0, 601)	125 (0, 413)	19 (0, 59)	17 (0, 52)
Low fruit intake
All	12,365 (5643, 21,961)	8290 (4090, 13,903)	8383 (3701, 15,208)	6048 (2946, 10,244)	3981 (1941, 6752)	2242 (1144, 3659)
Oral cavity, pharynx	791 (0, 2320)	324 (0, 941)	573 (0, 1674)	246 (0, 715)	218 (0, 645)	77 (0, 226)
Nasopharynx	1889 (0, 5636)	1015 (0, 2962)	1413 (0, 4184)	765 (0, 2230)	475 (0, 1452)	249 (0, 731)
Larynx	305 (0, 888)	211 (0, 608)	286 (0, 832)	194 (0, 560)	18 (0, 55)	16 (0, 48)
Lung, bronchus and trachea	9379 (5643, 13,115)	6740 (4090, 9390)	6109 (3701, 8517)	4842 (2946, 6738)	3269 (1941, 4598)	1898 (1144, 2652)
Red meat
All	6122 (60, 15,562)	3649 (34, 9081)	4350 (58, 10,906)	2691 (33, 6633)	1771 (24655)	958 (12447)
Esophagus	1590 (0, 4189)	1299 (0, 3427)	1321 (0, 3438)	1105 (0, 2883)	269 (0, 751)	193 (0, 543)
Stomach	1870 (60, 3709)	1270 (34, 2532)	1322 (58, 2595)	887 (33, 1749)	548 (21114)	383 (1782)
Colorectum	2661 (0, 7662)	1079 (0, 3121)	1706 (0, 4872)	697 (0, 2000)	954 (0, 2789)	381 (0, 1121)
Excess bodyweight
All	6659 (3243, 11,095)	2985 (1357, 4768)	3226 (1595, 5445)	1872 (821, 3056)	3433 (1647, 5649)	1112 (535, 1711)
Esophagus	629 (330, 928)	512 (269, 756)	510 (268, 752)	424 (223, 625)	119 (61, 176)	88 (45, 130)
Stomach	735 (409, 1061)	495 (275, 716)	494 (276, 713)	324 (180, 467)	240 (133, 347)	171 (95, 248)
Colorectum	1700 (1369, 2031)	678 (546, 811)	1026 (827, 1225)	403 (325, 482)	673 (542, 805)	274 (221, 328)
Pancreas	115 (29, 201)	101 (26, 176)	68 (17, 118)	60 (15, 104)	47 (12, 83)	41 (10, 72)
Liver	778 (29, 1527)	668 (24, 1311)	649 (25, 1273)	552 (20, 1083)	129 (4253)	116 (3228)
Gallbladder	94 (26, 162)	48 (13, 82)	52 (14, 89)	25 (7, 43)	42 (11, 72)	22 (6, 38)
Kidney	273 (243, 302)	76 (68, 84)	186 (166, 206)	54 (48, 60)	86 (77, 96)	21 (19, 24)
Thyroid	532 (0, 1629)	9 (0, 30)	164 (0, 500)	4 (0, 12)	368 (0, 1128)	5 (0, 17)
Corpus uteri	975 (618, 1331)	153 (97, 209)	/	/	975 (618, 1331)	153 (97, 209)
Ovary	334 (0, 707)	136 (0, 288)	/	/	334 (0, 707)	136 (0, 288)
Breast	416 (186, 646)	80 (36, 125)	/	/	416 (186, 646)	80 (36, 125)
Prostate	74 (0, 564)	22 (0, 175)	74 (0, 564)	22 (0, 175)	/	/
Diabetes
All	12,530 (6929, 18,177)	7300 (4685, 9931)	7518 (4636, 10,446)	5194 (3551, 6852)	5011 (2292, 7731)	2106 (1133, 3078)
Esophagus	269 (186, 352)	222 (153, 290)	214 (148, 280)	180 (125, 236)	54 (37, 71)	41 (28, 54)
Stomach	332 (51, 613)	236 (37, 436)	156 (16, 295)	106 (11, 201)	176 (35, 317)	130 (26, 234)
Colorectum	2229 (1155, 3304)	956 (494, 1417)	1336 (713, 1959)	563 (300, 826)	893 (441, 1345)	392 (193, 591)
Pancreas	351 (172, 530)	310 (152, 469)	263 (164, 363)	232 (145, 319)	87 (8166)	78 (7149)
Liver	5204 (4055, 6352)	4594 (3579, 5609)	4265 (3324, 5207)	3718 (2897, 4539)	938 (730, 1145)	876 (682, 1070)
Gallbladder	220 (32, 408)	116 (16, 215)	124 (18, 230)	63 (9116)	96 (13, 178)	53 (7, 98)
Kidney	225 (127, 323)	73 (41, 105)	149 (81, 216)	51 (28, 74)	76 (46, 107)	22 (13, 30)
Urinary bladder	180 (94, 266)	70 (37, 104)	150 (79, 222)	57 (29, 84)	29 (15, 44)	13 (7, 20)
Thyroid	732 (414, 1049)	21 (12, 30)	193 (89, 296)	7 (3, 11)	538 (325, 752)	14 (8, 19)
Corpus uteri	1066 (304, 1828)	199 (61, 338)	/	/	1066 (304, 1828)	199 (61, 338)
Ovary	192 (87, 297)	89 (41, 138)	/	/	192 (87, 297)	89 (41, 138)
Breast	861 (245, 1476)	194 (56, 333)	/	/	861 (245, 1476)	194 (56, 333)
Prostate	663 (0, 1373)	213 (0, 442)	663 (0, 1373)	213 (0, 442)	/	/
Epstein–Barr virus
All	9129 (8711, 9502)	4745 (4529, 4938)	6722 (6416, 6995)	3561 (3399, 3705)	2406 (2294, 2506)	1184 (1130, 1232)
Nasopharynx	9027 (8623, 9386)	4708 (4497, 4895)	6665 (6367, 6930)	3537 (3379, 3677)	2361 (2255, 2455)	1171 (1118, 1217)
Hodgkin lymphoma	101 (87, 116)	37 (31, 42)	57 (49, 65)	23 (20, 27)	44 (38, 51)	13 (11, 15)
Human papillomavirus
All	8861 (8274, 9552)	2874 (2598, 3209)	1597 (1177, 2108)	764 (549, 1032)	7263 (7097, 7444)	2109 (2049, 2177)
Oral cavity, pharynx	1475 (1149, 1844)	597 (465, 746)	1061 (827, 1326)	452 (352, 566)	413 (322, 517)	144 (112, 180)
Larynx	437 (266, 671)	299 (182, 459)	409 (249, 628)	275 (167, 422)	28 (17, 42)	24 (14, 36)
Cervix uteri	6622 (6584, 6660)	1879 (1868, 1889)	/	/	6622 (6584, 6660)	1879 (1868, 1889)
Vulva, vagina	199 (173, 223)	62 (54, 70)	/	/	199 (173, 223)	62 (54, 70)
Penis	126 (101, 152)	35 (28, 43)	126 (101, 152)	35 (28, 43)	/	/
Helicobacter pylori
All	2740 (1355, 4126)	1908 (943, 2872)	1819 (899, 2739)	1244 (615, 1873)	920 (455, 1386)	663 (328, 999)
Stomach	2740 (1355, 4126)	1908 (943, 2872)	1819 (899, 2739)	1244 (615, 1873)	920 (455, 1386)	663 (328, 999)
Hepatitis B virus
All	12,181 (11,398, 12,965)	10,726 (10,037, 11,416)	9958 (9318, 10,598)	8658 (8102, 9215)	2223 (2080, 2366)	2068 (1935, 2201)
Liver	12,181 (11,398, 12,965)	10,726 (10,037, 11,416)	9958 (9318, 10,598)	8658 (8102, 9215)	2223 (2080, 2366)	2068 (1935, 2201)
Hepatitis C virus
All	381 (253, 510)	336 (223, 449)	312 (207, 416)	271 (180, 362)	69 (46, 93)	64 (43, 86)
Liver	381 (253, 510)	336 (223, 449)	312 (207, 416)	271 (180, 362)	69 (46, 93)	64 (43, 86)
Clonorchis sinensis
All	326 (0, 736)	287 (0, 648)	267 (0, 602)	232 (0, 523)	59 (0, 134)	55 (0, 125)
Liver	326 (0, 736)	287 (0, 648)	267 (0, 602)	232 (0, 523)	59 (0, 134)	55 (0, 125)

Note: “All” refers to attributable cases of the specific risk factor in total cancer.

The primary cause of cancer incidence and mortality was smoking, which contributed to 28,001 incident cases and 19,960 deaths, respectively. The number of incident cases in males (26,610) explained by smoking was approximately 19 times higher than that in females (1391). However, the number of incident cases attributable to second‐hand smoking was found to be higher in females (2122) than in males (1089). By cancer type, lung cancer had the highest number of smoking‐attributable cases. Another behavioral factor, alcohol, was the second most significant contributor to incident cancer (15,114) and the third most significant contributor to cancer deaths (10,167). The number of gastrointestinal cancer cases attributable to alcohol consumption was found to be significantly higher than that of other cancers, with the greatest number of cases being attributed to liver cancer (Table 2 and Figures 3, S7, and S12).

FIGURE 3.

Number of incident cancer and cancer death attributable to studied risk factors (individually and all combined) in Guangdong, 2019. (A) incident cancer; (B) cancer death.

For all infection‐related cancers, the PAF of incident cancer reached 73.7%, which ranged from 96.9% for cervical cancer to 44.2% for penile cancer. By risk type, HBV was the infectious agent with the largest attributable proportion for cancer, followed by EBV, HPV, and Hp. Notably, HPV was the leading contributor to cancers in females, accounting for 6.0% (7263) of incident cases and 4.8% (2109) of deaths (Figures 1, 3, S7, and S12 and Tables 1 and 2).

Among the dietary factors in the analysis, low fruit intake had the largest PAF, contributing to 12,365 incident cases and 8290 deaths. Additionally, the number of incident cancer cases attributable to diabetes was the third highest of all risk factors (12,530), and the number of deaths was the fifth highest (7300) (Figure 3 and Table 2).

3.3. Age‐specific proportion of cancer attributable to risk factors

The attributable cancer burden varied across age groups, with the proportion of overall cancer attributable to potentially modifiable risk factors demonstrating an increasing and then decreasing trend with age. We recorded the highest PAFs among those aged 40–69 years, and lowest among those aged 20–29 years. Furthermore, the patterns of age‐specific PAF differed between risk factors. The trends in PAF changes with age for most individual risk factors were consistent with the overall trend, except for EBV, which demonstrated the highest among those aged 20–39 years. Additionally, PAFs for low fruit intake, physical inactivity, second‐hand smoking, and Hp were positively associated with age, while PAFs for low vegetable intake were inversely associated with age (Table S7).

4. DISCUSSION

In 2019, 34.4% of incident cancer and 44.5% of cancer deaths could be attributable to potentially modifiable risk factors. The highest attributable proportion was observed for cervical cancer, while the largest attributable number of incident cancer cases and deaths was recorded for lung cancer. Of all the risk factors evaluated, behavioral factors accounted for the highest PAF of incident cancer, followed by infectious agents, dietary factors, and metabolic factors.

We calculated the incidence PAFs (mortality PAFs) by adding up the number of attributable incident cancer (cancer death) cases at each cancer site and then dividing by the total number of incident cancer (cancer death) cases. Therefore, the differences between incidence PAFs and mortality PAFs for overall cancers primarily stem from the proportion of cancer types with different PAFs among overall cancer. Highly fatal cancers are often associated with clear risk factors—for example, smoking for lung cancer and alcohol/HBV for liver cancer. In our study, liver, stomach, and esophageal cancers—the leading causes of cancer deaths—had PAFs exceeding 50%. Conversely, cancers with high incidence but lower lethality, such as thyroid and breast cancers, showed lower PAFs (3.5% and 8.3%, respectively). Therefore, we observed that mortality PAFs were higher than incidence PAFs in our study.

In this study, smoking and alcohol consumption were identified as the primary contributors to the cancer burden. Tobacco is classified as a Group 1 carcinogen by the IARC; the major established pathway by which tobacco smoke causes cancer is the formation of covalent bonds between carcinogens and DNA, resulting in the production of DNA adducts and irreversible mutations in critical genes of somatic cells. ³¹ , ³² , ³³ In 2019, tobacco use accounted for a large share of the cancer burden, with more than 1 in 2 male lung cancer cases attributable to smoking and 1 in 10 female lung cancer cases attributable to second‐hand smoking. Guangdong had made significant progress in tobacco control over the past two decades, with a decline in smoking prevalence. The proportion of cancer attributable to smoking in 2019 was considerably lower than that in 2014, indicating the efficacy of the current tobacco control measures. ²³ Nevertheless, Guangdong, the most populous province, was still facing a high absolute cancer burden attributable to smoking, exacerbated by the aging population.

However, in contrast to the decline in smoking prevalence, alcohol consumption was still on the rise. The lag effect of alcohol‐related harms indicates that the cancer burden attributable to alcohol consumption is likely to increase in the future. Therefore, the implementation of comprehensive anti‐smoking laws and the strengthening of health education on the dangers of alcohol are essential. The carcinogenic mechanism of alcohol has also been clarified. Ethanol can induce inflammation and oxidative stress, while its metabolite, acetaldehyde, can cause DNA damage and disrupt DNA methylation. ³⁴

Besides the above unhealthy behavioural factors, infectious agents remain important contributors to cancer in Guangdong. HBV causes close to half of all liver cancer cases. Hepatitis viruses exert their carcinogenic effects directly through their proteins and indirectly by inducing chronic inflammation and tissue damage. ³⁵ Commendably, Guangdong is among the first provinces to have initiated early prevention and treatment of viral hepatitis, and the mother‐to‐child transmission rate of HBV is far below the WHO target. ³⁶ , ³⁷ , ³⁸ However, due to the high migration rate and the large foreign population in Guangdong, the overall population in Guangdong still has a higher HBV infection rate than the national average. Furthermore, 1.2% of liver cancers were attributable to Clonorchis sinensis infection. The prevalence of Clonorchis sinensis in Guangdong was the highest in the entire country, particularly within the Pearl River Delta (more than 10%), which may be related to the unique dietary habit of consuming raw freshwater fish there. ³⁹

The association between cervical cancer and HPV is clear, with approximately 96.9% of cervical cancers being attributable to HPV. Consequently, expanding the population coverage of the HPV vaccine and then reducing the HPV prevalence can efficiently mitigate the risk of cervical cancer. However, the HPV vaccine coverage was limited nationally, due to the high costs and exclusion from the National Immunization Programme (NIP). In 2022, Guangdong began offering free HPV vaccines to eligible females, boosting the vaccination rate for females aged 9–45 from 6.8% in 2021 to 15.7% in 2022, a leading level in the whole country but still far short of the WHO's 2030 target of vaccinating 90% of girls by age 15. ⁴⁰ , ⁴¹

Notably, the incidence of nasopharyngeal cancer in Guangdong is among the highest worldwide, ranking sixth among all cancers. ¹⁶ , ⁴² , ⁴³ The results demonstrate that 80% of nasopharyngeal cancer can be attributed to EBV. The high PAF was associated not only with a high infection rate of EBV in the population, but also with the specific high‐risk EBV strains prevalent in Guangdong. ⁴⁴ , ⁴⁵ Vaccination is the most effective measure to prevent infection; however, a vaccine for EBV is not yet available. The unexplained cases may be related to region‐specific dietary habits, such as the consumption of salted fish, preserved meat, and other cured foods.

Meanwhile, China is currently experiencing a rapid urbanization period, resulting in substantial shifts in dietary habits and lifestyle patterns. The consumption of cereals and low‐fat mixed dishes has been replaced by a dietary pattern dominated by high‐energy foods. Previous studies have reported a decline in vegetable consumption, alongside an increased intake of animal products, processed foods, and ultra‐processed foods that are high in energy, fat, sugar, and salt (HEFS) among the Chinese population. ⁴⁶ , ⁴⁷ , ⁴⁸ , ⁴⁹ , ⁵⁰ Guangdong, the province with the highest GDP in the country, has seen more pronounced changes. The WHO strongly recommends an intake of at least 400 g of vegetables and fruits per day for adults; however, the surveillance report showed that the average daily fruit intake of residents was only 66 g in Guangdong, which was far below the recommendation. ⁵¹ Furthermore, the consumption of red meat was seven times higher than the recommended intake of the Planetary Health Diet. ⁵² Cooking red meat at high temperatures generates carcinogenic compounds, while vegetables and fruits are rich in various anticancer phytochemicals. Moreover, dietary factors can influence cancer risk by affecting body weight and insulin metabolism. ⁵³ Consequently, it is a necessity to guide residents to gradually adopt a healthy dietary structure.

Moreover, urbanization has led to more convenient transportation and modern technological infrastructure, significantly reducing agricultural, occupational, and domestic physical activity. ⁵⁴ , ⁵⁵ , ⁵⁶ Additionally, the widespread use of electronic devices has reduced recreational physical activity and increased sedentary behavior. ⁵⁷ , ⁵⁸ , ⁵⁹ While the contribution of physical inactivity to the overall cancer burden was negligible, there is evidence that physical activity helps maintain a healthy weight and optimal metabolic levels, which also affect the cancer risk. ⁶⁰ , ⁶¹ , ⁶² Overall, if the entire population adhered to a healthy lifestyle (adequate intake of vegetables and fruits [more than 400 g/day of vegetables and more than 300 g/day of fruits], low intake of red meat [less than 22.5 g/day], sufficient physical activity [a weekly MET value of more than 600], a normal weight [BMI in the range of 18–24], and absence of diabetes), there would be a 13.0% reduction in incident cancer and a 10.2% reduction in cancer deaths. Furthermore, the analysis revealed that colorectal cancer was the second most prevalent cancer, suggesting that the cancer spectrum in Guangdong is progressively transitioning to cancers associated with rapid social and economic changes. ⁶³ In the future, the aforementioned risk factors related to social development will assume greater importance in attributing the cancer burden.

There are discrepancies in the PAF among different age groups for a given risk factor, as well as variations in the age‐specific PAF patterns for various risk factors. This trend is influenced by not only the joint effects of exposures at each birth cohort, but also the dominant age of the risk factor‐associated cancers. Compared to the country as a whole, Guangdong has a younger population structure, which allows us to focus more on risk factors with high PAF in the younger population, such as EBV. Understanding the age pattern of PAFs will allow appropriate cancer intervention strategies to be targeted at specific age groups, thereby maximizing the use of limited healthcare resources.

We made a comparison between our results and those of related published studies. In 2014, the PAF of overall cancer in Guangdong was 49.7%. ²³ As expected, the attributable proportion of cancer burden was lower in this study, which may be related to the observed decline in the prevalence of infection and smoking. The previous studies showed that the PAFs for the Chinese population in 2007 and 2012 were 54.1% and 64.5%, respectively, which were higher than the results of this study. ¹³ , ¹⁵ This may be because the above studies focused on the 10 cancers with the highest mortality, while this study included all cancers. It is also possible that the attributable cancer burden in Guangdong differs from that of the entire country. Furthermore, we found that the PAFs in Guangdong Province were similar to those reported in several developed areas, including the United States, Japan, Canada, and the Eastern Mediterranean region. ²¹ , ⁶⁴ , ⁶⁵ , ⁶⁶ This pattern reflects Guangdong's ongoing epidemiological transition from infection‐ and poverty‐related cancers to those associated with affluence, aging, and urban lifestyle changes, although the comparison may be affected by variations in the research year.

This study has several strengths. First, it is the first study to systematically evaluate the role of potentially modifiable risk factors on cancer burden across various age groups, sexes, and geographical regions in Guangdong, which helps to adequately understand the status of cancer attribution and to target cancer prevention and control to specific populations. Second, the cancer burden and prevalence of behavioral, dietary, and metabolic risk factors used in this study are real‐world data from high‐quality, representative population‐based surveillance. Finally, we conducted systematic meta‐analyses of exposure to infectious factors to better reflect the actual situation in Guangdong, whereas most previous studies used published prevalence information from global studies as a proxy. Nevertheless, it is subject to several limitations. First, most of the exposure data in our study was obtained from population‐based surveillance. Although such data is representative and collected by uniformly trained investigators, it primarily relies on self‐reports, which are subject to recall and reporting biases. Second, due to the unavailability of exposure information for risk factors such as ultraviolet radiation, air pollution, and industrial exposures, we focus exclusively on potentially modifiable behavioral, metabolic, and infectious risk factors. Additionally, we assume that the individual risk factors are independent of each other except for excess body weight and diabetes, which may lead to the overestimation of some PAFs. Finally, due to the lack of high‐quality and long‐term prospective cohort studies in Guangdong, most of the RR values selected in this study were from the Chinese population, and the same RR was used to estimate the risks of incidence and mortality.

5. CONCLUSION

This study systematically evaluated the cancer burden attributable to 15 potentially modifiable risk factors in Guangdong in 2019, which indicated that approximately one‐third of cancer incidence and two‐fifths of cancer deaths could be prevented by avoiding exposure to these risk factors. Smoking, alcohol consumption, and infections are major risk factors and should be a priority for cancer prevention and control.

AUTHOR CONTRIBUTIONS

Xiaolan Wen: Writing – original draft; writing – review and editing; methodology; validation; visualization. Yu Liao: Data curation; investigation; project administration. Jiayue Li: Visualization; software; writing – review and editing. Qian Zhu: Methodology; software; validation. Xinmei Lin: Writing – review and editing; visualization; writing – original draft. Bingfeng Han: Writing – review and editing; project administration. Li Li: Methodology; writing – review and editing. Ru Chen: Writing – review and editing; methodology. Ruilin Meng: Investigation; data curation. Ni Xiao: Investigation; data curation. Xueyan Zheng: Investigation; data curation. Xiaojun Xu: Investigation; data curation. Dejian Zhao: Investigation; data curation. Yue Gao: Investigation; data curation. Liming Pu: Data curation; investigation. Ye Wang: Project administration; funding acquisition; supervision; resources. Wenqiang Wei: Writing – review and editing; funding acquisition; methodology; formal analysis; supervision; resources. Shaoming Wang: Conceptualization; writing – review and editing; funding acquisition; methodology; project administration; formal analysis; supervision; resources.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Data S1. Supporting Information.

Wen X, Liao Y, Li J, et al. Over one‐third of cancer cases and two‐fifths of cancer deaths in southern China are preventable: Insights from the latest representative population‐based cancer registry data and risk factor survey. Int J Cancer. 2026;158(4):909‐923. doi: 10.1002/ijc.70114

DATA AVAILABILITY STATEMENT

All source code is publicly available on GitHub (https://github.com/Wen-Xiaolan/Xiaolan-Wen.git). Due to institutional policy restrictions, individual‐level data cannot be publicly shared. Researchers interested in accessing the data used in this study may contact the corresponding author for further information.