Fraction of cancer incidence and mortality attributable to dietary factors in Korea from 2015 to 2030

Article information

Epidemiol Health. 2025;47.e2025065

Publication date (electronic) : 2025 December 8

doi : https://doi.org/10.4178/epih.e2025065

Hyun Jeong Cho¹, Jin Young Yoo², Ga-Eun Yie², An Na Kim², Soseul Sung^3,4,5, Sungji Moon^3,4,6, Youjin Hong^3,4,7, Sangjun Lee^3,4,7, Inah Kim⁸, Kwang-Pil Ko^3,9, Sun-Seog Kweon¹⁰, Jung Eun Lee^,2,11, Sue K. Park^,3,4,7

¹Division of Prevention, National Cancer Center Institute for Cancer Control, Tokyo, Japan

²Department of Food and Nutrition, Seoul National University College of Human Ecology, Seoul, Korea

³Department of Preventive Medicine, Seoul National University College of Medicine, Seoul, Korea

⁴Cancer Research Institute, Seoul National University, Seoul, Korea

⁵Department of Biomedical Sciences, Seoul National University Graduate School, Seoul, Korea

⁶Interdisciplinary Program in Cancer Biology, Seoul National University College of Medicine, Seoul, Korea

⁷Integrated Major in Innovative Medical Science, Seoul National University Graduate School, Seoul, Korea

⁸Department of Occupational and Environmental Medicine, Hanyang University College of Medicine, Seoul, Korea

⁹Clinical Preventive Medicine Center, Seoul National University Bundang Hospital, Seongnam, Korea

¹⁰Department of Preventive Medicine, Chonnam National University Medical School, Hwasun, Korea

¹¹Research Institute of Human Ecology, Seoul National University, Seoul, Korea

Correspondence: Jung Eun Lee Department of Food and Nutrition, Seoul National University College of Human Ecology, 1 Gwanak-ro, Gwanak-gu, Seoul 08826, Korea E-mail: jungelee@snu.ac.kr

Co-correspondence: Sue K. Park Department of Preventive Medicine, Seoul National University College of Medicine, 103 Daehak-ro, Jongno-gu, Seoul 03080, Korea E-mail: suepark@snu.ac.kr

Received 2025 May 26; Accepted 2025 November 21.

Abstract

GRAPHICAL ABSTRACT

Key Message

In 2020, dietary factors accounted for a proportion of the cancer burden, with high intake of salted vegetables and low intake of non-starchy vegetables and fruits showing the largest contributions among dietary factors examined. These findings suggest the need for continued assessment of dietary patterns and their potential relevance in public health discussions on cancer prevention.

INTRODUCTION

Cancer is one of the leading causes of death worldwide, with 10 million cancer deaths reported in 2020 [1]. In Korea, cancer ranked as the first cause of death, with an age-standardized incidence rate of 262.2 per 100,000 in 2020 [2]. Newly diagnosed cancer cases are predicted to rise from 19.3 million in 2020 to 28.4 million in 2040 [1].

Diet plays an important role in modifying cancer risk. The World Cancer Research Fund (WCRF) report found that processed meat has been associated with a higher colorectal cancer risk, while dietary fiber and whole grains have been associated with lower risk [3]. The International Agency for Research on Cancer (IARC) classified processed meat as carcinogenic and red meat as probably carcinogenic [4]. The Global Burden of Disease (GBD) study estimated that dietary risks accounted for approximately 5% of global cancer disability-adjusted life years in 2019 [5].

Estimating the cancer burden attributable to diet is critical for effective prevention programs and public health planning. Because dietary patterns differ across countries, country-specific estimates are necessary. In Korea, salted vegetable intake remains high and dietary patterns have changed in recent decades, with intake measuring around 136.6 g/day in male and 90.4 g/day in female in 2015. Meanwhile, red and processed meat intake increased from roughly 48-76 g/day to 51-83 g/day for red meat and from 3-5 g/day to 8-13 g/day for processed meat between 2005 and 2015, based on our analysis of Korean National Health and Nutrition Examination Survey (KNHANES) data. These trends underscore the need for Korea-specific assessments. Previous studies have examined the burden of cancers attributable to dietary factors in the United States [6], the United Kingdom [7], France [8], Germany [9], Australia [10], Brazil [11], Japan [12], and China [13]. However, in Korea, although several reports have assessed other risk factors, studies estimating the preventive fraction of diet remain limited [14-20]. Most previous Korean work has focused on exposures such as excess body weight, hormone and reproductive factors, or carcinogenic drugs, and diet has not been comprehensively addressed. Therefore, we aimed to estimate the proportion of cancer incidence and mortality attributable to dietary factors among the Korean population from 2015 to 2030, with the goal of providing evidence to support cancer prevention and public health strategies.

MATERIALS AND METHODS

RESULTS

Population-attributable fraction of cancer

In 2020, dietary factors accounted for 6.08% of cancer incidence (8.43% for male and 3.45% for female) and 5.70% of cancer mortality (7.93% for male and 2.08% for female) (Table 1, Supplementary Material 5). Dietary risk-increasing factors contributed 2.25% of incidence (3.13% for male and 1.28% for female), while protective dietary factors accounted for 3.82% (5.30% for male and 2.18% for female). For cancer mortality, the respective contributions were 2.31% (2.76% for male and 1.59% for female) and 3.39% (5.17% for male and 0.49% for female). By 2025 and 2030, PAFs for cancer incidence decreased, whereas PAFs for mortality remained largely stable.

Table 1.

PAFs¹ of cancer attributable to dietary factors in Korea from 2015 to 2030

From 2015 to 2030, PAFs for high red meat intake increased slightly from 0.08% to 0.10%, and processed meat from 0.01% to 0.08%. PAFs for salted vegetables decreased substantially from 3.00% to 1.17%, and the PAF for low intake of non-starchy vegetables and fruits decreased from 2.22% to 1.92%. Dietary fiber PAFs peaked at 1.90% in 2020 and then declined to 1.54% in 2025 and 1.61% in 2030. For deaths, PAFs for dietary fiber increased to 1.29% in 2030, and PAFs for low non-starchy vegetables and fruits increased to 2.84% in 2030. In 2020, stomach cancer had the highest diet-related PAF, accounting for 24.61% of cancer incidence and 24.27% of cancer deaths (Table 2, Supplementary Material 6). Colorectal cancer followed, with PAFs of 18.38% for incidence and 18.57% for deaths, while nasopharyngeal cancer accounted for 6.97% and 7.43%, and aerodigestive cancers for 5.12% and 5.46%. From 2015 to 2030, PAFs for nasopharyngeal and stomach cancers declined, whereas colorectal cancer PAFs increased from 18.75% to 20.03%. The PAF for aerodigestive cancers remained stable. Among attributable cancer cases related to dietary factors, stomach cancer accounted for 44.3% of incident cases and 37.4% of deaths (Figures 1 and 2). The distribution varied modestly by sex: 47.0% for male and 35.5% for female for incident cases, and 34.8% for male and 52.4% for female for deaths.

Table 2.

PAFs¹ of diet-related cancer attributable to all dietary factors in Korea from 2015 to 2030

Figure 1.

The population-attributable fraction of cancer cases attributable to dietary factors and the proportion of specific cancers among all diet-related cancer cases in Korea, 2020. (A) Male cancer cases. (B) Female cancer cases. (C) Total cancer cases.

Figure 2.

The population-attributable fraction of cancer deaths attributable to dietary factors and the proportion of specific cancers among all diet-related cancer deaths in Korea, 2020. (A) Male cancer deaths. (B) Female cancer deaths. (C) Total cancer deaths.

When estimating PAFs attributable to all combined dietary factors, we conducted sensitivity analyses using Asian and global RRs (Table 3). The PAFs in 2020 were 4.99% for incident cases and 4.21% for deaths when applying Asian RRs, and 3.76% for incident cases and 5.09% for deaths when applying global RRs. These values were approximately 1-2 percentage points lower than estimates based on Korean RRs (6.08% for incidence and 5.70% for deaths), reflecting modest variation across RR sources and supporting the robustness of the findings. These small differences may partially reflect the limited number of studies contributing to the Asian or Korean RRs. Sensitivity analyses using 2000-2015 intake data produced similar patterns, with slightly higher PAFs under a 20-year latency assumption and comparable values under a 10-year latency, consistent with the long-term nature of diet–cancer associations (Supplementary Material 7).

Table 3.

PAFs of cancer attributable to dietary factors in Korea in 2020 based on Korean, Asian, and global RR

DISCUSSION

In 2020, dietary factors accounted for 6.08% of cancer incidence and 5.70% of cancer deaths in Korea, with greater contributions observed in male (8.43% for incidence and 7.93% for deaths) than in female (3.45% for incidence and 2.08% for deaths). Stomach and colorectal cancers showed the highest proportions of diet-attributable cases and deaths, driven primarily by high intake of salted vegetables and low intake of non-starchy vegetables and fruits. These results are broadly consistent with international estimates of cancer burden attributable to diet. Our findings were comparable to those reported for all diet-related cancer incidence in the United States (5.2%) [6], United Kingdom (9.2%) [7], France (5.4%) [8], Germany (7.8%) [9], Australia (5.4%) [10], and Brazil (5.1%) [11] (Figure 3).

Figure 3.

International comparison of the population-attributable fraction (PAF) of cancer cases and deaths attributable to dietary factors. (I) indicates incidence and (M) indicates mortality.

The 2 major dietary contributors to cancer burden in Korea were high intake of salted vegetables and low intake of non-starchy vegetables and fruits, which is consistent with findings from the Eastern Mediterranean region [35]. That study identified low fruit intake as the dietary factor with the highest PAF (6.0% for male and 3.5% for female), followed by low vegetable intake (4.9% for male and 2.7% for female), high salt intake (2.0% for male and 0.9% for female), processed meat (0.2% for both sexes), and red meat (0.2% for male and 0.1% for female). Across other multi-factor dietary burden studies, the dietary factor with the highest PAF for all cancer incidence varied: whole grains in the United States (1.81%) [6], dietary calcium in Canada (Alberta) (0.9%) [36] and Brazil (0.9%) [11], red and processed meat in the United Kingdom (2.7%) [37], and dietary fiber in France (1.4%) [8] and Germany (3.3%) [9].

When comparing our estimates for salted vegetables (2.12% for incidence and 1.78% for deaths) with PAFs for salt intake in 30countries lacking estimates specifically for salted vegetables, our PAFs were higher than those reported in Japan (1.6% for incidence and 1.4% for deaths) [12], Canada (Alberta) (0.2% for incidence) [36], the United Kingdom (0.5% for incidence) [37], Germany (0.3% for incidence) [9], and the Eastern Mediterranean region (2.0% for male and 0.9% for female for incidence) [35]. The higher PAFs observed in Korea likely reflect the notably high intake of salted vegetables, which reached 145.9 g/day in 2005. In Japan, estimated intakes of pickled vegetables (salted and seasoned) in 2007 ranged from 4.6 g/1,000 kcal/day to 11.7 g/1,000 kcal/day [38]. Additionally, about 10 g/day of salted vegetable intake was estimated in China in 2002 [39]. The particularly high Korean PAFs may therefore be partly explained by the frequent consumption of kimchi, a major contributor to dietary salt and not widely eaten in Western populations.

Several studies have examined the burden of cancers attributed to low intake of vegetables and fruits. Our 2020 estimates of 1.92% for incidence and 2.34% for deaths were slightly higher than estimates from Japan (1.3% for incidence and 1.4% for deaths) [12], France (1.9% for incidence) [8], and the United States (1.3% for incidence) [6]. However, our results were lower than estimates from China (15.0% for incidence and 16.6% for deaths) [13] and the Eastern Mediterranean region (10.9% for male and 6.2% for female for incidence) [35]. The relatively low intake levels of non-starchy vegetables and fruits in our population (340.5 g/day), compared with the optimal intake range recommended by the GBD 2017 study (490-730 g/day), likely contribute to the importance of this factor in Korea’s cancer burden. Korea’s PAFs were similar to those of other high-income countries such as Japan, France, and the United States, and lower than in regions where fruit and vegetable intake remains limited. These differences may reflect variations in income levels, climate, and agricultural systems that influence food availability.

The estimated burden of cancers attributable to low dietary fiber intake (1.90% for incidence and 1.04% for deaths) fell within the range reported by other countries: 0.9% in the United States [40], 0.7% in Canada (Alberta) [41], 0.8% in Brazil [11], 3.3% in the United Kingdom [7], 1.4% in France [8], 3.3% in Germany [9], and 2.3% in Australia for incident cases [42], as well as 1.7% in Australia for deaths [10]. Our estimates may be partly explained by the relatively high prevalence (42.4%) of dietary fiber intake below the optimal range of 19-28 g/day recommended by the GBD 2017 study.

In our study, the PAFs for red meat and processed meat were 0.10% and 0.02% for incidence, and 0.49% and 0.02% for deaths, respectively. These estimates were generally lower than those from other countries, where PAFs ranged from 0.4% to 0.6% for red meat and 1.0% to 2.1% for processed meat in the United States [6], France [8], and Germany [9]. The PAFs for combined red and processed meat intake were reported as 2.7% in the United Kingdom [37], and 2.3% in Australia [43]. The lower PAFs in our study reflect the comparatively lower intakes of these meats: 61.6 g/day of red meat and 4.2 g/day of processed meat in Korea (2005), compared with 100.5 g/day and 43.5 g/day in the United States (2011-2014) [6], 68.6 g/day for red meat and 36.5 g/day for processed meat in France (in 2006) [8], 36.9 g/day for red meat and 45.7 g/day for processed meat in Germany (in 2008-2011) [9], 121 g/day (male) and 64 g/day (female) for red and processed meat in Australia (in 1995) [43], and 118 g/day (male) and 69 g/day (female) for red and processed meat in the United Kingdom (in 2000-2001) [37]. For cancers attributable to high intake of salted fish, the PAF was very small (0.01%), reflecting the low intake level of salted fish (average 3.2 g/day in 2005). Given the limited number of studies evaluating cancer burden from salted fish consumption, further research is needed.

The PAF for cancer in 2020 highlighted substantial contributions from high salted vegetable intake and low non-starchy vegetable and fruit intake. In Korea, site-specific cancer trends differ, with stomach cancer showing a steady decline, whereas colorectal and breast cancers have continued to rise [44]. These trends provide important context for interpreting the projected PAFs for dietary factors. Projections to 2030 indicated a decrease in the PAF for high salted vegetable intake, while only minimal change was expected for the PAF associated with low intake of non-starchy vegetables and fruits. Individual food items, such as kimchi and other fermented vegetables, warrant further investigation to clarify their potential role in cancer risk, given their complex balance of possible beneficial and adverse effects. Continued efforts to reduce salted vegetable intake and to increase vegetable consumption therefore remain essential for cancer prevention. Moreover, although high intakes of processed meat contributed relatively little to the cancer PAF in Korea in 2015, projections suggested a steady increase through 2030. This emphasizes the need for ongoing surveillance of processed meat intake as part of broader cancer prevention strategies.

As strengths of our study, we used nationally representative data to estimate the prevalence and average intake levels of dietary factors. We also derived pooled RRs by incorporating additional findings from available Korean cohort studies. In addition, we applied a widely used Monte Carlo simulation approach to account for statistical uncertainty. However, several limitations should be noted.

First, dietary intake based on a single 24-hour recall may not adequately represent usual intake or long-term dietary patterns and may introduce non-differential misclassification. Second, several dietary factors—such as dairy products, calcium supplements, and coffee—were not included, potentially leading to an underestimation of the diet-related cancer burden. Third, the limited number of Korean cohort studies constrains the robustness of pooled RRs and the generalizability of the findings to the entire Korean population. Fourth, our PAF estimates assumed independent effects of dietary factors, which may underestimate potential combined or synergistic effects. Fifth, excluding studies with high heterogeneity may reduce bias but may further limit generalizability. Sixth, applying a uniform 15-year latency period across all cancers may not capture site-specific latency differences. Seventh, PAF estimates may vary by age, as both dietary exposures and cancer risks differ across age groups. Finally, projections beyond 2020 relied on extrapolating historical trends and therefore may not reflect unexpected changes in screening behaviors, treatment practices, or dietary patterns.

In conclusion, an estimated 6.08% of cancer incidence and 5.70% of cancer deaths in Korea in 2020 were attributable to dietary factors, with high salted vegetable intake and low non-starchy vegetable and fruit intake as major contributors. Projections for 2030 suggest a decline in the PAF associated with high salted vegetable intake and little change in the PAF related to low non-starchy vegetable and fruit intake, though these estimates may be conservative because they rely on single-day dietary assessments. These findings underscore the importance of continued efforts to reduce salted vegetable intake, increase vegetable consumption, and monitor processed meat intake to strengthen cancer prevention. They also suggest that enhancing nutrition education, refining dietary guidelines, and exploring supportive policy measures could be valuable directions for future public health efforts.

Supplementary materials

Article information Continued