Human authored, evidenced-based, 7 min read

According to data from the World health organisation (WHO), cancer rates have increased world-wide ⁽¹⁾, including Gibraltar ⁽²⁾. Cancer is a leading cause of mortality, accounting for nearly one in six deaths. The WHO predicts over 35 million new cancer cases may be expected by 2050 ⁽³⁾, and that four in ten cancer case may be prevented highlighting the needs for effective preventative measures. Physical activity, and dietary patterns that prioritise fruit, vegetables, and fibre-containing foods, and avoid processed meat, are recommended to help reduce global cancer burden ⁽⁴⁾, whereas increased intake of ultra-processed foods (UPFs) may increase the risks (figure 1).

New study published
A newly published large population study by the University of Oxford ⁽⁵⁾, has found that vegetarian diets are were associated with a lower risk of various cancers, specifically; pancreas, breast, prostate, kidney, and multiple myeloma, although pooled analysis found they were also associated with a higher risk of squamous cell carcinoma of the oesophagus.

How the study was conducted
Participants were split into five groups: meat eaters, poultry eaters, pescatarians, vegetarians, and vegans. Follow-up surveys were carried out at a median of four to 14 years after baseline. Cancer cases were identified through linkage to cancer registries and defined using WHO’s International Classification of Diseases. Data was harmonised for 1,817,477 participants cross nine prospective studies spanning four countries, comprising of:

• Meat eaters 1,645,555 (90.5%, red and/or processed meat)
• Poultry eaters 57,016 (3.1%)
• Pescatarians 42,910 (2.4%)
• Vegetarians 63,147 (3.5%)
• Vegans 8849 (0.5%)

Sophisticated statistical methods were used to determine the associations. Multivariable Cox regression estimated cohort-specific hazard ratios (HRs) and 95% confidence intervals, with results combined using meta-analysis – PS: apologies for the brief moment of statistical jargon! Compared with meat eaters, the risk of cancers in vegetarians were:

• 21% lower risk of pancreatic cancer (HR 0.79, 0.65–0.97)
• 9% lower risk of breast cancer (HR 0.91, 0.86–0.97)
• 12% lower risk of prostate cancer (HR 0.88, 0.79–0.97
• 28% lower risk of kidney cancer (HR 0.72, 0.57–0.92)
• 31% lower risk of multiple myeloma (HR 0.69, 0.51–0.93)

Vegetarians had nearly double the risk of squamous cell carcinoma of the oesophagus compared to meat-eaters (HR 1.93, 1.30–2.87), but no statistically significant differences in risk for colorectal, stomach, liver, lung (in never smokers), endometrial, ovarian, mouth and pharynx, or bladder cancers, or non-Hodgkin lymphoma, leukaemia, and oesophageal adenocarcinoma, meaning that the risk did not differ with regards to meat-eaters. View the HR’s by clicking the drop-down toggle below for figures 1, 2, 3.

CLICK TO VIEW THE POOLED HAZARD RATIOS FOR VARIOUS CANCERS Warning: only view if you understand statistics

Fig. 1: Pooled hazard ratios for cancers of the gastrointestinal tract in poultry eaters, pescatarians, vegetarians and vegans, relative to meat eaters.

Results were only reported for diet groups with ≥10 incident cases across all cohorts. Pooled multivariable-adjusted hazard ratios and 95% confidence intervals. The models were stratified by sex and by region or method of recruitment. Covariates in the multivariable-adjusted models were: living with a partner (yes, no), educational status (less than secondary/high school, secondary/high school or equivalent, university degree or equivalent), ethnic group (Asian, Black, Hispanic, White, other), study and sex-specific height categories (women in UK and USA cohorts: <160, 160–164.9, ≥165 cm; women in Asian cohorts: <150, 150–154.9, ≥155 cm; men in UK and USA cohorts: <175, 175–179.9, ≥180 cm; men in Asian cohorts: <163, 163–167.9, ≥168 cm), cigarette smoking history (never, previous, current <10 cigarettes/day, current 10–19 cigarettes/day, current ≥20 cigarettes/day, current unknown number of cigarettes), tobacco chewing (in CARRS-1 only; never, previous, current), physical activity (highly active, moderately active, inactive), alcohol intake (0.0, 0.1–9.9, 10.0–19.9, ≥20.0 g/day), history of diabetes (yes, no), parity (nulliparous, parous), ever used hormone replacement therapy (yes, no), and BMI (<20.0, 20.0–22.4, 22.5-24.9, 25.0–29.9, ≥30.0 kg/m²). For all variables, a further category of unknown was included for participants with missing data. ACC adenocarcinoma, SCC squamous cell carcinoma.

Fig. 2: Pooled hazard ratios for cancers of the reproductive system in poultry eaters, pescatarians, vegetarians and vegans, relative to meat eaters.

Results were only reported for diet groups with ≥10 incident cases across all cohorts. Pooled multivariable-adjusted hazard ratios and 95% confidence intervals. The models were stratified by region or method of recruitment. Covariates in the multivariable-adjusted models were: living with a partner (yes, no), educational status (less than secondary/high school, secondary/high school or equivalent, university degree or equivalent), ethnic group (Asian, Black, Hispanic, White, other), study and sex-specific height categories (women in UK and USA cohorts: <160, 160–164.9, ≥165 cm; women in Asian cohorts: <150, 150–154.9, ≥155 cm; men in UK and USA cohorts: <175, 175–179.9, ≥180 cm; men in Asian cohorts: <163, 163–167.9, ≥168 cm), cigarette smoking history (never, previous, current <10 cigarettes/day, current 10–19 cigarettes/day, current ≥20 cigarettes/day, current unknown number of cigarettes), tobacco chewing (in CARRS-1 only; never, previous, current), physical activity (highly active, moderately active, inactive), alcohol intake (0.0, 0.1–9.9, 10.0–19.9, ≥20.0 g/day), history of diabetes (yes, no), parity (nulliparous, parous), ever used hormone replacement therapy (yes, no), and BMI (<20.0, 20.0–22.4, 22.5-24.9, 25.0–29.9, ≥30.0 kg/m2). For breast, endometrial, and ovarian cancers, the models were further adjusted for age at menarche (≤10 years, 11–12 years, 13–14 years, ≥15 years), parity and age at first birth combined (nulliparous, and parity and age at first birth grouped as: 1–2 and <25 years, 1–2 and 25–29 years, 1–2 and ≥30 years, 1–2 and unknown, ≥3 and <25 years, ≥3 and 25–29 years, ≥3 and ≥30 years, ≥3 and unknown), menopausal status (pre-menopausal, post-menopausal), and ever used oral contraceptives (yes, no). For prostate cancer, the models were further adjusted for history of prostate antigen screening (yes, no). For all variables, a further category of unknown was included.

Fig. 3: Pooled hazard ratios for cancers of the urinary tract and blood in poultry eaters, pescatarians, vegetarians and vegans, relative to meat eaters.

Results were only reported for diet groups with ≥10 incident cases across all cohorts. Pooled multivariable-adjusted hazard ratios and 95% confidence intervals. The models were stratified by sex and by region or method of recruitment. Covariates in the multivariable-adjusted models were: living with a partner (yes, no), educational status (less than secondary/high school, secondary/high school or equivalent, university degree or equivalent), ethnic group (Asian, Black, Hispanic, White, other), study and sex-specific height categories (women in UK and USA cohorts: <160, 160–164.9, ≥165 cm; women in Asian cohorts: <150, 150–154.9, ≥155 cm; men in UK and USA cohorts: <175, 175–179.9, ≥180 cm; men in Asian cohorts: <163, 163–167.9, ≥168 cm), cigarette smoking history (never, previous, current <10 cigarettes/day, current 10–19 cigarettes/day, current ≥20 cigarettes/day, current unknown number of cigarettes), tobacco chewing (in CARRS-1 only; never, previous, current), physical activity (highly active, moderately active, inactive), alcohol intake (0.0, 0.1–9.9, 10.0–19.9, ≥20.0 g/day), history of diabetes (yes, no), parity (nulliparous, parous), ever used hormone replacement therapy (yes, no), and BMI (<20.0, 20.0–22.4, 22.5-24.9, 25.0–29.9, ≥30.0 kg/m²). For all variables, a further category of unknown was included.

CLICK TOGGLE AT THE TOP TO CLOSE THIS WINDOW

 

 

 

 

Cancer risks with other diets
Pescatarians had lower risks of colorectal (HR 0.85, 0.77–0.93), breast (HR 0.93, 0.88–0.98) and kidney cancers (HR 0.73, 0.58–0.93), whilst poultry eaters were found to have a lower risk of prostate cancer (HR 0.93, 0.88–0.98) compared to meat eaters. The study however did not find any statistically significant differences in risk for a variety of other cancers, including stomach, liver, lung, ovarian, and mouth cancers compared to meat-eaters

Associations in vegans
Vegans, on the other hand, had a statistically significantly higher risk of colorectal (bowel) cancer compared with meat eaters (HR 1.40, 1.12–1.75). For the other cancers analysed, there was no evidence that cancer risk differed between vegans and meat eaters. However, for some less common cancers there were too few vegan cases to draw reliable conclusions, which may partly explain these unexpected findings; therefore, further research is required. These results are interesting when considered alongside other recent studies showing that adopting a vegan or vegetarian diet from a typical Western diet may reduce the risk of premature mortality from non-communicable diseases by 18–21% ⁽⁷⁾. In addition, increasing consumption of plant-based foods (figure 3) has been associated with substantial environmental benefits, including reductions in dietary carbon footprint of around 46–51% ⁽⁷⁾.

Lower calcium intakes in vegans?
Let’s reflect for a moment as to why vegans may be associated to a higher incidence of colorectal cancers. Adherence to some vegetarian and vegan diets may lead to lower intakes of certain nutrients, such as dietary calcium ⁽¹⁰⁾, particularly where diets rely heavily on ultra-processed plant-based alternatives that may be relatively lower in calcium ⁽⁸⁾, which may have potentially contributed to the higher colorectal cancer risk observed among vegans in the current study, although causal mechanisms remain uncertain. More broadly, and as a side note, these findings may illustrate that while shifts toward plant-based diets may reduce the environmental impact of food systems, such changes do not automatically guarantee optimal nutritional or health outcomes without careful dietary planning.

Does dairy calcium reduce colorectal cancer risks?
For example, a recent systematic review and meta-analysis which included 74 studies involving 7,356 vegan, 51,940 vegetarian, and 107,581 omnivorous participants ⁽¹⁰⁾, suggested that vegans’ calcium intakes were lower than in vegetarians and omnivores, emphasising the need for vegans to monitor their calcium status. Other large population studies including 42,778 women in the United Kingdom over 16.6 years ⁽⁹⁾, found dairy consumption was inversely associated (more dairy, less risk) with the risk of colorectal, colon, and rectal cancers, which may suggest dairy products may help protect against colorectal cancer, arguably driven by increased dairy calcium. The reasons why are plausible, and happy to explain this to clients if you decide to work with me when signing up to the PERSONALISED NUTRITION COURSE that I offer.

Strength and weaknesses
A strength of the study was the lack of strong evidence of marked heterogeneity (variance) between studies despite the wide geographical spread of the populations, which increases the veracity and generalisability of the outcomes. The study also included large datasets followed over long periods. Weaknesses were that the data used in this study derived from groups of people who were recruited at least 10 years ago, and in some cases as far back as the 1980s, and eating habits change over time. However, the researchers factored this in to account by conducting sensitivity analyses. Vegan participants were also considerably under-represented in the study participants which could unfairly skew the outcomes in these diet populations. There may also have been some degree of misclassification of diet groups at baseline, and the researchers did not assess the duration of dietary adherence before participants joined the participants, although where data were available in individual participants the majority of vegetarians had followed their diet for at least several years.

Caution warranted on the outcomes
In contrast, in a previous blogpost I recently discussed how the associations of meat consumption and the risk of DEMENTIA may have been overstated, whilst a recent observational study from McMaster University involving 16,000 adults ⁽⁶⁾, suggested that moderate animal meat intakes were not adversely associated with all-cause, cardiovascular disease, or cancer-related mortality risk. Furthermore, the current study also did not find any statistically significant differences in risk for a variety of other cancers, including stomach, liver, lung, ovarian, and mouth cancers when comparing other diets to meat-eaters. Therefore, caution is warranted when interpreting cancer risk associations with meat intake, as meat-eating may also be associated with unhealthier and more sedentary behavours, as well as increased intakes of processed foods, which could amplify the said risks.

Summary and conclusions
There are several simple and plausible mechanistic explanations that could account for many of the findings in this study. It is in this area specifically that my nutrition and clinical expertise can provide a lot of nuance and help provide clarity and understanding for clients wishing to follow certain diets in a safe manner to address specific health concerns or optimise their health, one of the major benefits of working with me long-term with many of my health programmes. Get in touch to work with me.

Contact me to work together