---
title: "Healthy lifestyle and pan-cancer risk: UK Biobank prospective analysis"
id: "british-journal-of-cancer-2-healthy-lifestyle-and-cancer-risk-a-pan-cancer-analysis-in-the-uk-biobank"
canonical_url: "https://medichelpline.com/clinical-feed/british-journal-of-cancer-2-healthy-lifestyle-and-cancer-risk-a-pan-cancer-analysis-in-the-uk-biobank"
content_type: "clinical_feed_article"
specialty: "Oncology"
source_name: "British Journal of Cancer"
source_url: "https://www.nature.com/articles/s41416-026-03613-9"
published_at: "2026-09-15T12:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Healthy lifestyle and pan-cancer risk: UK Biobank prospective analysis
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/british-journal-of-cancer-2-healthy-lifestyle-and-cancer-risk-a-pan-cancer-analysis-in-the-uk-biobank
- **Specialty:** [Oncology](https://medichelpline.com/clinical-feed/oncology.md)
- **Primary Source:** British Journal of Cancer
- **Source URL:** [Original Journal Publication](https://www.nature.com/articles/s41416-026-03613-9)
- **Published At:** 2026-09-15T12:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- This pan-cancer analysis used data from 287,745 UK Biobank participants followed for a median of 11.5 years to evaluate associations between a composite **healthy lifestyle** score (HLS) and incidence of total and 27 site-specific cancers. - The HLS assigned one point each for **non-smoking**, moderate alcohol intake, a healthy diet (top 40% of a diet score), adequate physical activity (≥500 MET-hours/week) and a healthy BMI (18.5–<25 kg/m2); HLS categories were low (0–1), medium (2–3) and high (4–5). - During follow-up 12,947 participants developed cancer; prostate, breast and colorectal cancer had the highest incidence rates in this cohort. - High versus low HLS was associated with a lower risk of overall cancer (adjusted HR ~0.71). High HLS associated with decreased risks for 13 of 27 cancer types, including oral, pharyngeal, laryngeal, oesophageal, stomach, colorectal, liver, pancreatic, lung, soft tissue sarcoma, kidney, bladder and postmenopausal breast cancer (HRs ranged from 0.12 to 0.79 in the abstract). - Population attributable fractions (PAFs) for incomplete adherence to a healthy lifestyle were significant for 14 cancer types; smoking and overweight contributed to the largest number of cancer types. - Analyses used Cox models with progressive covariate adjustment, multiple imputation for missing data, correction for multiple testing (Benjamini-Hochberg), and PAFs estimated with bootstrap replications. - Subgroup analyses by age (<60 vs ≥60) and sex were performed for cancers with significant inverse associations. - The authors conclude that promotion of **smoking cessation** and **weight control** may be prioritized as cancer prevention strategies given their association with more cancer types than other lifestyle components. - Methods, covariate adjustments, cancer ascertainment windows and full supplementary details (tables, figures, exact site-specific HRs and PAF numeric values) are reported in the original article and supplementary materials; some numeric specifics beyond those summarized in the abstract are reported in supplementary tables referenced by the authors.
## Clinical Analysis & Structured Key Points
## Your privacy, your choice We use essential cookies to make sure the site can function. We also use optional cookies for advertising, personalisation of content, usage analysis, and social media, as well as to allow video information to be shared for both marketing, analytics and editorial purposes. By accepting optional cookies, you consent to the processing of your personal data - including transfers to third parties. Some third parties are outside of the European Economic Area, with varying standards of data protection. See our [privacy policy](https://www.nature.com/info/privacy) for more information on the use of your personal data. Manage preferences for further information and to change your choices. Accept all cookies Reject optional cookies [Skip to main content](https://www.nature.com/articles/s41416-026-03613-9#content) Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript. Advertisement [ ![British Journal of Cancer](https://media.springernature.com/full/nature-cms/uploads/product/bjc/header-76ac17391903a21e28f15ff5df31f5fa.svg) ](https://www.nature.com/bjc) * [ View all journals ](https://www.nature.com/siteindex) * [ Saved research ](https://www.nature.com/saved-research) * [ Search ](javascript:;) ## Search Search articles by subject, keyword or author Show results from All journals This journal Search [ Advanced search ](https://www.nature.com/search/advanced) ### Quick links * [Explore articles by subject](https://www.nature.com/subjects) * [Find a job](https://www.nature.com/naturecareers) * [Guide to authors](https://www.nature.com/authors/index.html) * [Editorial policies](https://www.nature.com/authors/editorial_policies/) * [Log in](https://idp.nature.com/auth/personal/springernature?redirect_uri=https://www.nature.com/articles/s41416-026-03613-9) * [ Content Explore content ](javascript:;) ## Explore content * [ Research articles ](https://www.nature.com/bjc/research-articles) * [ Reviews & Analysis ](https://www.nature.com/bjc/reviews-and-analysis) * [ News & Comment ](https://www.nature.com/bjc/news-and-comment) * [ Current issue ](https://www.nature.com/bjc/current-issue) * [ Collections ](https://www.nature.com/bjc/collections) * [Follow us on X ](https://twitter.com/BrJCancer) * [Sign up for alerts ](https://journal-alerts.springernature.com/subscribe?journal_id=41416) * [ RSS feed ](https://www.nature.com/bjc.rss) * [ About the journal ](javascript:;) ## About the journal * [ Journal Information ](https://www.nature.com/bjc/journal-information) * [ Open access publishing ](https://www.nature.com/bjc/open-access) * [ About the Editors ](https://www.nature.com/bjc/editors) * [ Contact ](https://www.nature.com/bjc/contact) * [ Special Issues ](https://www.nature.com/bjc/special-issues) * [ For Advertisers ](https://www.nature.com/bjc/advertising) * [ Subscribe ](https://www.nature.com/bjc/subscribe) * [ Publish with us ](javascript:;) ## Publish with us * [ For Authors & Referees ](https://www.nature.com/bjc/authors-and-referees) * [ Language editing services ](https://authorservices.springernature.com/go/sn/?utm_source=For+Authors&utm_medium=Website_Nature&utm_campaign=Platform+Experimentation+2022&utm_id=PE2022) * [Open access funding](https://www.nature.com/bjc/open-access-funding) * [Submit manuscript ](https://mts-bjcancer.nature.com/) * [ Sign up for alerts ](https://journal-alerts.springernature.com/subscribe?journal_id=41416) * [ RSS feed ](https://www.nature.com/bjc.rss) 1. [nature](https://www.nature.com/) 2. [british journal of cancer](https://www.nature.com/bjc) 3. [articles](https://www.nature.com/bjc/articles?type=article) 4. article Healthy lifestyle and cancer risk: a pan-cancer analysis in the UK Biobank prospective cohort [ Download PDF ](https://www.nature.com/articles/s41416-026-03613-9.pdf) [ Download PDF ](https://www.nature.com/articles/s41416-026-03613-9.pdf) * Article * [Open access](https://www.springernature.com/gp/open-science/about/the-fundamentals-of-open-access-and-open-research) * Published: 15 September 2026 Epidemiology - Cancer Disparities # Healthy lifestyle and cancer risk: a pan-cancer analysis in the UK Biobank prospective cohort * [Jie Ding](https://www.nature.com/articles/s41416-026-03613-9#auth-Jie-Ding-Aff1-Aff2)[1](https://www.nature.com/articles/s41416-026-03613-9#Aff1),[2](https://www.nature.com/articles/s41416-026-03613-9#Aff2), * [Ben Schöttker](https://www.nature.com/articles/s41416-026-03613-9#auth-Ben-Sch_ttker-Aff1)[1](https://www.nature.com/articles/s41416-026-03613-9#Aff1), * [Hermann Brenner](https://www.nature.com/articles/s41416-026-03613-9#auth-Hermann-Brenner-Aff3) [ORCID: orcid.org/0000-0002-6129-1572](https://orcid.org/0000-0002-6129-1572)[3](https://www.nature.com/articles/s41416-026-03613-9#Aff3) & * … * [Michael Hoffmeister](https://www.nature.com/articles/s41416-026-03613-9#auth-Michael-Hoffmeister-Aff1) [ORCID: orcid.org/0000-0002-8307-3197](https://orcid.org/0000-0002-8307-3197)[1](https://www.nature.com/articles/s41416-026-03613-9#Aff1) Show authors [_British Journal of Cancer_](https://www.nature.com/bjc) (2026) [Cite this article](https://www.nature.com/articles/s41416-026-03613-9#citeas) [ Save article ](https://www.nature.com/articles/s41416-026-03613-9/save-research?_csrf=VkXP9ndycgVwL60X4yDlF_iz1fQ8pHqk) [ View saved research ](https://www.nature.com/saved-research) ## Abstract ### Background Research on healthy lifestyles across various cancers is limited. We conducted a pan-cancer analysis to evaluate the associations of healthy lifestyle with cancer risk. ### Methods Data from 287,745 UK Biobank participants were analysed. A healthy lifestyle score (HLS) was calculated based on smoking, alcohol, diet, physical activity and BMI. Cox models were used to estimate 27 cancer risks, and population attributable fractions (PAFs) were calculated based on incomplete adherence to a healthy lifestyle. ### Results During a median 11.5-year follow-up, 12,947 participants developed cancer. Adhering to 4–5 vs 0–1 healthy lifestyle behaviours was associated with decreased risks of 13 of the 27 cancer types: oral, pharyngeal, laryngeal, oesophageal, stomach, colorectal, liver, pancreatic, lung, soft tissue sarcoma, kidney, bladder and postmenopausal breast cancer (HRs from 0.12 to 0.79). Significant PAFs were observed for 14 cancer types in relation to lower HLS. Smoking and overweight were associated with the largest number of cancer types. ### Conclusions These findings extend evidence on associations between a healthy lifestyle and lower risks of both common and less-common cancers. Smoking and overweight were associated with more cancer types than the other components, suggesting that smoking cessation and weight control may be cancer prevention strategies to be prioritised. ### Explore related subjects Discover the latest articles and news in related subjects. * [Cancer epidemiology](https://www.nature.com/subjects/cancer-epidemiology) * [Risk factors](https://www.nature.com/subjects/risk-factors) ## Introduction Cancer remains a critical public health issue, with an estimated 28.4 million cases globally by 2040 [[1](https://www.nature.com/articles/s41416-026-03613-9#ref-CR1 "Deo S, Sharma J, Kumar S. GLOBOCAN 2020 report on global cancer burden: challenges and opportunities for surgical oncologists. Ann Surg Oncol. 2022;29:6497–500.")]. Lifestyle factors, such as smoking, physical activity, diet, alcohol consumption and BMI, have been identified as major modifiable risk factors for cancer. Due to the diverse aetiological mechanisms underlying different cancer types [[2](https://www.nature.com/articles/s41416-026-03613-9#ref-CR2 "Schneider G, Schmidt-Supprian M, Rad R, Saur D. Tissue-specific tumorigenesis: context matters. Nat Rev Cancer. 2017;17:239–53.")], it is crucial to identify which cancers are affected by lifestyle factors to inform targeted prevention strategies. Cancer registration data from 1993 to 2018 across the UK showed that the age-standardised incidence rates were relatively stable. However, the incidence rates for some less common cancers, such as liver, oral and melanoma skin cancer, have shown notable increases [[3](https://www.nature.com/articles/s41416-026-03613-9#ref-CR3 "Shelton J, Zotow E, Smith L, Johnson SA, Thomson CS, Ahmad A, et al. 25 year trends in cancer incidence and mortality among adults aged 35-69 years in the UK, 1993-2018: retrospective secondary analysis. BMJ. 2024;384:e076962.")]. Despite this trend, not much is known about the associations of a healthy lifestyle with these less common cancers. The Third Expert Report by the World Cancer Research Fund (WCRF) in 2018 reported strong associations between single lifestyle factors and several cancers [[4](https://www.nature.com/articles/s41416-026-03613-9#ref-CR4 "WCRF/AICR. Diet, nutrition, physical activity and cancer: a global perspective: a summary of the Third Expert Report. World Cancer Research Fund International; 2018; London, UK.")], especially the more common cancers, but the evidence is limited or inconsistent for other cancers. Additionally, most pan-cancer analyses focused on only a combined healthy lifestyle [[5](https://www.nature.com/articles/s41416-026-03613-9#ref-CR5 "Byrne S, Boyle T, Ahmed M, Lee SH, Benyamin B, Hyppönen E. Lifestyle, genetic risk and incidence of cancer: a prospective cohort study of 13 cancer types. Int J Epidemiol. 2023;52:817–26."), [6](https://www.nature.com/articles/s41416-026-03613-9#ref-CR6 "Malcomson FC, Parra-Soto S, Ho FK, Lu L, Celis-Morales C, Sharp L, et al. Adherence to the 2018 World Cancer Research Fund \(WCRF\)/American Institute for Cancer Research \(AICR\) Cancer Prevention Recommendations and risk of 14 lifestyle-related cancers in the UK Biobank prospective cohort study. BMC Med. 2023;21:407.")], while the contributions of individual lifestyle factors to the association have not been thoroughly investigated. In the current study, we used a previously benchmarked healthy lifestyle score (HLS) in the large UK Biobank cohort to comprehensively investigate the joint impact of healthy lifestyle factors on the risk of total and 27 types of cancer [[7](https://www.nature.com/articles/s41416-026-03613-9#ref-CR7 "Carr PR, Weigl K, Jansen L, Walter V, Erben V, Chang-Claude J, et al. Healthy lifestyle factors associated with lower risk of colorectal cancer irrespective of genetic risk. Gastroenterology. 2018;155:1805–15.e5.")], as well as to estimate the population attributable fractions (PAFs) of incomplete adherence to a healthy lifestyle. ## Methods ### Study population The UKB is a prospective cohort of over 500,000 participants, aged 37–73, recruited in England, Wales, and Scotland from 2006 to 2010 [[8](https://www.nature.com/articles/s41416-026-03613-9#ref-CR8 "Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12:e1001779."), [9](https://www.nature.com/articles/s41416-026-03613-9#ref-CR9 "Collins R. What makes UK Biobank special? Lancet. 2012;379:1173–4.")]. Among the UK Biobank participants enroled at baseline, we excluded participants who withdrew from follow-up, who were pregnant, those with prevalent cancer (except non-melanoma skin cancer) at recruitment, without necessary data to constitute all of the considered lifestyle factors, and with follow-up time less than 2 years, leaving 287,745 participants in the analysis (Supplementary Fig. [1](https://www.nature.com/articles/s41416-026-03613-9#MOESM1)). This study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline [[10](https://www.nature.com/articles/s41416-026-03613-9#ref-CR10 "von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology \(STROBE\) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61:344–9.")] (Supplementary Table [1](https://www.nature.com/articles/s41416-026-03613-9#MOESM1)). ### Assessment of healthy lifestyle behaviours We used the established HLS to evaluate the combined associations of healthy lifestyle factors [[7](https://www.nature.com/articles/s41416-026-03613-9#ref-CR7 "Carr PR, Weigl K, Jansen L, Walter V, Erben V, Chang-Claude J, et al. Healthy lifestyle factors associated with lower risk of colorectal cancer irrespective of genetic risk. Gastroenterology. 2018;155:1805–15.e5."), [11](https://www.nature.com/articles/s41416-026-03613-9#ref-CR11 "Ding J, Schöttker B, Brenner H, Hoffmeister M. Thirteen simple lifestyle scores and risk of cancer, cardiovascular disease, diabetes, and mortality: prospective cohort study in the UK Biobank. Int J Cancer. 2025;157:2495–505.")]. The definition of individual healthy lifestyle factors used the predefined scoring approach of the established HLS, including non-smoking, moderate alcohol intake, a healthy diet score, adequate physical activity, a healthy BMI, and assigning one point to each healthy behaviour. Non-smoking was defined as never having smoked regularly or having been a former smoker with fewer than 30 pack-years. Moderate alcohol intake was defined as consuming alcohol ≤12 g/day for women and ≤24 g/day for men. The participants in the highest 40% of a healthy diet score distribution were considered to have a healthy diet (Supplementary Table [2](https://www.nature.com/articles/s41416-026-03613-9#MOESM1)). Adequate physical activity was defined by achieving 500 MET-hours/week for walking, moderate activity, or vigorous activity. A healthy BMI was defined as >18.5 to <25 kg/m2. The HLS was categorised into low (0–1), medium (2–3) and high (4–5). ### Cancer-related outcomes The outcome of interest was the first incident cancer diagnosis identified through linkage to cancer and death registries. We calculated the follow-up time from the date of recruitment to the date of cancer diagnosis, the date of loss to follow-up, the date of death, or the end of follow-up (31 December 2020 for England, 30 November 2021 for Scotland and 31 December 2016 for Wales), whichever occurred first. To enable meaningful analyses, we included cancers with at least 100 cases in the study population only (for International Classification of Diseases, Tenth Revision (ICD-10) codes, see Supplementary Table [3](https://www.nature.com/articles/s41416-026-03613-9#MOESM1)). We carried out separate analyses for premenopausal and postmenopausal breast cancer. ### Statistical analysis Missing data were imputed by multiple imputation using the Markov-Chain Monte Carlo method (_N_ = 5 imputed datasets, Supplementary Table [4](https://www.nature.com/articles/s41416-026-03613-9#MOESM1)). Baseline characteristics for each HLS category were summarised as means (with standard deviation, SD) for continuous variables and numbers (percentages) for categorical variables. Cox proportional hazards regression was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between HLS categories and cancer risk, with the low HLS group as the reference group. Model 1 was adjusted for age and sex. Model 2 was additionally adjusted for ethnicity, region, Townsend Deprivation Index, household income, education, history of cancer screening, family history of cancer (prostate, breast, bowel and lung cancer), hypertension, diabetes and height (Supplementary Methods [1](https://www.nature.com/articles/s41416-026-03613-9#MOESM1)). We further adjusted for additional cancer-specific covariates (Supplementary Table [5](https://www.nature.com/articles/s41416-026-03613-9#MOESM1)). To assess the association between individual lifestyle factors as defined in the HLS and cancer risk, we additionally adjusted for all other individual lifestyle factors. We used the Benjamini-Hochberg procedure to correct for multiple comparisons in the analyses [[12](https://www.nature.com/articles/s41416-026-03613-9#ref-CR12 "Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B Stat Methodol. 1995;57:289–300.")]. We calculated the variance inflation factor (VIF) to assess the potential inflation of the variance of regression coefficients due to multicollinearity. All VIF values of covariates were below 5 in the models. PAFs and 95% CIs were calculated for 0–3 vs 4–5 points on the HLS and for the individual unhealthy lifestyle behaviours, using the graphPAF R package (PAF_calc_discrete() function) with 1000 bootstrap replications [[13](https://www.nature.com/articles/s41416-026-03613-9#ref-CR13 "Ferguson J, O’Connell M. Estimating and displaying population attributable fractions using the R package: graphPAF\[J\]. European journal of epidemiology. 2024;39:715–42.")]. Details about the calculation of the PAF are shown in Supplementary Methods [2](https://www.nature.com/articles/s41416-026-03613-9#MOESM1). We ranked the site-specific cancers according to their estimated PAFs and quantified the number of cancer types for which each unhealthy lifestyle behaviour was significantly attributable (i.e. the PAF CIs did not include 0). Subgroup analyses were conducted by age (<60 vs ≥60 years) and sex (men vs women) to examine the associations between HLS groups and site-specific cancer in the fully adjusted models, restricted to cancer types with significant inverse associations with HLS in the main analysis. Interaction tests were conducted by including a cross-product term with the main effect terms in the models. All statistical tests were two-sided, and statistical significance was defined as _P_ < 0.05 or 95% CIs excluding 1. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) and R version 4.4 (R Foundation). ## Results Among 287,745 participants, 12,947 participants were diagnosed with cancer during a median follow-up time of 11.5 years. Prostate cancer in men had the highest incidence (7195, 4.31 per 1000 person-years), followed by breast cancer in women (4533, 2.82 per 1000 person-years). Among cancers affecting both sexes, colorectal cancer was the most frequently diagnosed cancer (3432, 1.03 per 1000 person-years) (Fig. [1](https://www.nature.com/articles/s41416-026-03613-9#Fig1)). **Fig. 1** ![Fig. 1](https://media.springernature.com/lw685/springer-static/image/art%3A10.1038%2Fs41416-026-03613-9/MediaObjects/41416_2026_3613_Fig1_HTML.png) [Full size image](https://www.nature.com/articles/s41416-026-03613-9/figures/1) Cancer counts and rate per 1000 person-years by cancer type. Table [1](https://www.nature.com/articles/s41416-026-03613-9#Tab1) presents the characteristics of all participants and participants stratified by HLS. The mean (SD) age was 56.0 (8.1) years. Participants with higher HLS were more likely to be women, have higher socioeconomic status and education level, be taller, have no hypertension or diabetes, have a lower family history of cancer and have undergone cancer screening. **Table 1 Baseline characteristics of all populations and by HLS groups.** [ Full size table](https://www.nature.com/articles/s41416-026-03613-9/tables/1) A decreased risk of overall cancer was observed with high vs low HLS both in model 1 (HR: 0.69, 95% CI: 0.67–0.72) and model 2 (0.71, 0.68–0.74) (Fig. [2](https://www.nature.com/articles/s41416-026-03613-9#Fig2) and Supplementary Table [6](https://www.nature.com/articles
## Related Clinical Research

- [Cancer type–specific chromatin regulator mutations linked to better survival after immune checkpoi](https://medichelpline.com/clinical-feed/medrxiv-17-cancer-type-specific-profiling-of-chromatin-regulator-mutations-identifies.md)
- [Predicting Cancer Immunotherapy Response Using Dynamic CD8+ T Cell Changes — Pan‑Cancer Retrospect](https://medichelpline.com/clinical-feed/frontiers-in-immunology-13-a-predictive-model-for-immunotherapy-efficacy-in-cancer-based-on-dynamic.md)
- [Age- and sex-specific links between metabolic syndrome and colorectal cancer risk in UK Biobank](https://medichelpline.com/clinical-feed/british-journal-of-cancer-0-age-and-sex-dependent-associations-of-metabolic-syndrome-with-colorectal-cancer.md)
- [Low-fat dietary pattern reduced dietary advanced glycation end-products in the Women’s Health Init](https://medichelpline.com/clinical-feed/british-journal-of-cancer-0-low-fat-dietary-pattern-and-dietary-advanced-glycation-end-products-intake-a.md)
- [Diabetes Mellitus and Cancer Risk: Epidemiology and Molecular Mechanisms](https://medichelpline.com/clinical-feed/pubmed-42364319.md) (DOI: 10.1016/j.bbrc.2026.154205)

## Navigation
- [← Back to Oncology Feed](https://medichelpline.com/clinical-feed/oncology.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.