UK Biobank

Study design and population

UK Biobank is a prospective study of ~ 500,000 UK adults aged 40–69 years at recruitment (including 229,000 men) established between 2006 and 2010 to study risk factors for disease. Details of the study protocol and information about data access are available online ( and elsewhere [10]. All individuals provided informed consent to participate and the study was approved by the National Information Governance Board for Health and Social Care and the National Health Service North West Multicentre Research Ethics Committee (reference number 06/MRE08/65). In brief, approximately 9.2 million people living within reasonable travelling distance (25 km) of 1 of the 22 assessment centres across England, Wales, and Scotland were identified from National Health Service (NHS) registers and invited to participate in the study, with a participation rate of 5.5% [11].

After excluding 9869 men with prevalent cancer (except C44: non-melanoma skin cancer), 1 man censored on entry day and 999 men with no adiposity measurements, the analyses included a total of 218,237 men (Additional File 1: Figure S1).

Assessment of adiposity and other predictor variables

At recruitment, participants provided detailed information on a range of sociodemographic, physical, lifestyle and health-related factors via self-completed touch-screen questionnaires and a computer assisted personal interview [11]. Anthropometric measurements (standing height, weight, waist and hip circumferences) were taken by trained research clinic staff at the assessment centre, while body mass index (BMI) was calculated and percentage body fat was estimated through bioimpedance measures [12].

UK Biobank imaging sub-cohort

In 2014, the UK Biobank imaging study re-invited a subsample of participants to undergo abdominal MRI and DXA, which has been detailed elsewhere [13, 14]. In brief, participants were scanned in a Siemens MAGNETOM Aera 1.5 T MRI scanner (Siemens Healthineers, Germany) using a 6-min dual-echo Dixon Vibe protocol, providing water-and-fat-separated volumetric information for fat and muscle. Body composition analyses for MRI images were performed using AMRA Profiler Research (AMRA Medical AB, Linköping, Sweden). DXA captures whole-body composition (e.g. bone, fat and lean mass) with no extensive additional processing and analysis. However, it is not possible to obtain direct compartmental volumetric measurements using this method, and therefore, regional volume estimates are obtained indirectly using anatomical models [13, 14]. By December 2021, imaging data on ~ 18,800 men were available. BMI, WC, hip circumference and body fat percentage were also re-assessed at the imaging visit.

Ascertainment of prostate cancer mortality

Our endpoint was prostate cancer as the underlying cause of death recorded on the death certificate (International Classification of Diseases Tenth revision codes: C61 [15]). Men were followed-up until 31 December 2020 for England and Scotland and 19 July 2020 for Wales. Mortality data were provided by NHS Digital for England and Wales and by the NHS Central Register and Information and Statistics Division for Scotland.

Statistical analysis in UK Biobank

Cross-sectional analyses of adiposity measurements

Pearson correlations between different anthropometric measurements were calculated. A subsample of men had both commonly used anthropometric measures of adiposity (i.e. BMI, body fat percentage, WC and WHR) and adiposity information (MRI and DXA) from the imaging visit. Multivariable linear regression adjusted for categories of age and height was used to estimate the mean differences in each MRI- and DXA-derived measure of body composition per 1-SD difference in the levels of each commonly used anthropometric measure of adiposity. We also used Pearson correlations to assess the associations between adiposity measures. Moreover, men were categorised into tenths of BMI, body fat percentage, WC and WHR, and multivariable linear regressions (adjusted for age and height) were conducted to calculate mean values for MRI and DXA. Moreover,

Prospective analyses

Cox proportional hazards models were used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for prostate cancer death, using age as the underlying time variable. Person-years were calculated from the date of recruitment to the date of death, loss to follow-up, or the censoring date, whichever occurred first. The proportional hazards assumption was examined using time-varying covariates and Schoenfeld residuals, and this revealed no evidence of deviation. Men were categorised into fourths of adiposity measurements based on the distribution in the cohort. We also modelled HRs per predefined increments and categories of the adiposity measurements: (i) BMI [per 5 kg/m2 increase, and as predefined World Health Organization (WHO) categories [16] (< 25, 25–29.9, and ≥ 30 kg/m2)]; (ii) body fat percentage (per 5% increase); (iii) WC [per 10 cm increase, and as predefined WHO categories [17] (< 94, 94–101.9, ≥ 102 cm)]; and (iv) WHR [per 0.05 unit increase, and as predefined WHO categories [17] (< 0.90, ≥ 0.90)]. Potential nonlinear associations between the anthropometric variables and prostate cancer mortality were evaluated using likelihood ratio tests comparing the model with the anthropometric variable entered as an ordered categorical (ordinal) variable to a model with the categorical variable treated as continuous, and no evidence of non-linearity was observed.

Adjustment covariates were defined a priori based on previous analyses by our group using UK Biobank data [18]. The minimally-adjusted models were stratified by geographical region of recruitment (ten UK regions) and age (< 45, 45–49, 50–54, 55–59, 60–64, ≥ 65 years) at recruitment. The fully adjusted model was further adjusted for Townsend deprivation score (fifths, unknown [0.1%]), ethnic group (white, mixed background, Asian, black, other, and unknown [0.6%]), height (< 170, 170–174.9, 175–179.9, ≥ 180 cm, and unknown [0.2%]), lives with a wife or partner (no, yes), cigarette smoking (never, former, current 1– < 15 cigarettes per day, current ≥ 15 cigarettes per day, current but number of cigarettes per day unknown, and smoking status unknown [0.6%]), physical activity (low [0–9.9 METs/week], moderate [10–49.9 METs/week], and high [≥50 METs/week], unknown [3.6%]), alcohol consumption (non-drinkers, < 1–9.9, 10–19.9, ≥20 g ethanol/day, unknown [0.5%]), diabetes (no, yes, and unknown [0.5%]) and history of PSA testing at recruitment (no, yes, unknown [5%]).

Sensitivity analyses

We also performed the following sensitivity analyses: excluding the first 5 years of follow-up. excluding men with BMI ≥ 27.5 kg/m2, excluding men with BMI ≥ 25 kg/m2, excluding extreme values of exposure variables (percentiles outside 1–99), excluding men < 50 years of age at recruitment, running the statistical analyses per 1 standard deviation (SD) increment, using the BMI-adjusted residuals of WC (or WHR, depending on which one is the exposure of interest) by regressing these variables in a linear model and using the residuals (that are statistically independent of BMI) as the exposures of interest.

Dose-response meta-analysis

We searched on PubMed, Embase, and Web of Science for prospective studies examining the associations of BMI, body fat percentage, WC and WHR with prostate cancer as the underlying cause of death, independently by two researchers up to 15 March 2021; please see details in the Additional File 1: Supplementary Methods [2, 6, 19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38], Tables S1-S3 and Figure S2) [6, 22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38]. We excluded reviews, abstract-only publications or editorials. When the same cohort study published more than one original article looking at these associations, the paper reporting the longest follow-up time was retained.

In the dose-response meta-analysis, we calculated the HR estimates in the studies that reported results for a different increment (e.g. per 1 SD increase) or from the categorical data using generalised least-squares [39] for the increments mentioned above (details in Additional File 1: Supplementary Methods). We then pooled study-specific log HRs to obtain a summarised effect size using a fixed effects model. The I2 statistic was used to assess heterogeneity across studies, and we assessed publication bias with funnel plots and Egger’s test.

Population-attributable risk

The number of prostate cancer deaths attributable to obesity as measured by BMI (population-attributable risk (PAR)) in the UK was calculated using the number of deaths in the UK in 2019, the estimate of relative risk from our dose response meta-analysis and information on the prevalence of obesity in English men aged 55–64 years in 2019 (as a surrogate for the UK; mean BMI 28.9 kg/m2) [40].

All analyses were performed using Stata version 14.1 (Stata Corporation, College Station, TX, USA), and figures were plotted in R version 3.2.3. All tests of significance were two-sided, and P-values < 0.05 were considered statistically significant.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.


This article is autogenerated using RSS feeds and has not been created or edited by OA JF.

Click here for Source link (