LAUSR

1 Wijeysundera DN, Pearse RM, Shulman MA, et al. Assessment of functional capacity before major non-cardiac surgery: an international, prospective cohort study. Lancet 2018; 391: 2631–40. 2 Barker L, Brown C. Logistic regression when binary predictor variables are highly correlated. Stat Med 2001; 20: 1431–42. sensible factors typically considered during preoperative assessment” and should arguably be part of any model for predicting 30-day death or myocardial infarction and excluding them might produce misleading re sults. Further, the sequential inclusion of highly correlated variables (eg, Duke Activity Status Index [DASI] and plasma concentrations of N-terminal pro-B-type natriuretic peptide [NT-pro-BNP]) might further reduce the predictive power of the regression model by increasing the mean square error of the estimates. Second, patients were categorised as having poor functional capacity when anaesthesiologists were “uncertain about patients’ functional capacity, typically due to preexisting conditions such as arthritis or peripheral arterial disease”. Patients scheduled for orthopaedic or vascular surgery constituted over a quarter of participants in the study, and this categorisation approach could have resulted in substantial misclassification of their functional capacity. This assumption is supported by the investigators’ data on the main study outcomes within strata defined by subjectively assessed functional capacity. Although the proportion of patients categorised as having moderate or good functional capacity who had events was commensurate with perceived risk, the proportion in those classified as having poor functional capacity was unexpectedly low. It is unclear whether excluding these patients might have produced different results. Moreover, the objectively assessed measures of functional capacity (DASI and NT-pro-BNP) were included as continuous variables, thereby increasing statis tical power, but without discrete categorisation (low, moderate, or good functional capacity), they are not useful in the clinical setting. Given these limitations, the results of this study should be considered inconclusive. It might be premature 3 Hooper L, Anderson AS, Birch J, et al. Public awareness and healthcare professional advice for obesity as a risk factor for cancer in the UK: a cross-sectional survey. J Public Health (Oxf) 2018; 40: 797–805. 4 Hidayat K, Du X, Shi BM. Body fatness at a young age and risks of eight types of cancer: systematic review and meta-analysis of observational studies. Obes Rev 2018; 19: 1385–94. 5 BBC Bitesize. GCSE AQA. Non-communicable diseases. https://www.bbc.com/bitesize/ guides/z372ng8/revision/2 (accessed March 18, 2019). 6 AQA. GCSE physical education. https://www.aqa.org.uk/subjects/physicaleducation/gcse/physical-education-8582 (accessed March 18, 2019). 7 Council for the Curriculum, Examinations and Assessment. Life and health sciences. http://www.rewardinglearning.org.uk/ microsites/life_health_sciences/gce/ specification/ (accessed March 18, 2019). 8 Council for the Curriculum, Examinations and Assessment. Home economics. http://www. rewardinglearning.org.uk/microsites/home_ economics/ (accessed March 18, 2019). 9 Pearson Edexcel. A levels. Biology A Salters-Nuffield from 2015. https:// qualifications.pearson.com/en/qualifications/ edexcel-a-levels/biology-a-2015.html (accessed March 18, 2019). 10 BBC Bitesize. GCSE OCR gateway. Cancer and cardiovascular disease—non-communicable. https://www.bbc.com/bitesize/guides/ ztqcmsg/revision/1 (accessed March 18, 2019). 11 BBC Bitesize. GCSE WJEC. Health and well being. https://www.bbc.com/bitesize/guides/ z3shycw/revision/3 (accessed March 18, 2019). 12 Renehan AG, Zwahlen M, Egger M. Adiposity and cancer risk: new mechanistic insights from epidemiology. Nat Rev Cancer 2015; 15: 484–98.