Risk prediction of bladder cancer among person with diabetes: A derivation and validation study

Aims This study aimed to devise and validate a clinical scoring system for risk prediction of bladder cancer to guide urgent cystoscopy evaluation among people with diabetes. Methods People with diabetes who received cystoscopy from a large database in the Chinese population (2009–2018). We recruited a derivation cohort based on random sampling from 70% of all individuals. We used the adjusted odds ratios (aORs) for independent risk factors to devise a risk score, ranging from 0 to 5: 0–2 ‘average risk’ (AR) and 3–5 ‘high risk’ (HR). Results A total of 5905 people with diabetes, among whom 123 people with BCa were included. The prevalence rate in the derivation (n = 4174) and validation cohorts (n = 1731) was 2.2% and 1.8% respectively. Using the scoring system constructed, 79.6% and 20.4% in the derivation cohort were classified as AR and HR respectively. The prevalence rate in the AR and HR groups was 1.57% and 4.58% respectively. The risk score consisted of age (18–70: 0; >70: 2), male sex (1), ever/ex‐smoker (1) and duration of diabetes (≥10 years: 1). Individuals in the HR group had 3.26‐fold (95% CI = 1.65–6.44, p = 0.025) increased prevalence of bladder than the AR group. The concordance (c‐) statistics was 0.72, implying a good discriminatory capability of the risk score to stratify high‐risk individuals who should consider earlier cystoscopy. Conclusions The risk prediction algorithm may inform urgency of cystoscopy appointments, thus allowing a more efficient use of resources and contributing to early detection of BCa among people planned to be referred.