Yin Cao

Background: Risk factors contributing to more than 10-fold increase in esophageal cancer in the last 50 years remain underexplored. We aim to examine the associations of sleep behaviors with esophageal adenocarcinoma (EAC) and squamous cell carcinoma (ESCC). Methods: We prospectively assessed the associations between sleep behaviors (chronotype, duration, daytime napping, daytime sleepiness, snoring, and insomnia) and EAC and ESCC risk in 393,114 participants in the UK Biobank (2006–2016). Participants with 0, 1, and ≥2 unhealthy behaviors, including sleep <6 or >9 h/d, daytime napping, and usual daytime sleepiness were classified as having a good, intermediate, and poor sleep. For EAC, we also examined interactions with polygenic risk score (PRS). Cox models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI). Results: We documented 294 incident EAC and 95 ESCC. Sleep >9 h/d (HR, 2.05; 95% CI, 1.18–3.57) and sometimes daytime napping (HR, 1.36; 95% CI, 1.06–1.75) were individually associated with increased EAC risk. Compared with individuals with good sleep, those with intermediate sleep had a 47% (HR, 1.47; 95% CI, 1.13–1.91) increased EAC risk, and those with poor sleep showed an 87% (HR, 1.87; 95% CI, 1.24–2.82) higher risk (Ptrend < 0.001). The elevated risks for EAC were similar within strata of PRS (Pinteraction = 0.884). Evening chronotype was associated with elevated risk of ESCC diagnosed after 2 years of enrollment (HR, 2.79; 95% CI, 1.32–5.88). Conclusions: Unhealthy sleep behaviors were associated with an increased risk of EAC, independent of genetic risk. Impact: Sleep behaviors may serve as modifiable factors for the prevention of EAC.

Background Racial/ethnic minority children and adolescents are more likely to have an advanced cancer diagnosis compared with non‐Hispanic Whites, which may relate to the lack of consistent health care access. This study aims to describe racial/ethnic disparities in cancer diagnosis stage among children and adolescents and assess whether health insurance mediates these disparities. Methods Data on individuals ≤19 years of age diagnosed with primary cancers from 2007 to 2016 were obtained from the Surveillance, Epidemiology, and End Results 18 database. Prevalence ratios (PRs) and 95% confidence intervals (CIs) for the association between race/ethnicity and cancer diagnosis stage were calculated using Poisson regression. Analyses addressing health insurance as a potential mediator were also performed. Results Compared with non‐Hispanic Whites, racial/ethnic minorities had a higher prevalence of a distant cancer diagnosis, with PRs of 1.31 (95% CI, 1.23–1.40) for non‐Hispanic Blacks, 1.14 (95% CI, 1.04–1.24) for non‐Hispanic Asian/Pacific Islanders, and 1.15 (95% CI, 1.09–1.21) for Hispanics. These associations were attenuated when adjusting for health insurance, with PRs of 1.24 (95% CI, 1.16–1.33) for non‐Hispanic Blacks, 1.11 (95% CI, 1.02–1.21) for non‐Hispanic Asian/Pacific Islanders, and 1.07 (95% CI, 1.01–1.13) for Hispanics. Any Medicaid or no insurance at diagnosis mediated 49%, 22%, and 9% of the observed association with distant stage in Hispanics, non‐Hispanic Blacks, and non‐Hispanic Asian/Pacific Islanders, respectively. Conclusions Disparities in cancer diagnosis stage in racial/ethnic minority children and adolescents may be partially explained by health insurance coverage. Further research is needed to understand the mechanisms.