Trajectories of endoscopic Barrett esophagus: Chronological changes in a community-based cohort

AIM To elucidate longitudinal changes of an endoscopic Barrett esophagus(BE), especially of short segment endoscopic BE(SSBE). METHODS This study comprised 779 patients who underwent two or more endoscopies between January 2009 and December 2015. The intervals between the first and the last endoscopy were at least 6 mo. The diagnosis of endoscopic BE was based on the criteria proposed by the Japan Esophageal Society and was classified as long segment(LSBE) and SSBE, the latter being further divided into partial and circumferential types. The potential background factors that were deemed to affect BE change included age, gender, antacid therapy use, gastroesophageal reflux disease-suggested symptoms, esophagitis, and hiatus hernia. Time trends of a new appearance and complete regression were investigated by Kaplan-Meier curves. The factors that may affect appearance and complete regression were investigated by χ~2 and Student-t tests, and multivariable Cox regression analysis. RESULTS Incidences of SSBE and LSBE were respectively 21.7% and 0%, with a mean age of 68 years. Complete regression of SSBE was observed in 61.5% of initial SSBE patients, while 12.1% of initially disease free patients experienced an appearance of SSBE. Complete regressions and appearances of BE occurred constantly over time, accounting for 80% and 17% of 5-year cumulative rates. No LSBE development from SSBE was observed. A hiatus hernia was the only significant factor that facilitated BE development(P = 0.03) or hampered(P = 0.007) BE regression. CONCLUSION Both appearances and complete regressions of SSBE occurred over time. A hiatus hernia was the only significant factor affecting the BE story.

Weighted quantile regression for longitudinal data using empirical likelihood

This paper proposes a new weighted quantile regression model for longitudinal data with weights chosen by empirical likelihood(EL). This approach efficiently incorporates the information from the conditional quantile restrictions to account for within-subject correlations. The resulted estimate is computationally simple and has good performance under modest or high within-subject correlation. The efficiency gain is quantified theoretically and illustrated via simulation and a real data application.

EMPIRICAL LIKELIHOOD FOR LONGITUDINAL PARTIALLY LINEAR MODEL WITH α-MIXING ERRORS

This paper considers large sample inference for the regression parameter in a partially linear regression model with longitudinal data and a-mixing errors. The authors introduce an estimated empirical likelihood for the regression parameter and show that its limiting distribution is a mixture of central chi-squared distributions. Also, the authors derive an adjusted empirical likelihood method which is shown to have a central chi-square limiting distribution. A simulation study is carried out to assess the performance of the empirical likelihood method.