Evaluation of Reporting Quality for Prediction Models of HBV-Related Hepatocellular Carcinoma by TRIPOD Statement

Abstract

Background: To assess the reporting quality for prediction models of hepatitis B virus (HBV) related hepatocellular carcinoma by the guideline of Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis Or Diagnosis (TRIPOD). Methods: We searched PubMed, Embase, and Web of science from inception to 1 June 2019, to identify potentially relevant scientific papers reporting original research. We assessed the reporting quality of eligible articles reporting prediction models of HBV-related hepatocellular carcinoma, according to the TRIPOD statement. Multivariate logistic regression model was used to analyze factors related to poor reporting quality. Results: A total of 38 papers were included in the final analysis. Only 10 (26.32%) out of 38 papers completed reporting for >75% items per publication. Among all the TRIPOD items, the poor reporting items were title (26.32% [10/38]), descriptions of outcome (24.24% [8/33]), predictors (15.15% [5/33]), sample size (0.00% [0/38]), missing data (i2.63% [1/38]), and risk groups (36.00% [9/25]) in the methods. The median proportion of completeness of reporting per publication was 70.56% (IQR: 63.33%-75.86%). Completeness of reporting per publication did not statistically significant in different year of publication, type of study, and study design. In the multivariate model, after adjustment of potential confounders, impact factor of published journal (OR=10.14, 95% CI: 1.66-61.84) and sample size (OR=22.77, 95% CI: 1.10-472.57) was independently associated with sufficient reporting. Conclusions: The reporting quality of the overall and some key items were not sufficient in studies focusing on developing and external validation of prediction models for HBV-related HCC development or prognosis. More attentions should be paid on the complete and transparent reporting on prediction models. Funding Statement: This article is funded by National Key Research and Development Project of China (No. 2018YFC1704400 and 2019YFC1710300) and the National Natural Science Foundation (No. 71874003, No. 71934002 and No. 81703240) and the Interdisciplinary Medicine Seed Fund of Peking University (No.BMU2018MX016). Declaration of Interests: The authors stated: None declared. Ethics Approval Statement: Not required.