Risk prediction with imperfect survival outcome information from electronic health records

Abstract

Readily available proxies for the time of disease onset such as the time of the first diagnostic code can lead to substantial risk prediction error if performing analyses based on poor proxies. Due to the lack of detailed documentation and labor intensiveness of manual annotation, it is often only feasible to ascertain for a small subset the current status of the disease by a follow-up time rather than the exact time. In this paper, we aim to develop risk prediction models for the onset time efficiently leveraging both a small number of labels on the current status and a large number of unlabeled observations on imperfect proxies. Under a semiparametric transformation model for onset and a highly flexible measurement error model for proxy onset time, we propose the semisupervised risk prediction method by combining information from proxies and limited labels efficiently. From an initially estimator solely based on the labeled subset, we perform a one-step correction with the full data augmenting against a mean zero rank correlation score derived from the proxies. We establish the consistency and asymptotic normality of the proposed semisupervised estimator and provide a resampling procedure for interval estimation. Simulation studies demonstrate that the proposed estimator performs well in a finite sample. We illustrate the proposed estimator by developing a genetic risk prediction model for obesity using data from Mass General Brigham Healthcare Biobank.

LATTE: Label-efficient incident phenotyping from longitudinal electronic health records

“…We next describe the construction of deep-learning models for incidence and for silver-standard label . Inspired by semi-supervised learning with a semi-parametric transformation model, 46 we let the models for gold-standard incidence and silver-standard label share the core visit representation learning component while allowing different prediction functions and : . We create in (1) a CR module to learn incident-indicative input concepts, (2) a VAN to highlight informative visits from background noises among other visits, and (3) the bidirectional gated recurrent unit (Bi-GRU) network for communication over time.…”