Using an optimized generative model to infer the progression of complications in type 2 diabetes patients

Abstract

Background: People live a long time in pre-diabetes/early diabetes without a formal diagnosis or management. Heterogeneity of progression coupled with deficiencies in electronic health records related to incomplete data, discrete events, and irregular event intervals make identification of pre-diabetes and critical points of diabetes progression challenging.

Methods: We utilized longitudinal electronic health records of 9298 patients with type 2 diabetes or prediabetes from 2005 to 2016 from a large regional healthcare delivery network in China. We optimized a generative Markov-Bayesian-based model to generate 5000 synthetic illness trajectories. The synthetic data were manually reviewed by endocrinologists.

Results: We build an optimized generative progression model for type 2 diabetes using anchor information to reduce the number of parameters learning in the third layer of the model from [Formula: see text] to [Formula: see text], where [Formula: see text] is the number of clinical findings, [Formula: see text] is the number of complications, [Formula: see text] is the number of anchors. Based on this model, we infer the relationships between progression stages, the onset of complication categories, and the associated diagnoses during the whole progression of type 2 diabetes using electronic health records.

Discussion: Our findings indicate that 55.3% of single complications and 31.8% of complication patterns could be predicted early and managed appropriately to potentially delay (as it is a progressive disease) or prevented (by lifestyle modifications that keep patient from developing/triggering diabetes in the first place).

Conclusions: The full type 2 diabetes patient trajectories generated by the chronic disease progression model can counter a lack of real-world evidence of desired longitudinal timeframe while facilitating population health management.

Keywords: Computer simulation; Diabetes mellitus, type 2; Disease progression model; Electronic health records; Probabilistic generative model.

Fig. 1

The outline of Wang et al.’s model, where $K$ is the number of disease stages, $M$ is the number of complications, and $N$ is the number ICD codes

Fig. 2

An illustration of before and after optimizing the observable bottom layer

Fig. 3

A Comparison of complications of 5000 virtual patients learned by our optimized model (a) as well as two representative patients retrieved by our medical experts (b & c). Note that for (a), we first use 35,210 encounters with 64,383 positive observations to learn our generative model. We next use this learned generative model to generate 5000 (that is a specified number) synthetic patient trajectories. Thus, 8.5, 13.0 19.3, 23.9, etc., were the mean progression years at stages I to V, respectively