Impact of missing values on the ability of the acute physiology and chronic health evaluation III and Japan risk of death models to predict mortality

Katsura Hayakawa; Shigehiko Uchino; Hideki Endo; Kazuki Hasegawa; Kazuya Kiyota

doi:10.1016/j.jcrc.2023.154432

Impact of missing values on the ability of the acute physiology and chronic health evaluation III and Japan risk of death models to predict mortality

J Crit Care. 2024 Feb:79:154432. doi: 10.1016/j.jcrc.2023.154432. Epub 2023 Sep 22.

Authors

Katsura Hayakawa¹, Shigehiko Uchino², Hideki Endo³, Kazuki Hasegawa⁴, Kazuya Kiyota⁴

Affiliations

¹ Department of Intensive Care Medicine, Toranomon Hospital, 2-2-2 Toranomon, Minato-ku, Tokyo 105-8470, Japan; Department of Emergency and Critical Care Medicine, Saitama Red Cross Hospital, 1-5 Shintoshin, Chu-o-ku, Saitama 330-8553, Japan. Electronic address: gene1982jp@gmail.com.
² Department of Anesthesiology and Intensive Care, Saitama Medical Center, Jichi Medical University, 1-847 Amanuma-cho, Omiya-ku, Saitama 330-0834, Japan.
³ Department of Healthcare Quality Assessment, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan.
⁴ Department of Emergency and Critical Care Medicine, Saitama Red Cross Hospital, 1-5 Shintoshin, Chu-o-ku, Saitama 330-8553, Japan.

PMID: 37742518
DOI: 10.1016/j.jcrc.2023.154432

Abstract

Purpose: This study assessed model performance of the Acute Physiology and Chronic Health Evaluation (APACHE) III and Japan Risk of Death (JROD) when degraded by the number and category of missing variables. We also examined the impact of missing data on predicted mortality for facilities with missing physiological variables.

Methods: We obtained data from the Japanese Intensive care PAtient Database (JIPAD). We calculated observed and predicted mortality rates using the APACHE III and JROD and the standardized mortality ratio (SMR) by the number and category of missing variables. Smoothed spline curves were calculated for the SMR to the missing proportion of the facility.

Results: A total of 61,357 patients from 57 ICUs were included between April 2015 and March 2019. The APACHE III and JROD SMRs increased as the number of missing values increased. The SMR in the APACHE III model was elevated in facilities with a larger proportion of missing in each of the APS categories, arterial blood gas, albumin, glucose, and bilirubin. Facilities with a high proportion of missing albumin data preserved their SMRs in only the JROD model.

Conclusion: An increased number of missing physiological variables resulted in falsely low predicted mortality rates and high SMRs.

Keywords: Intensive care unit; Missing; Risk prediction model; Standardized mortality ratio.

MeSH terms

APACHE
Albumins*
Hospital Mortality
Humans
Intensive Care Units*
Japan / epidemiology

Substances

Albumins