Skip to main content
. Author manuscript; available in PMC: 2024 Jul 5.
Published in final edited form as: Harv Data Sci Rev. 2024 Jan 31;6(1):10.1162/99608f92.9d86a749. doi: 10.1162/99608f92.9d86a749

Table B1.

Summary of missingness for the study population of 2,289 patients. Predictors with no missing values were not included in this table for conciseness. Predictors with greater than 30% missing data were excluded from the analysis. Missing values of predictors included in the analysis were imputed by mean (for continuous variables) or mode (for categorical variables).

Characteristic Number (%) Missing How Handled
Diastolic Blood Pressure (Invasive from Arterial Line) 1,806 (78.90%) Excluded
Systolic Blood Pressure (Invasive from Arterial Line) 1,806 (78.90%) Excluded
Religion 831 (36.30%) Excluded
Marital Status 753 (32.90%) Excluded
Temperature 436 (19.05%) Imputed
Oxygen Saturation 402 (17.56%) Imputed
Respiratory Rate 400 (17.47%) Imputed
Heart Rate 393 (17.17%) Imputed
Respiratory Support 350 (15.29%) Imputed
Diastolic Blood Pressure (Non-Invasive from Cuff) 162 ( 7.08%) Imputed
Systolic Blood Pressure (Non-Invasive from Cuff) 162 ( 7.08%) Imputed
Affluence Quartile 126 ( 5.50%) Imputed
Disadvantage Quartile 126 ( 5.50%) Imputed
Ethnic Immigration Quartile 126 ( 5.50%) Imputed
Education Quartile 126 ( 5.50%) Imputed
Body Mass Index 96 ( 4.19%) Imputed
Preferred Language1 13 ( 0.57%) Excluded
1

Preferred language was additionally excluded as a predictor, as 97% of the non-missing cases were English speaking.