Comparison of deep learning with regression analysis in creating predictive models for SARS-CoV-2 outcomes

Table 1 Summary of demographics, comorbidities, symptoms, and outcomes of 398 patients admitted with SARS-CoV-2 in a West London population, during March 1–April 23, 2020

	Number (%)	Chi-Square	p value
Demographics
Age (mean)	63.2 years
Sex (M)	223 (56.0%)	3.54
Comorbidities
Cardiac failure	22 (5.5%)	79.56	< 0.005
Cerebrovascular event	29 (7.3%)	24.14	< 0.005
Chronic kidney disease	33 (8.3%)	49.25	< 0.005
Chronic liver disease	6 (1.5%)	5.915	0.015
Chronic lung disease	84 (21.1%)	6.01	0.01
Diabetes	104 (26.1%)	4.39	0.036
Hypertension	147 (36.9%)	17.88	< 0.005
Ischaemic heart disease	47 (11.8%)	37.63	< 0.005
Obesity	15 (3.8%)	0.015	0.903
Symptoms
Abdominal pain	40 (10.1%)	2.6	0.11
Collapse	37 (9.3%)	55.5	< 0.005
Confusion	59 (14.8%)	117.35	< 0.005
Cough	247 62.1%)	3.05	0.081
Diarrhoea and vomiting	105 (26.4%)	4.64	0.031
Dyspnoea	223 (56.0%)	15.88	< 0.005
Fever	216 (54.3%)	0.12	0.73
Myalgia	68 (17.1)	0.002	0.97
Olfactory change	36 (9.0%)	1.11	0.29
Length of stay (median)	5 days
Number of days of symptoms prior to admission (median)	7 days
Outcome
Death	95 (23.9%)

The association of each predictor with death following log rank analysis (reported with the Chi-square statistic) is shown

ISSN: 1472-6947