Rule-based and machine learning algorithms identify patients with systemic sclerosis accurately in the electronic health record

Table 1 Characteristics of SSc cases and non-cases in the training set

Characteristics	SSc cases (n = 86)	Non-cases (n = 94)	p value¹
Age, years, mean ± standard deviation	68 ± 14	59 ± 20	< 0.01
Female, n (%)	71 (83%)	84 (89%)	0.19
White, n (%)	65 (76%)	70 (75%)	0.86
Number of counts of the SSc ICD-9² code (710.1), mean ± standard deviation	10 ± 16	2 ± 5	< 0.01
Number of counts of the SSc ICD-10-CM³ codes (M34*), mean ± standard deviation	6 ± 7	2 ± 8	< 0.01
Years of follow-up⁴, mean ± standard deviation	7 ± 6	10 ± 7	< 0.01

¹Mann-Whitney U test for continuous variables and chi-square test for categorical variables
²ICD-9 International Classification of Diseases, Ninth Revision
³ICD-10-CM International Classification of Diseases, Tenth Revision, Clinical Modification
⁴Years of data available in the electronic health record from first to last ICD-9 and/or ICD-10-CM codes for any conditions

ISSN: 1478-6362