Table 2.

Sociodemographic and clinical characteristics for subgroups of asthma by NLP-PAC and NLP-API.

	NLP-PAC+/ NLP-API+ (n=1614)	NLP-PAC+ only (n=954)	NLP-API+ only (n=105)	NLP-PAC−/ NLP-API− (n=5523)	Total (n=8196)	P value
Age at the last follow-up date, years, mean (SD)	12.2 (3.1)	12.2 (3.1)	11.8 (3.2)	11.6 (3.2)	11.8 (3.2)	<0.001
Male, n (%)	959 (59)	521 (55)	58 (55)	2664 (48)	4202 (51)	<0.001
Ethnicity, n (%)						0.007
White	1289 (80)	779 (82)	77 (73)	4414 (80)	6559 (80)
Black	93 (6)	53 (6)	8 (8)	244 (4)	398 (5)
Hispanic	50 (3)	27 (3)	7 (7)	233 (4)	317 (4)
Asian	69 (4)	26 (3)	5 (5)	270 (5)	370 (5)
Others	101 (6)	60 (6)	7 (7)	314 (6)	482 (6)
HOUSES* at birth in the lowest quartile, n (%)	343 (21)	191 (20)	23 (22)	971 (18)	1528 (19)	0.004
Overweight†, n (%)	281 (17)	179 (19)	20 (19)	N/A	480 f (18)	0.62‡
Maternal smoking history during pregnancy, n (%)	131 (8)	88 (9)	12 (11)	105 (2)	336 (4)	0.52
Family history of asthma, n (%)	586 (36)	135 (14)	36 (34)	826 (15)	1583 (19)	<0.001
Well-child visit per year, mean (SD)	0.95 (0.31)	0.90 (0.29)	0.96 (0.36)	0.93 (0.36)	0.93 (0.34)	<0.001

The percentage of each variable was calculated with the number of each group (column %).

*HOUSES: individual-level housing-based socioeconomic status measure in quartile.

†Overweight at asthma index date (for children age 2 years or more, body mass index-for-age at or above 85% used (https://www.cdc.gov/obesity/childhood/defining.html)); for those age less than 2 years, weight-for-lengths at or above 95%; (https://www.cdc.gov/nccdphp/dnpao/growthcharts/who/using/).

‡Overweight for NLP-PAC⁻/NLP-API⁻ group is not available; p values were calculated among asthmatic group, n=2673 (NLP-PAC⁺/NLP-API⁺, NLP-PAC⁺ only and NLP-API⁺ only).

NLP, natural language processing; NLP-API, NLP algorithms for Asthma Predictive Index; NLP-PAC, NLP algorithms for Predetermined Asthma Criteria.