Table 2.

Association between allergic disease severity and speech disorder in NSCH 2007/8

	History of speech disorder
	No (n = 77,196)	Yes (n = 2676)
Allergic disease severity	Freq	Freq	% Prevalence	Crude OR (95% CI)	p-value	Adjusted OR (95% CI)	p-value
Current asthma
None	70,106	2177	3.3 (3.0–3.6)	1.00	–	1.00	–
Mild	5129	286	6.3 (5.0–7.7)	1.99 (1.56–2.55)	<0.0001	1.58 (1.20–2.08)	0.001
Moderate	1525	142	11.8 (7.6–15.9)	3.93 (2.60–5.92)	<0.0001	2.29 (1.54–3.41)	<0.0001
Severe	251	55	28.6 (14.2–43.1)	11.82 (5.79–24.12)	<0.0001	5.70 (2.36–13.78)	0.0001
Hay fever
None	62,525	1893	3.3 (3.0–3.7)	1.00	–	1.00	–
Mild	10,239	453	4.9 (3.9–5.9)	1.50 (1.18–1.90)	0.0009	1.39 (1.07–1.79)	0.01
Moderate	3762	257	8.4 (6.4–10.4)	2.67 (2.02–3.53)	<0.0001	2.08 (1.46–2.97)	<0.0001
Severe	474	65	10.6 (6.6–14.6)	3.44 (2.23–5.30)	<0.0001	1.54 (0.90–2.66)	0.12
Food allergy
None	73,563	2370	3.5 (3.2–3.8)	1.00	–	1.00	–
Mild	1782	124	5.3 (3.7–6.8)	1.52 (1.09–2.11)	0.01	1.21 (0.83–1.75)	0.32
Moderate	992	102	16.9 (9.7–24.1)	5.59 (3.32–9.39)	<0.0001	3.48 (1.97–6.16)	<0.0001
Severe	738	71	10.9 (5.8–15.9)	3.33 (1.96–5.68)	<0.0001	2.02 (1.03–3.97)	0.04

Binary logistic regression models were constructed with history of speech disorder as the binary dependent variable. The independent (explanatory) variable was severity of asthma, hay fever, or food/digestive allergy (mild/moderate/severe). Prevalence odds ratios (OR) and 95% confidence intervals (95% CI) were determined. Multivariate logistic regression models were created that included allergic disease severity, history of eczema, age, gender, race/ethnicity, Hispanic origin, household income, birthplace in the United States, highest level of education in the household, insurance coverage, and outpatient healthcare utilization in the past year (all categorical) as independent variables. Adjusted prevalence odds ratios (aOR) and 95% confidence intervals (95% CI) were determined. Bold values indicate P < 0.05.