Parents and adolescents preferences for asthma control: a best-worst scaling choice experiment using an orthogonal main effects design

Table 4 Regression coefficients representing preferences of parents and adolescents for attributes of asthma control

Attribute	Parents (n = 52)				Adolescents (n = 44)
Attribute	Beta	Lower CI	Upper CI	P-value	Beta	Lower CI	Upper CI	P-value
Night Time Symptoms
None	2.09	1.64	2.55	<0.00001	1.91	1.40	2.42	<0.00001
3 days per week	−0.18	−0.61	0.24	0.40	−0.25	−0.73	0.23	0.32
5 days per week	−0.90	−1.33	−0.46	<0.00001	−1.15	−1.62	−0.69	<0.00001
Wheezing or tightening of chest
None	1.88	1.42	2.34	<0.00001	1.43	0.92	1.94	<0.00001
Manageable	0.68	0.23	1.12	0.00290	0.55	0.05	1.04	0.03000
Bothersome	−0.87	−1.29	−0.45	<0.00001	−1.15	−1.61	−0.70	<0.00001
Changing Medication
None	1.02	0.58	1.47	<0.00001	1.26	0.76	1.77	<0.00001
More doses/add medication	0.11	−0.33	0.56	0.62	−0.03	−0.52	0.47	0.92
Add oral steroids for 5 days	−0.20	−0.66	0.26	0.39	0.02	−0.49	0.53	0.94
Emergency room visits
None	1.82	1.37	2.28	<0.00001	1.90	1.39	2.41	<0.00001
4 per year	−1.14	−1.55	−0.73	<0.00001	−1.17	−1.62	−0.71	<0.00001
10 per year	−2.15	−2.58	−1.73	<0.00001	−1.83	−2.28	−1.39	<0.00001
Physical activity limitations
None	1.31	0.85	1.77	<0.00001	2.17	1.66	2.68	<0.00001
2 per month	−0.14	−0.60	0.32	0.54	−0.29	−0.78	0.21	0.26
10 per month	−1.14	--	--	--	−1.97	--	--	--

Conditional logit regression was used to analyse the choice data. The coefficients for all attribute levels were estimated relative to the reference level of 10 physical activity limitations per month. The estimated coefficients represent respondents’ average preference weights for a given attribute level relative to the reference level. R²(0) Goodness of fit for model with no intercept 0.2591 for parent model and 0.2804 for adolescent model