Effects of autism spectrum disorders on outcome in teenage-onset anorexia nervosa evaluated by the Morgan-Russell outcome assessment schedule: a controlled community-based study

Table 5 Morgan-Russell subscale C (mental state); dose–response analysis

Dose	AN study 2				AN study 3				AN study 4				Analysis for time trend doubly ordered r × c contingency tables Linear-by-linear association test	Exact Monte Carlo
Dose	Response category				Response category				Response category				Test Statistic	P (99% CI)
	0	4	8	12	0	4	8	12	0	4	8	12
Never ASD	0	4	12	18	0	5	9	20	0	6	8	20	0	1 (0.9995 to 1)
ASD × 1-3	0	3	1	6	0	1	4	5	0	2	3	5	0	1 (0.9995 to 1)
ASD × 4	1	3	0	2	1	3	1	1	2	1	3	0	−0.561	0.69 (0.67 to 0.7)
Dose–response analysis
Doubly ordered r × c contingency tables	Linear-by-linear association test				Linear-by-linear association test				Linear-by-linear association test				Stratified r × c contingency tables Cochran-Mantel-Haenszel test
Test statistic	−1.452				−2.56				−2.787				−4.342
Exact Monte Carlo P	0.15				0.008				0.007				0.0001
99% CI	0.14 to 0.16				0.005 to 0.01				0.005 to 0.009				0 to 0.0004

AN anorexia nervosa, AN study 2 second anorexia nervosa examination, AN study 3 third anorexia nervosa examination, AN study 4 fourth anorexia nervosa examination, ASD autism spectrum disorders, never ASD individuals who had never been assigned an ASD diagnosis, ASD × 1–3 those who had been assigned an ASD diagnosis in at least one study, but not at all four studies, ASD × 4 those with an ASD diagnosis at all four examinations, CI confidence interval, r × c rows and columns.

ISSN: 2040-2392