Table 3.

Sensitivity analysis: Associations between RAP (defined as 3+ episodes of abdominal pain in the last year) and fasting to control weight at age 16

	Unadjusted		Adjusted Model A		Adjusted Model B		Fully adjusted Model C
	OR (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p
Persistent RAP (3+) at 7 and 9 years (complete cases, n=3001)	1.82 (1.43, 2.33)	<.001	1.49 (1.15, 1.92)	.002	1.50 (1.16, 1.94)	.002	1.50 (1.16, 1.94)	.002
Persistent RAP (3+) at 7 and 9 years (imputed dataset, n=8041)	1.53 (1.23, 1.92)	<.001	1.27 (1.00, 1.62)	.051	1.29 (1.02, 1.64)	.040	1.28 (1.01, 1.63)	.041
Analyses restricted to individuals without DAWBA anxiety diagnosis at age 7
Persistent RAP (3+) at 7 and 9 years (complete cases, n=2928)	1.85 (1.44, 2.37)	<.001	1.50 (1.16, 1.95)	<.001	1.52 (1.17, 1.97)	.001	1.52 (1.17, 1.97)	.002
Persistent RAP (3+) at 7 and 9 years (imputed dataset, n=7791)	1.58 (1.27, 1.97)	<.001	1.31 (1.05, 1.65)	<.001	1.34 (1.07, 1.67)	.011	1.33 (1.05, 1.67)	.021

Model A = adjusted for gender;

Model B = adjusted for gender, maternal social class and maternal education;

Fully adjusted Model C = adjusted for gender, maternal social class, maternal education, maternal anxiety and maternal depression

RAP = Recurrent Abdominal Pain defined as 3+ episodes of stomach pain in the last 12 months