Mediating Roles of Psychological Distress, Insomnia, and Body Image Concerns in the Association Between Exercise Addiction and Eating Disorders

Figures & data

Table 1 Characteristics of the Study Participants (N=2088)

	Mean±SD or n (%)
Age (in years)	15.3±3.7
Sex
Males	989 (47.4)
Females	1099 (52.6)
Fathers’ years of education	9.8±4.1
Mothers’ years of education	7.4±3.9
BMI (kg/m^2)	21.1±5.3
Currently smoking (Yes)	176 (8.4%)
Psychological distress^a	24.1±8.9
Exercise addiction^b	16.9±5.6
Insomnia^c	9.6±3.4
Sleep quality^d	7.3±2.4
Body image concern^e	49.3±16.7
Eating disorder^f	25.2±10.6

Notes: ^aMeasured using Depression, Anxiety, and Stress Scale-21 (DASS-21). ^bMeasured using the Exercise Addiction Inventory youth version (EAI-Y), ^cMeasured using the Insomnia Severity Index (ISI), ^dMeasured using the Pittsburgh Sleep Quality Index (PSQI), ^eMeasured using the Body Image Concern Inventory (BICI), ^fMeasured using the Eating Attitudes Test - 26 Item (EAT-26).

Table 2 Pearson Correlation Matrix of the Variables of Interest

	1	2	3	4	5	6
1. Psychological distress^a	1	0.35	0.48	0.49	0.54	0.41
2. Exercise addiction^b		1	0.15	0.12	0.30	0.17
3. Insomnia^c			1	0.45	0.40	0.31
4. Sleep quality^d				1	0.46	0.35
5. Body image concern^e					1	0.38
6. Eating disorder^f						1

Notes: All p-values < 0.01. ^aMeasured using Depression, Anxiety, and Stress Scale-21 (DASS-21). ^bMeasured using the Exercise Addiction Inventory youth version (EAI-Y), ^cMeasured using the Insomnia Severity Index (ISI), ^dMeasured using the Pittsburgh Sleep Quality Index (PSQI), ^eMeasured using the Body Image Concern Inventory (BICI), ^fMeasured using the Eating Attitudes Test - 26 Item (EAT-26).

Table 3 Models of the Effect of Adolescents’ Exercise Addiction on Eating Disorder with Mediators of Insomnia, Sleep Quality, Psychological Distress and Body Image Concern

	Unstand. Coeff.	SE or (Bootstrapping SE)	t-value or (Bootstrapping LLCI)	p-value or (Bootstrapping ULCI)
Total effect of exercise addiction on eating disorder	0.234	0.030	7.706	0.001
Direct effect of exercise addiction on eating disorder 5th	0.096	0.028	3.440	0.001
Direct effect of exercise addiction on mediators
Psychological distress 2nd	0.382	0.055	6.934	<0.001
Insomnia 3rd	0.189	0.028	6.741	<0.001
Sleep quality 4th	0.131	0.024	5.356	<0.001
Body image concerns 1st	0.859	0.076	11.252	<0.001
Indirect effect of EAI-Y on eating disorder
Total indirect effect	0.139	(0.020)	(0.099)	(0.178)
Through psychological distress 2nd	0.036	(0.010)	(0.019)	(0.056)
Through insomnia 3rd	0.031	(0.008)	(0.017)	(0.047)
Through sleep quality 4th	0.019	(0.006)	(0.008)	(0.032)
Through body image concern 1st	0.053	(0.011)	(0.032)	(0.076)
Indirect effect comparisons
Psychological distress vs body image concern	−0.017	(0.015)	(−0.045)	(0.012)
Insomnia vs body image concern	−0.021	(0.012)	(−0.046)	(0.003)
Sleep quality vs body image concern	−0.034	(0.012)	(−0.059)	(−0.010)
Psychological distress vs insomnia	0.005	(0.012)	(−0.019)	(0.028)
Psychological distress vs sleep quality	0.017	(0.011)	(−0.004)	(0.040)
Insomnia vs sleep quality	0.012	(0.009)	(−0.004)	(0.030)

Notes: Age, gender, adolescents’ BMI, mothers’ and fathers’ years of education were adjusted for the model. 1st, 2nd, 3rd, 4th, and 5th represent the ranks in which exercise addiction is associated with the other variables.

Abbreviations: Unstand. Coeff., unstandardized coefficient; LLCI, lower limit in 95% confidence interval; ULCI, upper limit in 95% confidence interval.

Figure 1 Mediation model with psychological distress, insomnia, sleep quality, and body image concern as mediators of the association between exercise addiction and eating disorders. *P<0.05.