Table 2.

Relationship between Household Material Hardships and Internal Locus of Control over the Prevention of Child Obesity (LOC-PCO)

Material Hardships		Internal Locus of Control (Continuous Score)
				Model 1a		Model 2b
		Mean (SD)	p-value	Betac	95% CI	Beta	95% CI
Food insecure	Yes No	18.52 (1.85) 19.21 (1.54)	<.001	.20	.07 to .17	.16	.04 to .15
Difficulty paying bills	Yes No	18.88 (1.78) 19.05 (1.61)	.41	−.04	−.08 to .03	−.04	−.08 to .03
Housing disrepair	Yes No	18.83 (1.76) 19.09 (1.60)	.07	.06	−.02 to .08	.05	−.02 to .08
Neighborhood stress	Yes No	18.46 (1.86) 19.05 (1.63)	.003	.09	.003 to .17	.08	−.01 to .16
		Low Internal Locus of Control (Categorical Variable)
		n (%) with low LOC	p-value	AOR	95% CI	AOR	95% CI
Food insecure	Yes No	64 (38.1) 67 (17.7)	<.001	2.75	1.77 – 4.27	2.38	1.50 – 3.77
Difficulty paying bills	Yes No	42 (27.6) 90 (22.1)	.18	.82	.51 – 1.32	.79	.48 – 1.30
Housing disrepair	Yes No	55 (29.4) 77 (20.7)	.03	1.45	.95 – 2.21	1.43	.92 – 2.22
Neighborhood stress	Yes No	18 (37.5) 114 (22.3)	.03	1.75	.90 – 3.40	1.60	.80 – 3.18

^aModel 1 uses regression models with all four material hardships (food insecurity, difficulties paying bills, housing disrepair and neighborhood stress) entered simultaneously into model.

^bModel 2 uses regression models with all four material hardships entered simultaneously into model as well as additional potential confounders including US born, maternal education, marital status, working status, first child, depressive symptoms, and pre-pregnancy BMI ≥ 25.

^cMultiple linear regressions were performed utilizing a log-transformation of LOC-PCO scores that had been reversed to account for left skewing. Higher Betas represent lower LOC-PCO scores.