Table 3.

Multivariable mixed models results for predictors of body weight controlling for selected demographic, behavioral, psychosocial and environmental factors, by gender, Seasonal Variation of Blood Cholesterol Study, Worcester, Massachusetts, 1994–1998 (N = 572)

	Men				Women
	Coefficient	s.e.	95% Confidence interval		Coefficient	s.e.	95% Confidence interval
Demographic
Age, cross-sectional effect	−0.0004	0.07	−0.14	0.13	0.03	0.08	−0.14	0.19
Age, longitudinal effect	0.10	0.21	−0.31	0.51	−0.20	0.20	−0.59	0.19
Education
Education: high school or vocational (less than high school as referent)	−1.24	4.07	−9.22	6.73	−6.10	3.83	−13.61	1.40
Education: some college or associate’s degree (less than high school as referent)	−0.97	3.96	−8.74	6.79	−8.52	3.94	−16.24	−0.80
Education: college or graduate degree (less than high school as referent)	−6.19	4.08	−14.17	1.80	−11.91	4.02	−19.80	−4.03
Anthropometric
Height (cm)	0.84	0.12	0.61	1.08	0.76	0.13	0.50	1.01
Lifestyle factors
Dietary intake
Daily caloric intake, cross-sectional effect	0.001	0.002	−0.003	0.005	0.01	<0.001	<0.001	0.01
Daily caloric intake, longitudinal effect	0.0002	0.0002	−0.0003	0.0003	0.0006	0.0002	0.0002	0.001
% of calories from fat, cross-sectional effect	0.21	0.19	−0.16	0.57	0.15	0.33	−0.49	0.79
% of calories from fat, longitudinal effect	−0.01	0.02	−0.05	0.04	−0.01	0.04	−0.08	0.06
% of calories from carbohydrates, cross-sectional effect	0.03	0.16	−0.28	0.34	−0.09	0.29	−0.66	0.48
% of calories from carbohydrates, longitudinal effect	−0.03	0.02	−0.07	0.02	−0.02	0.03	−0.09	0.04
% of calories from protein, cross-sectional effect	0.84	0.34	0.17	1.51	0.63	0.50	−0.34	1.61
% of calories from protein, longitudinal effect	−0.06	0.03	−0.12	0.00	−0.02	0.04	−0.10	0.06
Dietary fiber intake (g/day), cross-sectional effect	−0.25	0.17	−0.58	0.09	−0.17	0.27	−0.70	0.36
Dietary fiber intake (g/day), longitudinal effect	0.00	0.02	−0.03	0.03	−0.02	0.02	−0.06	0.01
Alcohol consumption (non-drinker as referent)	0.16	0.26	−0.34	0.66	−0.35	0.30	−0.93	0.23
Physical activity in METs-h/da
Leisure-time physical activity, cross-sectional effect	−0.84	0.38	−1.59	−0.10	−1.10	0.54	−2.15	−0.05
Leisure-time physical activity, longitudinal effect	0.00	0.02	−0.05	0.05	−0.05	0.03	−0.10	−0.002
Occupational physical activity, cross-sectional effect	−0.30	0.14	−0.57	−0.02	−0.12	0.27	−0.41	0.65
Occupational physical activity, longitudinal effect	0.00	0.01	−0.03	0.02	−0.02	0.02	−0.06	0.01
Household physical activity, cross-sectional effect	−0.13	0.22	−0.57	0.30	0.33	0.33	−0.31	0.97
Household physical activity, longitudinal effect	−0.02	0.01	−0.04	0.01	−0.02	0.02	−0.06	0.01
Current Smoker (non-smoker as referent)	−1.23	0.38	−1.98	−0.48	−1.27	0.59	−2.43	−0.10
Psychosocial factorsb
Beck anxiety score, cross-sectional effect	0.15	0.26	−0.35	0.66	0.22	0.24	−0.24	0.69
Beck anxiety score, longitudinal effect	0.08	0.02	0.03	0.12	0.01	0.02	−0.02	0.04
Beck depression score, cross-sectional effect	0.19	0.22	−0.25	0.62	0.20	0.21	−0.21	0.63
Beck depression score, longitudinal effect	0.00	0.02	−0.04	0.04	−0.02	0.02	−0.06	0.02
Season of the year
Spring season (winter as referent)	−0.40	0.15	−0.68	−0.11	−0.18	0.13	−0.44	0.08
Summer season (winter as referent)	−0.59	0.19	−0.96	−0.22	−0.34	0.17	−0.67	−0.02
Fall season (winter as referent)	−0.30	0.15	−0.59	−0.01	−0.29	0.13	−0.56	−0.03
Constant	−74.24	26.26	−125.71	−22.78	−62.96	37.90	−137.24	11.32

The final model included the following variables: age (years), education category (with high school or less as the referent), marital status (with married or living with partner as the referent), height (cm), total caloric intake (kcal per day), percentage of calories from fat, percentage of calories from carbohydrates, percentage of calories from protein, total fiber intake (grams per day), alcohol consumption category (with non drinker as the reference), physical activity (in four domains: leisure-time, occupational, household and total; in MET/h/day), smoking status (with non smoker as the referent), Beck Anxiety Inventory score, Beck Depression Inventory score, and season of the year.

^aMETs = metabolic equivalents; i.e., one MET equals energy expenditure at rest.

^bIf only anxiety or depression scores were included separately in the final model; in both of these models there was a significant relationship between either anxiety or depression with baseline body weight. However, when both anxiety and depression scores were included together in the model, there was no cross-sectional relationship between ether depression or anxiety, although anxiety remained as a significant predictor of weight gain over time, but only among men. There was a significant correlation between anxiety and depression scores in this study population r² = 0.81.