TABLE 3.

Meal and snack consumption self-reported in a 24-h recall from National Health and Nutrition Examination Survey (NHANES) I to NHANES 1999–2002¹

	NHANES I	NHANES II	NHANES III	NHANES 1999–2002	P for trend
Reported breakfast (%)
All	89 ± 0.6	88 ± 0.6	82 ± 0.8	82 ± 0.6	< 0.0001
Men	88 ± 0.8	87 ± 0.7	80 ± 1.0	80 ± 0.7
Women	90 ± 0.7	88 ± 0.8	84 ± 0.8	84 ± 1.0
Energy from breakfast (%)2,3
All	17.0 ± 0.2	16.5 ± 0.2	16.2 ± 0.2	16.6 ± 0.2	0.3
Men	16.8 ± 0.3	16.4 ± 0.3	16.0 ± 0.3	15.6 ± 0.3	0.006
Women	17.2 ± 0.3	16.7 ± 0.3	16.4 ± 0.2	17.5 ± 0.4	0.6
Energy from evening foods (%)2–4
All	45.9 ± 0.3	47.1 ± 0.3	46.0 ± 0.4	44.2 ± 0.4	< 0.0001
Men	45.8 ± 0.4	47.1 ± 0.4	46.5 ± 0.5	46.0 ± 0.6	0.9
Women	46.0 ± 0.4	47.0 ± 0.4	45.6 ± 0.4	42.5 ± 0.4	< 0.0001
Reported snack consumption (%)2,3
All	90 ± 0.4	88 ± 0.4	87 ± 0.6	86 ± 0.6	< 0.0001
Men	91 ± 0.7	89 ± 0.5	87 ± 0.9	86 ± 0.8	< 0.0001
Women	89 ± 0.6	87 ± 0.6	88 ± 0.7	87 ± 0.9	0.2
Number of snacking episodes2,3
All	2.3 ± 0.04	2.2 ± 0.04	2.3 ± 0.04	2.1 ± 0.04	0.001
Men	2.5 ± 0.05	2.4 ± 0.05	2.3 ± 0.06	2.2 ± 0.04	< 0.0001
Women	2.2 ± 0.04	2.1 ± 0.04	2.2 ± 0.04	2.1 ± 0.04	0.9
Energy from snacks (%)2,3
All	19.7 ± 0.3	20.0 ± 0.3	21.5 ± 0.3	20.8 ± 0.2	0.001
Men	20.4 ± 0.5	20.6 ± 0.3	21.8 ± 0.5	21.0 ± 0.4	0.2
Women	19.2 ± 0.3	19.4 ± 0.3	21.3 ± 0.3	20.6 ± 0.5	0.007
Snack foods and beverages (g/snacking episode)5
All	289 ± 5	333 ± 7	328 ± 4	303 ± 5	0.4
Men	338 ± 8	400 ± 11	392 ± 7	364 ± 8
Women	244 ± 4	270 ± 7	266 ± 4	244 ± 5
Energy from snacks (kcal/snacking episode)5
All	185 ± 4	199 ± 3	230 ± 4	234 ± 4	< 0.0001
Men	221 ± 6	247 ± 5	279 ± 6	277 ± 4
Women	152 ± 3	154 ± 3	184 ± 3	192 ± 4

¹Estimates were derived from regression models with each variable in the table as a continuous or binary outcome; independent variables were sex (in models for all), age, age², race (white, black, other), education (<12 y, 12 y, >12 y), smoking status (never, former, current), any weekly leisure-time physical activity (yes, no), self-reported chronic diseases (diabetes, heart disease, hypertension) (yes, no), BMI (continuous), and survey (NHANES I, NHANES II, NHANES III, NHANES 1999–2002) as trend. The estimates are adjusted $\overline{x}\pm \text{SE}$ or % ± SE from weighted regression models that included respondents with complete covariate information (total n = 37 530; 17 173 men and 20 357 women).

²Interaction of sex by survey was significant, P < 0.05. Sex-specific P values are not given for variables for which the sex-by-survey interaction was not significant.

³Population average includes those who reported no breakfast, no evening food intake, or no snack.

⁴Evening foods were those reported eaten at or after 1700.

⁵Limited to those with complete covariate information among snack reporters (total n = 31 869; 14 540 men and 17 329 women).