OBJECTIVE

To (i) identify the major temporal patterns of energy intake among adults; (ii) examine the association between employment status and the patterns; and (iii) examine the association between dietary quality and the patterns.

DESIGN

Secondary analysis based on the cross-sectional population-based nutrition survey in Taiwan, 2005-2008. Based on energy intake levels at six time intervals of a day derived from 24 h recall data, we applied cluster analysis to identify major temporal patterns of energy intake. Self-reported employment status was categorized into six groups: full-time, part-time, no job, student, homemaker and retired. Multinomial logistic regression models were fitted to test the association between temporal patterns of energy intake and employment groups.

SETTING

Non-institutionalized community dwellers.

SUBJECTS

Non-pregnant adults (≥19 years old) with total energy intake between 2092 and 20920 kJ/d (500 and 5000 kcal/d; n 4508).

RESULTS

Five major patterns were identified, which can be seen as the traditional meal pattern and its variants. About 20 % of adults had the traditional pattern. The most prevalent pattern was the delayed morning meal pattern (33 %), which had lower Ca and P intakes than the traditional pattern. About 14 % of adults had the delayed lunchtime pattern, which had lower protein, PUFA, fibre, Ca, P, vitamin D and vitamin E intakes than the traditional. Adjusted prevalence of the delayed lunchtime pattern was highest among full-time students (34 %), followed by part-time workers (24 %), and was lower in retired (8 %), homemakers (11 %) and full-time employed adults (12 %).

CONCLUSION

Adults' temporal patterns of energy intake, which varied with their employment status, affected their dietary quality.