The epidemiology of morningness/eveningness: influence of age, gender, ethnicity, and socioeconomic factors in adults (30-49 years).
Level 4 - case-series / case-control
Cross-sectional observational population survey
PubMed 16461986 · doi:10.1177/0748730405283154
What was done
A modified Horne and Ostberg Morningness/Eveningness Questionnaire (MEQ) was mailed to 5,000 New Zealand adults aged 30 to 49 years randomly selected from electoral rolls. A total of 2,526 completed questionnaires were analyzed (55.7% response rate). The authors compared the original Horne and Ostberg (1976) cutoffs against revised cutoffs for working adults from Taillard et al. (2004) and analyzed associations with age, gender, ethnicity, socioeconomic deprivation, work schedule, and self-rated health.
What was found
By original Horne and Ostberg cutoffs, 49.8% were morning types and 5.6% were evening types; by Taillard cutoffs, 24.7% were morning types and 26.4% were evening types. Controlling for ethnicity, gender, and socioeconomic deprivation, adults aged 30 to 34 were more likely to be definitely evening type (OR = 1.59, p < 0.05) and less likely to be morning type (moderately morning OR = 0.59, p < 0.01; definitely morning OR = 0.59, p < 0.05) compared to those aged 45 to 49. Night workers were more likely to be definitely evening type (OR = 1.49, p = 0.05), and unemployed individuals were less likely to be moderately morning type (OR = 0.64, p < 0.05). Evening types were 2.5 times more likely to report poor or fair general health than morning types (p < 0.01). Chronotype was largely independent of ethnicity, gender, and socioeconomic status.
Why it matters
This study shows that standard MEQ cutoffs substantially overestimate morningness in working-age adults and provides population-level evidence linking evening chronotype to poorer general health independently of socioeconomic status.
Limits
The cross-sectional design cannot establish causality between chronotype, work schedules, and health status. The 55.7% response rate introduces potential non-response bias. The sample was restricted to adults aged 30 to 49 in New Zealand, and all health and sleep preferences were self-reported without objective circadian biomarkers.
Cited by
- partial Population chronotypes are roughly distributed as one-third morning types, one-third evening types, and one-third intermediate types.