The pervasive effect of youth self-report of hunger on depression over 6 years of follow up

Lynn McIntyre, Xiuyun Wu, Cynthia Kwok, Scott B. Patten

Research output: Contribution to journalArticleResearchpeer-review

16 Citations (Scopus)


Purpose: We used longitudinal data to clarify the association between self-report of hunger and subsequent depression risk among youth and young adults, accounting for other risk factors. Methods: Youth self-report of ever experiencing hunger data were collected from cycles 4–6 of the National Longitudinal Survey of Children and Youth cohort of Canadian youth 16 years and older (n = 4139). Data on depressive symptoms (CES-D 12) were collected over three cycles (2004–2009, cycles 6–8). We used multivariable regression based on generalized estimating equations (GEE) to examine prior youth hunger on later depression risk, adjusting for time-stable, time-varying, and lagged variables (e.g., depressive symptoms in previous cycle), thereby clarifying the temporal relationship. Results: The prevalence of youth hunger experience and depression risk reached 5.9 and 15.0%, respectively. The adjusted odds ratio of depression for participants reporting hunger was 2.31 (95% CI 1.54, 3.46) and changed little [2.17 (95% CI 1.29, 3.67)] after accounting for previous CES-D 12 scores, suggesting a temporal relationship in which hunger contributes to depression risk. Unlike never-hungry youth, depression in ever-hungry youth remained comparatively elevated over time. Conclusions: Our models support an independent and temporal relationship between youth self-report of hunger and depression in adolescence and young adulthood.

Original languageEnglish
Pages (from-to)537-547
Number of pages11
JournalSocial Psychiatry and Psychiatric Epidemiology
Issue number5
Publication statusPublished - May 2017
Externally publishedYes


  • Depression
  • Generalized estimating equation (GEE)
  • Hunger
  • Longitudinal
  • Youth

Cite this