Àá½Ã¸¸ ±â´Ù·Á ÁÖ¼¼¿ä. ·ÎµùÁßÀÔ´Ï´Ù.
KMID : 0614420120400030063
Mental Health & Social Work
2012 Volume.40 No. 3 p.63 ~ p.86
Trajectories of Korean Adults` Depressive Symptoms and its Associations with Smoking and Drinking: Using Latent Class Growth Modeling Implications for an Integrative Mental Health Care System in Korea
Jun Jin-A

Park Hyun-Yong
Sohn Sun-Ju
Abstract
Using latent class growth modeling(LCGM) and 4-year`s of longitudinal Korean Welfare Panel Study(KWPS, 2006-2009) data, this study examined various trajectories of Korean adults` depressive symptoms and simultaneous smoking and drinking behaviors. Study sample included 11,677 individuals aged 19 years or older at Wave I and those with 3+ years of longitudinal information on respondents` perceived levels of depressive symptoms measured by the 11-item Center for Epidemiologic Studies Depression Scale(CES-D). Data were analyzed via LCGM for classification purposes into subgroups that presented similar change patterns of depressive symptoms over time as varying trajectories. Korean adults` perceived levels of depressive symptoms were classified as the following three trajectories: consistently low(n=9,284, 79.5%), decreasing (n=1,081, 9.3%), and increasing depressive symptoms(n=1,312, 11.2%). Binary logistic regressions were further performed to examine smoking and/or drinking behaviors of each trajectory. Findings showed that Korean adults whose depressive symptoms have increased over time are more likely to smoke(OR=1.90, 95% CI=1.42-2.55) or engage in hazardous/harmful drinking (OR=1.28, 95% CI=1.04-1.57), compared to those whose depression level stayed relatively low over time. Implications include the need to develop and sustain an integrative care system in Korea and to also certify and educate current and prospective mental health practitioners as licensed addiction specialists in tailoring to the needs of Korean mentally ill population.
KEYWORD
latent class growth modeling(LCGM), longitudinal, trajectories, depressive symptoms, smoking, drinking
FullTexts / Linksout information
Listed journal information
ÇмúÁøÈïÀç´Ü(KCI)