Y Joo1. 1. Department of Environmental Planning, Environmental Planning Institute, Seoul National University, #220, 1 Gwanak-ro, Gwanak-gu, Seoul 151-742, Republic of Korea. Electronic address: yooju@yahoo.com.
Abstract
OBJECTIVES: This study analyzed the spatiotemporal pattern and spatial diffusion of elderly suicide by age cohort, in Korea. STUDY DESIGN: The research investigated the elderly suicide rates of the 232 municipal units in South Korea between 2001 and 2011. METHODS: The Gi* score, which is a spatially weighted indicator of area attributes, was used to identify hot spots and the spatiotemporal pattern of elderly suicide in the nation during the last 10 years. The spatial Markov matrix and spatial dynamic panel data model were employed to identify and estimate the diffusion effect. RESULTS: The suicide rate among elderly individuals 75 years and older was substantially higher than the rate for those between 65 and 74 years of age; however, the spatial patterns of the suicide clusters were similar between the two groups. From 2001 to 2011, the spatial distribution of elderly suicide hot spots differed each year. For both age cohorts, elderly suicide hot spots developed around the north area of South Korea in 2001 and moved to the mid-east area and the mid-western coastal area over 10 years. The spatial Markov matrix indicates that the change in the suicide rate of one area was affected by the suicide rates of neighbouring areas from the previous year, which suggests that suicide increase in one area inflates a neighbouring area's suicide rate over time. Using a spatial dynamic panel data model, elderly suicide diffusion effects were found to be statistically significant for both age cohorts even after economic and demographic indicators and a time variable are included. For individuals 75 years and older, the diffusion effect appeared to be larger. CONCLUSIONS: This study demonstrates that elderly suicide can spread spatially over time in both age cohorts. Thus, it is necessary to design a place-based and age-differentiated intervention policy that precisely considers the spatial diffusion of elderly suicide.
OBJECTIVES: This study analyzed the spatiotemporal pattern and spatial diffusion of elderly suicide by age cohort, in Korea. STUDY DESIGN: The research investigated the elderly suicide rates of the 232 municipal units in South Korea between 2001 and 2011. METHODS: The Gi* score, which is a spatially weighted indicator of area attributes, was used to identify hot spots and the spatiotemporal pattern of elderly suicide in the nation during the last 10 years. The spatial Markov matrix and spatial dynamic panel data model were employed to identify and estimate the diffusion effect. RESULTS: The suicide rate among elderly individuals 75 years and older was substantially higher than the rate for those between 65 and 74 years of age; however, the spatial patterns of the suicide clusters were similar between the two groups. From 2001 to 2011, the spatial distribution of elderly suicide hot spots differed each year. For both age cohorts, elderly suicide hot spots developed around the north area of South Korea in 2001 and moved to the mid-east area and the mid-western coastal area over 10 years. The spatial Markov matrix indicates that the change in the suicide rate of one area was affected by the suicide rates of neighbouring areas from the previous year, which suggests that suicide increase in one area inflates a neighbouring area's suicide rate over time. Using a spatial dynamic panel data model, elderly suicide diffusion effects were found to be statistically significant for both age cohorts even after economic and demographic indicators and a time variable are included. For individuals 75 years and older, the diffusion effect appeared to be larger. CONCLUSIONS: This study demonstrates that elderly suicide can spread spatially over time in both age cohorts. Thus, it is necessary to design a place-based and age-differentiated intervention policy that precisely considers the spatial diffusion of elderly suicide.
Authors: Miriam Marco; Antonio López-Quílez; David Conesa; Enrique Gracia; Marisol Lila Journal: Int J Environ Res Public Health Date: 2017-07-06 Impact factor: 3.390