Socioeconomic status as an effect modifier of the association between built environment and mortality in elderly Hong Kong Chinese: A latent profile analysis.
Built environment
Cohort study
Latent profile analysis
Mortality
Journal
Environmental research
ISSN: 1096-0953
Titre abrégé: Environ Res
Pays: Netherlands
ID NLM: 0147621
Informations de publication
Date de publication:
04 2021
04 2021
Historique:
received:
11
12
2020
revised:
28
01
2021
accepted:
29
01
2021
pubmed:
7
2
2021
medline:
22
4
2021
entrez:
6
2
2021
Statut:
ppublish
Résumé
Previous studies have focused on associations between individual built environment (BE) characteristics and mortality, and found the BE-mortality associations differed by socioeconomic status (SES). Different individual BE characteristics may have different impacts on health and thus could interact. Combinations of BE characteristics may be a better approach to explore the BE-mortality associations. This study aimed to investigate the associations of BE pattern with mortality in a prospective cohort of elderly Hong Kong Chinese (Mr. OS and Ms. OS Study), and assess whether the BE-mortality association differed by SES. Between 2001 and 2003, 3944 participants aged 65-98 years at baseline were included in the present analysis. BE characteristics were assessed via Geographic Information System. Data on all-cause and cause-specific mortality were obtained from the Hong Kong Government Death Registry. Latent profile analysis was used to derive BE class, and the Cox proportional hazards model was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). Three BE classes were identified. During a total of 53276 person-years of follow-up, 1632 deaths were observed. There were no significant associations of BE class with all-cause and cause-specific mortality. However, we found the associations of BE class with all-cause mortality were modified by SES. In comparison with Class 3 (characterized by greater green space), HRs (95%CIs) were 0.72 (0.54, 0.97) for Class 1 (characterized by greater commercial land use) and 0.77 (0.64, 0.94) for Class 2 (characterized by greater residential land use) among low-SES participants. The associations were stronger among high-SES participants, with 0.55 (0.33, 0.89) for Class 1 and 0.68 (0.48, 0.97) for Class 2. In contrast, Class 2 (HR 1.18, 95%CI 1.01-1.39) had a higher mortality risk compared with Class 3 among middle-SES participants. Our findings provide new evidence on the role of SES as an effect modifier of BE pattern and mortality. BE pattern has a varied effect on mortality risk for different SES groups.
Sections du résumé
BACKGROUND
Previous studies have focused on associations between individual built environment (BE) characteristics and mortality, and found the BE-mortality associations differed by socioeconomic status (SES). Different individual BE characteristics may have different impacts on health and thus could interact. Combinations of BE characteristics may be a better approach to explore the BE-mortality associations.
OBJECTIVES
This study aimed to investigate the associations of BE pattern with mortality in a prospective cohort of elderly Hong Kong Chinese (Mr. OS and Ms. OS Study), and assess whether the BE-mortality association differed by SES.
METHODS
Between 2001 and 2003, 3944 participants aged 65-98 years at baseline were included in the present analysis. BE characteristics were assessed via Geographic Information System. Data on all-cause and cause-specific mortality were obtained from the Hong Kong Government Death Registry. Latent profile analysis was used to derive BE class, and the Cox proportional hazards model was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs).
RESULTS
Three BE classes were identified. During a total of 53276 person-years of follow-up, 1632 deaths were observed. There were no significant associations of BE class with all-cause and cause-specific mortality. However, we found the associations of BE class with all-cause mortality were modified by SES. In comparison with Class 3 (characterized by greater green space), HRs (95%CIs) were 0.72 (0.54, 0.97) for Class 1 (characterized by greater commercial land use) and 0.77 (0.64, 0.94) for Class 2 (characterized by greater residential land use) among low-SES participants. The associations were stronger among high-SES participants, with 0.55 (0.33, 0.89) for Class 1 and 0.68 (0.48, 0.97) for Class 2. In contrast, Class 2 (HR 1.18, 95%CI 1.01-1.39) had a higher mortality risk compared with Class 3 among middle-SES participants.
CONCLUSIONS
Our findings provide new evidence on the role of SES as an effect modifier of BE pattern and mortality. BE pattern has a varied effect on mortality risk for different SES groups.
Identifiants
pubmed: 33548297
pii: S0013-9351(21)00124-9
doi: 10.1016/j.envres.2021.110830
pii:
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
110830Informations de copyright
Copyright © 2021 Elsevier Inc. All rights reserved.