Background: To address the age-friendliness of living environment in cities, the World Health Organization (WHO) launched the "Age-friendly cities" (AFC) initiative in 2005. To date, however, no universal standard tool for assessing age-friendliness of a community has been agreed.
Methodology: Two quantitative studies on AFC conducted in two Hong Kong districts-Sha Tin and Tuen Mun-were compared. A total of 801 residents aged ≥50 years were interviewed using structured questionnaires based on the WHO's AFC criteria. District-wide differences in age-friendliness were compared on the basis of eight domain scores. Multiple linear regression was used to examine associations with demographic and socio-economic characteristics. The provision of services and amenities was also compared to help explain the difference in domain scores.
Results: Variations in mean domain scores were observed in both districts. Sha Tin showed significantly lower scores in outdoor spaces and buildings, transportation, social participation, respect and social inclusion, civic participation and employment, communication and information, as compared with Tuen Mun. Although a significantly higher score on the housing domain was observed in Sha Tin, differences in community and health services domains were insignificant. Socio-demographic factors, such as age group, gender, area of residence, type of housing, experience of elderly care, employment status, self-rated health and income, were associated with domain scores. However, variations in services and amenities provision appeared not to be strongly associated with district-wide difference in domain scores.
Conclusions: District differences in public opinions towards age-friendly characteristics were observed in this study. Except for two of the eight domains, Sha Tin had significantly lower scores than Tuen Mun. Some socio-demographic indicators seemed predictive to the differences. Paradoxically, Sha Tin had better services and infrastructure and higher socio-economic status, but lower age-friendliness. This warrants detailed research on psychosocial factors that may influence residents' perceptions of local environments.