Data Sheet 1_Defining health and lifestyle characteristics of the age 50+ population: cluster analysis of data from the PROTECT study.pdf

IntroductionThe proportion of older people in the world is increasing and evidence suggests that older adults interact differently with products. Understanding this change is necessary to develop products that satisfy this cohort’s needs. Chronological age is typically used to segment older consumers however, given the diversity of ageing, a multi-dimensional approach considering other factors contributing to this behavior change is important. Using data from the PROTECT study in the UK, this research aimed to identify clusters of older people with distinct characteristics and investigate whether chronological age was fundamental in defining these groups.
MethodsTwelve variables, covering measures related to physical capabilities, mental health and lifestyle choices, were derived from the baseline questionnaire data from the PROTECT study and subjected to a k-means cluster analysis. Subsequent analyses investigated the association between participants’ cluster membership and other key variables.
ResultsCluster analysis identified 8 unique clusters of older adults differentiated on factors such as physical health (physical activity, pain, BMI and sleep quality), mental health (cognitive decline, depression and anxiety) and lifestyle (social events, puzzle and technology use and vitamin intake). Age was considered to be an important contributory factor to some clusters however did not explain all differences observed between the groups.
DiscussionOur findings indicate that in addition to chronological age, health and lifestyle variables are important in defining the unique characteristics of different clusters of those in the 50+ cohort. Future research should consider the multi-dimensional nature of ageing when conducting research with older consumers.

【引言】全球老年人口占比持续攀升，现有研究证据表明，老年群体与产品的交互方式存在显著差异。开发契合该队列人群需求的产品，亟需深入理解这一变化趋势。当前老年消费者细分通常以实足年龄（chronological age）为依据，但由于衰老进程存在显著个体多样性，综合考量其他影响行为改变的因素、采用多维研究方法显得尤为重要。本研究依托英国PROTECT研究（PROTECT study）的相关数据，旨在识别具有独特特征的老年人群聚类簇，并探究实足年龄是否为划分这些群体的核心依据。

【研究方法】本研究从PROTECT研究的基线问卷数据中提取了12项变量，涵盖身体机能、心理健康与生活方式选择相关指标，并对其开展k均值聚类分析（k-means cluster analysis）。后续分析进一步探究了参与者的聚类归属与其他核心变量之间的关联。

【研究结果】聚类分析共识别出8个独特的老年人群聚类簇，其群体差异体现在身体健康（身体活动水平、疼痛状况、身体质量指数（BMI）与睡眠质量）、心理健康（认知衰退、抑郁与焦虑）以及生活方式（社交活动、益智游戏使用、技术使用情况与维生素摄入情况）等维度。研究发现，实足年龄是部分聚类簇形成的重要影响因素，但无法完全解释群体间的所有差异。

【讨论】本研究结果表明，针对50岁及以上队列人群的不同聚类簇，除实足年龄外，健康与生活方式变量同样是界定其独特特征的重要因素。未来开展老年消费者相关研究时，应充分考虑衰老进程的多维性特征。