Assaad Sarah, Hayat Shabina, Brayne Carol, Zaninotto Paola, Steptoe Andrew
Department of Epidemiology and Public Health, Institute of Epidemiology and Health Care, University College London, London, UK.
Department of Behavioural Science and Health, Institute of Epidemiology and Health Care, University College London, London, UK.
Alzheimers Dement. 2025 Apr;21(4):e70169. doi: 10.1002/alz.70169.
Understanding cognitive decline trajectories is crucial for dementia prevention, as many cases go undetected. Identifying participation biases in such studies is essential for data validity.
We examined non-participation correlates in the Healthy Cognitive Ageing Project (HCAP), a sub-study of the English Longitudinal Study of Ageing (ELSA). We compared sociodemographic and health characteristics of invited, interviewed, and non-interviewed individuals, and assessed the impact of sample weights.
Of 1778 ELSA members invited in 2018, 1273 (72%) participated. Participants were similar to the invited sample in sociodemographics but were younger, had fewer daily living difficulties, and had better cognition. Non-participation was linked to difficulties in daily living (odds ratio 1.78), dementia (1.55), and psychiatric conditions (1.34). Weighted analyses highlighted differences in disability and cognition.
Non-participation in cognitive studies is not random, lowering response and retention rates, and requiring adjustments to data analysis beyond the use of weights.
We compared the sociodemographics of invited, interviewed, and non-interviewed individuals. We used sample weights to assess differences in participants' characteristics. We found non-participation linked to daily living difficulties, dementia, and psychiatric conditions.
了解认知衰退轨迹对于痴呆症预防至关重要，因为许多病例未被发现。识别此类研究中的参与偏差对于数据有效性至关重要。
我们在健康认知老龄化项目（HCAP）中研究了不参与的相关因素，该项目是英国老龄化纵向研究（ELSA）的一项子研究。我们比较了受邀、接受访谈和未接受访谈个体的社会人口统计学和健康特征，并评估了样本权重的影响。
在2018年受邀的1778名ELSA成员中，1273人（72%）参与了研究。参与者在社会人口统计学方面与受邀样本相似，但更年轻，日常生活困难更少，认知能力更好。不参与与日常生活困难（优势比1.78）、痴呆症（1.55）和精神疾病（1.34）有关。加权分析突出了残疾和认知方面的差异。
不参与认知研究并非随机现象，这会降低应答率和留存率，并且除了使用权重之外，还需要对数据分析进行调整。
我们比较了受邀、接受访谈和未接受访谈个体的社会人口统计学特征。我们使用样本权重来评估参与者特征的差异。我们发现不参与与日常生活困难、痴呆症和精神疾病有关。
