Music, Leisure and Dementia Risk: Clarifying What is Really Being Measured

We read with interest the article by Jaffa and colleagues, ‘What is the association between music-related leisure activities and dementia risk? A cohort study,’ based on the ASPREE/ALSOP cohort [1]. The authors report that older adults who ‘always’ listened to music and those who regularly engaged in both music listening and playing had substantially lower risks of dementia and cognitive impairment than those who rarely or never engaged. While this work adds to the literature on leisure activities and cognitive health [2, 3], we wish to highlight several conceptual issues that are important for interpretation and for the design of future interventions. First, the exposure measured may reflect a broader construct than ‘music listening’ itself. Participants reported how often they listened to music or the radio and how often they played a musical instrument [1]. This self-reported behaviour likely captures multiple intertwined elements: the motivation and habit of engaging in leisure activities, the ability to secure discretionary time, sufficient physical and mental capacity to enjoy such activities, and a longstanding preference for music as a hobby. In other words, the variable may be indexing a general pattern of active, self-directed leisure engagement rather than the specific acoustic or aesthetic properties of music. The observed association might therefore reflect the benefits of a structured, stimulating lifestyle more than the effect of ‘listening to music’ in isolation. Second, music engagement may act as both a marker and a mediator of lifelong social and economic advantage. Education, occupational complexity, income, social networks and access to cultural resources are all known to shape cognitive reserve and dementia risk [2, 3]. Although the authors adjusted for years of education and an area-level socioeconomic index, these variables only partially describe the life-course accumulation of social and cultural capital. Individuals who frequently listen to or play music may be those who have enjoyed greater educational opportunities, richer social connections and more cognitively stimulating environments throughout life. In that case, music engagement could function as a proxy or mediator for these upstream determinants rather than as an independent, modifiable risk factor. Even sophisticated multivariable models may not fully disentangle residual confounding and mediation in this context. Third, the heterogeneity of ‘music listening’ is not addressed, particularly with respect to genre, loudness and voluntariness. Psychological and physiological responses can differ markedly between self-selected, pleasant music and sound that is experienced as intrusive or aversive. Self-selected music is often used to promote relaxation and stress recovery [4], whereas unwanted or excessively loud sound may be perceived as noise and could contribute to stress or hearing damage. Hearing loss, in turn, has been identified as an important and potentially modifiable dementia risk factor [5]. In the present study, individuals who report that they ‘always’ listen may include those who attentively listen to preferred music, those who keep the radio or television on mainly as background sound, and those passively exposed to environmental audio. Collapsing these distinct experiences into a single exposure category risks obscuring important effect modification by musical preference, listening context and auditory health. Taken together, these considerations suggest that the reported hazard ratios should be interpreted with caution. We agree that the findings are broadly consistent with prior work linking cognitively and socially engaging leisure activities with lower dementia risk [2, 3]. However, they do not yet show that prescribing music listening will prevent dementia. Before translating such associations into recommendations, it would be helpful to describe music-related behaviours in more detail, to model potential mediators such as social engagement and broader leisure patterns, and to test in interventional or quasi-experimental designs whether enjoyable, self-selected music offers benefits beyond other activities. We commend Jaffa and colleagues for drawing attention to music as a potentially accessible, low-cost strategy to support brain health in ageing populations. Clarifying what is really being measured—and for whom, and under what conditions music is beneficial—will be essential for designing equitable, evidence-based interventions that move beyond correlation towards causally informed prevention. Ryuichi Minoda Sada conceptualised this letter, wrote the initial draft, conducted critical revisions and approved the final version. This work was conducted as part of ‘The Nippon Foundation–The University of Osaka Infectious Disease Prevention.’ I am grateful to Dr. Tadao Okada (Department of Family Medicine, Tesshoukai kameda Family Clinic Tateyama, Chiba, Japan) for sharing the article on social media and for his insightful comments. The author has nothing to report. The author declares no conflicts of interest. No new data were created or analysed in this study.