Background: The objective of this study is to explore the bidirectional, longitudinal associations between self-reported sensory functions (hearing/vision) and cognitive functioning among older adults in Sweden and Denmark during the period 2004–2017.

Methods: The study is based on data from The Survey of Health, Ageing and Retirement in Europe and consists of 3164 persons aged 60 to 93 years. Within-person associations between sensory and cognitive functions were estimated using random intercept cross-lagged panel models.

Results: The results indicated that cognitive and sensory functions were associated within their respective domains over time. The results on the bidirectional associations between sensory functions and cognition over time showed weak and statistically non-significant estimates.

Conclusion: Our study showed no clear evidence for cross-lagged effects between sensory functions and cognitive functioning. Important to note, however, is that using longitudinal data to estimate change within persons is a demanding statistical test and various factors may have contributed to the absence of conclusive evidence in our study. We discuss several of these factors.