In a pre-registered UK Biobank cohort (n=158,686) with ~9 years follow-up, poorer speech-in-noise performance (digit triplet test) showed a directionally positive but statistically imprecise association with incident Parkinson's disease dementia (101 cases), with exploratory categorical analyses…

If confirmed in larger cohorts, hearing impairment could serve as an early, potentially modifiable risk marker for Parkinson's disease dementia, informing screening and prevention strategies, but current findings are exploratory and underpowered.

UNLABELLED: Hearing impairment is implicated as a risk factor for the incidence of all-cause dementia. However, few studies have explored the relationship between hearing loss and specific rarer dementia subtypes. Therefore, this pilot study sought to explore whether hearing impairment, estimated using a speech-in-noise test (the digit triplet test, DTT), is a risk factor for Parkinson’s disease dementia (PDD) incidence. We conducted a pre-registered prospective cohort study using UK Biobank data from 158,686 individuals (72,015 males; 86,671 females), who underwent the DTT, and were free from PDD at the point of assessment. A Cox Proportional Hazard model, controlling for age, sex and educational attainment was conducted. Over a median follow-up of 8.93 years (SD = 2.29), 101 cases of incident PDD were observed. While the hazard ratio were in the direction of a positive relationship between baseline DTT and PDD incidence, imprecision in estimates leaves it uncertain as to whether risk was influenced by speech-reception-threshold (HR = 2.39, 95% CI: 0.75,7.63; p = .141), and findings should be treated as exploratory. An exploratory analysis, in which speech-reception-thresholds were categorised according to UK Biobank norms, showed that ‘insufficient’ (p = .024), but not ‘poor’ (p = .236) hearing increased PDD risk. The observed HR directionally aligns with the previously observed positive relationship between hearing loss and all-cause dementia incidence. However, further studies employing alternative datasets with more PDD cases are required to substantiate these findings. If further supported, these findings may have clinical and public health implications. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00221-026-07282-1.