Shift work was associated with elevated dementia risk. Definitive guidelines for sleep duration among shift workers have not been proposed. We aimed to identify sleep durations associated with elimination of excess dementia risk in shift workers.285,213 dementia-free UK Biobank participants at baseline, aged 38-71 years, were enrolled between 2006 and 2010 and followed up through 2022 in this cohort study. Cox proportional hazards models were used to examine the associations between shift work, sleep duration, and risk of dementia.The 285,213 participants included 49,079 shift workers and 236,134 non-shift workers. Over a median follow-up of 13.8 years, 1887 dementia cases were documented. Current shift workers had significantly higher dementia risk than non-shift workers (hazard ratio [HR] 1.26; 95% CI 1.11-1.42). However, this excess risk was eliminated in shift workers with 8 h of sleep (HR 1.02; 95% CI 0.80-1.29). Analysis of shift work frequency indicated that "sometimes" and "usually/always" shift work were associated with increased dementia risk compared to that of non-shift workers, but excess dementia risk was eliminated in members of either frequency group receiving 8 h of sleep ("sometimes", HR 1.05; 95% CI 0.75-1.48; "usually/always", HR 0.98; 95% CI 0.70-1.35). Both "non-night shift" and "night shift" workers showed increased dementia risk compared to non-shift workers. Workers with 8 h of sleep mitigated the excess risk (HR 1.13; 95% CI 0.84-1.53 and HR 0.86; 95% CI 0.59-1.26, respectively).8-h sleep may eliminate excess dementia risk among shift workers, suggesting a potentially effective dementia prevention guideline for shift workers.© 2024. Springer-Verlag GmbH Germany, part of Springer Nature.