Aim: We used register-based data to estimate the effect of all-type dementia on road traffic accidents (RTAs) risk, combined with comorbidities or sedative medicines, among non-institutionalized older people in Denmark. Methods: The source population was all residents in Denmark aged 65 years and older, alive as of January 1, 2008 (n = 853,228). Cases were those who had any type of RTA in 2009–2014. Each case was matched for age, sex and geographic location to 4–6 controls. All-type dementia was ascertained using the International Classification of Diseases version 10 (ICD-10) diagnosis supplemented with prescribed medicine records. Eight chronic diseases were selected to assess comorbidities. Four types of medicines were categorized as sedative medicines for analysis. Conditional logistic regression with adjustment for education and marital status as well as either the number of comorbidities or sedative medications use was performed using STATA software. Results: Older people with dementia had lower RTAs risk compared to their controls (odds ratio = 0.43, 95% confidence interval (0.32–0.60), p < 0.001). Significant interaction was observed between dementia and the number of comorbidities for RTAs estimation. Conclusions: The significantly lower RTAs risk for older individuals with dementia observed in our study may be due to people with dementia living at home having a lower frequency of outdoor activities; that is, less exposure to traffic. However, this, together with the interaction between dementia and comorbidities as well as sedative medications, should be investigated further.