We would like to comment some methodological issues that arise from the paper of Tedrus et al. 1 and that could compromise some of their results. The use of CERAD (Consortium to Establish a Registry for Alzheimer' Disease) is indicated to Alzheimer's disease or eventually to other forms of cortical dementia, being unsuitable for subcortical dementia such as Parkinson' s disease dementia. As an example of what we intend to say is that CERAD battery is sensitive to capture limbic amnesia, apraxia, agnosia and naming deficits, which are relatively common in Alzheimer's disease but rarer in subcortical forms of dementia, and, in the other hand, the same battery is less sensitive to detect some executive dysfunction which are very common in Parkinson's disease dementia². According to that bias, prevalence of dementia and mild cognitive impairment among Tedrus et al. paper may be even higher. Another methodological issue is that, as reported else-where³, many antiparkinsonian agents may cause cognitive and psychiatric symptoms, mainly those with anti-cholinergic effects. Besides that, another antiparkinsonian drugs (excluding this time the anti-cholinergic ones) may cause psychiatric symptoms (sleep disturbance, inapetence, concentration difficulties, disturbed thoghts) which also can cause secondary cognitive impairment 4 , thus explaining, at least in part, some discrepancies between the prevalence of cognitive deficits in the parkinsonian group when compared with controls. According to that bias, prevalence of dementia and mild cognitive impairment among Tedrus et al. paper may be even lower. Finally, we would like to comment on the difficulties related to the diagnosis of functional impairment in PD as an adjunctive to the diagnosis of dementia in this condition. Because motor features of PD frequently jeopardize daily activities, this fact can impact the Pfeffer questionnaire scores, making the diagnosis of dementia more probable.

No abstract