Loading...
Please wait, while we are loading the content...
Similar Documents
Author ' s response to reviews Title : Framingham Stroke Risk Profile and Poor Cognitive Function : A Population-Based Study
| Content Provider | Semantic Scholar |
|---|---|
| Author | Lang, Iain A. Huppert, Felicia Langa, Kenneth M. |
| Copyright Year | 2008 |
| Abstract | and discussion. After adjusting for sex and age, the magnitude of this effect drops by nearly a factor of 5 to .088. Thus, this result probably primarily reflects the fact that older people score worse on global cognition. We do not agree that this is misleading. Indeed it would be strange if our results did not suggest that age accounted for a large proportion of the variance given that age is known to be strongly associated with both cognitive function and stroke. We stress in the abstract and elsewhere that the association was attenuated with adjustment, though the fact that the association remained significant suggests that potentially modifiable risk factors are also important. New analyses with the FSRP components following your suggestions support this conclusion (p.9, para.3). 7. Are all items from ELSA, including atrial fibrillation, assessed with self-report or were medical records available for validation? Cognitive function was directly measured and other items were based on self-reported diagnosed conditions, which we acknowledge as a limitation (p.11, para.3). Minor Essential Revisions 1. On page 9, 1st sentence of the last paragraph, which cognitive domain is the beta estimate referring to? And why is it <=? These results are now presented in Table 3 following your suggestion (p.24). 2. It probably won’t make much of a difference, but given the range of social data available in ELSA, why have you chosen to control for such a limited set of potential confounders (e.g., why not income and wealth, if they are available)? We control for education and socioeconomic status which are highly correlated with income and wealth. We encounter problems with collinearity in our models if we introduce too many variables that are highly similar. It should also be noted that previous studies have incorporated a more limited set of covariates, and our study is the first to adjust for socioeconomic status. And what is the motivation for selecting the covariates you did choose? CESD score in particular seems to me potentially causally subsequent to the FSRP. We selected variables that are reliably measured in ELSA and known to be associated with levels of cognitive function (p.8, para.1). We also took into account which variables were included as covariates in previous studies in order to ensure that our results were reasonably comparable. Adjusting for depression is common practice in cognitive research, though we acknowledge that we may have effectively over adjusted our final models given the similarity in risk factors for poor cognitive function and stroke (p.11, para.2). The fact that the association remained significant is testament to its robustness. 3. Repeated reference to “probabilities of stroke” instead of “FSRP score” is confusing. We amended a sentence which refers specifically to FSRP scores (p.12, para.2). However, FSRP scores correspond to the 10-year probability or risk of incident stroke (p.6, para.1), and we therefore refer to stroke risk or 10-year stroke risk throughout. 4. How was the subsample of 7716 ELSA participants who provided FSRP information selected? Or do you just mean that the other 4,015 ELSA participants had missing data? Each wave of the Health Survey for England (HSE) is an independent nationally representative sample, and the focus of each wave changes each year. Extensive information on cerebrovascular risk factors has been collected in some waves (e.g. 1998) whereas in other waves with a different emphasis this data was not collected, or was collected for a random subsample of participants. Our response to reviewer 2 (Matti Viitanen) This is an interesting article describing the association with stroke risk factor load and cognitive decline. Background: The authors claim that risk factors for incident stroke such as smoking status and diabetes predispose older adults to cerebral white matter pathology, and biological aging is partly attributable to effects of cerebrovascular disease. Please explain what you mean with cerebral white matter pathology. The point we were making in this introductory statement is that risk factors for incident stroke predispose older adults to subclinical cerebrovascular disease in addition to clinical stroke. We have revised this section to clarify this point (p.3, para.1). I do not agree that biological aging is attributable to effects of cerebrovascular disease â## please explain. This section is taken from the commentaries of Seshadri and others who argue that biological aging of the brain is partly attributable to aging of the cerebrovascular circulation and the effects of these vascular changes on the brain. We have amended this section to make this clearer (p.3, para.1). Methods: With which examinations the stroke and dementia were excluded and how were stroke and dementia defined. Were the patients with mild cognitive impairment excluded? Were the silent cerebral infarctions included? Those with a self-reported diagnosis of stroke or dementia were excluded, and we acknowledge the use of selfreported diagnosed conditions as a limitation (p.11, para.3). Participants with mild cognitive impairment and silent cerebral infarctions were included, indeed it was essential to include these individuals, and we argue that subclinical cerebrovascular pathology is likely to account for the observed association between stroke risk and cognitive function. In the discussion asymptomatic cerebral infarction should be silent cerebral infarction. We have revised this sentence accordingly (p.13, para.2). Conclusion: â##Further research is necessary to establish whether 10-year stroke risk can be used to predict whether individuals will become demented in the futureâ#¦â## I do not understand what you want to say â## dementia is only a stage of cognitive decline when the cognitive decline interferes significantly with work or usual social activities or relationships with others â## already mild cognitive deterioration interferes with work. Dementia definition works well in Alzheimerâ##s disease but in cerebrovascular diseases where the patients have executive dysfunction and relatively intact memory the definition is almost useless especially in the epidemiological studies. We have revised our conclusions following your feedback, and do not now make reference to the prediction of dementia (p.14, para.1). |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://static-content.springer.com/openpeerreview/art:10.1186%2F1471-2377-8-12/12883_2007_174_AuthorComment_V3.pdf |
| Alternate Webpage(s) | http://www.biomedcentral.com/imedia/9913170681882983_comment.pdf |
| Alternate Webpage(s) | http://www.biomedcentral.com/imedia/9324572571782627_comment.pdf |
| Alternate Webpage(s) | https://static-content.springer.com/openpeerreview/art:10.1186%2F1471-2377-8-12/12883_2007_174_AuthorComment_V2.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
