Author(s) This pre-registration is currently anonymous to enable blind peer-review. It has one author.
Pre-registered on 2022/06/28 - 09:47 AM (PT)
1) Have any data been collected for this study already? It's complicated. We have already collected some data but explain in Question 8 why readers may consider this a valid pre-registration nevertheless.
2) What's the main question being asked or hypothesis being tested in this study? We hypothesise that:
1) Brain Age Gap (BAG) can be calculated in middle aged individuals.
2) Cross-sectionally, higher BAG is associated with worse cognitive performance in middle-aged individuals.
3) Longitudinal change in BAG over a two year period is positively associated with longitudinal deterioration in cognitive performance in middle aged individuals.
4) BAG is positively associated with: certain modifiable risk factors from the 2020 Lancet Commission on dementia prevention (hypertension, obesity, alcohol, hearing impairment, head injuries); genetic risk of Alzheimer's disease (APOE4); and PET biomarkers of amyloid deposition in middle aged individuals.
5) BAG is not associated with education level.
3) Describe the key dependent variable(s) specifying how they will be measured. Brain Age Gap (BAG) will be estimated from T1-weighted MRI head scans using publicly available code (https://github.com/james-cole/brainageR).
4) How many and which conditions will participants be assigned to? See analyses below.
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. Testing Hypothesis 2 - cross-sectional association between BAG and cognition:
Linear regression models will be fitted with: baseline cognitive score as the dependent variable; baseline BAG as an independent variable; while controlling for age, age-squared (PMID 30405393), biological sex, age*sex (PMID 31837193), scanning site, and years of education.
Testing Hypothesis 3 - longitudinal association between BAG and cognition:
Linear regression models will be fitted with: change in cognitive score between baseline and follow-up as the dependent variable; change in BAG between baseline and follow-up as an independent variable; while controlling for baseline BAG, baseline cognitive score, age at baseline, age-squared (PMID 30405393), biological sex, age*sex (PMID 31837193), scanning site, and years of education.
Testing hypothesis 4 - associations between BAG and risk factors:
Linear regression models will be fitted with baseline BAG as the dependent variable, and various putative variables associated with dementia as independent variables in separate models. This will include one model with modifiable risk factors of dementia as independent variables (hypertension, obesity, alcohol, hearing impairment, head injuries); one model with genetic risk of Alzheimer's disease as the independent variable (APOE4); and one model with amyloid PET deposition as the independent variable. Models will control for age, age-squared (PMID 30405393), biological sex, age*sex (PMID 31837193), scanning site, and years of education.
Testing hypothesis 5 - association between BAG and education:
A linear regression model will be fitted with: baseline BAG as the dependent variable; years of education as an independent variable; while controlling for age, age-squared (PMID 30405393), biological sex, age*sex (PMID 31837193) and scanning site.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. Data will not be excluded.
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. A previous paper assessed the association between BAG and cognition in middle aged individuals (PMID 31822815) and based on the observed effect size (standardised beta coefficient=0.2), a sample size of 358 participants would be needed to produce 80% power (7 parameters in model, alpha 0.05) and a 600 patient sample would produce power of 97%. The sample included in this study will have approximately 600 participants.
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) This is a pre-registered analysis of data from a larger observational study that has already been/is being collected.
