Modelling longitudinal cognitive test data with ceiling effects and left skewness
Bibliographic record
Abstract
Cognitive tests such as the Mini Mental State Examination (MMSE) may result in data with discrete and skewed distributions that necessitate proper statistical models for valid inference. We review different longitudinal approaches to model cognitive decline data in older individuals and provide recommendations for model choice and result interpretation. We used data from the Alzheimer’s Disease Neuroimaging Initiative study and focused on MMSE scores as response variable collected on up to four visits over a two-year period in older individuals (mean age 73 years). At baseline individuals were classified as having Alzheimer’s disease (AD), early or late mild cognitive impairment, subjective memory concern, or being cognitively normal. We considered generalized additive models for location, scale and shape (GAMLSS) with binomial/beta-binomial response distribution and parametric/non-parametric random effects, selected the best model and used graphs for illustration. Binomial model with non-parametric random intercept and slope fit the data the best according to the Bayesian Information Criterion. The three-way interaction between time, age and diagnostic group was statistically significant suggesting that AD individuals had the steepest cognitive decline among all groups, especially in younger individuals. Furthermore, males and APOE4 carriers had worse cognitive performance, while more educated people had better cognitive performance compared to less educated. Various plots are used to illustrate and aid in interpretation of the results. GAMLSS are an appropriate class of models providing interpretable results for repeatedly measured cognitive test data. We recommend that they are used more widely, accompanied by effect estimation, statistical testing and visualizations for illustration.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.071 | 0.131 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.007 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".