Search tips
Search criteria

Results 1-2 (2)

Clipboard (0)

Select a Filter Below

more »
Year of Publication
Document Types
1.  Binomial regression with a misclassified covariate and outcome 
Statistical methods in medical research  2012;10.1177/0962280212441965.
Misclassification occurring in either outcome variables or categorical covariates or both is a common issue in medical science. It leads to biased results and distorted disease–exposure relationships. Moreover, it is often of clinical interest to obtain the estimates of sensitivity and specificity of some diagnostic methods even when neither gold standard nor prior knowledge about the parameters exists. We present a novel Bayesian approach in binomial regression when both the outcome variable and one binary covariate are subject to misclassification. Extensive simulation results under various scenarios and a real clinical example are given to illustrate the proposed approach. This approach is motivated and applied to a dataset from the Baylor Alzheimer’s Disease and Memory Disorders Center.
PMCID: PMC3883897  PMID: 22421539
Misclassification; Bayesian inference; latent class model; sensitivity; specificity; Alzheimer’s disease
2.  Staging Dementia Using Clinical Dementia Rating Scale Sum of Boxes Scores 
Archives of neurology  2008;65(8):1091-1095.
The Clinical Dementia Rating Scale Sum of Boxes (CDR-SOB) score is commonly used, although the utility regarding this score in staging dementia severity is not well established.
To investigate the effectiveness of CDRSOB scores in staging dementia severity compared with the global CDR score.
Retrospective study.
Texas Alzheimer's Research Consortium minimum data set cohort.
A total of 1577 participants (110 controls, 202 patients with mild cognitive impairment, and 1265 patients with probable Alzheimer disease) were available for analysis.
Main Outcome Measures
Receiver operating characteristic curves were generated from a derivation sample to determine optimal cutoff scores and ranges, which were then applied to the validation sample.
Optimal ranges of CDR-SOB scores corresponding to the global CDR scores were 0.5 to 4.0 for a global score of 0.5, 4.5 to 9.0 for a global score of 1.O, 9.5 to 15.5 for a global score of 2.0, and 16.0 to 18.0 for a global score of 3.0. When applied to the validation sample, κ scores ranged from 0.86 to 0.94 (P <.001 for all), with 93.0% of the participants falling within the new staging categories.
The CDR-SOB score compares well with the global CDR score for dementia staging. Owing to the increased range of values, the CDR-SOB score offers several advantages over the global score, including increased utility in tracking changes within and between stages of dementia severity. Interpretive guidelines for CDR-SOB scores are provided.
PMCID: PMC3409562  PMID: 18695059

Results 1-2 (2)