Comparison of models for analyzing two-group, cross-sectional data with a Gaussian outcome subject to a detection limit
| Autor: | Ryan E. Wiegand, Charles E. Rose, John M. Karon |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Statistics and Probability
Anti-HIV Agents Epidemiology Gaussian Normal Distribution HIV Infections 01 natural sciences Article Statistical power 010104 statistics & probability 03 medical and health sciences symbols.namesake 0302 clinical medicine Bias Health Information Management Limit of Detection Raltegravir Potassium Statistics Humans Tobit model 030212 general & internal medicine 0101 mathematics Probability Mathematics Detection limit Likelihood Functions Cross-sectional data Viral Load Outcome (probability) Confidence interval Regression Cross-Sectional Studies Multivariate Analysis HIV-1 symbols RNA Viral Software |
| Zdroj: | Statistical Methods in Medical Research. 25:2733-2749 |
| ISSN: | 1477-0334 0962-2802 |
| DOI: | 10.1177/0962280214531684 |
| Popis: | A potential difficulty in the analysis of biomarker data occurs when data are subject to a detection limit. This detection limit is often defined as the point at which the true values cannot be measured reliably. Multiple, regression-type models designed to analyze such data exist. Studies have compared the bias among such models, but few have compared their statistical power. This simulation study provides a comparison of approaches for analyzing two-group, cross-sectional data with a Gaussian-distributed outcome by exploring statistical power and effect size confidence interval coverage of four models able to be implemented in standard software. We found using a Tobit model fit by maximum likelihood provides the best power and coverage. An example using human immunodeficiency virus type 1 ribonucleic acid data is used to illustrate the inferential differences in these models. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|