Comparison of models for analyzing two-group, cross-sectional data with a Gaussian outcome subject to a detection limit

Details

• Alternative Title:
  Stat Methods Med Res
• Personal Author:
  Wiegand, Ryan E. ; Rose, Charles E. ; Karon, John M.
• Description:
  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.
• Subjects:
  Anti-HIV Agents Article Bias Cross-Sectional Studies HIV-1 HIV Infections Humans Likelihood Functions Limit Of Detection Multivariate Analysis Normal Distribution Probability Raltegravir Potassium Regression RNA, Viral Software Statistical Power Viral Load

  More +
• Source:
  Stat Methods Med Res. 25(6):2733-2749
• Pubmed ID:
  24803511
• Pubmed Central ID:
  PMC6880228
• Document Type:
  Journal Article
• Funding:
  CC999999/ImCDC/Intramural CDC HHS/United States
• Volume:
  25
• Issue:
  6
• Collection(s):
  CDC Public Access
• Main Document Checksum:
  urn:sha256:49d0ce0968b7f5d9d40ab46e9cbec013771c9e4b7b254288e3706542120c4f8f
• Download URL:
  https://stacks.cdc.gov/view/cdc/83429/cdc_83429_DS1.pdf
• File Type:
  [PDF - 974.86 KB ]