Skip to main content

Table 2 Summary of accuracy of delay-to-diagnosis predictions for the simulation analysis of 2000 patients without truncation of observation times. Our methodology was applied using either uniform priors or an exponential survival model for diagnosis delays, and CD4 back-estimation was used for comparison

From: Estimation of delay to diagnosis and incidence in HIV using indirect evidence of infection dates

  Full biomarker model, uniform prior Full biomarker model, exponential survival model CD4 back-estimation
Absolute error
Mean (years) 2.22 1.04 2.12
Lower quartile (years) 0.79 0.37 0.67
Median(years) 1.73 0.75 1.54
Upper quartile(years) 3.24 1.40 2.98
Mean squared error (years2) 8.39 2.15 8.40
Bias (mean error) (years) 1.97 –0.02 0.68
Coverage of 95% CrIa 89.5 94.1 NA
  1. acoverage of 95% credibility intervals for posterior distribution of the diagnosis delay in individual patients (relative to known true value in simulation)