Table 3 Likelihood ratio tests assessing if adding additional variables significantly improves the Cox model fit

From: Neutralizing and binding antibodies are a correlate of risk of COVID-19 in the CoVPN 3008 study in people with HIV

Variables

Added Variables

P-valuea

Baseline covariates

M0 nAb-ID50 BA.4/5

0.004

Baseline covariates + M0 nAb-ID50 BA.4/5

M0 IgG N Index

0.66

Baseline covariates + M0 nAb-ID50 BA.4/5

M0 IgG Spike BA.4/5

0.32

Baseline covariates + M0 nAb-ID50 BA.4/5

M0 IgG N Index &M0 IgG Spike BA.4/5

0.53

Baseline covariates + M0 nAb-ID50 BA.4/5

Peak nAb-ID50 BA.4/5

0.45

Baseline covariates + M0 nAb-ID50 BA.4/5

Peak IgG Spike BA.4/5

0.29

Baseline covariates + M0 nAb-ID50 BA.4/5

Peak nAb-ID50 BA.4/5 &Peak IgG Spike BA.4/5

0.47

  1. Variables: predictors included in the restricted model. Added variables: predictors further added to the restricted model.
  2. nAb-ID50, 50% inhibitory serum dilution neutralizing antibody titer.
  3. aTwo-sided P-values are based on the partial-likelihood ratio test64 implemented by the function regTermTest in the survey package. The P-values tested the null hypothesis that the coefficients of added variables (one or more) all equal 0.
Back to article page