Skip to main content
. Author manuscript; available in PMC: 2015 Dec 15.
Published in final edited form as: J Acquir Immune Defic Syndr. 2014 Dec 15;67(5):499–507. doi: 10.1097/QAI.0000000000000348

Table 2.

ITT generalized linear mixed effects modeling with non-normal outcome distributions (negative binomial) and AR(1) covariance structure comparing intervention to control.

Variable Estimate Standard Error df t p
Number of events without condom with any partner
  Intercept 0.6239 0.1891 6 3.30 0.0164
  Condition (Intervention) 0.08902 0.2646 6 0.34 0.7480
  Time effect in Intervention Arm −0.4462 0.05162 6628 −8.64 <.0001
  Time effect in Control Arma −0.2210 0.05098 6628 −4.33 <.0001
  Time by condition (Intervention) −0.2241 0.07351 6628 −3.05 0.0023
Number of events without condom with partners perceived to be HIV negative or HIV status unknown partners
  Intercept 0.01043 0.1353 6 0.08 0.9411
  Condition (Intervention) 0.2293 0.1895 6 1.21 0.2717
  Time effect in Intervention Arm −0.7212 0.05328 6628 −13.54 <.0001
  Time effect in Control Arma −0.3124 0.04604 6628 −6.79 <.0001
  Time by condition (Intervention) −0.4088 0.07042 6628 −5.81 <.0001
a

Derived by model where intervention arm served as referent group. Intervention condition=1; Control condition=0.