Skip to main content

Table 2 Mean differences in neurocognitive outcomes between children taking protease inhibitor and non-protease inhibitor-based antiretroviral therapy

From: Neurocognitive function among HIV-infected children on protease inhibitor -based versus non-protease inhibitor based antiretroviral therapy in Uganda: a pilot study

Outcome

Antiretroviral therapy regimen

Mean difference (95% CI)

Hommel’s method corrected

Non-PI based (n = 42)

PI based (n = 34)

P Value

KABC - II, mean ± SE

 Sequential processing

−0.05 (0.20)

0.85 (0.23)

− 0.90 (− 1.58, − 0.23)

0.05

 Simultaneous processing

− 0.34 (0.13)

0.08 (0.15)

− 0.42 (− 0.86, 0.02)

0.19

 Learning

−1.06 (0.22)

− 0.78 (0.26)

− 0.28 (− 1.02, 0.47)

0.46

 Planning

− 0.31 (0.16)

0.15 (0.19)

− 0.46 (− 1.02, 0.09)

0.22

 Mental processing index

− 0.95 (0.22)

− 0.41 (0.25)

− 0.55 (− 1.29, 0.19)

0.29

TOVA, mean ± SE

 Omission errors

0.30 (0.19)

0.10 (0.22)

0.20 (− 0.44, 0.84)

0.53

 Commission errors

0.64 (0.20)

0.05 (0.23)

0.59 (−0.07, 1.26)

0.24

 Response time total

0.27 (0.19)

−0.39 (0.21)

0.66 (0.04, 1.29)

0.15

 Response time variability

0.80 (0.23)

−0.11 (0.26)

0.90 (0.13, 1.67)

0.10

 D′ prime

−0.64 (0.22)

− 0.18 (0.25)

−0.45 (−1.18, 0.27)

0.43

  1. Abbreviations: PI Protease Inhibitor, SE Standard error
  2. Age-adjusted z-scores were computed using community control children from another study as the reference population and all analyses were adjusted for WHO stage at ART initiation, co-trimoxazole prophylaxis, child born before term, socioeconomic status, MUAC, and age at start of ART.
  3. After adjusting for multiple testing using the Hommel’s method, none of the mean difference was statistically significant.