Skip to main content
. Author manuscript; available in PMC: 2018 Sep 1.
Published in final edited form as: J Acquir Immune Defic Syndr. 2017 Sep 1;76(1):33–42. doi: 10.1097/QAI.0000000000001467

Table 4.

Prevalence of Low 25(OH)D and High PTH and Differences in Levels by HIV status, and by Use of Efavirenz and Tenofovir

Prevalence of Low 25(OH)D
or High PTH
Difference in 25(OH)D
or PTH Levels

Prevalence Ratio (95%CI)
P value
Difference (95%CI)
P value

Exposure Unadjusted Adjusted1,2 Unadjusted Adjusted1
25(OH)D ≤ 20 ng/mL 25(OH)D (ng/mL)
Model 1A Model 1B2 Model 7A Model 7B
PHIV vs. PHEU 1.24 (0.99,1.54) 1.00 (0.81,1.24) −0.77 (−2.05,0.51) 0.48 (−0.69,1.65)
0.06 0.98 0.24 0.42
Model 2A Model 2B Model 8A Model 8B
PHIV EFV+ vs. PHEU3 1.52 (1.14,2.03) 1.30 (0.98,1.74) −2.83 (−4.93, −0.72) −1.94 (−3.88,0.01)
0.01 0.07 0.008 0.05
PHIV EFV− vs. PHEU3 1.17 (0.93,1.48) 0.95 (0.76,1.18) −0.31 (−1.64,1.01) 0.99 (−0.23,2.21)
0.17 0.63 0.643 0.11
Model 3A Model 3B Model 9A Model 9B
PHIV TDF+ vs. PHEU3 1.09 (0.79,1.51) 0.77 (0.56,1.06) −1.20 (−3.16,0.75) 1.02 (−0.87,2.92)
0.61 0.10 0.23 0.29
PHIV TDF− vs. PHEU3 1.28 (1.02,1.6) 1.06 (0.86,1.31) −0.65 (−1.99,0.70) 0.37 (−0.85,1.59)
0.03 0.58 0.35 0.55

PTH > 65 pg/mL PTH (pg/mL)
Model 4A Model 4B Model 10A Model 10B
PHIV vs. PHEU 3.80 (1.52,9.51) 3.17 (1.25,8.06) 7.10 (4.17,10.02) 5.26 (2.35,8.17)
0.004 0.01 <0.001 <0.001
Model 5A Model 5B Model 11A Model 11B
PHIV EFV+ vs. PHEU3 1.13 (0.22,5.69) 0.85 (0.17,4.56) 5.44 (0.61,10.28) 3.37 (−1.85,8.60)
0.88 0.89 0.03 0.21
PHIV EFV− vs. HEU3 4.40 (1.75,11.04) 3.67 (1.44,9.36) 7.27 (4.11,10.44) 5.54 (2.47,8.61)
0.002 0.006 <0.001 <0.001
Model 6A Model 6B Model 12A Model 12B
PHIV TDF+ vs. PHEU3 6.79 (2.54,18.12) 5.50 (1.95,15.49) 14.22 (9.11,19.32) 11.65 (6.47,16.83)
<0.001 0.001 <0.001 <0.001
PHIV TDF− vs. PHEU3 2.92 (1.13,7.6) 2.64 (1.01,6.94) 4.80 (1.69,7.91) 3.71 (0.61,6.82)
0.03 0.05 0.002 0.02

Abbreviations: PHIV-Perinatally HIV-infected; PHEU-perinatally HIV-exposed uninfected; TDF-tenofovir; EFV-efavirenz, 25(OH)D – 25 hydroxy-vitamin D.

1

For the outcome 25(OH)D, the models were adjusted for age, black race, ancestry markers, latitude as a continuous variable, and season. For the outcome PTH, models were adjusted for age, black race, and ancestry markers.

2

Model 1B, 2B, and 3B were fit using the Poisson link and the robust standard error.

3

In sensitivity analyses including children with ARV data just prior to the 25(OH)D date, and not on the 25(OH)D date, the results did not change for the TDF or EFV analyses.