Table 2. Characteristics of 20 studies included in the meta-analysis of the relationship between baseline CD4 count and first year mortality on antiretroviral therapy.
Study, Year# | Country | N | MedianBaselineCD4* | Number ofCD4 strata | LowestCD4Strata | HighestCD4Strata | Total % Mortality | Total Loss to Follow up** | Reported Follow up Time (months) | |
1 | Barth, 2008 | South Africa | 606 | 67 | 4 | 0–50 | >200 | 18 | 15 | 12 |
2 | Bisson, 2008 | Botswana | 410 | 81 | 0 | N/A | N/A | 8 | 21 | 11 |
3 | Bourgeois, 2005 | Cameroon | 109 | 150 | 2 | 0–50 | >50 | 8.3 | 2.8 | 16 |
4 | Brinkhof, 2009 | Multiple | 10714 | 87–131 | 5 | 0–25 | >200 | 8.8 | 7.1–31.7 | 24 |
5 | Bussman, 2008 | Botswana | 633 | 67 | 2 | 0–50 | >50 | 8.9 | 17.3 | 12 |
6 | Calmy, 2006 | Multiple | 6961 | 89 | 5 | 0–15 | >200 | 10 | 4.8 | 24 |
7 | Cornell, 2009 | South Africa | 2023 | 102 | 4 | 0–50 | >150 | 4.9 | 6.2 | 12 |
8 | Culbert, 2007 | Congo, D.R. | 494 | 124 | 0 | N/A | N/A | 5.4 | 7.9 | 12 |
9 | Etard, 2006 | Senegal | 404 | 128 | 3 | 0–50 | >200 | 11.7 | 1.7 | 12 |
10 | Geng, 2008 | Uganda | 3628 | 95 | 0 | N/A | N/A | 1.3 | 23 | 11 |
11 | Lowrance, 2009 | Rwanda | 3194 | 141 | 0 | N/A | N/A | 4.6 | 4.9 | 12 |
12 | Marazzi, 2008 | Multiple | 3456 | 166* | 3 | 0–50 | >500 | 7.5 | 1.2 | 12 |
13 | Mzileni, 2008 | South Africa | 2380 | – | 4 | 0–50 | 151–200 | 7.8 | 14.1 | 30 |
14 | Mutevedzi, 2010 | South Africa | 5179 | 116 | 0 | N/A | N/A | 3.7 | 10.9 | 12 |
15 | Ojikutu, 2008 | South Africa | 298 | 65 | 4 | 0–20 | >100 | 16 | 7.4 | 8 |
16 | Seyler, 2003 | Ivory Coast | 101 | 135 | 2 | 0–50 | >50 | 9.9 | 0 | 24 |
17 | Stringer, 2006 | Zambia | 11776 | 143 | 4 | 0–50 | >350 | 7.1 | 2 | 18 |
18 | Toure, 2008 | Ivory Coast | 10211 | 123 | 4 | 0–50 | >150 | 15 | 19 | 32 |
19 | Weidle, 2002 | Uganda | 342 | 73 | 2 | 0–50 | >50 | 16 | 24 | 12 |
20 | Zachariah, 2006 | Malawi | 1507 | 123 | 2 | 0–50 | >50 | 10.3 | 11.7 | 12 |
Full citations given in Appendix S2 in File S1.
Denotes mean not median.