Table 1.
Characteristics of study communities
| Country | Triplet | Community number | Adult HIV prevalence (%) 1 | HIV-infected on ART (%) 2 | Population size 3 | Men circumcised (%) |
|---|---|---|---|---|---|---|
| Zambia |
1 |
1 |
16 |
23 |
42,898 |
16 |
| |
1 |
2 |
13 |
29 |
33,297 |
17 |
| |
1 |
3 |
17 |
15 |
38,081 |
7 |
| |
2 |
4 |
19 |
30 |
60,222 |
17 |
| |
2 |
5 |
17 |
18 |
45,234 |
12 |
| |
2 |
6 |
19 |
32 |
34,623 |
8 |
| |
3 |
7 |
16 |
13 |
129,221 |
8 |
| |
3 |
8 |
15 |
22 |
166,251 |
8 |
| |
3 |
9 |
16 |
25 |
124,284 |
19 |
| |
4 |
10 |
25 |
24 |
31,629 |
14 |
| |
4 |
11 |
18 |
27 |
55,011 |
21 |
| |
4 |
12 |
16 |
38 |
41,615 |
14 |
| South Africa |
5 |
13 |
19 |
35 |
34,096 |
87 |
| |
5 |
14 |
19 |
35 |
21,386 |
Data unavailable |
| |
5 |
15 |
19 |
35 |
38,059 |
Data unavailable |
| |
6 |
16 |
15 |
37 |
72,544 |
Data unavailable |
| |
6 |
17 |
18 |
28 |
37,084 |
Data unavailable |
| |
6 |
18 |
14 |
36 |
44,821 |
53 |
| |
7 |
19 |
11 |
25 |
36,009 |
Data unavailable |
| |
7 |
20 |
11 |
25 |
82,953 |
Data unavailable |
| 7 | 21 | 12 | 18 | 45,067 | Data unavailable |
1Estimated from ZAMSTAR 2010 TB/HIV prevalence survey, for all Zambian communities, with age standardisation to the age structure of prevalence survey participants and assuming 50% of the adult population are men. For Western Cape communities, source of HIV prevalence data varies by triplet. For Triplet 5, community 13 was included in the ZAMSTAR trial and the ZAMSTAR 2010 TB/HIV prevalence survey data are used, as for Zambia. HIV prevalence is then assumed to be the same in communities 14 and 15. For communities 16, 17, 19, 20, and 21, sub-district level data on antenatal clinic (ANC) prevalence were used, with the assumption that adult HIV prevalence is 80% of the ANC prevalence value. Community 18 was included in the ZAMSTAR trial and the ZAMSTAR 2010 TB/HIV prevalence survey data are used.
2Estimated from ZAMSTAR 2010 TB/HIV prevalence survey data, for all Zambian communities. The number of HIV-positive adults among prevalence survey participants was estimated, separately for men and women, as the age-standardised HIV prevalence multiplied by the number of survey participants. The proportion of HIV-positive individuals on ART was then calculated as (number self-reported on ART)/(estimated number of HIV-positive survey participants), and assuming that 50% of the adult population are men. For Western Cape communities, data were used from October 2012 on (a) the number of individuals aged >15 years old on ART – measured either at community or sub-district level, (b) population size among individuals >15 years old – measured using census data either at community or sub-district level, and (c) HIV prevalence estimates. The number of HIV-positive individuals aged >15 years old was estimated as HIV prevalence × community (or sub-district) population size. The proportion of HIV-positive individuals on ART was then calculated as (number of individuals >15 years old on ART)/(estimated number of HIV-positive individuals aged >15 years old).
3Population size – for Zambia, based on 2001 census data; for Western Cape, based on 2011 census data.