Table 4 Overview of findings per study reporting on equity in ART adherence

Boyles 2011 *** Observational [30]

To determine the factors predicting loss to follow-up and mortality in a public-sector HIV and ART programme in rural South Africa

Rural area: Elliotdale/Xora area of Mbhashe sub-district in Eastern Cape province, combined public/donor program

Jan 2005 – Sept 2009

a. HIV + patients that loss to follow up (n = 117 (6.5%)),

Multiple Cox proportional hazard regression

Sex (not associated): females and males have same risk of being loss-to-follow-up: HR female: 1.42 [95% CI 0.90-2.23, p = 0.134] Age (associated, younger (<25 yrs) < older (25–50 yrs)): younger people have significant higher risk to loss-to-follow-up: HR <25 yrs (compared to 25–50 yrs): 1.87 [95% CI: 1.15-3.05, p = 0.012] Severity of disease (associated, ≥ 200 CD4 < <200 CD4): higher CD4 cell count significantly increases risk to loss-to-follow-up: 50–199 CD4 (referent); HR 0–49 CD4: 1.00 [95% CI: 0.61-1.64, p = 0.019]; HR ≥ 200 CD4: 1.74 [95% CI 1.09-2.78, p = 0.019]

b. HIV + patients that do not loss to follow up (n = 1686). Both groups are patients enrolled in clinics of Madwaleni HIV wellness and ART program including adherence counseling and home visits (i.e. Madwaleni Hospital, its 7 primary healthcare feeder clinics and a community based outreach program): tested HIV+, ART naïve at time of study enrollment, >19 years, initiated ART (CD4 < 200 CD4), could be follow for at least 3 months (n = 1803)

Orrell 2003 *** Observational [31]

To determine adherence of an indigent African HIV-infected cohort initiating ART to identify predictors of incomplete adherence and virologic failure

Urban area: Cape Town, Western Cape province, university of Cape Town clinic

Jan 1996 – May 2001

a. Patients discontinued 48 weeks of ART (n = 47),

T-test (age, VL, CD4 cell count), X² test (gender, socioeconomic status)

Sex (not associated): no significant difference in percentage female between those discontinued (40.4%) and completed (43.4%) 48 weeks of ART [p = 0.7] Age (associated, younger < older): those discontinued ART before 48 weeks were significantly younger (31 yrs) than those completed (34.1 yrs) [p <0.005] Severity of disease (contradicting results, CD4 cell count associated and WHO stages not associated): those discontinued ART before 48 weeks had significantly lower mean CD4 cell count (197) than those completed (268) [p < 0.01] / those discontinued before 48 weeks ART had a significantly higher VL (5.71 log₁₀) than those completed (5.49 log₁₀) [p <0.05] / no significant difference in percentage WHO stage 3 or 4 between those discontinued (49.2%) and completed (38.2%) 48 weeks of ART [p = 0.2] Socio-economic status (not associated): no significant difference in the percentage of patients with low socio-economic status in the group that discontinued (36.2%) and completed (43.6%) 48 weeks of ART [p = 0.4]

b. Patients that completed 48 weeks of ART (n = 242). Both groups are from Cape Town AIDS Cohort (CTAC): HIV + patients, presenting at University of Cape Town HIV clinics (referred by health care workers in the public sector of the wider Cape town area, mainly serving indigent populations), were ART naïve and eligible for adherence monitoring

Kranzer 2010 *** Observational [27]

To investigate the frequency and risk factors of defaulting treatment and identify factors associated with subsequent return to care in a long-term treatment cohort in South Africa

Peri-urban: township in Cape Town, Western Cape province, public clinic

Mar 2004 - Dec 2009

a. HIV + patients that defaulted ART (n = 291),

Multivariate Poisson regression

Sex (associated, men < women): compared to women, men have a significant increased risk to default ART treatment, HR men: 1.51 [95% CI: 1.18-1.93, p < 0.01] Age (not associated): no significant association between age and defaulting treatment, compared to younger age (≤30 years), HR > 30 years: 0.90 [95% CI: 0.70-1.15, p = 0.40] Severity of disease (contradicting results, CD4 cell count associated and WHO stages not associated): higher CD4 cell count increases significantly risk for defaulting treatment, ≤100 CD4 (referent); 101–200 CD4: HR 1.32 [95% CI: 0.99-1.76, p = 0.06], CD4 > 200 HR: 1.39 [95% CI 1.02-1.91, p = 0.04]. No significant difference in the risk of defaulting treatment being in WHO stage 3/4 or 1/2, HR stage 3/4: 1.14 [95% CI: 0.85-1.53, p = 0.37]

b. HIV + patients that not defaulted ART (n = 863). Both groups are from patients presenting at public-sector primary care clinic (single ART server in the area), >15 years, started ART (until 2007 < 350 CD4 cells (NIH research study), after 2007 < 200 CD4 cells (provincial ART program) (n = 1154)

Fatti 2010 *** Retrospective cohort study [29]

To compare clinical, immunological and virological outcomes between rural and urban children on ART in a large cohort from multiple public health facilities in four provinces of South Africa

Rural and urban: areas in Western Cape, KwaZulu-Natal, Eastern Cape and Mpumalanga province, public clinics supported by NGOs

Nov 2003 – Mar 2008

a. Children on ART that loss to follow up (n = 179),

Multivariable Cox proportional hazards regression

Sex (not associated): gender is not associated with risk of LTFU: HR male: 1.1 [95% CI: 0.82-3.12, no p value reported, 1.0 falls within CI] Age (not associated): younger children (<2 yrs) are as likely to LTFU than older children (>2 yrs): > 2 yrs (referent); HR 1–2 yrs: 1.61 [95% CI: 0.96-2.68, no p value reported, 1.0 in CI > 0.90]; HR < 1 yr: [1.81, 95% CI: 0.94-3.64, no p value reported, 1.0 in CI] Severity of disease (contradicting results, CD4 cell count not associated and WHO stages associated): severe clinical status is associated with risk LTFU: HR severe clinical status: 1.47 [95% CI: 1.03-2.12, no p value reported, 1.0 not within CI]/ severe immunodeficiency was associated with risk LTFU: HR severe immunodeficiency: 0.81 [95% CI: 0.52-1.24, p value not reported, 1.0 in CI Area of living (associated, rural/urban < urban/urban < rural/rural): patient in rural areas visiting clinics in urban areas are more likely to LTFU than patients from rural areas visiting rural clinics and patients in urban areas visiting urban clinics: rural (referent); HR urban: 1.14 [95% CI: 0.57-2.24]; HR rural/urban 2.85 [95% CI, 1.41-5.79] [p = 0.004]

b. Children on ART that do not loss to follow up (n = 2153). Both from retrospective cohort of children, (<16 yrs, ART naïve), enrolled in 44 routine public healthcare facilities (7 rural, 33 urban/12 secondary level hospitals, 32 primary health care clinics) supported by a NGO, used electronic data collection systems for patient monitoring. Children were divided in 3 groups a) urban residence and urban ART facility attended (urban group, n = 1727); rural residence and rural facility attended (rural group, n = 228); and rural residents attending urban facilities (rural/urban group, n = 377)

Cornell 2009 ** Observational [28]

To investigate the impact of gender and income on survival and retention in a South African public sector ART programme

Urban: Nyanga township, outskirts of Cape Town, Western Cape province, public clinics supported by NGOs

Sept 2002 – Apr 2007

a. HIV + patients that loss to follow up (n = 137),

Proportional hazards regression models

Sex (not associated): gender is not associated with risk to LTFU: HR men: 1.38, [95% CI: 0.94-2.03, p = 0.100] Age (not associated): no significant difference between age and risk to LTFU: HR age: 0.98 [95% CI 0.96-1.00, p = 0.102] Severity of disease (contradicting results, CD4 cell count associated and WHO stages not associated): patients with CD4 cell count <50 have higher risk to LTFU than CD4 cell count 50–150, but a similar risk as CD4 > 150: CD4 < 50 (referent); HR CD4 51–100: 0.62 [95% CI: 0.37-1.05, p = 0.077]; HR CD4 101–150 [0.57, 95% CI: 0.33-1.00, p = 0.049]; HR CD4 > 150: 1.01 [95% CI: 0.64-1.59, p = 0.971]/ WHO stage has no association with risk to LTFU: WHO stage I & II (referent); HR stage III: 0.78 [95% CI: 0.50-1.21, p = 0.274] HR stage IV: 0.75 [95% CI 0.75 (0.44-1.28), p = 0.294] /VL was not significantly associated with risk to LTFU: HR RNA level <5 log10 copies/ml (referent); >5 log: 1.13 [95% CI: 0.78–1.64, p = 0.520] Employment (associated, no income < income): patient with no income have a increased risk to LTFU: HR with income: 0.53 [95% CI: 0.37-0.77, p = 0.002]

b. HIV + patients that do not loss to follow up (n = 2059). Both groups from Gugulethu clinic patient cohort that receive adherence counseling including home visits, >15 years, ART naïve, WHO stage IV or CD4 < 200 (n = 2196)