Author,
Year, country
|
Objective
|
Study design, setting and participants
|
Participants’ age
|
Males (%)
|
Summary of findings
|
Study limitation
|
Burnett
2018,
Uganda [31]
|
To determine whether TB treatment outcomes and antiretroviral initiation of TB/HIV-coinfected patients are associated with TB/HIV integration status at rural health facilities.
|
Cross-sectional study involving 296 patients (117 under non-integrated care, and 179 under integrated care) in 14 rural health facilities drawn from all regions of Uganda
|
33+
|
57.8
|
TB/HIV integration was associated with lower mortality but not with antiretroviral initiation, TB treatment success, lost to follow-up, or failure.
|
Several sources of bias could not be controlled especially because of the retrospective data collection with many missing variables
|
Schwartz
2013, Botswana [32]
|
To determine if starting anti-tuberculosis treatment at clinics without co-located HIV clinics would delay time to highly active antiretroviral therapy initiation and be associated with lower survival compared to starting anti-tuberculosis treatment at clinics with on-site HIV clinics.
|
Historical cohort study of 439 patients (259 from clinics without and 80 from clinics with on-site HIV) in public clinics in Greater Gaborone (urban area)
|
-integrated:
35.5+
-non-integrated:
35+
|
55.2
|
Mortality did not differ between clinics without or with on-site HIV clinics nor did median time to antiretroviral initiation.
|
Lack of generalizability to rural areas, some bias due to missing data, some confounders could not be controlled for
|
Gandhi
2009,
South Africa[19]
|
To assess adherence to integrated TB and HIV treatment via concurrent home-based care
|
Prospective cohort study of 119
TB/HIV-coinfected patients in rural KwaZulu-Natal with once-daily antiretroviral therapy concurrently with TB therapy by home-based,
modified directly observed therapy.
|
34.0*
|
43.7
|
This model of integration resulted in excellent TB and HIV outcomes including adherence.
|
No randomisation; outcome measurement bias; small sample size
|
Herce
2018,
Zambia[33]
|
To assess the feasibility of integrating HIV care and treatment into TB clinics and to evaluate the effects of the intervention on timely linkage to HIV care, early ART uptake, and TB treatment outcomes.
|
Quasi-experimental study of routine program data for 473 patients to estimate outcomes before (June 1, 2010—January 31, 2011: 248 patients) and after (August 1, 2011—March 31, 2012: 225 patients) the intervention in 2 TB clinics in Lusaka (urban area)
|
-integrated:
33.3*
-non-integrated:
33.7*
|
56.7
|
Integration using a ‘one-stop shop’ model increased linkage to HIV care, rates of early ART initiation, and TB treatment success among patients with HIV-associated TB in Lusaka, Zambia
|
Substantial missing data on TB treatment outcomes; also, important confounders were not adjusted for
|
Conradie
2013,
South Africa[34]
|
To compare tuberculosis outcomes of patients who were co-infected with tuberculosis and HIV and attending an integrated tuberculosis/HIV clinic, to those of patients who were co-infected with tuberculosis and HIV who continued to receive tuberculosis treatment at their local clinic.
|
Historical cohort study involving data on 98 patients: 42 in the tuberculosis/HIV integrated clinic and 56 who received their tuberculosis treatment at the local clinic in Breed valley subdistrict (both rural and urban areas were included)
|
33.7*
|
45.0
|
There was a significantly better tuberculosis outcome in the cohort receiving integrated treatment
|
Unmatched groups; use of historical data
|
Schulz
2013,
South Africa[35]
|
To compare the outcomes of coinfected patients starting antiretroviral treatment in a tuberculosis hospital who received different models of care.
|
Prospective cohort study of 271 co-infected patients receiving either integrated care or the vertical care in a rural setting
|
-integrated:
35.4*
-non-integrated: 34.6*
|
39.9
|
Patient outcomes were better when TB and HIV care was received from the same service provider at the same visit (integrated care)
|
Outcomes occurring after 6 months were not assessed; poor quality of routine data used; purposive selection of study site increased risk of selection bias.
|
Hermans
2012, Uganda[36]
|
To evaluate antiretroviral therapy initiation and tuberculosis treatment outcomes before (2007) and after (2009) the implementation of an integrated TB/HIV clinic
|
Historical cohort study of 712 patients: routinely collected data for 366 patients initiating TB treatment in 2009 and with TB register data for 346 patients in 2007 in a large clinic in Makerere Uganda (urban area)
|
-integrated:
36.8*
-non-integrated: 37.0*
|
48.7
|
Treatment integration improved tuberculosis treatment outcomes and led to earlier initiation of antiretroviral therapy
|
use of routinely collected data with issues of missing data and limitations in outcome ascertainment;
|
Kerschberger
2012, South Africa[37]
|
1) to compare the time from the start
of tuberculosis treatment to antiretroviral therapy initiation in TB/HIV co-infected adults before and after complete integration of TB and HIV services
2) to describe a ‘‘one stop shop’’ model of fully integrated TB and HIV services.
|
Historical review of data on 188 co-infected patients in a primary care clinic in a South African township (urban area)
|
35+
|
49.5
|
Complete integration increased the rate of antiretroviral initiation and shortened time to antiretroviral initiation.
|
no randomization leading to increased risk of sampling
|
Ndagijimana
2015, Rwanda[38]
|
To evaluate one-stop TB-HIV services in
Rwanda by comparing the TB treatment outcomes before and after their implementation
|
Historical (before-and-after) study in 12 public facilities in rural and urban areas of Rwanda involving 888 patients: 413 post-intervention cases and 475 pre-intervention cases
|
------
|
------
|
One-stop integrated services were operational in all the facilities. TB treatment outcomes after
the intervention were comparable before the intervention.
|
Poor documentation: missing data and outcome ascertainment; limited control for confounders; few facilities were enrolled.
|
Owiti
2015,
Kenya[39]
|
To assess the uptake and timing of cotrimoxazole preventive therapy and antiretroviral treatment as well as anti-tuberculosis treatment outcomes
among HIV-infected TB patients before and after the introduction of integrated TB-HIV care.
|
Before-and-after study of routine data on 797 patients: 347 pre-integration (March-October 2010) and 450 post-integration (March-October 2012) cases in 17 rural public health facilities in Western Kenya
|
-integrated:
34+
-non-integrated: 35+
|
51.6
|
Integration of TB and HIV services enhanced
uptake and reduced delay in instituting CPT and ART in rural health facilities.
|
Difficult to compare outcomes before and after introduction of individual models presented in study. Lack of comparisons of biochemical data such as viral load. The before-and-after design may have introduced temporal trends that are not controlled for.
|
Jack
2004,
South Africa[20]
|
To determine the feasibility and effectiveness of integrating antiretroviral therapy into existing tuberculosis directly observed therapy programs
|
Prospective study of 20 co-infected patients in an urban tuberculosis clinic
|
31*
|
25.0
|
Integrating the programmes was feasible, led to high treatment success and facilitated monitoring of antiretroviral treatment outcomes
|
lack of generalizability: small sample size and only one facility used
|
|
*= Mean
+=Median
|