Title: Integration of HIV and TB services:
• a: Does ART provided at the TB clinic result in better outcomes than referring people with TB and HIV for ART in specialized HIV clinics?
• b: Does TB diagnosis and/or TB treatment at specialized HIV clinics result in better outcomes than referring people living with HIV to TB clinics for TB diagnosis and/or TB treatment?
Contents
1. PICO question ... 1
2. Search strategy ... 2
3. Flow diagram of screening process ... 3
4. Evidence summaries ... 3
4.1. Observational studies ... 3
5. Quality of evidence ... 9
5.1. PICO a: Does ART provided with TB treatment at the TB clinic result in better outcomes than referring people with TB and HIV for ART in specialized HIV clinics? ... 9
5.2. PICO b: Does TB diagnosis and/or TB treatment at specialized HIV clinics result in better outcomes than referring people living with HIV to TB clinics for TB diagnosis and/or TB treatment? .... 10
6. Bibliography of included studies ... 10
7. Excluded studies with reasons ... 15
1. PICO questions
a
Provision of HIV treatment and care at TB clinicsP People in countries with a high burden of TB and HIV coinfection
I ART and HIV care services provided with TB diagnosis and treatment at the TB clinics C Provision of ART and HIV care services separately from TB diagnosis and treatment
O HIV incidence, transmission, mortality, morbidity, access, retention and other outcomes to be noted in the protocol
b
Provision of TB treatment and care at HIV clinicsP People in countries with a high burden of TB and HIV coinfection I TB diagnosis and treatment at HIV care clinics
C Referral of PLHIV to TB clinics for diagnosis and treatment of TB
O HIV incidence, transmission, mortality, morbidity, access, retention, and other outcomes to be noted in the protocol
2. Search strategy
3. Flow diagram of screening process
4. Evidence summaries 4.1.Observational studies
4.1.1.ART provided at the TB clinic (where TB treatment is also provided) compared to referring people with TB and HIV for ART in specialized HIV clinics [PICO a]
Outcome: ART uptake “after” the TB clinic began providing ART (versus “before”). See Fig. 2 and 3. Twelve before-and-after studies show increased uptake after ART is started to be provided in TB clinics. The evidence was downgraded due to possible confounding by other interventions and events (such as changes in ART initiation criteria) during the study time frame. One study (Kerschberger 2012) adjusted for other factors influencing ART initiation, such as CD4 count. The study by van Rie (2008) directly addresses the comparison of interest. Two studies (Bygrave 2010, Kerschberger 2012) describe a variation on the intervention; instead of the TB clinic providing ART, a one-stop shop1 provides both TB and HIV diagnosis and treatment. In several studies, imprecision is an issue due to small numbers of patients with events. All but one (Bygrave 2010) of these studies report increase in ART uptake after the TB clinic began providing ART. Very low quality evidence from
1Médecins Sans Frontières (MSF), the technical partner in the Kerschberger et al. (2012) and Bygrave et al. (2010) studies, defines one-stop shops as clinics in which people with HIV and TB visit one clinician on the same day, in the same room, with a single patient chart and drug delivery point.
Records after duplicates removed (n = 461)
Full-text articles assessed for eligibility (n = 19) + 11 (identified
from cross-references)
= 29
Full-text articles excluded (n = 20)
Studies included (abstracts and full-
text articles) (n = 23) Additional records
identified through cross-referencing other sources
(n = 10)
Citations excluded based on details in abstracts
(n = 442)
Additional conference abstracts
identified (n = 13) (all
included)
• One study (van Rie 2008) in the Democratic Republic of the Congo found higher proportions of people with TB and HIV initiating ART after TB clinics began providing ART, compared with before, when people with TB and HIV were referred to an ART centre (RR 2.86, 95% CI 2.06–
3.97).
• Two studies (Pevzner 2011, Howard 2012) using national surveillance data found higher proportions of people with TB and HIV receiving ART2 in the years after TB clinics began to provide ART, compared to before this service delivery change. Pevzner et al. (2011) (Rwanda) found a RR of 4.84 (95% CI 4.34–5.41) comparing years after TB clinics began providing ART to before this service delivery change. Howard (2012) (Kenya) found a RR of 1.59 (95% CI 1.56–
1.62) comparing a similar change in service delivery.
• In one study (Kerschberger 2012), people with TB and HIV attending a nurse-run one-stop shop in South Africa were two times more likely to initiate ART (adjusted HR 1.6, 95% CI 1.11–2.29) compared to those attending the same primary care clinic in the year before when they were referred to an ART clinic in a separate facility.
• In one study (Bygrave, 2010), nurse-managed “one-stop shops” were established in rural primary health care clinics in Lesotho; before then, people with TB and HIV were referred to a district hospital for ART (and to initiate TB treatment). The proportion of eligible patients receiving ART two years after the “one-stop shops” were introduced was significantly decreased when compared to one year after this was introduced (RR 0.89, 95% CI 0.81–0.98).
• One study (Morse 2012) in Zambia found a higher proportion of people with TB and HIV
initiating ART after the TB clinics in two primary health centres began to provide ART compared to historical controls when patients were referred to a separate HIV clinic in the same health facilities (RR 2.27, 95% CI 1.45–3.56).
• Two studies (Chifundo 2010, Huerga 2010) found higher proportions of people with TB and HIV receiving ART after the TB clinic in a district hospital began providing ART compared to before when people with TB were referred to the separate HIV clinic within the same district hospital.
Chifundo (2010) (Malawi) found an RR of 1.72 (95% CI 1.54–1.94) when comparing one year after the TB clinic began providing ART to the year the intervention was introduced. Huerga (2010) (Kenya) found an RR of 5.40 (95% CI 3.37–7.94) when comparing the year after to the year before the intervention was introduced.
• One study (Phiri 2011) in Malawi found higher proportions of people with TB and HIV receiving ART during the first year when TB clinic began to provide ART, compared to the year before when their patients were referred within the same facility to the HIV clinicians to initiate ART (RR 1.16, 95% CI 1.09–1.23).
• In one study (Mugo, 2009) in Kenya, the proportion of people with TB enrolled on ART increased 45% after the TB clinic in the district hospital began to provide ART, but no denominators were available to make any statistical inference.
• One study (Odhiambo 2012) in Kenya found an increase in the percentage of people with TB receiving ART after the province implemented ranges of service integration models, but no denominators were available to make any statistical inference. (These data are presumably included in the national surveillance data in Howard (2011).)
• One study (Kambale, 2010) in the Democratic Republic of the Congo found that the number of people with TB receiving ART increased after TB clinics began providing ART, but
denominators are not specified.
2 At any time before TB treatment is completed, whether ART naive or already receiving ART at the time of TB diagnosis.
Outcome: ART uptake when provided by an HIV clinic collocated with the TB clinic (versus referral to a geographically separated HIV clinic). See Fig. 3. The quality of evidence in these studies is very low for several reasons. These are more indirect than the studies above; both the collocated and the geographically separated models required people with TB to be referred to specialized HIV clinics. In fact, the intervention model in these studies is used as the comparison in six of the studies listed above (Chifundo, Huerga, Morse, Mugo, Phiri and Kerschberger). In contrast to the before-and-after studies, the comparisons in these studies are made during the same calendar period so they are less subject to events over time. However, there is a risk of bias in the studies that did not adjust for other factors influencing ART uptake (such as CD4 count) or service enhancements in the intervention (besides the co-location of HIV and TB clinics). Imprecision is an issue in several of the studies. The results of these four studies are inconsistent; two found that ART uptake improved, but two found no statistically significant differences in ART uptake.
• In one study (Louwagie 2012) in South Africa, eligible people with TB attending a facility with collocated TB and HIV clinics were more than twice as likely to start ART during TB treatment than patients referred to start ART in geographically separated HIV clinics (subhazard ratio 2.49, 95% CI 1.06–5.88).
• In one study (Muvuma 2012) in Zambia, ART uptake was higher in people with TB and HIV attending health facilities with visiting (mobile) ART services than in those referred to start ART during treatment in geographically separated HIV clinics during the same time period (OR 8.82, 95% CI 0.56–138), although this estimate is not significant with a very wide confidence interval.
• One study (Schwartz, 2012) in Botswana found that patients of TB clinics with collocated HIV clinics were no more likely to initiate ART during TB treatment than patients from clinics without collocated HIV clinics (RR 0.95, 95% CI 0.75–1.20). (This comparison is calculated for this review based only on patients with ART initiation data.)
• One study (Ikeda, 2012) compared ART uptake in people with TB newly diagnosed with HIV in a regional TB hospital in Guatemala after an ART clinic was added versus before, when people with TB and HIV were referred elsewhere for ART. The proportion starting ART was
significantly higher once ART was available on site (OR 9.62, 95% CI 5.26–17.57). (The time frame for ART initiation relative to TB treatment is not specified.)
Outcome: timeliness of ART initiation among people who began ART during TB treatment. See Fig.
3 and 4.
• In one study (Phiri 2011), during the first year TB officers in an HIV and TB clinic in Malawi began to provide ART, people with TB and HIV were more likely to start ART during the first two months of TB treatment compared to before when people with TB were referred down the hall to HIV care clinic (RR 2.22, 95% CI 1.93 to 2.56).
• One study (Schwartz 2012), there was no difference in the proportion starting ART within 60 days of TB treatment initiation, when patients started TB treatment in facilities with collocated HIV clinics were compared to those referred for ART in geographically separate HIV clinics (RR 0.83, 95% CI 0.56 to 1.24). Since baseline CD4 count was associated with time to ART initiation, authors also looked at median time to ART initiation in those with CD4 counts of 100 cells/mm3 or less, and found no difference by clinic type.
• In one study (Lawn 2011) in South Africa, patients with CD4 counts lower than 50 cells/mm3 starting TB treatment in an ART clinic were more likely to start ART within four weeks after TB treatment initiation than patients referred from TB clinics (RR 4.91, 95% CI 1.98 to 12.2 for the
most recent calendar period). In a competing risk regression analysis, the time from TB diagnosis to start of ART was substantially shorter for patients with TB diagnosed in the ART clinics, compared to patients referred from TB clinics (adjusted sub-hazard ratio 0.53, 95% CI 0.43 to 0.66).3
Outcome: referral or enrolment into HIV care. See Fig. 3. The two studies that reported on these process indicators provide very low quality evidence due to study bias and indirectness. In addition, Morse has very few cases (imprecision).
• One study (Louwagie 2012) in South Africa, eligible people with TB attending a facility with collocated TB and HIV clinics were more likely to be referred for HIV-related care than in geographically separated HIV clinics (RR 2.75, 95% CI 2.28–3.32).
• One study (Morse 2012) in Zambia found a higher proportion of people with TB and HIV enrolling in HIV care after the TB clinics in two primary health care centres began to provide ART, compared to historical controls, when patients were referred to a separate HIV clinic in the same health centres (RR 1.65, 95% CI 1.18–2.32).
Outcome: mortality. See Fig. 2 and 3. The quality of evidence is very low and was downgraded for imprecision (few patients), indirectness (only van Rie (2008) reports the same intervention and comparator as the PICO question), and study limitations including differential follow up, and no
adjustment for confounders (van Rie 2008, Ikeda 2012). Given these differences, it is not surprising that the three studies found inconsistent results.
• One study (Schwartz 2012) in Botswana found no significant difference in mortality during TB treatment among people with HIV and TB with CD4 counts ≤250 cell/mm3 who started outpatient TB treatment in TB clinics with or without collocated HIV clinics (RR 1.27, 95% CI 0.58–2.76).
• One study (van Rie 2008) in the Democratic Republic of the Congo found a lower proportion of people living with HIV died during TB treatment compared to the people with TB and HIV registered before the clinics began providing ART (RR 0.36, 95% CI 0.17–0.77).
• One study (Ikeda 2012) evaluated the impact of adding an ART clinic to a regional TB hospital in Guatemala. Compared to before, when people with TB and HIV were referred elsewhere for ART, the OR for death at 50 weeks of follow up was 0.08 (95% CI 0.05–0.15).
Outcomes during TB treatment: TB treatment success.4 See Fig. 2 and 3. Two studies report only one TB treatment outcome (success) provide very-low-quality evidence due to indirectness of the comparator and/or intervention model and the lack of adjustment for other factors affecting mortality among people living with HIV (such as ART and baseline CD4 count) or other interventions during the study period that were likely to have contributed to improved outcomes.
• In one study (Bygrave, 2010) in Lesotho, the proportion with TB treatment success was not significantly higher in the cohort of people with TB and HIV registered two years after the
3 Reciprocal of the published article, since this systematic review compares patients starting TB treatment in the HIV clinic to those referred from TB clinics.
4 TB treatment success includes patients with documented cure (acid-fast bacilli smear-negative sputum) and patients completing their course of treatment without documentation of cure.
creation of a one-stop shop compared to one year after (RR 1.08, 95% CI 0.96–1.21).
• In one study (Huerga 2010), the proportion with TB treatment success was not significantly higher in the cohort of patients receiving ART starting TB treatment four to five years after the TB clinic began to provide ART compared to the cohort starting TB treatment (and ART) two years after (RR 0.93, 95% CI 0.83–1.03). Similarly, there was no difference in TB treatment success among people with TB and HIV not receiving ART (RR 1.06, 95% CI 0.88–1.27).
All TB treatment outcomes. See Fig. 3. The two studies that reported on TB treatment outcomes provide very-low-quality evidence due to indirectness, lack of adjustment for confounders such as receiving ART (Kaplan 2012) or CD4 count at ART initiation (Dube 2010) and imprecision.
• In one study (Kaplan 2012),5 people with TB and HIV registered in facilities providing both TB and ART were less likely to achieve TB treatment success when compared to patients who were referred off site for ART (RR of TB treatment success 0.90, 95% CI 0.88–0.92). There was no difference in TB treatment failure for people with TB and HIV registered in facilities providing both TB treatment and ART compared to those providing TB treatment with ART offsite (RR of TB treatment failure 1.15, 95% CI 0.79–1.67). Further, patients in facilities providing both TB treatment and ART were more likely to die or default from TB treatment than patients referred off site for ART (RR of death 1.26, 95% CI 1.04–1.52; RR of defaulting 1.19, 95% CI 1.02–1.39).
• One study (Dube 2010) found no differences in TB treatment outcomes for patients receiving ART after mobile ART services were provided on site, compared to before, when patients were referred off site to receive ART (RR for treatment success 1.21, 95% CI 0.98–1.51; RR for TB treatment failure or death 0.38, 95% CI 0.13–1.07; RR for defaulting 0.61, 95% CI 0.90–4.12).
4.1.2 TB diagnosis and/or treatment at HIV clinics compared to referring people living with HIV to TB clinics for TB diagnosis and/or TB treatment [PICO b]
Outcome: proportion of people living with HIV screened for TB. See Fig. 6.
• One study (Howard 2009) in six African countries found that the proportion of people living with HIV screened for TB using a symptom questionnaire was higher in HIV facilities with TB treatment on site than in HIV sites that had to refer for TB treatment off site (RR 1.36, 95% CI 1.32–1.40). The evidence is very low quality due to risk of bias (weighted proportions were used to calculate the proportion screened for TB).
Outcome: ART uptake. See Fig. 7.
This evidence is very low quality due to imprecision and possible confounding by other interventions during the study periods that accompanied the reorganization of services. This study also provides indirect evidence, since the comparisons are different from the PICO question.
• One study (Hermans 2012) in Uganda found a lower proportion of people with TB and HIV starting ART one year after the creation of a TB and HIV unit within the HIV service compared to before when people with TB were referred to a discrete TB clinic (also within the HIV service)
5 One of this study’s 13 facilities providing both TB treatment and ART is the same clinic described in Brown (2011).
(RR 0.86, 95% CI 0.75–0.99). The authors attribute this decrease to clinicians appropriately giving priority to patients with lower CD4 counts when faced with an ART shortage. Patients with CD4 counts ≥250 cells/mm3 were less likely to receive ART after the creation of a TB and HIV unit compared to before (RR 0.40, 95% CI 0.21–0.76).
Outcome: timeliness of ART initiation. See Fig. 7.
• One study (Hermans 2012) in Uganda found that a higher proportion of people with TB and HIV started ART during the intensive phase of TB treatment among those starting TB treatment one year after a TB and HIV unit was created within the HIV service compared to before when people with TB were referred to a discrete TB clinic (RR 2.62, 95% CI 1.90–3.58). This evidence is downgraded for imprecision, indirectness as well as possible confounding (including providers learning during the study period of new evidence supporting earlier ART).
Outcomes of TB treatment. See Fig. 7 and 8. This is very-low-quality evidence. Since the intervention in Bento (2010) and Cerda (2011) are home care models, these provide indirect evidence. Evidence was also downgraded for possible confounding (directly observed therapy in Bento (2010)), by other
interventions during the study period (all three studies) and differential ascertainment of death (Hermans 2012).
• In one study (Hermans 2012) in Uganda, one year after the creation of a TB and HIV unit within the HIV service, there was no difference in TB treatment success for people living with HIV started on ART (RR 1.04, 95% CI 0.92–1.17) and no difference in death or loss to follow-up (RR 0.83, 95% CI 0.49–1.38) compared to before when people with TB were referred to a discrete TB clinic (also within the HIV service). By contrast, these outcomes improved for people with TB not receiving ART after the creation of the TB and HIV unit compared to before when they were referred to a discrete TB clinic (RR for treatment success 1.59, 95% CI 1.07–2.36; RR for death or loss to follow up 0.65, 95% CI 0.49-0.87)
• One study (Bento 2010) in Brazil found no difference in TB treatment success (RR 1.03, 95% CI 0.81–1.30) and lower defaulting (RR 0.42, 95% CI 0.24–0.76) but increased mortality (RR 3.33, 95% CI 0.98–11.37) among people with TB and HIV receiving weekly home visits (delivering TB medicines and ART for administration by a caregiver) compared to patients who picked up their TB medicines and ART from a hospital outpatient department for self-administration.
• In one study (Cerda 2011) in Peru, patients receiving directly observed TB treatment and supervised ART at home had a lower hazard of dying or defaulting from treatment (adjusted HR 0.34, 95% CI 0.12–0.98) than people with TB and HIV referred to district health centres for directly observed TB treatment. Patients receiving directly observed TB treatment and supervised ART at home had a nonsignificant increase in likelihood to achieve cure (adjusted RR 1.28, 95%
CI 0.90–1.83) than people with TB and HIV referred to district health centres for directly observed TB treatment.
Outcomes: resource utilization. See Fig. 8.
• In one study (Cerda 2011) in Peru, patients receiving directly observed TB treatment and supervised ART at home used fewer total hospital days per person per year of observation than people with TB and HIV referred to district health centres for directly observed TB treatment (adjusted incidence rate ratio 0.37, 95% CI 0.14–0.99). This study provides very-low-quality evidence due to indirectness and confounding by other interventions implemented along with the home delivery model.
5. Quality of evidence
5.1. PICO a: Does ART provided with TB treatment at the TB clinic result in better outcomes than referring people with TB and HIV for ART in specialized HIV clinics?
For ART uptake, the quality of the evidence is very low. The most direct evidence comes from the 11 studies comparing ART uptake after versus before ART began being provided by TB clinics (Fig. 2, top of Fig. 3). However, only Kerschberger et al. (2012) adjusted for other patient factors influencing ART initiation (such as CD4 count). In most studies, the change in service delivery was one part of a package of additional interventions to improve TB and HIV service coordination. There was also possible confounding by events during the study period, such as the release of new guidelines to initiate ART sooner after TB treatment initiation and among patients with higher CD4 counts. Several studies had small numbers of patients, leading to imprecise measurements. This group of studies reports fairly consistent increases in ART uptake after the TB clinic began providing ART.
The three studies comparing collocated clinics to geographically separate clinics (bottom of Fig. 3) during one period (Louwagie 2012, Muvuma 2012, Schwartz 2012) avoid the problem of confounding by events over calendar time and also adjusted for several confounders. However, the evidence from these studies is less direct, since both the collocated and geographically separated clinics require people with TB to be referred to specialized HIV clinics (Fig. 3). The results of these three studies are less consistent; only Louwagie (2012) found a statistically significant improvement in ART uptake. Ikeda (2012) also
compared ART uptake in collocated versus geographically separate clinics but used an after versus before design without adjustment for confounding by calendar events, other interventions during the time frame or patient characteristics.
For timeliness of ART initiation, the quality of the evidence is very low. The three relevant studies do not directly address the question; the intervention was referral to a collocated HIV clinic (Schwartz 2012) or an HIV clinic providing TB treatment (Lawn 2011), and the comparator was referral within the same facility (Phiri 2011). Given three different comparisons (Fig. 3, 4), inconsistent results would be expected.
Two of the three studies found higher proportions of people with TB and HIV initiating ART within the first one to two months, but Schwartz (2012) found no difference.
For TB treatment outcomes, the quality of the evidence is very low. Three studies (Schwartz (2012), van Rie (2008) and Ikeda (2012)) reported mortality in people with TB and HIV (without information on other TB treatment outcomes). The evidence was downgraded for imprecision (few patients), indirectness (only van Rie (2008) reports the same intervention and comparator as the PICO question), and study limitations including differential follow-up, and no adjustment for confounders (van Rie 2008, Ikeda 2012). Given these differences, inconsistent results would be expected.
Two studies reported no significant change in TB treatment success (Bygrave 2010, Huerga 2010). The quality of evidence is very low due to indirectness and the lack of adjustment for other factors affecting mortality among people living with HIV (such as ART and CD4 count) or other interventions during the study period that were likely to have contributed to improved outcomes.
The two studies that reported on all TB treatment outcomes provide very low quality of evidence due to indirectness, lack of adjustment for confounders such as receiving ART (Kaplan 2012) or CD4 count (Dube 2010) and imprecision.
We found no studies that explored such outcomes as viral suppression, immune recovery, immune reconstitution inflammatory syndrome, adverse events or retention on ART during or after TB treatment.
5.2.PICO b: Does TB diagnosis and/or TB treatment at specialized HIV clinics result in better outcomes than referring people living with HIV to TB clinics for TB diagnosis and/or TB treatment?
For TB screening, the quality of the evidence is very low due to risk of bias.
For ART uptake and the timeliness of ART initiation, the quality of the evidence is also very low due to imprecision, indirectness and possible confounding by other interventions during the study periods that accompanied the reorganization of services.
For TB treatment outcomes among people living with HIV, the quality of the evidence was downgraded to very low for indirectness, confounding by other interventions and differential ascertainment of death.
The evidence for resource utilization was also very low quality, due to indirectness and confounding by other interventions.
We found no studies that reported outcomes of viral suppression, immune recovery, retention in ART or timeliness of TB diagnosis.
6. Figures
Fig 1. Does ART provided with TB treatment at the TB clinic result in better outcomes than referring HIV-infected TB patients for ART in specialized HIV clinics?
Fig 2. After versus before
TB HIV TB
Provides ART Refers for ART
Fig 3. After versus before
Fig 4. ART initiation among people who began ART during TB treatment TB
HIV TB
Provides ART
Refers for ART Kambale,* Pevzner, Howard, van
Rie
* “before” model not specified
TB/HIV
Same nurse provides TB treatment and ART
Kerschberger, Bygrave,Odhiambo**
**A few % of settings used “partially integrated” or
“collocated” models.
HIV TB
TB staff provides ART Refers for ART
Chifundo, Huerga, Morse, Mugo, Phiri
TB HIV
Fig 5. Does TB diagnosis and / or TB treatment at specialized HIV clinics result in better outcomes than referring HIV-infected patients to TB clinics for TB diagnosis and/or treatment?
Fig 6. 159 HIV sites versus 22 HIV sites that refer that treat TB HIV-infected patients in care at HIV
clinic; found to have TB; start ART and TB treatment
HIV-infected TB patients referred from TB Clinics to specialized HIV clinic to start ART
HIV
TB
Lawn
TB
Referrals for ART
HIV HIV TB
Treat and/or diagnose TB Refer for TB treatment and/or diagnosis
Fig 7. After creation of TB / HIV unit versus before
Fig 8. HIV team visits patient’s home to provide TB treatment versus patient travels to TB clinic for TB treatment
HIV
Howard, 2009
TB HIV
Treat TB
Refer for TB treatment
HIV
HIV
TB
Hermans
Refers for TB treatment TB/HIV
7. Bibliography of included studies Peer-reviewed articles
1. Bento C, Pedroso E. Assessment of the effectiveness of a home-based care program for patients coinfected with tuberculosis and human immunodeficiency virus after discharge from a reference hospital in south-eastern Brazil. Braz J Infect Dis 2010; 14(6): 594-600.
2. Cerda R, Munoz M, Zeladita J, Wong M, Sebastian JL, Bonilla C, Bayona J, Sanchez E, Arevalo J, Caldas A, Shin S. Health care utilization and costs of a support program for patients living with the human immunodeficiency virus and tuberculosis in Peru. Int J Tuberc Lung Dis 2011; 15: 363-368.
3. Hermans SM, Castelnuovo B, Katabira C, Mbidde P, Lange JM, Hoepelman AI, Coutinho A, Manabe YC. Integration of HIV and TB services results in improved TB treatment outcomes and earlier prioritized ART initiation in a large urban HIV clinic in Uganda. J Acquir Immune Defic Syndr 2012;
60: e29-35.
4. Howard A, Gasana M, Getahun H, Harries A, Lawn S, Miller B, Nelson L, Sitienei J, Coggin W.
PEPFAR support for the scaling up of collaborative TB/HIV activities. JAIDS 2012; 60(3): S136- 144.
5. Huerga H, Spillane H, Guerrero W, Odongo A, Varaine F. Impact of introducing human immunodeficiency virus testing, treatment and care in a tuberculosis clinic in rural Kenya. Int J Tuberc Lung Dis 2010; 14(5): 611-15.
6. Kerschberger B, Hilderbrand K, Boulle A, Coetzee D, Goemaere E, De Azevedo V, Van Cutsem G.
The effect of complete integration of HIV and TB services on time to initiation of antiretroviral therapy: a before-after study. PLoS ONE 2012; 7(10):e46988.
7. Lawn SD, Campbell L, Kaplan R, Little F, Morrow C, Wood R. Delays in starting antiretroviral therapy in patients with HIV-associated tuberculosis accessing non-integrated clinical services in a South African township. BMC Infectious Diseases 2011; 11:258.
8. Louwagie G, Girdler-Brown B, Odendall R, Rossouw T, Johnson S, Van der Walt M. Missed opportunities for accessing HIV care among Tshwane tuberculosis patients under different models of care. Int J Tuberc Lung Dis 2012; 16(8): 1052-58.
TB
Bento,* Cerda
*organization of TB and HIV services in outpatient department is not specified
9. Pevzner E, Vandebriel G, Lowrance D, Gasana M, Finaly A. Evaluation of the rapid scale-up of collaborative TB/HIV activities in TB facilities in Rwanda, 2005-2009. BMC Public Health 2011;
11:550.
10. Phiri S, Khan PY, Grant AD, Gareta D, Tweya H, Kalulu M, Chaweza T, Mbetewa L, Kanyerere H, Weigel R, Feldacker C. Integrated tuberculosis and HIV care in a resource-limited setting: experience from the Martin Preuss centre, Malawi. Trop Med Int Health 2011; 16:1397–403.
Conference abstracts
1. Bygrave H, Trivino L, Makakole L. TB/HIV integration: lessons learned from implementation of a TB/HIV “one stop shop” at primary health care clinics in rural Lesotho. AIDS 2010 – XVIII International AIDS Conference: Abstract no. MOAB0301.
2. Chifundo K, Wandwalo E, Kanyerere H, et al. What is the best model of TB/HIV service delivery?
Experience from Malawi. AIDS 2010 – XVIII International AIDS Conference: Abstract no. MOPE0858.
3. Dube C, Kayama N, Miyano S, et al. Step forward to health system strengthening: the impact of scaling up of ART services on TB services in rural settings, Zambia. AIDS 2010 – XVIII International AIDS Conference: Abstract no. THAE0104.
4. Howard AA, Saito S, Nash D, Flam R, Elul B, Scardigli A, Oyeledun B, Cunningham A, Hoos D, and El-Sadr W. On-Site location of TB services is associated with TB screening of HIV-infected patients at enrollment in HIV care programs in 6 sub-Saharan African countries. 2009 CROI. #590.
5. Ikeda J, Page K, Hudes E, Barrios R, López Tellez CA, Hearst N. HIV and TB and Integration reduces mortality among the indigenous population in rural Guatemala. AIDS 2012.
6. Kambale A, Okiata JP, Okenge A, et al. Impact of the expansion of NTP-based HIV care on TB outcomes in DR Congo. UNION. 2010. PS-100688-14.
7. Kaplan R, Caldwell J, Bekker L-G, Jennings K, Lombard C, Enarson D, Beyers N, Wood R.
Provision of ART in TB facilities in Cape Town South Africa: impact on TB treatment outcomes.
UNION. 2012. Abstract # OP-147-16.
8. Morse J, Luhanga D, Harris J, Musople R, Besa S, Nhandu V, Samungole G, Kancheya N. Integrated TB/ART clinics in Lusaka, Zambia: an evaluation of enrollment into HIV care and early initiation of antiretroviral therapy in TB/HIV co-infected patients. UNION. 2012. Abstract #PC-545-17.
9. Mugo P, Gachengo J, Muguongo R, et al. Integrating TB and HIV care services: experience from a rural district hospital in Kenya. UNION. 2009. PS-94524-07.
10. Muvuma S, Miyano S, Ishikawa N, Sikaze I, Moyo C, Kapyata H, Changala M, Msiska C, Syakantu G. Poor linkages between TB and HIV services affects the quality of care; a retrospective cohort study of TB/HIV patients from HIV testing to ART initiation in a rural setting in Zambia. AIDS 2012.
11. Odhiambo J, Gondi J, Miruka F, Muttai H, Cain K, Ochanda B, Sitienei JK, Nganga L, Soti D.
Models of TB-HIV integration and accomplishments in Nyanza Province, Kenya. UNION 2012. PC- 542-17.
12. Schwartz A, Tamuhia N, Steenhoff A, Nkakana K, Letlhogile R, Chadborn T, and Bisson G.
Outcomes among HIV+ adults with active pulmonary TB treated in clinics with and without on-site HIV clinics—a retrospective cohort study: Botswana. 2012 CROI. #928.
13. Van Rie A, Roger I, Edmonds A, et al. Implementation of primary health care ART model for HIV co-infected TB patients improves treatment outcomes. UNION. 2008. TS-82281-19.
8. Excluded studies with reasons
Reference Reason for exclusion
(missing element population, intervention, comparator) Abdool Karim SS, Naidoo K, Grobler A, Padayatchi N, Baxter C,
Gray A, Gengiah T, Nair G, Bamber S, Singh A, Khan M, Pienaar J, El-Sadr W, Friedland G, Abdool Karim Q. Timing of initiation of antiretroviral drugs during tuberculosis therapy. N Engl J Med 2010;
362: 697-706.
No comparison for site of ART.
(C)
Atkins S, Lewin S, Jordaan E, Thorson A. Lay health worker- supported tuberculosis treatment adherence in South Africa: an interrupted time-series study. Int J Tuberc Lung Dis 2011; 15:1, 84- 89, i.
No description of HIV status or ART. (P, I)
Beversluis D, Musoke R, Mandima P, Nyamapfeni P, Chipato T, Mukasa JB, Byamugisha J, Mugerwa R, Morrison C, Salata RA.
Incidence of major clinical outcomes in a cohort of Ugandan and Zimbabwean women with HIV-1 infection followed from
seroconversion. International Journal of STD and AIDS 2012; 23:
111-114.
No details of where TB diagnosis and/or treatment were provided in either country. (I, C)
Bonnet MMB, Pinoges LLP, Varaine FFV, Oberhauser BBO, O’Brien DDO, Kebede YYK, Hewison CCH, Zachariah RRZ, Ferradini LLF. Tuberculosis after HAART initiation in HIV-positive patients from five countries with a high tuberculosis burden. AIDS 2006 20: 1275-1279.
No comparison of TB diagnosis and/or treatment in HIV clinic versus TB clinics. (I, C)
Brust JC, Shah NS, Scott M, Chaiyachati K, Lygizos M, van der Merwe TL, Bamber S, Radebe Z, Loveday M, Moll AP, Margot B, Lalloo UG, Friedland GH, Gandhi NR. Integrated, home-based treatment for MDR-TB and HIV in rural South Africa: an alternate model of care. Int J Tuberc Lung Dis 2012;16(8):998-1004.
Multidrug-resistant TB service providing ART (but no
comparator of referral to HIV service for ART). (C)
Burgess AL, Fitzgerald DW, Severe P, Joseph P, Noel E, Rastogi N, Johnson WD Jr, Pape JW. Integration of tuberculosis screening at an HIV voluntary counselling and testing centre in Haiti. AIDS
2001;15(14):1875-9.
Study describes cough screening in a voluntary counselling and testing centre with referral for TB evaluation. There is no comparison to another model of TB diagnostic service delivery.
(C) Chan AK, Njala J, Kanyerere H, van Lettow M. Improved update of ART following integration of TB and HIV services in a district in southern Malawi. 5th IAS Conference on HIV Pathogenesis and Treatment [Abstract no. CDD062] 2009.
We included a more recent abstract (Chifundo 2010) presenting more detailed data on the same hospital. Kanyerere is an author on both.
Coetzee D, Hilderbrand K, Goemaere E, Matthys F, Boelaert M.
Integrating tuberculosis and HIV care in the primary care setting in South Africa. Trop Med Int Health 2004; 9: A11-15.
No comparison of ART in TB vs.
HIV clinic to address PICO 4a.
No comparison of TB diagnosis and/or treatment in HIV versus TB clinics to address PICO 4b.
(I,C) Desvarieux M, Hyppolite PR, Johnson WD, Pape JW. A novel
approach to directly observed therapy for tuberculosis in an HIV- endemic area. American Journal of Public Health 2001; 91: 138-141.
No comparison of TB diagnosis and/or treatment in TB clinic.
(C) Durovni B, Cavalcante SC, Saraceni V, Vellozo V, Israel G, King
BS, Cohn S, Efron A, Pacheco AG, Moulton LH, Chaisson RE,
No description of diagnosis or treatment of active TB. (I)
Golub JE. The implementation of isoniazid preventive therapy in HIV clinics: the experience from the TB/HIV in Rio (THRio) study.
AIDS 2010; 24(Suppl 5):S49-56.
Eibach D, et al. Decentralized integrated TB/HIV program in an unstable environment (Nimba county, Liberia). AIDS 2006 – XVI International AIDS Conference [Abstract no. THPE0192].
No description of how the
“integrated” care model worked;
no comparison is provided. (I, C) Farmer P, Leandre F, Mukherjee J, Gupta R, Tarter L, Kim JY.
Community-based treatment of advanced HIV disease: introducing DOT-HAART (directly observed therapy with highly active antiretroviral therapy). Bull World Health Organ 2001; 79: 1145- 1151.
Patient population had HIV infection but not TB. (P)
Friedland G, Churchyard GJ, Nardell E. Tuberculosis and HIV coinfection: current state of knowledge and research priorities. J Infect Dis 2007;196(Suppl 1):S1-3.
Review article
Gandhi NR, Moll AP, Lalloo U, Pawinski R, Zeller K, Moodley P, Meyer E, Friedland G; Tugela Ferry Care and Research (TFCaRes) Collaboration. Successful integration of tuberculosis and HIV treatment in rural South Africa: the Sizonq’oba study. J Acquir Immune Defic Syndr 2009;50(1):37-43.
No comparison is provided. (C)
Gasana M, Vandebriel G, Kabanda G, Tsiouris SJ, Justman J, Sahabo R, Kamugundu D, El-Sadr WM. Integrating tuberculosis and HIV care in rural Rwanda. Int J Tuberc Lung Dis 2008;12(3 Suppl 1):39-43.
Compares before/after for provider initiated HIV testing, but not ART. (I)
Harris JB, Hatwiinda SM, Randels KM, Chi BH, Kancheya NG, Jham MA, Samungole KV, Tambatamba BC, Cantrell RA, Levy JW, Kimerling ME, Reid SE. Early lessons from the integration of tuberculosis and HIV services in primary care centres in Lusaka, Zambia. Int J Tuberc Lung Dis 2008;12(7):773-9.
Compared separate HIV and TB clinics before and after referral mechanisms were strengthened.
There is no comparison across delivery models. (I, C)
Hilderbrand K, et al. From TB/HIV services coordination to TB/HIV service integration: different model of service delivery experimented in Khayelitsha, South Africa. 5th IAS Conference on HIV
Pathogenesis and Treatment [Abstract no. TUPED118] 2009.
We included a more recent (Brown, 2011) abstract by 5 of these authors from Khayelitsha Jack C, Lalloo U, Karim QA, Karim SA, El-Sadr W, Cassol S,
Friedland G. A pilot study of once-daily antiretroviral therapy integrated with tuberculosis directly observed therapy in a resource- limited setting. Journal of Acquired Immune Deficiency Syndromes 2004; 36: 929-934.
No comparison of referral to HIV clinics for ART. (C)
Kwange SO, Budambula NL. Effectiveness of anti-tuberculosis treatment among patients receiving highly active antiretroviral therapy at Vihiga district hospital in 2007. Indian J Med Microbiol.
2010;28(1):21-5.
There is no comparison of TB clinics providing ART versus referring to specialized HIV clinics for ART to address PICO 4a. (I, C)
Médecins Sans Frontières. Khayelitsha annual activity report 2008–
2009: providing HIV/TB care at the primary health care level.
February 2010.
We included a more recent abstract (Brown 2011) that provides more details on service delivery model and patient outcomes.
Miti S, Mfungwe V, Reijer P, Maher D. Integration of tuberculosis treatment in a community-based home care programme for persons living with HIV/AIDS in Ndola, Zambia. Int J Tuberc Lung Dis
Did not specify HIV status of the patients; provides only an estimate of 70–80%. (P)
2003;7(9 Suppl 1):S92-8.
Njozing NB, Miguel SS, Tih PM, Hurtig AK. Assessing the accessibility of HIV care packages among tuberculosis patients in the Northwest Region, Cameroon. BMC Public Health 2010;10:129.
No description or comparison of different models for providing ART. (I, C)
Odhiambo J, Kizito W, Njoroge A, Wambua N, Nganga L, Mburu M, Mansoer J, Marum L, Phillips E, Chakaya J, De Cock KM.
Provider-initiated HIV testing and counselling for TB patients and suspects in Nairobi, Kenya. Int J Tuberc Lung Dis 2008;12(3 Suppl 1):63-8.
This study does not address PICO 4a because there is no comparison of TB clinics providing ART versus referral to HIV service. It does not speak to PICO 4b because there is no comparison of HIV services diagnosing or treatment TB versus referral to TB clinics. (I, C)
Okot-Chono R, Mugisha F, Adatu F, Madraa E, Dlodlo R, Fujiwara P. Health system barriers affecting the implementation of
collaborative TB-HIV services in Uganda. Int J Tuberc Lung Dis 2009;13(8):955-61.
Provides no data comparing service delivery models. (I,C)
Roger I, et al. Integration of family planning, HIV and TB treatment at primary health care centres in Kinshasa, Democratic Republic of the Congo. AIDS 2008 – XVII International AIDS Conference [Abstract no. WEPE0805].
Provides no comparison to TB nurses providing ART. (C) Note that we included an abstract (van Rie) by four of the same authors presented the same year (2008) at the UNION conference.
Sackoff JE, Torian LV, Frieden TR. TB prevention in HIV clinics in New York City. Int J Tuberc Lung Dis 2001; 5: 123-128
No comparison. (C) Sweeney S, Obure CD, Maier CB, Greener R, Dehne K, Vassall A.
Costs and efficiency of integrating HIV/AIDS services with other health services: a systematic review of evidence and experience.
Sexually Transmitted Infections 2012; 88: 85-99.
Review of literature.
van Engelgem I, Telfer B, Kizito W, Ombeka VO, Raganoud M, Zachariah R. High acceptability of cotrimoxazole and antiretroviral treatment among HIV infected tuberculosis patients offered
integrated services in three health centres in Kibera, Nairobi.
Médecins Sans Frontières, 2008.
Integration model not described.
No comparison is reported. (I, C)
van Lettow M, Chan AK, Ginsburg AS, Tweya H, Gareta D, Njala J, Kanyerere H, Phiri S, Idana I. Timing and uptake of ART during treatment for active tuberculosis in HIV co-infected adults in Malawi. Public Health Action 2011; 1(1): 6-9.
No comparison of alternative models of care. (C)
Vassall A, Seme A, Compernolle P, Meheus F. Patient costs of accessing collaborative tuberculosis and human immunodeficiency virus interventions in Ethiopia. Int J Tuberc Lung Dis
2010;14(5):604-10.
No description of model of HIV and TB services is provided, nor comparison of outcomes by how care is organized. (I, C)
Wannasorn P. Integrated TB and HIV services: an approach for quality improvement of care: AIDS 2006 – XVI International AIDS Conference [Abstract no. THPE0201].
Note that there was no abstract with this number in the 4th IAS conference; we found the same author and abstract number in IAC 2006 (with a different title and additional authors).
There is no description or comparison of models for ART provision, so doesn’t address PICO 4a. A multidisciplinary team began to provide comprehensive HIV care including “intensive TB
screening for HIV-infected patients,” but there is no report of where TB diagnosis is confirmed now or before the multidisciplinary team was instituted, so this does not address PICO 4b. (I, C) Wiktor SZ, Sassan-Morokro M, Grant AD, Abouya L, Karon JM,
Maurice C, Djomand G, Ackah A, Domoua K, Kadio A, Yapi A, Combe P, Tossou O, Roels TH, Lackritz EM, Coulibaly D, De Cock KM, Coulibaly IM, Greenberg AE. Efficacy of trimethoprim- sulphamethoxazole prophylaxis to decrease morbidity and mortality in HIV-1-infected patients with tuberculosis in Abidjan, Côte d’Ivoire: a randomised controlled trial. Lancet
1999;353(9163):1469-75.
Both arms received unsupervised TB treatment from TB clinics (no comparison to HIV clinic—
therefore does not address PICO 4b).
Neither arm received ART (therefore does not address PICO 4a). (I, C)
World Health Organization. Report of a “lessons learnt” workshop on the six ProTEST pilot projects in Malawi, South Africa and Zambia. 2004.
No ART provided (so does not address PICO 4a). No
information on TB diagnosis in HIV clinic versus TB clinic (so does not address 4b). (I, C) Zachariah R, Harries AD, Manzi M, Gomani P, Teck R, Phillips M,
Firmenich P. Acceptance of anti-retroviral therapy among patients infected with HIV and tuberculosis in rural Malawi is low and associated with cost of transport. PLoS One 2006;1:e121.
No comparison of TB clinic providing ART versus referral to HIV clinic. (I, C)
Zwarenstein M, Fairall LR, Lombard C, Mayers P, Bheekie A, English RG, Lewin S, Bachmann MO, Bateman E. Outreach
education for integration of HIV/AIDS care, antiretroviral treatment, and tuberculosis care in primary care clinics in South Africa:
PALSA PLUS pragmatic cluster randomised trial. BMJ 2011;
342:d2022
No comparison of TB diagnosis and/or treatment at HIV clinics versus referral to TB clinics. (I, C)
