University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Counseling and testing of TB patients for HIV: evaluation of three implementation models in Kinshasa, DR Congo
Global Health Sciences Literature Digest
Published November 03, 2008
Journal Article

Van Rie A, Sabue M, Jarrett N, et al. Counseling and testing of TB patients for HIV: evaluation of three implementation models in Kinshasa, DR Congo. Int J Tuberc Lung Dis 2008;12(3):S73-8.


To evaluate the uptake of provider-initiated HIV counseling and testing (CT) for tuberculosis (TB) patients in three different models of implementation

Study Design

Program evaluation of three referral models of provider-initiated HIV counseling and testing for TB patients: referral to a off-site freestanding voluntary counseling and testing (VCT) center (Model 1), on-site referral to a CT center at the primary health care center housing the TB clinic (Model 2), and routine HIV CT by the TB clinic nurse (Model 3).


Three TB clinics within primary health care centers in Kinshasa, the Democratic Republic of Congo (DRC), with an annual caseload of 320-700 cases (medium-sized clinics), evidence of pre-existing HIV CT activities, and a clinic director who expressed interest in the project. The DRC has the eleventh highest estimated new TB cases yearly and an incidence rate of 356 per 100,000 population. Adult (age 15-49 years) HIV prevalence was estimated at 4.3% and HIV prevalence among TB patients at 20%.


Between August 2004 and June 2005, all individuals aged ≥18 months who were diagnosed with TB disease and registered at one of the project clinics were eligible for inclusion in the study. The mean age was 31 years. Approximately half (48.2%) of the patients were female, 22.7% were diagnosed with extra-pulmonary TB, and 11.8% reported a previous history of TB treatment.


Assignment of one of the three models to each of the project TB clinics was based on pre-existing HIV CT infrastructure. For the purpose of this study, HIV counseling was provided by 10 health care workers, including 6 research nurses (all three models), 2 VCT staff (Model 1), one counselor/nurse (Model 2), and one TB nurse (Model 3). The six research nurses rotated monthly between clinics. All health care workers providing HIV CT had received the same 2-week training before the start of the study. The TB nurses also were trained in HIV CT referral and primary care for TB-HIV co-infection (excluding antiretroviral treatment [ART]). A laboratory technician at each clinic was trained in rapid HIV tests, which were performed at the laboratory of the primary health care center or VCT site. HIV CT was offered to all patients at time of TB diagnosis, and for those who refused, after 1 and 2 months of TB treatment or at any time upon patient request. Patients infected with HIV were offered cotrimoxazole at a dose of 480 mg twice daily for adults and 6-8 mg/kg for children, provided there were no contraindications. Cotrimoxazole was distributed weekly at the TB clinics during the intensive phase of TB treatment and monthly thereafter. Given its scarcity in the DRC, ART was not provided by the project. Data on HIV pre-test counseling date, date and result of HIV test, post-test counseling date, and data on cotrimoxazole prophylaxis uptake and adherence were collected from a newly developed TB-HIV register (Model 3, HIV CT by TB nurse) or from project HIV CT referral forms or cotrimoxazole register (Models 1 and 2). Data from HIV CT referral forms collected weekly were validated against CT log books and laboratory registers. Test results were communicated to the TB clinic staff, unless the patient objected.

Primary Outcomes

Acceptance of HIV CT, HIV infection, and factors associated with HIV infection


The proportion of patients accepting HIV pre-test counseling was significantly lower (68.5%) at the clinic with referral to a freestanding VCT center compared with 94.8% at the clinic with on-site referral and 97.7% at the clinic where CT was conducted by the TB nurse (P<0.001). Among the 136 patients who did not accept HIV CT at TB diagnosis, 11 accepted within the first month of treatment and 1 at a later date. Only 6 patients (0.5%) who had accepted pre-test counseling refused to be tested. Almost all (1,022, 96.7%) of those tested received their result during post-test counseling. Among the 35 patients who did not receive their HIV test result, 26 had been transferred to another TB clinic, one had died soon after HIV CT, and 6 had defaulted from TB treatment. The overall HIV prevalence among the 1,088 patients with a known HIV result was 18.8% (95% confidence interval [CI]: 16.5-21.2). HIV prevalence was higher at the routine CT site (21.9%) compared with the two other sites (15.8% vs. 15.4%; P=0.008). HIV prevalence was higher among women (adjusted prevalence odds ratio [aPOR], 1.91), among patients with recurrent TB (aPOR, 2.74), and among those with extra-pulmonary TB (aPOR, 1.97), and was significantly lower among adolescents (aPOR, 0.25). Cotrimoxazole prophylaxis was initiated in 89.8% of HIV patients coinfected with TB. Reasons for not initiating treatment at the TB clinic included patient transfer (n=8), contraindication (n=5), patient refusal (n=2), and provision of HIV care and ART at an HIV clinic (n=5). Of those who initiated cotrimoxazole prophylaxis at the TB clinic, 88.4% were still on it at the end of 8 months of TB treatment or at time of death, 14 had defaulted from both TB and cotrimoxazole treatment, 16 had been transferred, 3 had stopped due to side effects, and data were missing on 4 others.


The authors conclude that a TB nurse or health care worker at the primary health care center offering provider-initiated routine HIV CT can result in a very high level of uptake of HIV testing.

Quality Rating

There is no widely accepted quality rating system for program evaluations such as this; however, some limitations were noted: 1) this was not a randomized clinical trial but a comparison of three clinics where different approaches to HIV CT were implemented. Clinics were purposefully selected and therefore not fully comparable and not representative of all TB clinics in Kinshasa; and 2) results were obtained in an operational research setting. That similar high acceptance rates of HIV CT can be achieved in routine TB clinic settings must be demonstrated. In addition, the overall 18.8% HIV prevalence may be an underestimate as patients known to be HIV infected who were attending the clinic were excluded. The prior estimate for the DRC was 21% HIV prevalence.(3)

In Context

This study demonstrated that the uptake of HIV CT by TB patients in Kinshasa was high (95-98%) when CT was performed at the TB clinic or at the primary health care centers housing the TB clinics and was significantly lower (68.5%) when TB patients were referred to a freestanding VCT center. Similar uptake rates (>90%) among TB patients also have been reported in other African countries, such as Malawi(2) and Côte d'Ivoire.(3)


  1. World Health Organization. WHO report 2007. Global Tuberculosis Control: Surveillance, Planning and Financing. WHO/ HTM/TB/2007.376. Geneva, Switzerland: WHO, 2007. Abstract not available.
  2. Chimzizi R, Gausi F, Bwanali A, et al. Voluntary counseling, HIV testing and adjunctive cotrimoxazole are associated with improved TB treatment under routine conditions in Thylolo District, Malawi. Int J Tuberc Lung Dis 2004 May;8(5):579-85.
  3. Abouya L, Coulibaly IM, Wiktor SZ, et al. The Côte d'Ivoire national HIV counseling and testing program for tuberculosis patients: implementation and analysis of epidemiologic data. AIDS 1998;12:505-12.