University of California, San Francisco Logo

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

Home > Global Health Literature Digest > Provider-Initiated
Provider-initiated HIV testing and counselling for TB patients and suspects in Nairobi, Kenya
Global Health Sciences Literature Digest
Published December 16, 2008
Journal Article

Odhiambo J, Kizito W, Njoroge A, et al. Provider-initiated HIV testing and counselling for TB patients and suspects in Nairobi, Kenya. Int J Tubercul Lung Dis 2008 Mar;12(Supp 1):S63-68.


To evaluate an integrated tuberculosis (TB)/HIV pilot program targeting patients with TB and suspected TB for provider-initiated testing and counseling (PITC) in a resource-constrained urban setting

Study Design

Program evaluation, described as an "evaluation of programmatic public health work that was not set up for the purposes of research," reporting programmatic data on testing uptake, HIV/TB diagnoses, and HIV treatment enrollment


Tuberculosis clinics in a resource-constrained setting serving more than half a million people in the eastern part of Nairobi, Kenya


Between December 10, 2003 and March 31, 2004, PITC was offered routinely to consecutive newly diagnosed TB patients (n=160). From April 2004 to December 2005, PITC was offered to all patients with suspected TB during their first clinic encounter before confirmation or exclusion of TB disease (n=5457).


Conducted by clinical officers (assistant physicians), PITC was offered routinely to consecutive, newly diagnosed TB patients between December 10, 2003 and March 31, 2004. After April 2004, the protocol was modified and PITC was provided by trained nurses and offered to all patients who had suspected TB during their first clinic encounter before confirmation or exclusion of TB disease. All participants were given pre-test information, including an explanation of the link between TB and HIV, the benefits of knowing HIV status for long-term care and support, and procedures for safeguarding confidentiality. Afterwards, each patient was asked to confirm their willingness to proceed with HIV testing and to provide informed consent. Clients who declined the HIV test were allowed to "opt out" and were assured of uncompromised care. Patients who accepted received two rapid tests conducted in parallel. Test results were provided within 20-30 minutes, and patients received post-test counseling, which included an explanation of the implications of the test result and guidance on TB/HIV prevention and treatment. Emotional support was provided as necessary to help patients cope with their HIV test results. After April 2004, nurses ordered sputum tests for all patients with suspected TB, and patients with confirmed TB disease or with an uncertain TB diagnosis were referred with their HIV test results to clinical officers. Regardless of TB status, all HIV-infected patients suspected of having TB were considered for cotrimoxazole (CTX) prophylaxis. Patients with HIV infection but with no chest X-ray features suggestive for TB disease were referred to on-site antiretroviral treatment (ART) clinics. Charts and registers from existing patients were updated to document both TB and HIV data, focusing on processes and outcomes. Community outreach and education were conducted to inform the public about the new PITC protocol.

Primary Outcomes

HIV testing uptake, HIV prevalence, TB diagnosis, CTX and ART enrollment


In the first 4 months, 160 TB patients were diagnosed. Thirty-one were referred from onsite VCT clinics with their HIV test results; 22 had positive test results for HIV infection. The remaining 129 patients were offered and accepted PITC, and 63 patients tested positive for HIV infection. The overall HIV prevalence among TB patients diagnosed in this period was 53%. Sixty-nine (81%) of co-infected patients were referred from HIV care, with 59 (69%) starting CTX prophylaxis and 35 (41%) starting ART.

In the following 21 months, 5,457 patients suspected for TB disease were tested for HIV. Of these, 881 were referred from VCT clinics; 4,576 were offered PITC, of whom 3,962 accepted (87%) and 614 declined. The overall HIV prevalence among patients with suspected TB diagnosed in the second period was 62%. A total of 2,595 (48%) of patients with suspected TB were diagnosed with TB disease (281 referred from VCT clinics; 2,314 offered PITC). Of the 2,314 patients with confirmed TB who were offered PITC, 1,912 accepted (83%) and 402 declined. Of the 2,862 patients with suspected TB who did not have confirmed TB disease, 600 were referred from VCT clinics and 2,262 were offered PITC; 2050 (91%) of these patients accepted testing and 212 declined. The prevalence of HIV among patients diagnosed with TB disease in this period was 61%, which is not statistically different from the 63% among patients without confirmed TB disease. Of the 614 patients with suspected TB who declined to take the HIV test, 402 (65%) were diagnosed with TB disease. Of the 2,988 HIV-infected patients, 2,283 (76%) were referred for routine CTX prophylaxis and 1,951 (65%) enrolled. Of 1,727 (58%) patients assessed for ART, 1,618 (94%) were found eligible and 1,441 (71%) initiated ART.


The authors conclude that the introduction of PITC is feasible and acceptable to both TB patients and patients with suspected TB in primary health care clinics; that PITC should be offered to all patients with suspected TB, and not just to those with confirmed disease; and that a high proportion of patients with suspected TB and patients co-infected with HIV go on to access additional care.

Quality Rating

This study was an evaluation of a pilot program and was not designed for the purpose of research. Although programmatic data supporting the implementation of PITC in TB clinic settings is presented, the results do not have the rigor necessary to provide conclusive evidence for the benefit of PITC in this context. The programmatic results presented here, however, provide an argument for further research on TB/HIV service integration through PITC.

In Context

Tuberculosis is the main cause of morbidity and mortality among HIV-infected people worldwide.(1) The high prevalence of HIV among TB patients supports the urgent need for identification of HIV status in this population. The joint UNAIDS/WHO document Guidance on Provider-initiated HIV Testing and Counselling in Health Facilities states: "Provider-initiated HIV testing and counselling presents an opportunity to ensure that HIV is more systematically diagnosed in health care facilities in order to facilitate patient access to needed HIV prevention, treatment, care and support services".(2) Routine provision of onsite PITC for patients suspected of having TB and patients who have confirmed TB has the potential to increase uptake of HIV testing, reduce stigma associated with HIV testing, and allow for targeted counseling and treatment, given the special needs of TB/HIV co-infected individuals.

The routine offer of PITC to patients who have TB or suspected TB on an opt-out basis in settings with generalized HIV epidemics, such as Kenya, represents a paradigm shift from VCT.(2) In October 2004, the Kenyan Ministry of Health issued clinical guidelines that establish PITC for TB patients as the standard of care.(3) These guidelines have been widely disseminated in support of national scale-up of HIV testing in TB clinics in Kenya.

Programmatic Implications

PITC for people with suspected TB appears to be a viable method of reaching this population en masse. Further research, designed to test the outcomes of PITC integrated into existing TB programs, must be conducted to provide rigorous evidence in support of this intervention. Cost-effectiveness of PITC in TB clinics must be assessed to determine its feasibility in diverse settings. Additionally, the infrastructural, staffing, and clinic capacity of each setting should be considered before implementing PITC. The effectiveness of PITC may vary in different settings with the prevalence of HIV infection among patients with suspected TB. If PITC is found to be effective, feasible, and cost-effective, promotion of PITC in national policy guidelines should be advocated.


  1. Mukadi DY, Maher D, Harries A. Tuberculosis case fatality rates in high HIV prevalence populations in sub-Saharan Africa. AIDS 2001;15:143-52.
  2. Joint United Nations Programme on HIV/AIDS /World Health Organization. Guidance on provider initiated HIV testing and counselling in health facilities. Geneva, Switzerland: UNAIDS and WHO, 2007.
  3. Kenya Ministry of Health. Guidelines for HIV testing in clinical settings. Nairobi, Kenya: MOH, 2004.