Health Surveillance & Disease Prevention & Control — Communicable Diseases: 
Acute Respiratory Infections - Anthrax - Antimicrobial Resistance - Campylobacter - Chagas -
Cholera - Dengue - Diarrheal/Enteric Diseases - Disease Outbreak News -
Emerging/Reemerging Diseases - Filariasis - Hantavirus - Influenza: Avian |  Pandemic |  Seasonal - International Health Regulations - Leishmaniasis - Leprosy - Malaria -
Neglected Diseases - Parasitic Diseases - Rabies - Research - Salmonella - Shigella - Smallpox -
Tuberculosis - West Nile Virus  - Yellow Fever  -
Health Analysis & Statistics - Chronic Diseases - Veterinary Public Health

Guidelines for HIV Surveillance among Tuberculosis Patients

(2^nd ed., World Health Organization TB/HIV Working Group of the Global Partnership to Stop TB and UNAIDS/WHO Working Group on Global HIV/AIDS/STI Surveillance, 2004)

WHO E-Book (38 pp, PDF)
1. Introduction
1.1. Background
1.2. Rationale for surveillance
1.3. Challenges to surveillance
2. Methods for surveillance of HIV among tuberculosis patients
2.1. An overview of different surveillance methods
2.2. Surveillance methods in different HIV prevalence settings
3. Methodological issues
3.1. Initial situation assessment
3.2. Case definitions
3.3. Population under surveillance
3.4. Sampling
3.5. Specimen selection
3.6. Data management
3.7. Programme responsibility
3.8. Resource considerations
3.9. Evaluation
4. Implementation
Annexes
1. Minimum data requirements from tuberculosis clinic settings where patients are routinely tested for HIV
2. Sample data collection from for use in HIV prevalence surveys or sentinel surveillance among tuberculosis patients
3. Options for the capture of data obtained from routine care on HIV prevalence among tuberculosis patients
4. 1994 WHO Guidelines for HIV surveillance among tuberculosis patients
5. Sample size determination
References

WHO TB Page
WHO HIV Infections Page

Stop TB

PAHO TB/HIV Workshops: 2004 | 2003
PAHO TB Page
PAHO AIDS and STI Page

Summary   |   Recommendations

These guidelines are addressed to the managers of national tuberculosis programs (NTPs) and national AIDS programs (NAPs), those people responsible for HIV surveillance, and public-health decision-makers at national and subnational level. They form part of the TB/HIV series of documents produced by the Stop TB Department in the World Health Organization and also of the "Second-Generation Surveillance" (SGS) series.

The main objective of these guidelines is to provide a framework for the methods to be used for measuring HIV prevalence among tuberculosis patients and to encourage implementation of HIV surveillance.

Surveillance of HIV among TB patients is being increasingly recognized as important, as the HIV epidemic continues to fuel the global TB epidemic. In many countries, HIV prevalence in TB patients is a sensitive indicator of the spread of HIV into the general population. Information on HIV levels in TB patients is essential to respond to the increasing commitment to provide comprehensive HIV/AIDS care and support, including antiretroviral therapy (ART), to HIV-positive TB patients.

The first edition of these guidelines, published in 1994, detailed one specific approach to determining HIV prevalence—through cluster sampling and unlinked anonymous seroprevalence surveys. The increasing availability of routine HIV-testing and counseling as an entry point to HIV/AIDS care for TB patients has highlighted the need for updated and broader guidelines.

WHO's 3 by 5 initiative, to reach 3 million HIV-infected people with antiretroviral therapy by the end of the year 2005, will further increase demand for HIV-testing among TB patients and for knowing the size of the burden of HIV associated TB. HIV-testing is the entry point for ART delivery, and this applies equally to patients with TB. Reliable HIV surveillance systems for TB patients and large-scale access to HIV-testing and -counseling services are cornerstones for effective TB/HIV collaboration.

This document outlines the three main methods for HIV surveillance among TB patients: data from the routine testing of TB patients for HIV; sentinel surveillance; and periodic (special) surveys. Selecting the appropriate surveillance strategy will depend on the existing surveillance system, the underlying HIV epidemic status in a country, and the status of ART implementation, as well as the overall TB situation. This document provides an overview of the principal issues to be considered by countries in strengthening their existing surveillance systems or developing new systems and increasing their utility.

Recommendations

At all levels of an HIV epidemic (low-level, concentrated, generalized), routine HIV-testing data—when available—should be used for surveillance purposes. These data can be calibrated by periodic (special) or sentinel surveys. In countries where HIV prevalence among TB patients is unknown, a seroprevalence survey should be undertaken as part of the initial assessment of the situation.

WHO recommends the following HIV surveillance methods, which vary according to the level of the HIV epidemic:

1. All countries with a generalized HIV epidemic (HIV prevalence consistently >1% in pregnant women) should aim to ensure that HIV counseling and testing are actively promoted and offered to all TB patients. Whenever possible, this should be done in conjunction with the provision of ART. The data obtained in this way can form the basis of a reliable surveillance system where high coverage (>80%) of testing among TB patients is achieved. One of the best systems for capturing this information is through a computerized TB notification system, which also captures information on HIV status. Periodic (special) surveys or sentinel surveys are also recommended, to calibrate the results of routine testing.
2. In countries with a concentrated epidemic (HIV prevalence consistently over 5% in at least one defined subpopulation, e.g. intravenous drug-users (IDUs); sex-workers (SWs), men who have sex with men (MSM), and less than 1% in pregnant women in urban areas), data from routine HIV counseling and testing of all TB patients should still form the basis for the surveillance. If this system is not yet in place, periodic (special) surveys or sentinel surveys are suitable alternatives.
3. In countries with a low-level HIV epidemic (HIV prevalence has not consistently exceeded 5% in any defined subpopulation) and where HIV-testing is not routinely offered to TB patients, periodic (special) surveys (at intervals of 2–3 years) or sentinel surveys should be conducted among TB patients.

Periodic sentinel serosurveillance for HIV in general is usually conducted among pregnant women (as a proxy for the general population) or among population groups with high-risk behavior, depending on the level of the epidemic. This surveillance is useful for monitoring the trends in HIV prevalence and can identify, at an early stage, areas where routine HIV counseling and testing of individuals with tuberculosis should be undertaken.

HIV tests other than on serum or blood—principally on gingival secretions—are available and being further developed (for example for sputum testing). WHO recommends that further work be done to improve the sensitivity, specificity and therefore the positive predictive value of HIV test on sputum. Until such further work is reported HIV-testing using sputum is only advisable if the HIV prevalence among TB patients is anticipated to be at least 10%.

It is intended to pilot these guidelines in several sites around the world in 2004. This will permit evaluation of their feasibility and answer questions regarding sputum-based HIV-testing under different conditions.