Skip to content
PaperFren

Infectious disease

Can easy TB diagnosis keep infectious TB low where HIV is common?

Corbett EL, Bandason T, Cheung YB, et al. · PLoS medicine · 2007

Open access · cc by · source: Europe PMC

HIV made new TB cases about twenty times more common, but because HIV-positive people got sick and were diagnosed quickly, most undiagnosed infectious TB found by screening was in HIV-negative workers.

Study at a glance

Design
Cohort — Two-year prospective workforce cohort (nested in a cluster-randomised HIV testing trial) measuring incident TB, followed by a cross-sectional TB culture screening survey of all remaining employees.
N
N=6440 · 6,440 employees of 22 Harare workplaces in the incidence cohort; 4,668 of them (874 HIV positive) were screened in the end-of-study prevalence survey.
Population
Adult employees (mostly male, middle-aged manual and factory workers) of small and medium enterprises in Harare, Zimbabwe, with 19% HIV prevalence and no routine antiretroviral therapy.
Outcome
TB incidence per 1,000 person-years, point prevalence of undiagnosed TB, risk factors for each, and estimated duration of infectious disease before diagnosis by HIV status.

Structured fields used in claim comparison tables when every cited study has a complete layer.

Key findings

New culture-positive TB occurred at 25.3 per 1,000 person-years in HIV-positive workers versus 1.3 in HIV-negative workers, an adjusted rate ratio of about 19. Yet at the end-of-study screen, smear-positive TB prevalence was only 1.3 per 1,000, lower than most African surveys, and many cases found were subclinical. HIV-positive TB was diagnosed much sooner (about 12 weeks of culture positivity versus 108 weeks for HIV-negative TB), so HIV accounted for 78% of incident but only 14% of prevalent culture-positive TB.

Methodology

Researchers enrolled the workforces of 22 companies in Harare and gave their clinics a trained nurse who investigated anyone with a persistent cough using sputum smears and cultures. They counted new TB cases over two years by HIV status, then screened every remaining employee with symptom questions and sputum culture to find TB that had not yet been diagnosed. Comparing prevalence with incidence let them estimate how long people had infectious TB before diagnosis.

Limitations

It is one workplace population that was mainly employed, male and middle-aged, so a healthy-worker effect may make prevalence look lower than in the general population. Only 27 prevalent cases were found, so risk-factor estimates and duration estimates are imprecise. Some incident cases may have been missed, and the effects of isoniazid preventive therapy and HIV care offered through the parent trial were not analysed. It also relied on culture and sensitive microscopy, which many programmes lack, so results may not hold where diagnosis is harder to reach.

How this study connects

Role on claims

Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.

  • SupportsHIV and Co-infectionconcept

    HIV multiplies TB risk roughly twentyfold.

    HIV sharply raises tuberculosis incidence: among 6,440 Harare workers, culture-positive TB occurred at 25.3 per 1,000 person-years in HIV-positive versus 1.3 in HIV-negative workers, an adjusted rate ratio of about 19.

    Evidence for the claim as stated.

  • SupportsHIV and Co-infectionconcept

    Prevalence surveys under-count HIV's contribution because HIV-TB progresses and is found quickly.

    HIV-positive TB was diagnosed much faster (about 12 versus 108 weeks of culture positivity), so HIV accounted for 78% of incident but only 14% of prevalent culture-positive TB in the same cohort.

    Evidence for the claim as stated.

  • SupportsHIV and Co-infectionconcept

    The TB cohort and the adolescent cohort measure HIV's toll in different ways: incidence and diagnostic delay in adults without routine ART versus survival and retention in children on ART. They are not comparable estimates of the same effect, and both are observational.

    Evidence for the claim as stated.

Open questions

Tensions this paper is part of

From concept pages' “where studies disagree.” Disagreement means the same question; scope means different assays, populations, or outcomes.

  • Scope difference — different assays, populations, or outcomes

    The TB cohort and the adolescent cohort measure HIV's toll in different ways: incidence and diagnostic delay in adults without routine ART versus survival and retention in children on ART. They are not comparable estimates of the same effect, and both are observational.

Related papers in this topic

Same topic cluster — not a recommendation engine.