Active Case Finding (ACF), used to identify people with TB disease, not presenting for their own symptoms,
is recommended, but rarely performed outside health faculties in high disease burden settings. Contacts of
a TB case are a key population at extreme risk of both prevalent and incident TB and likely should be included
in ACF strategies. However there is little evidence to support this. Moreover, there is no data describing
efficacy of contact tracing where the index case is a young child with TB, both because of great challenges
in diagnosing pediatric TB and because young children are considered epidemiologic “dead-ends”. Our pilot
study of contact tracing in hospitalized, children with TB, found high rates of previously undiagnosed HIV and
TB. Using sentinel children with TB, we plan to assess the impact of a home-based ACF, HIV and TB testing,
home-initiation of therapy, with linkage to treatment on improving HIV and TB free survival in the household.
Hospitalized children (=5 years) diagnosed with TB will be allocated to either our intervention – a
combination of several already proven strategies - or the standard of care plus arm. Our intervention will, at
baseline screen caregivers and other household contacts for HIV, and TB disease and infection, and initiate
appropriate treatment in the household, and thereafter link individuals to care and treatment at local health
facilities. The current standard of care is to test the child with TB for HIV and then passively refer for ART
initiation if HIV-infected. Little or nothing is currently done for close contacts of the index child with TB.
However, for all TB-positive children randomized to the non-intervention arm, we will test the
mother/caregiver at the hospital for both HIV and TB, then conduct a baseline household assessment with
referral for care as appropriate. Outcomes will be measured in all households, at final 27 month follow-up
household visit to test all contacts for HIV and TB. We hypothesize that our home-based intervention will
significantly improve TB- and HIV-free survival in the entire household because of early and improved rates
of initiation, and adherence to HIV and TB treatment, and TB preventive treatment3-5. We will also measure
and compare viral loads in all HIV-infected household members at the end of follow up.
TB commonly co-exists with severe lower respiratory tract infections (LRTI),6 but about 47% of
hospitalised children with culture-confirmed TB are not diagnosed with TB whilst in hospital. Moreover, there
is an absence of data describing the TB and HIV burden in the household contacts of children with LRTI.
Thus it is critically important to evaluate whether targeting children with TB cases - a minority of admissions
- or, including those with LRTI – the majority of pediatric admissions - is worthwhile. We will therefore also
assess prevalence and incidence of TB and HIV in household contacts of children admitted to
hospital with LRTI.
Incremental cost effectiveness will be calculated to compare the yield of undiagnosed or untreated
HIV and TB infection at baseline; and overall survival, without incident HIV infection and/or TB disease.