The yield of a tuberculosis household contact investigation in two regions of Ethiopia.

SETTING: Amhara and Oromia regions, Ethiopia.

OBJECTIVE: To determine the yield of a household contact investigation for tuberculosis (TB) under routine programme conditions.

DESIGN: Between April 2013 and March 2014, TB clinic officers conducted symptom-based screening for household contacts (HHCs) of 6015 smear-positive TB (SS+ TB) index cases. Based on quarterly reported programme data, we calculated the yield in terms of number needed to screen (NNS) and number needed to test (NNT).

RESULTS: Of 15,527 HHCs screened, 6.1% had presumptive TB (8.5% in Oromia vs. 3.9% in Amhara). All forms of TB and SS+ TB were diagnosed in respectively 2.5% (Oromia 3.9% vs. Amhara 1.2%) and 0.76% (Oromia 0.98% vs. Amhara 0.55%) of contacts. The NNS to detect a TB case all forms and SS+ TB was respectively 40 and 132. The NNT to diagnose a TB case all forms and SS+ TB was respectively 2.4 and 8. Of 1687 eligible children aged <5 years, 323 were started on isoniazid preventive therapy.

CONCLUSIONS: The yield of the household contact investigation was over 10 times higher than the estimated prevalence in the general population; household contact investigations can serve as an entry point for childhood TB care.