Abstract:
Background: Globally, 40% of all tuberculosis (TB) cases, 65% paediatric cases and 75%
multi-drug resistant TB (MDR-TB) cases are missed due to underreporting and/or under
diagnosis. A recent Kenyan TB prevalence survey found that a significant number of TB
cases are being missed here. Understanding spatial distribution and patterns of use of TB
diagnostic tests as per the guidelines could potentially help improve TB case detection by
identifying diagnostic gaps.
Methods: We used 2015 Kenya National TB programme data to map TB case
notification rates (CNR) in different counties, linked with their capacity to perform
diagnostic tests (chest x-rays, smear microscopy, Xpert MTB/RIF®, culture and line
probe assay). We then ran hierarchical regression models for adults and children to
specifically establish determinants of use of Xpert® (as per Kenyan guidelines) with
county and facility as random effects.
Results: In 2015, 82,313 TB cases were notified and 7.8% were children. The median
CNR/100,000 amongst 0-14yr olds was 37.2 (IQR 20.6, 41.0) and 267.4 (IQR 202.6,
338.1) for ≥15yr olds respectively. 4.8% of child TB cases and 12.2% of adult TB cases
had an Xpert® test done, with gaps in guideline adherence. There were 2,072 microscopy
sites (mean microscopy density 4.46/100,000); 129 Xpert® sites (mean 0.31/100,000);
two TB culture laboratories and 304 chest X-ray facilities (mean 0.74/100,000) with
variability in spatial distribution across the 47 counties. Retreatment cases (i.e. failures,
relapses/recurrences, defaulters) had the highest odds of getting an Xpert® test compared
to new/transfer-in patients (AOR 7.81, 95% CI 7.33-8.33). Children had reduced odds of
getting an Xpert® (AOR 0.41, CI 0.36-0.47). HIV-positive individuals had nearly twice
the odds of getting an Xpert® test (AOR 1.82, CI 1.73-1.92). Private sector and higherlevel hospitals had a tendency towarrds lower odds of use of Xpert®.
Conclusions: We noted under-use and gaps in guideline adherence for Xpert® especially
in children. The under-use despite considerable investment undermines costeffectiveness of Xpert®. Further research is needed to develop strategies enhancing use
of diagnostics, including innovations to improve access (e.g. specimen referral) and
overcoming local barriers to adoption of guidelines and technologies.