Losses in the Sputum Specimen Referral Cascade in Mpulungu District, Zambia: A Cross-Sectional Study.

Sputum specimen referral cascades in resource-limited settings are characterized by losses of specimens, resulting in delays in tuberculosis (TB) diagnosis. Mpulungu District Health Office in Zambia conducted a quantitative based cross-sectional study using both primary and secondary data to identify points at which loss of specimens occurred in the sputum referral cascade. Primary data were collected through observations and interviews with 22 TB service providers. Secondary data were collected through examination of patient files and presumptive TB and laboratory registers to retrospectively track sputum specimens referred by ten health centers from April to September 2018. Proportions of specimens/laboratory results at every stage of the referral cascade were calculated using Epi Info v7. Only 49 (23%) out of 209 sputum specimens completed the referral cascade. The remaining 160 (76%) were lost at various stages of the referral cascade. The largest loss (51%) occurred between the release of laboratory results by the diagnostic facility and their receipt at referring facilities. Barriers included an inadequate number of staff oriented in sputum specimen referral, negative staff attitudes, and lack of specimen packaging material and specimen transportation. The district health office should strengthen the sputum specimen referral system by providing transport and specimen packaging material and by training staff in sputum collection transportation and tracking.

Feasibility of district wide screening of health care workers for tuberculosis in Zambia.

BACKGROUND: Many health care workers (HCWs) are at increased risk for tuberculosis (TB). The World Health Organization (WHO) recommends screening HCWs for TB in high burden settings but this is often not implemented in countries with a high TB incidence. We assessed the feasibility of TB screening among HCWs, including participation rate and yield, as part of a project introducing facility specific TB interventions. METHODS: This study had a cross-sectional design. HCWs (including paid staff and community volunteers) from 13 clinics and two hospitals in the Ndola district of Zambia participated. HCWs were screened by a designated person in their own facility. The agreed screening algorithm for HCWs included annual symptom screening, with sputum smear, culture (or Xpert) and chest x-ray offered to HCWs with at least one TB symptom, i.e. those with presumptive TB. RESULTS: A total of 1011 out of 1619 (62%) staff and 71 out of 138 (51%) community volunteers were screened within one year, total 1082/1757 (62%). Five percent (52/1082) of those screened were presumptive TB patients. Seventy-three percent (38/52) of presumptive TB patients received all diagnostic tests according to the agreed algorithm. Eighteen out of 1757 staff and volunteers combined were diagnosed with TB within a calendar year, showing a notified TB incidence of 1%. At least five of them were diagnosed during the screening appointment (0.5% of those screened). One of the 18 HCWs died of TB. Seventy-six percent (822/1082) of screened HCWs indicated that they already knew their HIV status. Screening was considered feasible if confidentiality can be guaranteed although challenges such as the time required for screening and sample transport were reported. CONCLUSIONS: It is feasible to conduct and implement screening programs for TB among HCWs in hospitals and clinics, and the notified incidence and yield is high. Advocacy is needed to educate managers and HCWs on the importance of screening and the implementation of locally relevant screening algorithms. It is essential to ensure access to TB infection control, diagnostics, treatment and confidential registration for HCW.