Introduction Mortality from advanced HIV disease (AHD) remains high, and
current strategies to promptly test people eligible for AHD screening are
insufficient. Task sharing for point of care (POC) testing utilizing lay
health workers (LHW) is recommended, however it is marginally practised in
many countries. This study sought to describe the feasibility and testing
outcomes of task-shared implementation of the AHD POC diagnostic tests
utilizing LHW and professional health care workers (HCW). Methods This was
a cross-sectional mixed-methods implementation study in seven primary and
three secondary health facilities in Mozambique and the Democratic
Republic of Congo (DRC). From March to November 2022, consenting HIV
positive adults eligible for AHD screening, were offered Visitect CD4
lateral flow assay (LFA), and or subsequently urinary Mycobacterium
tuberculosis lipoarabinomannan antigen (TB LAM) and cryptococcal antigen
(CrAg) tests. The primary study outcome was the proportion of testers (LHW
and HCW) who found it ‘easy’ to integrate the 3 POC tests within their
routine work together with their opinions on the value of task shared AHD
POC testing. Results A total of 1542 patients were screened for AHD by 35
LHW (34 counsellors, and 1 lay educator) and 45 professional HCW (28
nurses and 9 clinical officers and 8 doctors). In the study period, LHW
conducted a median number of 27 [IQR: 16 - 34] Visitect CD4 LFA tests,
whereas nurses, clinical officers together with doctors conducted 19 and
11 respectively. Visitect CD4 LFA increased CD4 testing by 10.7% in
Munhava (Beira) and 22.9% in CHK (DRC), complementing existing CD4 testing
instruments. Among testers who completed the feasibility survey, nearly
sixty percent of testers (25/42; LHW in particular) found it easy to
integrate AHD POC testing within their routine workflow. The prevalence of
AHD was 39.2% (604/1542). A total of 34% (146/430) and 5.4% (22/407) of
patients tested positive for urine TB LAM and plasma CrAg respectively. Of
these, 82.2% (120/146) and 36.4% (8/22) had a documented therapeutic
intervention. The median time for completing the Visitect CD4 LFA and
conveying the results to the clinician was 59 minutes [IQR: 48 – 71].
Conclusion Task-shared integrated testing for AHD at POC among LHW and
professional HCW, is feasible and can improve access to AHD testing.
However, as POC testing responsibilities become shared, documentation of
testing activities could increase in complexity and can be easily
fragmented, especially when there is limited supervision. Nevertheless,
LHW are well suited for POC testing due to limited availability and higher
clinical workload of other HCW. # Feasibility and testing outcomes of task-shared implementation of
advanced HIV disease point-of-care tests in Mozambique and DRC Dataset
DOI: [10.5061/dryad.j0zpc86w4](
doi.org) ##
Description of the data and file structure ### Visitect_MSF_MozDRC.csv The
dataset contains variables from multi-country testing for advanced HIV
disease. Variables: * studyid=> this is the unique ID of the patient
in the study * country=> this is the country where the study took
place --values: DRC Democratic Republic of Congo, Mozambique *
Sex=>this is the sex of the patient, --values: M=males; F=female *
Reason_CD4=> is the reason why the CD4 testing was conducted for the
patient --values: ART initiation, missing CD4 count result, hospital
admission, return after loss to follow up, high HIV viral load, Other. *
IPD_OPD=>Inpatient department and outpatient department *
Visitect_Result=> this is a binary result of greater or less than 200
as per the CD4 lateral flow assay test ---values: <200= less than 200;
>200= greater than 200 * TBLAM_Result=> this is a binary
(positive/negative) result from the urinary TB LAM lateral flow assay
test ---values: Pos=positive; Neg=negative * TB_Prophylaxis=> this is
TB prophylaxis (yes/no) * TB_txmnt=> TB treatment (yes/no) * CrAg
result=> this is a binary (positive/negative) result from cryptococcal
antigen lateral flow test, ---values: Pos=positive; Neg=negative * CM
txmnt=> Cryptococcal menengitis treatment ## Human subjects data All
data has been anonymised and there is no personally indentifiable
information. Patients consented to the publication of de-identified data.