Effects of implementation of a community health worker programme on HIV testing, treatment initiation, and patient retention in a rural South African District
Date
2019
Authors
Naidoo, Nireshni
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Introduction
This study evaluated the fidelity of implementation (FOI) of the South African community-based HIV
programme (CBHP), which is part of its primary healthcare re-engineering (PHC) strategy to improve
public healthcare delivery including human immunodeficiency virus (HIV) testing, treatment initiation
and patient retention. FOI is the degree to which an intervention was implemented as prescribed.
Assessment of FOI can improve health programmes. This study evaluated FOI of the South African
CBHP in a rural district to strengthen service delivery and to inform policy design.
Methods
This PhD research study was conducted in 12 wards of the Mopani district of Limpopo province, South
Africa from May 2016 through September 2017. The contextual framework for implementation
fidelity by Carroll and colleagues was used to assess FOI of the CBHP using quantitative and qualitative
methods. Three elements of this framework were evaluated in three separate studies: adherence,
potential modifiers to adherence, and outcomes.
1) Adherence was assessed in a community survey of 900 randomly selected households by
measurement of coverage, frequency, duration, and content of CHW visits in comparison to the
appropriate guidelines.
2) We used an exploratory study to assess potential modifiers: in-depth interviews were conducted
with community members, team leaders, community leaders, and social workers whereas focus
group discussions were conducted with CHWs and facility nurses. Thematic content analysis was
conducted and the Consolidated Framework for Implementation Research (CFIR) was used to
interpret the findings. The CFIR consists of five domains related to the intervention or programme,
including 1) the intervention characteristics, 2) the outer setting, 3) the inner setting, 4)
characteristics of individuals, and 5) the implementation process. The qualitative data collected
was also used to assess contextual factors that affect CBHP implementation.
3) Third, we assessed the outcomes element defined as the impact of HIV services delivered by the
CBHP on the overall HIV programme including case identification, adherence support to
individuals on ART and tracing of individuals lost to the ART programme. This was done through
collection of service provision data from the daily activity register of CHWs and verification of the
personal identifiers in these registers against routine data sources at the facilities.
Lastly, the Carroll framework was revised by assessing the individual elements of the framework in
relation to our findings from three studies and other available literature. We then propose an overall
adapted conceptual framework for implementation fidelity.
Results
In the community survey, with regards to adherence, good content (67%) and duration (100%) of CHW
visits was reported, but there was room for improvement of the CBHP coverage (47%) and frequency
(39%) of visits. Individuals reporting a CHW visit were more likely to know their HIV status than those
not visited (OR = 2.0; 95% CI 1.06–3.8; p = 0.032). Among those visited by the CHW discussion of HIV
was associated with knowing the HIV status (OR = 2.2; 95% CI 1.02–4.6; p = 0.044); in particular for
women (OR = 2.9; 95% CI 1.5–5.4; p = 0.001).
In the second study, we used the CFIR framework to assess for potential modifiers to implementation
of the CBHP. In terms of intervention characteristics, community members generally valued the HIV
services provided. However, the varying needs impacted on programme implementation. Outer
setting challenges include inability to meet the need of patients as a result of stigma, nondisclosure of
HIV status, and social factors. In terms of the inner setting, CHWs were grateful for the equipment and
training received but expressed the need for better support from management and the provision of
additional resources. With regard to characteristics of the implementers, the CHWs expressed the
desire for further training despite reporting having sufficient knowledge to conduct their HIV work.
Finally, in terms of the implementation process, the importance of relationship building between
CHWs and community members was emphasised.
Six critical contexts affecting the implementation of the CBHP were identified in our data analysis.
These contexts provide the background to the programme, and vary in the extent to which they can
be influenced by management and policy makers. Contexts include: 1) geographic, 2) social, cultural,
and economic, 3) local community, 4) personal and inter-personal relationships, 5) local governance
and authority, and 6) local organisational.
In the third study, with regard to outcomes, CHWs assessed 13% of the study population for HIV risk;
only 290 (2%) were referred for HIV testing services of whom 213 (73%) went to the facility. Adherence
support to individuals on antiretroviral therapy (ART) was provided during 5 657 visits; only one
individual was referred for complications. CHWs managed to trace 452/864 (52%) of individuals lost
to the HIV programme; of these 241 (53%) were (re)initiated on ART.
Bringing together all the results of these studies and combined with available literature, we adapted
the framework for assessment of implementation fidelity to be more feasible for evaluation of the
South African CBHP. The following modifications/adaptations were made: a) the core and modifiable
components should be identified a priori to allow for the precise measurement of adherence and b)
additional potential moderators such as organisational readiness to implement the intervention,
contextual factors, and stakeholder involvement were included. The element outcome was retained.
Discussion and conclusions
This PhD study demonstrates high value of using a systematic approach to evaluate the CBHP using a
conceptual framework. Implementation science is a relatively novel field. This unique study assessed
FOI of a fully operational rural CBHP in the context of HIV. Evaluating the fully operational CBHP
enabled the assessment of the impact of CHW activities on programme outcomes. Various positive
attributes, barriers and challenges were identified that can be used to strengthen the programme.
Some of these reflect in the new policy published in 2018, but several gaps still remain as discussed in
the general discussion section of this thesis. The adapted conceptual framework for implementation
fidelity provides an important, feasible approach to guide and strengthen FOI evaluations of CBHPs
across Africa. The actionable findings will inform policy makers to optimise the CBHP and contribute
to ending the HIV epidemic in South Africa.
Description
A thesis submitted to the Faculty of Health Sciences, University of the Witwatersrand,
Johannesburg, in fulfillment of the requirements for the degree
of
Doctor of Philosophy
Johannesburg, April 2019