thesis_access.pdf (5.25 MB)
Download file

Are Women Free To Opt Out? Implementation Fidelity Of The ‘Opt-Out’ Hiv Testing For Pregnant Women In Ghana

Download (5.25 MB)
posted on 27.06.2021, 21:06 by Isaac Amankwaa

In 2008, Ghana adopted WHO/UNAID’s provider-initiated opt-out HIV testing policy and integrated it into all maternal services. The intervention’s central principle was that women are free to choose whether or not to test for HIV (Consent), assured of Confidentiality, Correct test results, Connection to care, and Counselling services( referred to as 5Cs). However, the weak healthcare infrastructure, low hospital staffing levels, hierarchical and paternalistic nursing and midwifery culture in sub-Saharan Africa were considered potential threats to achieving rights-based testing. Despite these concerns, much mainstream HIV testing research had focused on outcome-related to report high HIV test uptake among women attending the antenatal clinic. However, the reported high testing uptake had not produced the desired impact, as many women testing positive for HIV did not enter care. To date, no process evaluation exists to explain these outcomes. The current study recognises the need for a careful examination of the delivery process. Therefore, it has aimed to evaluate the antenatal clinic-based opt-out HIV testing programme’s implementation fidelity to explain the observed outcomes.

Employing a mixed-methods design and guided by Carroll’s seminal conceptual framework of implementation fidelity, the study collected quantitative and qualitative data from 12 antenatal clinics in Ghana. Adherence was measured quantitatively through brief facility surveys, healthcare provider and pregnant women self-reports and structured observation of counselling sessions at the antenatal clinic. Interviews with key informants, healthcare providers and women, and the keeping of field notes provided qualitative data. Descriptive statistical analysis of the quantitative data was used to describe the sample and antenatal clinic characteristics. To calculate fidelity scores, percentage means and standard deviation(SD) of components delivered were used. Qualitative data were analysed using framework analysis, aided by NVIVO data analysis software.

Routine testing of women for HIV was widely available in all the 12 antenatal clinics, and testing among pregnant women was high (98.1%). Many healthcare providers were, however, unaware of the opt-out approach for offering HIV test. Instead of group pre-test discussions, many clinics delivered information about HIV through individual pre-test counselling. Adherence to the core principles of consent, confidentiality, counselling, and connection to care was low (38%) for direct observation, moderate (54%) for pregnant woman self-reports and moderately high (78.9%) for healthcare provider self-reports. Implementation of the opt-out intervention at the health facilities was fraught with challenges due to the complex nature of the opt-out intervention, lack of facilitation of intervention delivery, beliefs about autonomy that were not in line with the intervention’s underlying principles, and antenatal contextual constraints. The outcome of this thesis is a proposed human rights framework supporting rights-based testing in the antenatal clinic. The framework provides a structured, comprehensive, and context-specific approach to support future rights-based interventions and research.

The study concludes that implementation fidelity was low to moderate for all the 5Cs of the opt-out intervention. Thus, in the context of this study, no claims can be made about the opt-out testing’s ability to increase HIV testing uptake as widely reported. The absence of impact in terms of linkage to care and other behavioural outcomes is best explained by the low implementation fidelity, poor facilitation, complex and unfamiliar intervention, and a misfit between demands of the intervention and realities of the antenatal clinic setting. The findings highlight the need for culturally appropriate HIV testing guidelines that incorporate shared or relational decision-making approaches acceptable to women. The findings also generate new insights into the need to make programmes more straightforward, engage healthcare providers, and offer supportive supervision to equip them with the skills and knowledge needed to implement such complex intervention.


Advisor 1

Skinner, Joan

Advisor 2

Maude, Robyn

Copyright Date


Date of Award



Victoria University of Wellington - Te Herenga Waka

Rights License

Author Retains Copyright

Degree Discipline

Public Policy

Degree Grantor

Victoria University of Wellington - Te Herenga Waka

Degree Level


Degree Name

Doctor of Philosophy

ANZSRC Type Of Activity code


Victoria University of Wellington Item Type

Awarded Doctoral Thesis



Alternative Language


Victoria University of Wellington School

School of Nursing, Midwifery, and Health Practice