Tobacco Cessation for People Living with HIV
Barriers and facilitators to delivering tobacco cessation interventions at HIV care clinics in Sub-Saharan Africa
LINK to Published Study Results
This study examined the factors that may support or hinder the delivery of tobacco cessation services within HIV care settings in Uganda and Zambia. Using the Consolidated Framework for Implementation Research (CFIR), the research team assessed implementation experiences across 16 HIV clinics participating in Quit4Life+. The study drew on site reports, health system assessments, interviews with healthcare providers, and focus groups with people living with HIV to identify barriers and facilitators at the patient, provider, clinic, health system, and broader contextual levels. The findings help identify what is needed to successfully integrate and sustain tobacco cessation services within HIV care and contribute to building implementation science capacity for cancer- and HIV-related research in sub-Saharan Africa.
Key Methods
Study Locations: The study was carried out at the 16 Quit4Life+ sites in Uganda and Zambia
Multiple data sources were reviewed throughout the Quit4Life+ implementation process to examine experiences relevant to the five principal CFIR domains.
Site visit reports (n=22) and standardized health systems assessments (n=4) were completed by study coordinators to document ongoing challenges in each study country. Additionally, KIIs (n=14) with clinic health workers and FGDs (n=55) with PLWH were conducted to explore more in-depth perspectives on barriers and facilitators to delivering and engaging with tobacco cessation support services within the selected HIV clinics17.
Data analysis
Identification of implementation factors
Qualitative data analysis followed an iterative, multi-stage process guided by the updated CFIR framework. The study principal investigator (HW) and country-level coordinators (KG, EM, MM) first conducted an initial review of all documentary data sources (Table 1). Preliminary themes were identified inductively from transcripts, focusing on emerging implementation experiences related to intervention delivery, operational challenges, barriers, and facilitators. Data were coded after interpretation and analysis using DeDoose Software V.9.05418. Two study investigators (KG, EM) independently coded a subset of transcripts, compared coding decisions, and refined initial themes. A third investigator (HW) reviewed discrepancies until consensus was reached.
Selected implementation themes (experiences) were reviewed during weekly virtual multisite meetings involving the full research teams from Uganda, Zambia, and the US. These meetings served to collaboratively validate experiences, discuss interpretation, and address any challenges. Detailed meeting notes were maintained and shared with study partners and relevant stakeholders, if necessary, for further confirmation and refinement.
Mapping of CFIR domains and constructs
Using a deductive qualitative approach, the identified implementation themes were subsequently mapped to the publicly available CFIR domains and associated constructs13. Factors mapped within the domains and constructs were further categorized as a barrier or facilitator to Quit4Life+ implementation. Qualitative analysis continued until thematic saturation was reached. Criteria for thematic saturation were defined under the following criteria: 1) no new CFIR constructs were identified after coding consecutive data sources, 2) no new barriers or facilitators emerged within previously identified domains/constructs, and 3) the same constructs were identified across multiple sources. The final analytic structure was discussed with the study team, adjusting to ensure accuracy and consistency across countries.
Through this process, the CFIR domains were operationalized for this study as the following: the innovation domain encompassed features of the text-message program, NRT delivery, and costs; the inner setting described organizational context and capacity within HIV clinics; the outer setting captured patient needs, community conditions, and external policy environments; the individuals domain reflected patient and provider attitudes; and the implementation process examined intervention planning, coordination, and execution. Results for selected domains and related constructs are presented visually, and with illustrative examples and supporting quotations.