Menu
Menu
Menu
Menu
Background: Clinical trials conducted in Africa often require substantial investments to support trial centres and public health facilities. Trial resources could potentially generate benefits for routine health service delivery but may have unintended consequences. Strengthening ethical practice requires understanding the potential effects of trial inputs on the perceptions and practices of routine health care providers. This study explores the influence of malaria vaccine trials on health service delivery in Ghana, Kenya and Burkina Faso. Methods: We conducted: audits of trial inputs in 10 trial facilities and among 144 health workers; individual interviews with frontline providers (n=99) and health managers (n=14); and group discussions with fieldworkers (n=9 discussions). Descriptive summaries were generated from audit data. Qualitative data were analysed using a framework approach. Results: Facilities involved in trials benefited from infrastructure and equipment upgrades, support with essential drugs, access to trial vehicles, and placement of additional qualified trial staff. Qualified trial staff in facilities were often seen as role models by their colleagues; assisting with supportive supervision and reducing facility workload. Some facility staff in place before the trial also received formal training and salary top-ups from the trials. However, differential access to support caused dissatisfaction, and some interviewees expressed concerns about what would happen at the end of the trial once financial and supervisory support was removed. Conclusion: Clinical trials function as short-term complex health service delivery interventions in the facilities in which they are based. They have the potential to both benefit facilities, staff and communities through providing the supportive environment required for improvements in routine care, but they can also generate dissatisfaction, relationship challenges and demoralisation among staff. Minimising trial related harm and maximising benefits requires careful planning and engagement of key actors at the outset of trials, throughout the trial and on its’ completion.
An initial partners’ meeting was held in Kintampo, Ghana in March 2012 to develop a common study methodology across all three sites to allow for inter-site descriptive comparisons. To describe the malaria vaccine trial inputs and explore their impact on the perceptions and practices of these health managers and health care providers we employed both quantitative (health facility and human resources audits) and qualitative (in-depth interviews and focus group discussion) methods. Data were collected between May 2012 and June 2013 by teams of trained research officers and fieldworkers in each country using the agreed common data collection techniques. Sampling procedures were purposive to ensure inclusion of people of different cadre, level of involvement in trial activities and demographic characteristics. The data sources, sampled facilities and characteristics of interviewees in the three countries are described in Table 3. MVT: Malaria vaccine trial a MVT facilities audited include a referral hospital (Ghana = 1), health centres (Ghana = 3; Kenya = 1), dispensaries (Kenya = 2), community clinics (Burkina Faso = 2) and a clinical trial facility (Burkina Faso = 1). In each country, all of the facilities (n = 10) involved in the MVT were included in the health facility audits. Each facility was visited by the research team and a structured checklist was used to collect data on the presence, functional status (on day of survey) and funding source of laboratory and clinical care equipment, as well as facility infrastructure (such as buildings and vehicles). The human resource audit (n = 144) covered all of the facilities involved in the MVTs and involved a purposefully selected sample of staff to represent the range and type/cadre of all staff present in each facility, and research centre staff involved in clinical trials. Data were collected on their primary and secondary responsibilities and whether these were linked to the malaria vaccine trial, and the type and nature of training received. In-depth interviews (IDIs) were undertaken with a range of health care providers and managers to investigate their experiences and perceptions of the impact of clinical trial activities on the quality of health care provided at the health facilities. Issues around MVT-MoH linkages and perceptions of post MVT impact were also explored. IDIs were held with senior investigators (n = 12), front-line health workers (n = 87) and health mangers (n = 14). All senior investigators with at least one year’s involvement in planning and implementation of the malaria vaccine trials were approached for interviews, and frontline health workers were purposefully selected based on their mechanism of employment (MoH staff involved in the trial, MoH staff not involved in the trial, and MVT employed staff) and their cadre (clinicians, public health officers, fieldworkers and support staff). We also approached health managers and policy makers at district, regional and national level with relevant health programme planning and implementation roles, such as Regional hospital mangers, District Health Management Team (DHMT) members and Maternal and Infant Health programme directors—see Table 3. In addition, in Kenya and Ghana focus group discussions (n = 9) were conducted with the fieldworkers (employed by the trial to conduct follow-up home visits and liaise between the health facility and community) in order to gather their views on how the trial inputs affected the functioning of the health facilities. Each site was responsible for their own data collection and management. For qualitative data, verbatim transcriptions and back translations were undertaken by the study team in each country. Across sites the data were managed in NVivo 8 and analysed using a framework approach [44, 45]. This process involved in-depth reading of transcripts to identify emerging themes across the datasets, developing a coding framework to code data, generating charts to summarise the data by categories, and data interpretation, to identify differences/similarities and provide explanations of the analysed data. The quantitative audit data were collated by the study teams in each country, and entered into Microsoft Excel to generate descriptive summaries. To facilitate cross-site synthesis and comparison, empty analysis tables were developed on the basis of an in-depth discussion of preliminary results from each site (one week workshop in Ghana). Each site then pulled out all data—qualitative and quantitative—to populate these tables. In a second workshop in Kenya, the data for each site were then collated and discussed. We employed the World Health Organisation (WHO) health systems building blocks to aid in the development of our framework for analysis and tables, and in the interpretation of data [46]. Ethical approvals were obtained from national ethics committees of the respective countries prior to the commencement of study activities (Ghana Health Service and Kintampo Health Research Center in Ghana; KEMRI National Ethical Review Committee in Kenya; Comité d’Ethique sur la Recherche en Santé (CERS) and Comité de Bioéthique Institutionnel du CNRFP (CIB/CNRFP) in Burkina Faso. Separate local ethics clearances were also obtained at each site. All data were anonymised, with access limited only to researchers. Written informed consent was obtained from all interviewees. Prior to any data collection, the malaria vaccine trial staff were briefed about our study, and permission was sought to conduct the work from health facility staff and opinion leaders. We gave feedback to malaria vaccine trial staff on emerging issues of importance and shared preliminary results with the MVT teams.
N/A
