i. Physical inputs to facilities.

Physical infrastructure. In Ghana, the Kintampo North Municipal hospital and one of the three health centres had previously been involved in other KHRC research activities. During the RTS,S phase III trial set-up, the children’s ward in the referral hospital and child welfare clinics in the three health centres were repainted and the patient waiting bay area expanded. In Kenya, the Kilifi County Hospital receives on-going support from the KWTRP but the 3 rural facilities selected for the RTS,S phase III had minimal prior engagement with the KWTRP research activities. At the time of MVT set-up, upgrades to these rural facilities included the construction of additional rooms, expansion of patient waiting bays, new roofing and repainting. In Burkina Faso, the Clinical Research Unit of Banfora (CRUB) has continued to receive support from pooled funds from the programme to centrally run its clinical trial activities. During MVT set-up, the CRUB was entirely refurbished and fully equipped by the CNRFP, although no upgrades were undertaken in the community clinics, which served as recruitment sites (Table 4).

Across all three countries the renovations undertaken to support the MVT were appreciated by frontline health staff. Some respondents said that the additional rooms provided extra working space that could be used during and after MVT activity hours, and that upgrades of existing structures (such as the children’s wards and patient waiting bays) created a conducive environment for providers and health facility users. In Kenya for instance, one health care provider said that during MVT set-up, the facility management requested the MVT team to have one of the newly constructed rooms converted to a facility delivery room, and this was perceived to contribute to improved maternal deliveries and staff performance.

…when the trial was starting I had a big problem with finding a place to do deliveries for pregnant mothers, and as government employees we are gauged on our performance annually on certain services…when the trial came we requested they construct for us a room…and it has helped because the community benefits and I am able to report to my employer… (IDI11/MoH with top-up/Kenya)

However, some of the respondents reported concerns linked to the physical location of MVT activities and capacity within facilities. For instance, in smaller facilities there were concerns of insufficient space for both trial and routine service delivery, potentially leading to over-crowding, as expressed by this respondent from Kenya:

…in the same location we [MoH staff] do pregnancy screening, they [MVT team] sometimes use the room for blinding [investigational product concealment] and administer vaccines. When a client needs VCT [voluntary counselling and testing] we [MoH staff] use the same room…and yet the same room is our [facility] delivery site… (IDI16/MoH with top-up/Kenya)

Another challenge articulated by some participants was that while there were efforts to ensure MVT activities were laid out to blend with existing structures not all were successful. For instance, in Burkina Faso there was a concern that the separate construction for MVT activities limited the access to these enhanced health services for community members not enrolled in the trial.

Transport infrastructure. Across all three countries MVTs purchased additional off-road vehicles (ranging between 2 and 9 cars) to facilitate MVT activities at the peripheral MVT health facilities. Although availability of MVT transport was mainly to facilitate the implementation of MVT activities, the majority of frontline providers across the three countries felt the presence and access to MVT vehicles played a big role in supporting routine services. For instance, transportation of essential drugs and other medical supplies, improved patient access to facilities (allowing for early diagnosis and treatment of illnesses) and assisted with patients’ referral and community outreach. These inputs were perceived to benefit MVT participants and non-participants alike.

Medical care support. Across all three countries, the MVTs channelled substantial funding into the purchase of medical equipment such as laboratory equipment, oxygen cylinders and examination couches (Table 5). The MVTs ensured there were regular supplies and access to essential drugs (such as paracetamol, antibiotics, antimalarials and cough syrup) at MVT sites and clinical equipment (such as thermometers, anthropometric scales, and stethoscopes) where needed for routine patient clinical care (Table 5). Audit data indicate that in Ghana and Burkina Faso, hospitals and clinical trial units received more clinical care equipment than peripheral health facilities; by contrast in Kenya, there was more support to dispensaries and health centres. This difference is likely to have been due to the different roles of these facilities in MVT related activities and the additional inputs to facilities to help meet MVT site standards.

Most MoH health providers appreciated the improved access to essential drugs provided by the MVTs. Some MoH staff said the availability of malaria diagnostic kits and training in their use help them adhere to treatment guidelines rather than treating malaria presumptively, while others suggested there had been a general reduction in malaria morbidity and mortality due to MVT activities.

…were it not for the study child [participant], you would have treated them clinically [presumptively]…but because we have been given kits [malaria RDTs], you test them and at least you feel that you have done something good. (IDI20/MoH with top-up/Kenya)

In addition, in the past most facilities had challenges handling emergency care, hence the availability of oxygen cylinders and gas donated by MVTs to peripheral and tertiary health facilities, was reported to enable clinicians help resuscitate sick patients. Furthermore, in Ghana the availability of digital x-rays was perceived as a major boost to the hospital and due to the presence of MVT activities, it became standard practice to screen all febrile children for malaria and bacteraemia on admission in all MVT facilities.

ii. Human resource inputs to facilities in each country.

Staffing. In all three countries additional full time staff were hired during the MVT period to assist with patient clinical care at the MVT peripheral health facilities (Table 6). These included study clinicians (doctors, nurses and clinical officers) who were hired to provide clinical care and administer MVT vaccines to study participants; and fieldworkers, who were employed from local communities where MVTs were conducted to assist in conducting follow-up home visits. In MVT facilities, existing MoH clinical staff were also employed on a locum basis, or provided salary top-ups to assist with the provision of clinical care, especially to study participants during and out-of facility working hours.

Across all three countries, the majority of respondents appreciated the additional health workers employed by MVTs, as they were perceived to contribute to a reduction in workload for MoH health workers. For instance, MVT clinicians supported with all paediatric patient consultation during busier clinic days in MVT primary care facilities or while the nurse in-charge/MoH clinicians were absent from the facility (while attending official meetings—all three countries, and during the nationwide health workers strike—Kenya only). In Kenya and Ghana, MVT clinicians regularly supported MoH clinicians with the administration of routine childhood vaccines alongside MVT vaccines.

…we [MoH staff] are over-stretched and we don’t have a lot of staff; so when they [MVT staff] go [to health facilities], they also assist with some of the routine aspects of the OPD [out-patient department] cases and other cases. We are able to take lessons and pass it on to the other [MoH] nurses… (IDI/MoH without top-up/Ghana)

The facility staff and some trial frontline staff also reported that MVT staff at MVT peripheral facilities provided supportive supervision, acted as role models and could easily be approached to seek clinical advice/consultation. Most of these respondents felt the good working relationship that existed between trial and MoH staff facilitated a good learning environment.

With the presence of CNRFP workers [nurses] in the facility, the behaviour of other health workers has changed. They are more attentive towards patients and are more careful in filling in [information in clinic] books, even when they are not those CRFs [case report forms] of CNRFP. (IDI44/MoH without top-up/Burkina Faso)

By contrast, across all three countries, some staff (among all cadres and types of employment) felt MVT activities had led to an increase in their workload, through adding new activities such as filling in MVT forms, having to work for extra hours, and performing extra chores without pay. The latter concerns were raised particularly by trial fieldworkers and MoH non-trial staff.

Training. In all three countries, the MVTs provided training in good clinical practice (GCP), good laboratory practice (GLP) and health research ethics (HRE). However, the training provided varied across the countries by cadre of MoH staff and according to level of involvement in trial related activities. For instance, in Ghana and Kenya, MoH clinical staff involved in trials were provided with training but non-clinical staff were excluded. By contrast, in Burkina Faso, all staff at the community clinics involved in the trial received training. In most instances, the majority of the MoH frontline providers had minimal prior research experience and the training provided by the MVTs were organised during site initiation and when new MoH staff were posted to MVT facilities. In general, access to training was perceived to be an individual benefit and a benefit for the patients through improvements in the quality of care provided.

All the staff involved with patient care in our health facility were trained on microscopy use…and how to perform RDTs [rapid diagnostic tests]. In district meetings, we [MoH staff in MVT facilities] are considered privileged… (IDI40/MoH with top-up/Burkina Faso)

However, there were mixed perceptions regarding the nature, practicality and relevance of the MVT related training. For instance, participants in Burkina Faso expressed concern that staff were trained on GCP but lack supportive supervision and in Kenya one of the respondents found the training ‘complex’ and would need to reconsider attending a refresher course. In Burkina Faso concerns were raised about the amount of training needed to handle new MVT purchased medical equipment (such as haematology analysers), particularly since high staff turn-over invariably meant further training for newly posted MoH staff.

iii. Health financing and perceptions of input.

MoH staff incentives. In all three countries, the MVTs provided salary top-up or allowances to existing MoH staff (including paediatricians, nurses, doctors) at peripheral and referral health facilities to support various aspects of MVT activities (see Table 6). For instance health managers provided oversight and support supervision to MoH staff in MVT facilities, while MoH clinicians provided care to study patients during and out-of work hours. However, the calculation of payment varied by country. In Ghana, it was based on number of hours worked and in Kenya, it was calculated as a fixed proportion of the monthly salary. In Burkina Faso, it was based on involvement in specific trial activities. There were also country variations in who was eligible for a salary top. For instance, in Ghana and Kenya only MoH staff providing clinical care were included while in Burkina Faso, all MoH staff, irrespective of their cadre/specialization were included.

… the project [MVT] pays monthly to the facility the amount corresponding to the total number of filled files. Then the in-charge redistributes this amount to all facility staff so that each staff gains something, even the guard and the essential drugs deposit staff. This [approach] seems more equitable to us [facility staff]. (IDI31/MoH with top-up/Burkina Faso)

The mechanism of rewarding staff generated mixed reactions. In all three countries, most MoH clinical staff receiving a salary top-up reported that their standards of living had improved (particularly compared to their MoH colleagues not involved in MVTs) and they felt motivated to perform better. These sentiments were echoed in interviews with trial staff.

Regularly giving them “something” [top-up] motivates MoH staff in the peripheral health facilities participating in CTs. Not only do they work well, but they also work regularly…the output is quite different… (IDI01/trial staff/Burkina Faso)

However, MoH frontline staff in all three countries also reported challenges with the MVT reward mechanisms. Some of the MoH clinical staff receiving a top-up felt their pay compared poorly to trial clinicians even though they performed similar roles. Many MoH staff not receiving a top-up were unhappy that they were excluded from this benefit despite providing back-up to their colleagues where needed.

Facility operational cost. The MVTs covered some of the operational costs incurred by the MVT peripheral health facilities including a monthly fuel and communication allowance in Ghana and Kenya, and an internet connection to the referral hospital in Ghana and the CRUB in Burkina Faso. These inputs improved the connectivity of the peripheral facilities, enhancing their ability to manage their drugs supplies and obtain support in clinical decision making. In Kenya, the dispensaries were able to charge the MVT for the user-fee payments that they would have otherwise received from the MVT participants when they visited the facility. Given the relatively high use of facilities by trial participants (encouraged by the MVT staff) these funds contributed to increased facility funding, allowing for the payment of support staff salaries and the purchase of supplies like gloves.

…so in one of the dispensaries…we [MVT and facility staff] had to agree that any participant seen will be recorded and pay their user fee on a monthly basis…so it’s a small income for the dispensary…and they use the money to pay subordinate staff and sustain services within that dispensary. (IDI24/trial staff/Kenya)

i. Equipment and infrastructure.

The participants were ambiguous about the potential post-trial benefits and challenges associated with the equipment and infrastructure brought to the facilities by the MVTs. Many were unsure of what would be left behind and, if equipment was left behind, who would be responsible for maintenance. In all three countries concerns were raised over MoH capacity (financial and technical) to handle and maintain equipment after trial closure.

…these days facilities don’t generate income and it would be very difficult for them to take good care…or provide maintenance to their structures….any money that comes to the system is programmed…so getting funds to maintain the clinical trial infrastructure would be difficult. (IDI/MoH without top-up/Ghana)

In addition, some MoH frontline staff suggested that before the closure of the trial consideration should be given to providing additional support such as facility upgrades, or donation of funds to a facility kitty to assist with facility operations post-trial.

ii. Staffing and employment.

In all three countries, fieldworkers and MoH clinicians reported concerns about loss of employment and earnings, with negative implications for staff morale and facility workload.

…there will be a very big gap. The cadre of professionals who were working here…like medical officers at dispensary level is something that rarely happens in our country…we get a lot of assistance from them [trial]… (IDI16/MoH with top-up/Kenya)

iii. Facility service delivery.

It was universally perceived that withdrawal of MVT staff and services (transport and drug supply) would lead to a significant deterioration in facility services including longer queues/waiting time, increased MoH staff workload, and a return to drug-shortages and problems with facility access.

When the study leaves our hands will be chopped! Because when we ran out of supplies, they [MVT] bring for us….if they leave we will suffer… (IDI20/MoH with top-up/Kenya)

Other dilemmas expressed especially by fieldworkers and MoH staff were how to handle community expectations. In particular, they felt participants and their families would face problems meeting their health care costs, and that this could impact on their treatment seeking patterns.

… Mothers will go back to their initial habits of delaying to seek care in facilities when their child is sick, because they lack money to buy drugs…when the trial ends and there is no free care, mothers will opt to stay at home when their children are sick. (IDI28/trial staff/Burkina Faso)