Table 4. Key features of the two FEAST trial sites.
| Site | Kilifi District Hospital (KDH), Kenya | Soroti Regional Referral Hospital (SRRH), Uganda |
| Size of paediatric wards | 42 beds | 62 beds |
| Experience conducting trials | Clinical trials conducted for over 20 years | First major research activity |
| Community engagement activities | Coordinated community engagement activities focusing onthe institution and (where appropriate) specific studies | No formalised community engagement strategies for research |
| Employment and trainingof staff | Most staff within pre-existing clinical research group | Most staff involved in the trial trained and recruited specifically |
| Refusal rates for trial | High relative to other sites | Low relative particularly to KDH |
| Hospital user charges andother costs (a) | National exemption of charges for under five year olds, but little policy adherence.Food is provided for patients and most basic consumables available at a cost. | National exemption of charges for all admissions, but little policy adherence.Relatives provide and cook food in hospital grounds, and purchase many consumables needed (e.g. gloves, intravenous lines, some medicines) from local shops. |
| Trial inputs into study site | Established clinical programme as a robust platform forclinical trials (b). Also, trial provided extra personnel foremergency triage and patient monitoring, additionaldiagnostic tests and service-wide training inemergency care. | Basic maintenance and painting of paediatric wards, emergencyand triage equipment and training for all staff, and employmentof additional personnel. |
| Trial benefits forindividuals (d) | Close observation, treatment of new illnesses identified during admission, and free treatment of minor illnesses post discharge up until the 28-day follow-up visit. | As with KDH, every trial participant was allocated a dedicated‘clinical trial pack’(c). |
(b) Includes substantial support to the hospital for medical personnel (doctors, clinical officers, nurses and ward assistants); paediatric drugs, devices and equipment. Research funds also support the construction and running of an 8-bed paediatric high dependency ward available to all paediatric admissions, regardless of research involvement.
(c) Including cannulae, syringes, infusion sets, antibiotics, anti-malarials and blood testing consumables. Not needed in KDH where such support is provided to all inpatients in HDU.
(d) The usual hospital admission fees for the participants were not waived during the trial, in an effort to ensure that parents did not feel obliged to join the trial to save money. However, in Soroti only a few patients in a semi-private room are charged fees by the hospital.