Publication
Background Preterm birth and resulting respiratory failure is a leading cause of newborn death- the majority of which occur in resource-constrained settings and could be prevented with bubble continuous positive airway pressure (bCPAP). Commercialized devices are expensive, however, and sites commonly use improvised devices utilizing 100% oxygen which can cause blindness. To address this, PATH and a multidisciplinary team developed a very lowcost bCPAP device including fixed-ratio oxygen blenders. Objective We assessed feasibility of use of the device on neonatal patients as well as the usability and acceptability of the device by healthcare workers. This study did not evaluate device effectiveness. Methods The study took place in a Ugandan level two unit. Neonates with respiratory failure were treated with the bCPAP device. Prospective data were collected through observation as well as likert-style scales and interviews with healthcare workers. Data were analyzed using frequencies, means and standard deviation and interviews via a descriptive coding method. Retrospectively registered via ClinicalTrials.gov number NCT05462509. Results Fourteen neonates were treated with the bCPAP device in October-December 2021. Patients were born onsite (57%), with median weight of 1.3 kg (IQR 1-1.8). Median Birth States the Hospital, treatment length was 2.5 days (IQR 2-6). bCPAP was stopped due to: improvement (83%) and death (17%). All patients experienced episodes of saturations >95%. Median time for device set up: 15 minutes (IQR 12-18) and changing the blender: 15 seconds (IQR 12-27). After initial device use, 9 out of 9 nurses report the set-up as well as blender use was "easy" and their overall satisfaction with the device was 8.5/10 (IQR 6.5-9.5). Interview themes included the appreciation for the ability to administer less than 100% oxygen, desire to continue use of the device, and a desire for additional blenders. Conclusions In facilities otherwise using 100% oxygen, use of the bCPAP device including oxygen blenders is feasible and acceptable to healthcare workers. Trial registration ClinicalTrials.gov, Identifier NCT05462509.