632 - Advancing neonatal jaundice screening with mHealth: a mixed-methods randomized controlled feasibility study in a resource-limited setting in Mexico

Publication Number: 2617.632

Gabriela Jimenez-Diaz, PICTERUS AS, Trondheim, Sor-Trondelag, Norway; Jesús Elizarrarás-Rivas, Mexican Social Security Institute (IMSS) / Universidad Autónoma “Benito Juárez” de Oaxaca (UABJO), Oaxaca de Juárez Oaxaca, Oaxaca, Mexico; Martina Keitsch, NTNU, Trondheim, Sor-Trondelag, Norway; Lobke M. Gierman, Picterus AS, Trondheim, Sor-Trondelag, Norway; Anna Marcuzzi, Norwegian University of Science and Technology, Trondheim, Sor-Trondelag, Norway; Jennifer J.. Infanti, Norwegian University of Science and Technology (NTNU), Trondheim, Sor-Trondelag, Norway

Background: Neonatal jaundice is a common and usually harmless condition, but 10% of cases can progress to severe forms that can cause irreversible brain damage or death if undetected and untreated. The prevalence of severe cases is higher in resource-limited settings due to limited access to reliable detection methods among other contributing factors.

Objective: To assess the feasibility of implementing Picterus Jaundice Pro, a smartphone-based detection tool for neonatal jaundice, in a resource-limited Mexican healthcare setting

Design/Methods: A mixed-methods approach was employed, integrating a pragmatic randomized controlled trial (RCT) with qualitative insights from semi-structured interviews, all conducted within a feasibility framework. In the RCT, newborns were randomly assigned to screening by visual assessment alone (control) or visual assessment plus Picterus Jaundice Pro (intervention). Feasibility outcomes included RCT derived measures - the proportion of positive screenings, successful referrals to emergency room (bilirubin ≥ 14.6 mg/dL [250 µmol/L]), treatment needed, and treatment received. Recruitment capability, device practicality, integration, acceptability, and protocol adherence were also assessed.

Results: From January to June 2024, 542 newborns were included (273 control, 269 intervention). Compared with control, the intervention arm showed higher proportion (proportional difference [95% Confidence Interval]) of positive screenings (0.299 [0.230-0.362]), successful referrals (0.112 [0.058- 0.166]), treatment needed (0.045 [0.008-0.082]), and treatment received (0.038 [0.003-.072]).
Interviewed participants considered Picterus Jaundice Pro user-friendly and well-integrated into workflows. Despite high protocol adherence, incomplete referrals due to logistical barriers and low perceived risk by parents occurred in 11% of cases.

Conclusion(s): Implementing Picterus Jaundice Pro yielded encouraging improvements in screening practices, underscoring its potential to contribute to advance neonatal care. Addressing systemic barriers and raising awareness among parents and healthcare workers is crucial for maximizing impact.

Table 1. Proportions of newborns with positive neonatal jaundice screening, successful referral to emergency room, treatment needed, and treatment received*
NNJ: neonatal jaundice, ER: emergency room, CI: confidence interval
* Proportions were calculated using the total number of newborns enrolled per group, as these outcomes were prespecified feasibility endpoints reflecting real-world screening and referral process.
1 Chi-square and Z-test to compare proportions
#Three participants with Picterus JP negative screening but positive visual assessment
**Two participants with Picterus JP negative screening but positive visual assessment


Taable 2. Key findings
Picterus JP: Picterus Jaundice Pro, ER: emergency room, NNJ: Neonatal Jaundice

Figure 1. Study flow chart illustrating enrolment, randomisation, allocation, and follow-up of participants.
ER: emergency room, TSB: total serum bilirubin, TSB positive: total serum bilirubin ≥ 14.6 milligrams per decilitre (250 micromoles per litre), Picterus JP: Picterus Jaundice Pro