mHealth to Screen for Neonatal Jaundice in Low-resource Mexican Settings: a Feasibility Study

Background Neonatal jaundice (NNJ) affects a significant percentage of newborns (60%-80%) within the first 48 to 72 hours of life. Due to increased production and decreased elimination in the newborns, bilirubin accumulates in the blood, causing neonatal hyperbilirubinemia (NHB). While most cases resolve without consequences, severe NHB affects approximately 1.1 million infants annually, necessitating close monitoring and treatment. If undetected, NHB can lead to brain damage and conditions such as acute bilirubin encephalopathy, impaired neurologic development, long-term disabilities (including audiologic, language processing, cognitive and visual-motor disturbances) and, in the worst cases, death.

Currently, reliable detection of NNJ depends on invasive or expensive methods, such as transcutaneous bilirubinometer (TcB) devices and total serum bilirubin (TSB) blood tests, which are limited in most low- and middle-income countries (LMICs) as well as in out-of hospital settings. As a result, detection relies on using visual assessment (VA), an unreliable and inaccurate screening method that leads to a higher prevalence of severe consequences in LMICs than in high-income countries (HICs). This results in preventable health inequities with significant social and economic costs.

In response to these challenges, a Norwegian multidisciplinary research group developed an accessible, affordable mobile health (mHealth) solution called Picterus Jaundice Pro (Picterus JP) to screen for NHB. Picterus JP consists of a smartphone app to capture and upload images of the newborn's skin, a calibration card for accurate colour calibration in the images and a server for image analysis and storage. The app has undergone validation in Norwegian hospitals and in pilot studies in Mexico and the Philippines, demonstrating significant positive correlation between Picterus JP's bilirubin values and TSB levels as well as higher accuracy over VA.

Digital health technologies such as Picterus JP have immense potential to enhance healthcare in resource-constrained areas, but such innovations face implementation challenges that can hinder their adoption, leading to slow, unreliable and unsustainable outcomes, particularly in LMICs, such as Mexico.

To get insights about low-resource settings in Mexico, particularly in the state of Oaxaca, an exploratory study on the field to identify the main barriers and facilitators to implement Picterus JP in urban and rural public health facilities was recetly conducted. Through non-participants observations and interviews deficiencies in the current neonatal jaundice detection process and the participants' desire for a more accurate method were identified. Participants tested Picterus JP on a dummy perceiving the app as easy to use, useful and compatible with the work routine, but barriers to adoption were identified, including internet deficiencies and costs.

The aim of this study is to evaluate the feasibility of introducing Picterus JP as a tool to screen for neonatal jaundice in low-income settings of Mexico by examining whether and how the app can be implemented and evaluated.

Feasibility studies are usually conducted prior to outcome-focused pilot studies or full-scale evaluation of an intervention effectiveness and are more focused in the intervention process than in the intervention outcomes. Aspects to evaluate in feasibility studies include: recruitment capability; appropriateness of data collection procedures and outcome measures; acceptability of the intervention and study procedures; adaptability, practicality and integration of the intervention; effectiveness of the intervention; and participant responses to the intervention.

Specific Objectives:

• Assess the recruitment capability.
• Assess the adaptability, practicality, and integration of Picterus JP
• Assess the acceptability of Picterus JP among health care workers and newborn parents.
• Describe the potential effectiveness of Picterus JP compared to the standard care.

Methodology:

Study design: feasibility RCT Study setting: Rural Hospital Bienestar No 36 in Tlacolula (HR36), Oaxaca, Mexico.

Participants:

• Health care workers providing neonatal care at the "preventive medicine unit" and emergency room in the Rural Hospital Bienestar No. 36, Tlacolula, Oaxaca, Mexico.
• Newborns attending the preventive medicine unit for metabolic screening and immunization.
• Parents of newborns screened with Picterus JP for neonatal jaundice

Sample size or rationale for determining a feasible sample size: no formal sample size calculation is needed for feasibility studies. We aim to recruit 6-8 health care workers, 10-12 newborn parents, and at least 200 newborns in each arm for the intervention study based on the sample size requiered for an RCT with 90 percent power and alfa = 0.05.

Procedures:

• The HCWs will be informed about the study and will receive training on the use of Picterus JP by the researcher prior to the start of the study.
• The baseline values of the study parameters will be determined by measuring them for 1 month prior to the start of the intervention.
• Newborn participants will be identified by the research assistant in the hospital waiting room when they come to the neonatal metabolic screening or immunization.
• Prior extensive information about the study, informed consent will be requested from the parents and/or guardians of the newborns.
• Once the informed consent letter has been signed by the parents or guardians (the signature of only one of them will suffice, that is, the mother or father or guardian) and two witnesses, the participants will be randomly assigned to either of the two study groups (intervention or control) before proceeding to the consultation room.
• In the intervention group, the HCW will use Picterus JP in addition to SC to screen for NNJ. To do the screening with Picterus, a paper calibration card will be placed on the newborn's chest and six images will be captured with a validated phone with the app downloaded, which will be analysed by the software to later show the bilirubin level measured.
• In the control group, the HCW will use only the SC to screen for NNJ.
• In case that Picterus JP shows a bilirubin value of 12 mg/dl or higher or the HCW suspects of severe jaundice through SC, the newborn will be referred to emergency room for further evaluation by a medical doctor.
• The data collection will be carried out for 6 months.
• HCWs who used the app will be interviewed by the main researcher at the end of the data collection to explore their experiences and perceptions towards the use and acceptance of Picterus JP.
• A questionnaire to explore attitudes towards the use of Picterus in their child will be applied to parents of newborns.

Data analysis:

The results will be analysed by descriptive statistics for outcomes 1,2 and 3c; by thematic analysis for outcomes 3a, 3b and 4; and by X2 tests to compare proportions of variables for outcome 5 (independent variable = introduction of Picterus as a tool to detect NNJ; dependent variables = number of newborns referred to laboratory for blood test, number of referred newborns with positive blood tests, number of newborns hospitalized with NNJ as main diagnosis and number of newborns receiving NNJ specific treatment).

Inclusion criteria

• Term healthy newborns
• Age 1-14 days
• Gestational age >37 weeks
• Birthweight > 2000 grams

Exclusion criteria

• Inborn diseases
• Transferred to pediatric wards for non-related neonatal jaundice treatment
• Have received phototherapy