dc.description.abstract | Background: Neonatal Jaundice (NNJ) is a common and often benign finding among newborn babies. In rare cases, untreated NNJ can lead to bilirubin associated neurological complications. Since these complications are preventable, many healthcare guidelines recommend improved NNJ screening. Currently the most common method for NNJ screening is a clinical visual assessment, but transcutaneous bilirubin (TcB) meters have better validity. While TcB meters have become popular in high income countries, they remain prohibitively expensive for many healthcare settings in low- and middle-income countries (LMICs). Smartphone based tools are an emerging technology that aim to provide reliable and affordable transcutaneous screening for jaundice. However, their cost-effectiveness in the context of an LMIC is not known.
Objective: To evaluate the cost-effectiveness of a smartphone application for neonatal jaundice screening compared to screening with a TcB meter in Pakistani context.
Methods: A cohort-based decision tree model is built to represent the pathway of a typical neonate in urban Pakistan with regards to jaundice screening, diagnosis and treatment. The model represents the comparison of screening with a TcB meter or smartphone application from a healthcare system perspective. Visual assessment by a clinician is built into each strategy. The tree is populated with estimates of pathway probabilities, costs and utilities from published literature as well as expert opinion. Two scenarios are created to account for uncertainties in the estimates of smartphone application’s validity as well as its expected market share. The base case scenario assumes that the cost per screening and market share of the smartphone application is the same as a TcB meter. The realistic scenario assumes that the smartphone application will be 30% less costly and achieve a 30% higher market share compared to TcB meters. Sensitivity analyses are performed for both scenarios.
Results: Results from the both the scenarios indicate that the use of the smartphone application could lead to a reduction in costs and an increase in effects compared to the TcB meter. However, the difference in effects in terms of Quality Adjusted Life Years (QALYs) is very small (<-0.0001 QALYs). Difference in clinical outcomes indicates that the smartphone application may help to decrease the rate of false negative screenings, hence reducing the need for emergency treatment for jaundice and the rate of adverse outcomes from jaundice. The results are sensitive to changes in the smartphone application’s validity and market share.
Conclusion: Smartphone application is cost-effective when compared to a TcB meter in Pakistani context given the assumptions of the cohort-based decision analytic model. | |