A decision support system for nurses caring for babies at risk of neonatal hypoglycemia was associated with a drop in the proportion of screened newborns developing low blood sugar, according to a mixed-methods study published in BMC Medical Informatics and Decision Making. Among 3,544 neonates screened across the study periods, hypoglycemia incidence fell from 30.14 percent (551 of 1,828) before the system was deployed to 24.94 percent (428 of 1,716) afterwards.
The Neonatal Hypoglycemia Nursing Decision Support System, or NHNDSS, was designed to help nurses make timely, evidence-based decisions about monitoring and managing low blood glucose in newborns.
The research team, led by Tiantian Wu and colleagues, evaluated the system using a quasi-experimental interrupted time series design with segmented regression analysis. This statistical approach allowed them to estimate both the immediate change in outcome levels after deployment and any longer-term shifts in trend. Data were collected weekly across 20 time points in each period: the pre-implementation window ran from June to October 2024 and the post-implementation window from December 2024 to April 2025.
The outcomes examined included hypoglycemia incidence, recurrence, transfers to the neonatal intensive care unit, length of hospital stay, adherence to breastfeeding guidance, and missed blood glucose tests. In addition to the quantitative series, the researchers conducted qualitative interviews with nurses after implementation to explore how usable the system was and how it fit into clinical workflows.
Segmented regression showed a significant decline in initial hypoglycemia incidence over time following deployment, with a trend coefficient of β3 = -0.014 (P < 0.001). Recurrent hypoglycemia initially rose, reaching 8.185 percent, but subsequently followed a significant downward trend (β3 = -0.517, P = 0.003). The average length of hospital stay initially increased but then declined significantly over time (β3 = -0.144, P < 0.001).
Missed blood glucose tests fell significantly over the post-implementation period (β3 = -0.308, P < 0.001), suggesting the system helped nurses keep up with the demanding monitoring schedule that hypoglycemia care requires. Two other outcomes showed little change: NICU transfers showed a non-significant trend (β3 = -0.149, P = 0.439), and the measured impact on breastfeeding guidance was also not statistically significant (β3 = 0.440, P = 0.279).
The qualitative interviews indicated that nurses acknowledged the value of the NHNDSS, while also suggesting improvements in interoperability with other hospital systems and in technical functionality. The authors concluded that the system supports evidence-based decision-making, improves neonatal hypoglycemia management, and has gained clinical acceptance among nursing staff.
Even so, the combination of interrupted time series modeling with nurse interviews offers a detailed picture of how a targeted nursing decision support tool performed in routine clinical practice, with the clearest signals appearing in reduced hypoglycemia incidence, fewer missed glucose checks, and shorter hospital stays over time.
Subject of Research: Development and interrupted time series evaluation of a neonatal hypoglycemia nursing decision support system
Article Title: Neonatal hypoglycemia nursing decision support system: mixed-methods development and evaluation using interrupted time series
Article References: Wu, T., Yi, Y., Du, X., Li, Q., Hao, Y., Zhou, T., & Tao, L. (2026). Neonatal hypoglycemia nursing decision support system: mixed-methods development and evaluation using interrupted time series. BMC Medical Informatics and Decision Making. https://doi.org/10.1186/s12911-026-03943-x
Image Credits: AI Generated
DOI: 10.1186/s12911-026-03943-x
Keywords: neonatal hypoglycemia, decision support system, nursing informatics, interrupted time series, segmented regression, NICU transfers, blood glucose monitoring, breastfeeding guidance, hospital length of stay, clinical decision-making, quasi-experimental study, nurse interviews
News Source: Harold Sullivan. (October 28, 2026). Nursing decision support tool linked to lower neonatal hypoglycemia rates in hospital study. Scienmag.
