This study protocol describes a type 1 hybrid effectiveness-implementation study evaluating a clinical decision support dashboard for neonatal pain management in a tertiary NICU. Using a pre- and post-implementation design, the study assesses whether the dashboard reduces the duration of inadequately managed neonatal discomfort during the first seven days of admission, while concurrently evaluating implementation outcomes such as acceptability, appropriateness, fidelity, and penetration. Quantitative EHR- and dashboard-derived data are complemented by qualitative interviews and observations to provide a comprehensive evaluation of both clinical impact and implementation processes.

This protocol outlines a single-center, type 1 hybrid effectiveness-implementation study evaluating an in-house developed clinical decision support dashboard designed to optimize pain management in the Neonatal Intensive Care Unit (NICU). The study is conducted at a level III-IV NICU and follows a pre-implementation (retrospective), mid-implementation, buffer, and post-implementation (prospective) design over approximately 26 months.

The primary effectiveness objective is to determine whether implementation of the dashboard reduces the duration of inadequately managed neonatal discomfort during the first seven days of admission. Discomfort is operationalized as the cumulative duration a neonate is classified as "red" by the dashboard prioritization algorithm, indicating the need for immediate clinical attention. Pre-implementation "red" classifications are generated retrospectively from electronic health record data using the same algorithmic criteria applied prospectively post-implementation.

The secondary objective is to evaluate the implementation process of the dashboard within routine NICU workflows. Implementation outcomes, based on the framework of Proctor et al., include acceptability, appropriateness, fidelity (implementation and effectiveness), and penetration. These outcomes are assessed using a mixed-methods approach, combining quantitative measures such as dashboard use metrics, prescribing patterns, and timing of pain assessments with qualitative data from semi-structured interviews and observations involving healthcare professionals and parents.

The comfort-dashboard is implemented using a phased ward-based approach supported by targeted training, local champions, and ongoing facilitation. The study poses minimal risk, as the dashboard supports but does not replace clinical decision-making, and no changes are made to core algorithm logic or outcome definitions during the study. By integrating effectiveness and implementation evaluations, this protocol aims to generate clinically relevant evidence while supporting reproducibility, transparency, and future scalability of digital decision support tools in neonatal care.

Inclusion Criteria:

Primary endpoint - Effectiveness evaluation (neonates):

• Admitted to the NICU of Erasmus MC-Sophia Children's Hospital during either the pre-implementation or post-implementation phase
• Admitted and discharged during the respective phase
• Parental consent for use of medical information provided upon admission

Secondary endpoints - Implementation process evaluation (healthcare professionals):

• Physician, nurse, or nurse specialist working in the NICU during the study period who has been exposed to TMD-C
• Management staff who has been exposed to TMD-C

Secondary endpoints - Implementation process evaluation (parents):

- Parent or guardian of an included neonate who was exposed to the dashboard during the samenzorgvisite (family-integrated ward round)

Primary endpoint - Effectiveness evaluation (neonates):

• Readmitted to the NICU during the same study phase
• Admitted to the NICU for less than 24 hours
• Discharged or transferred before 7 days of admission without complete pain management data available

Secondary endpoints - Implementation process evaluation:

- None

Individual participant data will not be proactively shared at this stage. Sharing will be reconsidered after study completion and publication. Requests for de-identified data may be considered on a case-by-case basis by the principal investigator.