Consensus for the Development of a New Early Warning Score for Predicting Patients' Clinical Deterioration in Angola:...

Area of Science:

• Healthcare Innovation
• Clinical Assessment Tools
• Global Health

Background:

• Early Warning Scores (EWSs) are crucial for assessing acute illness severity but lack standardized adaptations for resource-limited settings.
• Existing EWSs often rely on multiple physiological parameters, outperforming single-parameter systems, yet optimal approaches remain debated.
• Angola requires cost-effective and context-specific EWSs to enhance patient care in its healthcare system.

Purpose of the Study:

• To identify physiological parameters suitable for integration into existing EWSs through the perspectives of Angolan healthcare experts.
• To lay the groundwork for developing a novel EWS tailored to Angola's unique healthcare context.
• To address the need for adaptable and cost-effective patient assessment tools in resource-constrained environments.

Main Methods:

• A three-round Delphi survey was conducted with a national panel of 25 Angolan physicians and nurses.
• Participants evaluated potential physiological parameters for EWS inclusion using a five-point Likert scale.
• Consensus was defined as achieving an ≥ 80% agreement rating from the expert panel.

Main Results:

• Consensus was reached on including standard parameters: systolic blood pressure, heart rate, respiratory rate, temperature, oxygen saturation, neurological status, and supplemental oxygen use.
• Experts agreed on incorporating additional parameters like seizures, jaundice, cyanosis, capillary refill time, and pain.
• Consensus supported excluding oxygen saturation and temperature measurements where oximeters and thermometers are unavailable.

Conclusions:

• Angolan experts successfully identified essential physiological parameters for adapting early warning scores.
• The study highlights the potential for a new aggregated EWS model tailored to the Angolan healthcare setting.
• Further validation is required to assess the impact of these suggested parameters on the new model's predictive capabilities.