Development of a Screening Model for Patients with Clinical Suspicion of Stroke at the Outpatient department, Mahasarakham Hospital

Abstract

Objectives : This developmental research aimed to 1. develop a screening model for patients with clinical suspicion of stroke in the outpatient department of Mahasarakham Hospital, 2. pilot-test the developed screening model, and 3. evaluate its effectiveness in terms of nurses' compliance with the model, user satisfaction, and clinical screening outcomes.

Methods : This developmental research was conducted after approval from the Human Research Ethics Committee of Mahasarakham Hospital. The study consisted of three phases. Phase I involved the development of a stroke screening model based on the BEFAST concept and the National Health and Medical Research Council (NHMRC) clinical practice guideline development framework. The model was validated by three experts, yielding a content validity index (CVI) of 0.83, and its reliability was tested using inter-rater reliability, resulting in a coefficient of 0.85. Phase II: Pilot Implementation of the Clinical Suspicion Screening Model and phase III: evaluated the effectiveness of the model among 20 outpatient nurses and 30 patients selected through purposive sampling. Outcomes included nurses’ compliance with the screening model, user satisfaction, accuracy of patient screening, identification of Fast-Track stroke cases, and incidence of patient deterioration while awaiting medical assessment.

Results : The developed screening model consisted of four categories. Nurses demonstrated a screening competency rate of 92.40% and reported a satisfaction rate of 97.5%, with a mean satisfaction score of 4.87 (SD = 0.42), indicating the highest level of satisfaction. The screening model successfully activated the fast-track alarm for patients presenting with clinical symptoms suggestive of stroke, accounting for 30.00% of cases. Among these patients, 3.33% were admitted to an inpatient ward, 10.00% were referred to the Emergency Department for urgent evaluation, and 16.67% were assigned priority queuing for expedited physician consultation. The screening error rate was 1.75%, which was lower than the 7.00% threshold established by the Ministry of Public Health. Furthermore, no incidents of clinical deterioration were observed among patients while awaiting examination after implementation of the screening model.

Conclusion : The developed screening model for patients with clinical symptoms suggestive of stroke demonstrated effectiveness, reflecting the benefits of having a clear model and utilizing tools that support more accurate symptom screening, thereby reducing errors in assessing patient severity.