This clinical trial aims to test the validity of a two-item swallowing screen and to examine the effects of the Swallowing and Oral-Care (SOC) Program on resumption of oral intake, incidence of penetration and aspiration, and incidence of pneumonia in adult patients who successfully extubated after ≥ 48 hours of endotracheal intubation.

Endotracheal intubation is life-sustaining, but it may contribute to postextubation dysphagia (PED) increasing the risk of penetration, aspiration, and aspiration pneumonia. Up to 84% of extubated patients had PED and approximately 60% penetrated and aspirated that can lead to aspiration pneumonia. The aims of this three-year, two-stage study are: 1) In the first stage, the investigators develop a two-item swallowing screen involving oral stereognosis and cough reflex test for predicting the resumption of oral intake and feeding-tube dependence by using a diagnostic accuracy study method; 2) In the second stage, the investigators conduct a randomized, open-label, controlled trial design to examine the effect of a once-daily, 7-day SOC Program on resumption of oral intake, incidence of penetration and aspiration, and incidence of pneumonia in adult patients who successfully extubated after ≥ 48 hours of endotracheal intubation.

Inclusion Criteria:

• Patient who is over 20 years old.
• Patient how has received over 48 hours endotracheal intubation and had been successfully extubated.

Exclusion Criteria:

• Patient who has a history of neuromuscular disease (e.g., parkinsonism or stroke) or head and neck deformities.
• Patient who has preexisting difficulty swallowing.
• Patient who has received a tracheostomy.
• Patient who were unable to follow verbal instructions.
• Patient who were on contact and droplet precautions (e.g., open tuberculosis)
• Patient who were receiving continuous noninvasive ventilation after extubation that precluded the delivery of an SOC intervention

To maintain allocation concealment, the investigators ensured that only the intervention nurse had access to the random sequence. Physicians and staff at the study sites were aware of a pending nursing-intervention study but were blinded with respect to the hypothesis, group allocation, specific SOC protocols, and study endpoints. Moreover, outcome assessors were blinded to the group assignments, ensuring an unbiased evaluation.