This retrospective observational study looked at 126 patients who had their first-ever stroke. The researchers examined routinely available clinical variables such as age, sex, stroke type, and specific scores like the modified Rankin Scale. They also included serum albumin levels and body mass index in their analysis. The goal was to see if these common data points could predict swallowing status without needing specialized tests like videofluoroscopic swallowing studies or fiberoptic endoscopic evaluations.
The study found that 88.9 percent of patients met a threshold of 3 on the Food Intake LEVEL Scale, while 57.9 percent met a higher threshold of 7. Decision curve analysis suggested the model using the higher threshold of 7 offered greater clinical utility. This means it might be more helpful for making decisions about patient care.
The main reason to be careful is that the study used only internal validation. The researchers recommend external validation before these models are widely adopted. This retrospective observational study shows a link between common variables and swallowing estimates but does not prove causation. Readers should understand that these findings may not apply to all settings until further testing confirms them.