Avoiding False Positives: Why Comprehensive Dysphagia Evaluation Matters

Swallow Screening: Why identifying who does NOT have dysphagia is just as important as identifying who does

BIG thanks to Dr. James Coyle PhD, CCC-SLP, BCS-S, for reviewing and providing feedback on this article!

Dysphagia can be an elusive little critter. It comes in many different forms, often pops up and disappears at a moment's notice, and can last for years without a clear solution. So then identifying it from the get-go is incredibly important… And incredibly difficult.  

Managing dysphagia involves decision-making processes that include a fascinating blend of quick "gut checks" and deep-dive standardized evaluations. The first of these clinical decisions is whether a patient needs only a quick screen or a full clinical assessment.

The Smoke Alarm

Imagine your kitchen. A screening is like your smoke alarm. It’s designed to be loud, fast, and maybe a little oversensitive. If you burn a piece of toast, it goes off. It doesn't tell you where the fire is or how to put it out; it just says something might be wrong and that you need to pay attention RIGHT now.

A full, comprehensive clinical assessment, on the other hand, is like calling in the fire marshal. They don't just listen for the beep; they look at the wiring, check the stove, and figure out exactly why the alarm went off so they can create a formal evidence-based plan to keep the house safe.

Red Light Green Light

You might assume that coughing while drinking water is a guaranteed sign of swallowing issues. However, using a single symptom as a pass/fail cutoff can cause significant issues. In fact, Martino et al. (2000) showed that of 39 clinical signs, only 5 are even moderately better than a coin toss at predicting whether someone is aspirating. 

I wouldn’t want my clinician tossing a coin to decide the next step in my care… Would you? 

Take coughing, for example. Coughing is an indicator of dysphagia. It’s just a weak one. Particularly if it’s by itself. The predictability for problems increases only when we obtain additional data points (I.e., more clinical findings). 

So, a cough is an important early warning, but it isn't a final diagnosis. 

In a sense, every data point from a swallowing evaluation can be seen as a mini-screen. Like playing a game of red light, green light. The patient is lethargic? Red light. They’re alert and responsive? Green light. The patient isn’t tolerating their secretions? Red light. They have good, strong vocal quality and a functional cough? Green light. The patient has a significant facial droop? Red light? The cranial nerve exam came back completely normal? Green light. 

Each data point by itself isn’t all that helpful. But many data points? That’s a comprehensive assessment. The more data points we have, the more informed our clinical judgment becomes. 

Sensitivity VS Specificity

Another study, this one by Hey et al. (2013), found that of 80 patients, a water swallow screen had 100% sensitivity. In other words, it caught every single person who was actually aspirating. Great. 100% is good, right? Yes, but it’s not the whole story. 

There’s also the specificity. This tells us if we correctly identified those who passed the screen as those who do NOT have dysphagia. And in this study, it was only 61%. The screener essentially "failed" 16 people who weren't aspirating at all.

While missing patients who actually have dysphagia might seem like a more serious issue, think about how someone might be impacted if identified as dysphagic despite NOT having dysphagia… 

They may be put on a modified diet, given thickened liquids, made NPO, and even considered for a feeding tube. They may lose weight, become dehydrated, and have their quality of life completely flipped upside down. 

You would hope these types of gaps become identified and filled quickly, but oftentimes they don’t, leaving a patient labeled as dysphagic for years without a proper follow-up assessment. 

So identifying those who have dysphagia is important, of course (sensitivity). But I would argue that identifying those who don’t have it is just as important (specificity).

Why the "Fail" Might Be a False Alarm

Why would a screen fail someone who is actually fine? It often comes down to the cutoff point.

Low Cutoff: If we say one tiny cough equals a "fail," we catch everyone at risk (high sensitivity), but we also end up treating many healthy people (too many false positives).

High Cutoff: If we wait for someone to turn blue before calling it a fail, we might miss people who really need help (leading to high false negatives).

Case Study

Consider a 61-year-old man who just had a medullary stroke. He failed a nursing swallow screen because he coughed during a water test. There are two scenarios: either we rely solely on the screen or we pursue a full, comprehensive SLP swallowing evaluation. 

Let’s see how these two scenarios play out…

Scenario A (The Smoke Alarm Only): Based solely on that one "fail," he is immediately put on a nasogastric tube and told he can't eat or drink anything by mouth (NPO). 

Scenario B (The Fire Marshal): An SLP performs a full clinical assessment. They notice he has a history of COPD, which causes a chronic cough anyway. They also see that his palate is drooping. But his mind appears to be 100%; he’ll likely be able to follow directions for compensatory strategies after an instrumental swallowing evaluation helps us dictate a plan of care. Instead of a permanent feeding tube, they recommend ongoing, serial assessments—checking him every 24 hours to see if he improves as he recovers from the stroke.

So the best approach is having a low(ish) cutoff, but having a follow-up, comprehensive, standardized assessment that can confirm (or deny) whether that person actually has dysphagia and needs management.

Conclusion

Ultimately, screening is about exclusion—ruling out healthy individuals so we can focus on those who are at-risk. Whereas a complete, thorough clinical assessment is about diagnosis and prognosis—understanding the "why" so we can predict what should come next.

As Dr. Coyle notes in his course linked below, while we can lower uncertainty with better testing, we can never eliminate it entirely. The goal is to move from a simple "pass/fail" mindset to a comprehensive understanding of the person behind the swallow. The better we understand them, the better we can care for them. 

In other words, the screen should be the beginning of a process. Not the end.

If you’re interested in learning more about this topic, check out Dr. Coyle’s course on Northern Speech Services: Pros And Cons Of Dysphagia Bedside Screening And Assessment