[Dysphagia] Swallowing screening

[eripley at yahoo.com (Irene Campbell-Taylor)]

The subject of screening tests arises from time to time on this mail list. I have recently had occasion to be involved in a critique of a proposed screening test for dysphagia in acute stroke. This immediately struck me as a problem since screening tests, by definition, seek to identify individuals who have a specific disease e.g. diabetes, not a symptom such as dysphagia. McKeown?s definition - any medical investigation that does not arise from a patient?s request for advice for a specific complaint. Another definition: ?The presumptive identification of cases of disease through tests that can be applied rapidly?. 

 I have the impression that there is a belief that screening can be made up ?on the fly? as it were by selecting a few items, strining them together and calling it a test. Even if we allow that screening is possible for a symptom (one dfinition includes the term ?defect?) it is essential to realize that any test requires that it be reliable i.e demonstrate specificity, sensitivity, positive predictive value and negative predictive value with a reasonable cut-off point.. Sensitivity refers to the proportion of false-negative results; true cases will be told incorrectly that they are free of the item. Specificity refers to the proportion of persons without the condition who correctly test "negative" when screened. A test with poor specificity will result in healthy persons being told they have the condition (false positives). An accepted reference standard ("gold standard") is essential to determining sensitivity and specificity, because it provides the means for distinguishing between
 "true" and "false" test results. Unfortunately, there is no universally accepted ?gold standard.?, VFSS notwithstanding. Please keep in mind that we are considering swallowing dysfunction, not aspiration.

The use of screening tests with poor sensitivity and/or specificity is of special significance because of the potentially serious consequences of false 'negative and false-positive results. Persons who receive false-negative results may experience important delays.

A proper evaluation of a screening test must include a determination of the likelihood of producing false-positive results. This is done by calculating the positive predictive value (PPV) of the test in the population of positive results that are correct (true positives). A test with low PPV can generate more false-positive than true-positive results, but this depends to a large extent on the ty pe of population in which it is used. The PPV increases and decreases in accordance with the prevalence of the target condition in the screened population. Thus, unlike sensitivity and specificity, the PPV is not a constant performance characterisitic of a screening test. If the target conditioin is sufficiently rare in the screened population, even tests with excellent sensitivity and specificity can have low PPV, generating more false-positive than true-positive results. The incidence of swallowing dysfunction in acute stroke hovers around the 50% mark, insufficient for accurate PPV.

 Reliability (reproducibility), the ability of a test to obtain the same result when repeated, is another important consideration in the evaluation of screening tests. An accurate test with poor reliability, whether due to differences in results obtained by different individuals (interobserver variation) or by the same observer (intraobserver variation), may produce results that vary widely from the correct value, even though the average of the results approximates the true value.

One also has to consider yield ? the amount of the variable detected relative to the effort to detect it. A reasonable cut-off point must be established ? when is enough , enough? This depends on the importance of missing a case. There is also the matter of bias and a number of aother variables. A "screening test" must be very carefully constructed. Fortunately, this isn't necessary as we already have one.Note that the authors refer to it as a "screening tool" not a test. Subtle difference but important.

The most reliable ?screening? for swallowing dysfuntion (and, in this case, aspiration)  in acute stroke is still the 3oz water swallowing test. 

Mari F, Matei M, Ceravolo MG, Pisani A, Montesi A, Provinciali L.Predictive value of clinical indices in detecting aspiration in patients with neurological disorders. J Neurol Neurosurg Psychiatry. 1997 Oct;63(4):456-60. 

The symptom "cough on swallowing" proved to be the most reliable in predicting the risk of aspiration, with 74% sensitivity and specificity, 71% positive predictive, and 77% negative predictive value. The standardised 3-oz test had a higher predictive potential than the clinical signs, but had low sensitivity. The association of cough on swallowing with the 3 oz test gave a positive predictive of 84%, and an negative predictive value of 78%. In cases where the clinical tests failed to detect any impairment, videofluoroscopy documented only a low risk (20%) for mild aspiration.  The association of two clinical items (such as history of cough on swallowing and 3 oz test positivity) provides a useful screening tool.

 Irene

 

 

 

 

 

 

 

 

 

 

 

 

 

 








Dr I Campbell-Taylor
Clinical Neuroscientist
Exclusive Distributor:
www.interactivetherapy.com