Using Classical Test Theory to Determine the Psychometric Properties of the Deglutition Handicap Index.
Renée Speyer, Reinie Cordier, Clara Bouix and 2 others
PMID 33515311WHAT IT FOUND
The DHI should be scored as one total, not as three separate emotional, functional and physical scores.
In 453 patients and 132 controls, a cut-off of 8 helped separate people with dysphagia from healthy controls.
Key findings
01The DHI should be used as one total score, not as separate emotional, functional and physical subscale scores.
02Using FEES or VFS as reference tests, a DHI Total score cut-off of 8 had sensitivity 92.7% and specificity 94.7%.
03Patients with dysphagia had significantly higher DHI Total scores than healthy controls (p < 0.001).
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study did not check reliability over time because participants were not measured more than once. It did not check responsiveness, meaning whether the score changes when swallowing improves or worsens. The analysis used Classical Test Theory only, and the authors say this is a first step, not a full psychometric evaluation. Classical Test Theory findings depend on the sample, so they may not transfer unchanged to other populations. There is no international consensus on the best way to interpret FEES or VFS as the reference standard. Healthy controls were relatives and caretakers of patients, not a general healthy sample. Severe cognitive impairments were excluded, so the findings do not cover that group. The recruited patient count was 453, but floor and ceiling analyses report 395 participants, so the analysed sample is not fully explained. The paper gives two different total-score ranges: 0 to 120 in the questionnaire description and 0 to 12 in interpretability, so the exact meaning of the cut-off of 8 is unclear from the supplied text. The three-domain factor model explained 41.6% of variance, below the 50 to 60% range the paper cites as expected.
Declared interests
The supplied text names the University of Oslo (including Oslo University Hospital) as funder. It does not provide an author conflict-of-interest disclosure list.
The easy way to misread this
Do not treat a DHI total score of 8 as a stand-alone diagnosis of dysphagia. The cut-off was checked against FEES or VFS in one French sample, and the authors say Classical Test Theory is only a first step.