Expanding the Evidence-base for Clinician-Graded Dysphagia Using Dynamic Imaging Grade of Swallowing Toxicity (DIGEST): Validation Across Non-head and Neck Cancer Oncology Populations.
Beatrice Manduchi, Carla L Warneke, Xiaohui Tang and 3 others
PMID 41848899WHAT IT FOUND
Two SLPs grading the same videofluoroscopy studies independently agreed closely across 386 patients with cancer outside the head and neck, and every step up the DIGEST scale tracked a worse pharyngeal impairment score.
Correlations with patient-reported swallowing were weaker than hypothesised.
Key findings
01Two raters working independently agreed substantially on DIGEST, DIGEST-S and DIGEST-E across the total sample (weighted kappa 0.68 to 0.71, p < 0.001).
02DIGEST grades tracked an established measure of pharyngeal impairment (MBSImP pharyngeal score: rs = 0.78 in the total sample, 0.68 to 0.86 across subgroups), and every pair of adjacent grades was statistically distinguishable, with median MBSImP scores of 5, 8, 9, 12 and 17 for grades 0 to 4.
03Agreement with patient-reported measures was weaker than the authors had predicted: DIGEST correlated -0.36 with the MDADI Physical score and -0.32 with the PSS-HN diet scale, against a pre-specified expectation of -0.4 to -0.7.
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
Sampling was deliberately balanced across cancer groups, so the spread of grades in this paper cannot be read as how common dysphagia is in these populations. Only three cancer types plus mixed in- and outpatient groups were studied; blood cancers and gastrointestinal cancers were not included, so nothing here tells you about them. The comparison measures (MDADI and PSS-HN) were built for head and neck cancer patients and may not capture what matters to people with other cancers. Correlations with how patients reported their own swallowing and eating were weak, and fell short of the level the authors had set in advance; in two subgroups the swallowing-quality-of-life correlation was not statistically significant. It is one centre's retrospective records; the clinical raters graded during routine care and were not blinded, so the two ratings in each pair were not made under the same conditions. One component of the reference impairment score was excluded from the main analysis because 22.5% of the data were missing, though a sensitivity analysis including it gave similar results.
Declared interests
The work was funded by the National Cancer Institute (R01CA271223), an NCI Cancer Center Support Grant, and the Cancer Prevention and Research Institute of Texas. The declaration text provided lists no competing interests. The study was carried out at MD Anderson Cancer Center, the site that developed DIGEST, and the research raters were trained against the tool's own developer.
The easy way to misread this
Do not read this as showing that a DIGEST grade captures how a patient is doing. Its correlations with patient-reported swallowing (-0.36) and with level of oral intake (-0.32) were weak and below what the authors had predicted, and the swallowing-quality-of-life correlation was not statistically significant in the thoracic and inpatient groups. The authors' summary describes the validity results as strong, which overstates correlations this size.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →