The TOR-BSST© was developed and validated by Dr. Martino of The Swallowing Lab, University Health Network, University of Toronto.
A literature search was conducted to identify all relevant publications on the psychometric properties of the TOR-BSST©. Four studies were identified.
The TOR-BSST© is a 5-item screening test to determine risk of dysphagia. The screening should be discontinued as soon as an individual fails an item.
No studies have reported on internal consistency of the TOR-BSST©.
No studies have reported on the test-retest reliability of the TOR-BSST©.
No studies have reported on the intra-rater reliability of the TOR-BSST©.
Martino et al. (2009) established inter-rater reliability of the TOR-BSST© in the first 50 patients with stroke enrolled, using intraclass correlation coefficient (ICC) and 95% confidence intervals (CI). Results indicated excellent test-retest reliability (ICC=0.92; CI, 0.85 to 0.96).
Martino et al. (2006) examined 24-hour inter-rater reliability of the TOR-BSST© item and total screen scores in a sample of 286 patients with stroke (acute, n=78; subacute/chronic, n=208), using kappa statistics. Results indicated moderate reliability for the total score, with a higher reliability early after training (k = 0.90). Item reliability ranged from poor to adequate; the item ‘water swallowing’ including both the 50-ml and sip achieved the highest item reliability (k=0.82; CI, 0.66-0.98).
Initial item generation for the TOR-BSST© resulted from systematic review of the accuracy and benefit of non-invasive bedside dysphagia screening tests with patients with stroke (see Martino, Pron & Diamant, 2000). Two measures were shown to be accurate predictors of dysphagia by videofluroscopic assessment (VFS) of aspiration, and a further two were considered to show promising (although inconsistent) predictive ability:
- Dysphonia/coughing during the 50mL Kidd water swallow test
- Impaired pharyngeal sensation
- Impaired tongue movement
- General dysphonia – voice before or voice after water intake
The final measure, general dysphonia, was defined as two sub-items (voice before, voice after).
Item reduction was then performed, whereby positive results across the 5 items were compared with the total score. The item ‘water swallow’ contributed 25% to the total positive score, indicating that this item was the most frequent single item to identify dysphagia. The item ‘tongue movements’ contributed 8% to the total positive score. The remaining items contributed less than 5% each to the total positive score, and so were considered for elimination on review of practical application as determined by expert Speech-Language Pathologists. These expert clinicians considered the item ‘pharyngeal sensation’ to be impractical due to difficulty differentiating from a gag reflex in the clinical setting.
Martino et al. (2014) conducted item descriptive analysis in the original sample of 311 patients with stroke from acute and rehabilitation settings. The TOR-BSST© was administered by trained nurses. Items were eliminated individually to evaluate the impact of each item on the total score. Results showed that the ‘water swallow’ item contributed most significantly to identification of dysphagia, identifying 42.7% of patients in the acute setting and 29.0% of patients in the rehabilitation setting.
No studies have reported on the concurrent validity of the TOR-BSST©.
Martino et al. (2009) examined predictive validity of the TOR-BSST© by comparison with gold standard VFS assessment identifying any abnormal swallow physiology including all severity. The randomized controlled diagnostic study design included four blinded Speech-Language Pathologists and 68 patients with stroke in acute and rehabilitation settings. Nine participants with stroke were eliminated when the TOR-BSST© and VFS assessments were performed more than 24 hours apart as per a priori criteria for patient flow. VFS assessment was used to confirm findings obtained by TOR-BSST© screening; clinicians rated the VFS images using three standardized scales: (1) Penetration Aspiration Scale; (2) Mann Assessment of Swallowing Ability (MASA) dysphagia subscore; and (3) MASA aspiration subscore. Across the entire sample of acute and rehab patients, results showed that 61% (n=36) of patients were confirmed by experts to have no dysphagia vs. 39% (n=23) with dysphagia. These results indicate high accuracy to predict dysphagia using the TOR-BSST©, where dysphagia is defined by aspiration and/or physiological abnormality on VFS.
No studies have reported on the convergent/discriminant validity of the TOR-SST©.
No studies have reported on the known-group validity of the TOR-BSST(c).
Martino et al. (2009) examined sensitivity of the TOR-BSST© by comparison with VFS assessment, in a sample of 68 patients with stroke in acute and rehabilitation settings. Nine patients were eliminated when the TOR-BSST© and VFS assessments were performed more than 24 hours apart. The TOR-BSST showed 91.3% sensitivity (CI, 71.9 – 98.7) and 66.7% specificity (CI, 49.0 – 81.4) among all patients. Sensitivity and specificity was 96.3% and 63.6% (respectively) among patients in an acute setting, and 80.0% and 68.0% (respectively) among patients in rehabilitation settings. The TOR-BSST© showed high negative predictive value of 93.3% and 89.5% in participants in acute and rehabilitation stroke settings, respectively.
Martino et al. (2014) conducted sensitivity analysis of the TOR-BSST© in the original sample of 311 patients with stroke from acute and rehabilitation settings. The TOR-BSST© was administered by trained nurses using the standard 10 teaspoons plus a sip of water. Positive screening occurred in 59.2% of patients in the acute setting (n=103) and 38.5% of patients in the rehabilitation setting (n=208).
Martino et al. (2014) further examined sensitivity of the TOR-BSST© when modifying administration according to water volume intake. Using the original sample from Martino et al. (2009), sensitivity was examined on administration of 1 to 10 teaspoons of water to determine the acceptable cut-point to identify dysphagia. Among all participants (n=311), sensitivity ranged from moderate to excellent for 5, 8 and 10 teaspoons of water (79%, 92%, 96% respectively). Among patients in the acute setting and rehabilitation settings, sensitivities were 84% and 75% (respectively) for 5 teaspoons of water, 93% and 92% (respectively) for 8 teaspoons, and 95% and 97% (respectively) for 10 teaspoons. Results indicate greater accuracy on administration of 10x 5mL teaspoons of water, as per the original assessment guidelines
Martino, R., Maki, E., & Diamant, N. (2014). Identification of dysphagia using the Toronto Bedside Swallowing Screening Test (TOR-BSST©): are 10 teaspoons of water necessary? International Journal of Speech-Language Pathology, 16(3), 193-8. https://www.ncbi.nlm.nih.gov/pubmed/24833425
Martino, R., Nicholson, G., Bayley, M., Teasell, R., Silver, F., & Diamant, N. (2006). Interrater reliability of the Toronto Bedside Swallowing Screening Test (TOR-BSST©) [Abstract]. Dysphagia, 21(4), 287-334. https://doi.org/10.1007/s00455-006-9044-5
Martino, R., Pron, G., & Diamant, N. (2000). Screening for oropharyngeal dysphagia in stroke: insufficient evidence for guidelines. Dysphagia, 15, 19-30. https://www.ncbi.nlm.nih.gov/pubmed/10594255
Martino, R., Silver, F., Teasell, R., Bayley, M., Nicholson, G., Streiner, D.L., & Diamant, N.E. (2009). The Toronto Bedside Swallowing Screening Test (TOR-BSST): Development and validation of a dysphagia screening tool for patients with stroke. Stroke, 40, 555-61. https://www.ncbi.nlm.nih.gov/pubmed/19074483