Functional performance tests are often not applicable to bedridden patients. The proposal to perform a physical test that requires little equipment, minimal training and simple execution in bed, expands the opportunity for evaluation and rehabilitation strategies for a variety of hospitalized patients, from bedridden to independent ones. Thus, the bed bridge test (BBT), and its time-limited and repetition-limited versions, may constitute a new functional test.

To test the clinimetric properties of BBT reliability and validity: 5 repetitions (BBT 5R) and 10 repetitions (BBT 10R), 30 seconds (BBT 30sec) and 60 seconds (BBT 60sec), in hospitalized patients.

Were included 92 patients eligible for the study performed in random order the BBT5rep, BBT 10rep, BBT 30sec and BBT 60sec repeated on two days with an interval of 48 hours. Validity was tested by correlation analysis between the Functional Status Score (FSS) scale, the sit-to-stand test (BBT) and the Short-Physical Performance Battery test (SPPB). With data from day 2, reproducibility was analyzed with the intraclass correlation coefficient (ICC), standard error of measurement (SME) and minimum detectable difference (DMD). Effect floor and ceiling were also tested.

Participants were 50.9±17.2 years old, 60% women and 66% with clinical condition. The test-retest ICC (95%) was good to excellent (BBT 5R CCI:0.89, 95%CI 0.84-0.93; BBT 10R CCI:0.92, 95%CI 0.88-0.95; BBT 30sec CCI:0.87, 95%CI0.80-0.91; and BBT 60sec CCI:0.88, 95%CI0.83-0.92). The concordances observed for the BBT 5R were EPM: 1.2 and DMD: 3.4; for BBT 10R EPM: 1.8 and DMD: 5.0, for TPL30sec EPM: 1.6 and DMD: 4.4; and for BBT 60sec EPM: 2.8 and DMD: 7.6. There were appropriate ceiling and floor effects for all versions. Content validity was observed by the weak association between the performance of the BBT versions and the performance on the FSS (r s =-0.27 to -0.37 and 0.29 to 0.36, p<0.05 for all). Construct validity was observed by the moderate association between the four versions of the BBT and the SPPB (r s =-0.63 to -0.58 and 0.43 to 0.53, p<0.05, for all). Criterion validity was observed by the moderate association between the four versions of the BBT and the BBT (r s =-0.48 to -0.58 and 0.64, p<0.05) for all.

All versions of the BBT showed good reproducibility, measurement error and validity measurement, with no ceiling or floor effect in hospitalized patients. The BBT versions can be a good alternative for the functional assessment of bedridden patients.

This study allows us to present suggestions for future studies. Thus, it is suggested to continue investigating whether the BBT can be used as a predictor of other outcomes.