Test-Retest Reliability of the Triangle Completion Test

The triangle completion test (TCT) is a clinical measure of spatial navigation that has been associated with vestibular function. There are limited data on its statistical reliability.

This study aimed to assess the test-retest reliability of the TCT.

Test-retest study of prospectively recruited participants.

Participants aged 60 years or older with no falls within the last 12 months and no history of vestibular dysfunction were recruited. Twenty participants were included, with a mean (standard deviation) age of 69.2 (6.25) years. Participants were equally divided among male and female participants (50.0 percent) and were predominantly White (80.0 percent).

Patients walked right triangles with leg lengths 1 meter (m) by 2 m by 2.23 m. Participants walked four triangles in the test trial (Trial One) and four triangles in the retest trial (Trial Two) without visual or auditory cues.

Endpoint error and angular deviation were measured for each triangle. Participants were retested three hours later. Test-retest reliability was assessed using intraclass correlation coefficients (ICC) and absolute reliability was assessed using the minimal detectable change (MDC).

In Trial One, participants had a mean endpoint error of 80.5 (53.4) centimeters (cm) and a mean angular deviation of 12.5 (4.82) degrees. In Trial Two, participants had a mean endpoint error of 64.4 (23.6) cm and a mean angular deviation of 14.4 (6.91) degrees. Endpoint error had poor reliability (ICC 0.05, 95 percent confidence interval [CI]: -1.41, 0.62), whereas angular deviation had fair reliability (ICC 0.72, 95 percent CI: 0.28, 0.89). The endpoint error had an MDC of 144.7 cm, and the angular deviation had an MDC of 7.13 degrees.

TCT angular deviation shows promise for good reliability, whereas endpoint error showed poor reliability. Further refinements to the TCT are needed to improve clinical feasibility and reliability.

These results call into question the feasibility and reliability of the TCT and highlight the need for protocol standardization and optimization.