Test-retest reliability and agreement of physical fall risk assessment tools in adults with intellectual disabilities

Physical decline and high rates of inactivity lead to an increased risk of falling in the intellectual disability (ID) population. It is important therefore to develop interventions to prevent falls and to develop valid and reliable assessment tools, which are suitable for use with people with ID. Targeting the most important fall risk factors such as strength, balance and gait measurement is important, yet there is a paucity of previous research on testing the feasibility and reliability of strength, balance and gait assessments with people with ID.

The aims of this study are (i) to describe the test-retest reliability and agreement [standard error of measurement (SEM)] of slightly adapted fall risk assessments and (ii) to describe the test-retest reliability and SEM of these tests in younger and older age groups and mild/moderate and severe/profound ID-level groups.

Residents of a German residential facility for people with ID were asked to take part. The study has a test-retest design, whereby all participants were tested twice, with 7 days in-between the first (T1) and second (T2) testing days. The 'timed up and go', '30-second chair stand', 'handgrip' and 'Romberg balance test' were all performed. Intraclass correlation coefficient (ICC) (2,1) values and SEM were calculated for the full sample, two age groups (split at the age of 60 years) and mild/moderate and severe/profound ID-level groups.

A total of 37 residents with a mean age of 59.3 years (standard deviation = 17.7) performed the tests on both testing days. Mainly moderate to excellent ICC values were found for all tests for the full sample and in all groups (0.59-0.97). Different SEM values were found for full sample and sub-groups.

Strength, balance and gait assessment tools, which are easy to use and understand, were found to be reliable in adults with ID. The SEM is most important for interpreting the real effects of an intervention. Further analyses will be required to gain more information about the SEM in different age groups or ID-level groups.