Factors that predict good Active Support in services for people with intellectual disabilities: A multilevel model.

The design was repeated cross-sectional, so it did not follow individual service users over time and cannot show what sustains good Active Support. The study reports associations, not proof that changing one factor will improve Active Support. Staff self-report gave limited information about training type, so the study could not test whether classroom plus on-the-job training mattered. Observer agreement for the Active Support Measure varied across time points and was not assessed for Time Point 3. Only 14 organizations were included, so organization-level findings rest on a small number of clusters. The sample was described as relatively able compared with previous studies, so results may not apply to services with more severe disability. Where a service was measured more than once, the highest mean Active Support time point was used, which may not represent usual service performance. Some services were excluded from the analysis because of missing data. Staff culture, often proposed as influential, was not measured.

The article states that the study was supported by the Australian Research Council. The supplied text does not give a separate author conflict-of-interest declaration, and it does not say that a commercial sponsor designed or wrote the study.

Do not read these factors as causes. The study was repeated cross-sectional and did not follow individual change over time, so it cannot show that improving practice leadership, training coverage, service size, or organization size will raise Active Support.