Deimplementing Untested Practices in Homecare Services: A Preobservational-Postobservational Design.

This was a census of homecare services, not a client trial, so it cannot show whether clients received better bathing equipment. There was no control group or randomization, so the before-and-after design cannot separate Algo from other changes in homecare over the same period. The intervention included Algo plus a reference manual, user guide, website, and training, so the study cannot separate which component influenced use. Answers came from telephone surveys of managers and clinicians, so they describe reported tool use, not audited practice. The authors note possible inaccurate responses, maturation bias, historical bias, and that the study does not identify what drove knowledge use. The paper describes validation with 74 elders, but the supplied text does not report agreement between non-occupational therapists and occupational therapists, so Algo's clinical accuracy cannot be judged. One service implemented Algo, and it had not used a local tool before, so the study does not show how to replace existing local tools. Findings are from Quebec homecare services in 2009 to 2013, so they may not apply to other systems or times.

The paper says financial and scientific support was available for the clinical occupational therapist to lead the BATH project, but it does not name the funder or declare conflicts of interest.

Do not conclude that Algo works because one service reported tangible benefits. That was one service, it had not used a local tool before, the project included a reference manual, user guide, website, and training, and the study did not report client outcomes.