Impact of Occupation-Based Groups on Occupational Performance and Satisfaction Outcomes: Pilot Study.

Small and underpowered: 21 people took part against a target of 40 needed for adequate power and 60 planned for allowing for dropouts. The authors say the results should be interpreted with caution. Not randomised. Groups were formed by who happened to be admitted in each time period, and different therapists delivered the early part of the control period, so the two groups could differ in ways nobody measured. There was no follow-up after discharge. Outcomes are people's own ratings of what they could do while still in hospital, so nothing is known about whether the gains held up at home. The control group had six people with neurological conditions against three in the group arm, and more people living alone. Neither difference was statistically significant, but the authors note that neurological and non-neurological rehabilitation can progress very differently. People had to be able to take part with supervision and to have a goal about everyday domestic tasks. Many patients approached declined because they had no such goal, so this sample may look different from a typical caseload. People with significant cognitive or communication problems were excluded, so the study says nothing about them. Therapists could not be blinded to what they were delivering, and the amount of actual practice within each session (how many tasks were done) was not recorded, so a difference in practice intensity cannot be ruled out. The inclusion criteria as printed say participants scored below 24/30 on the MMSE and below 13/20 on the short aphasia screen, which does not match who was actually recruited: the groups averaged 28.6/30 and 19.3/20, and the authors state elsewhere that people with significant cognitive or communication deficits were excluded.

The authors declared no conflicts of interest. The study received no specific grant from any public, commercial or not-for-profit funding agency. The first author was supported by an Australian Government Research Training Program Scholarship and a 2022 Graduate Scholar Award from the Golden Key Society, and Canberra Health Services provided in-kind support, including covering the cost of an on-site research assistant.

Do not read this as proof that group therapy works as well as one-to-one therapy, or as proof that it is worse. The study was not randomised, had only 21 participants against a target of 40, and found no significant difference in either direction, so it cannot show equivalence. Equally, do not read the figures on who reached a clinically significant change (82% against 50% for performance) as evidence that individual therapy was better. Those percentages come from a small, non-randomised sample and were not tested for a difference.