A repeated measures design was used to assess the acceptability and effectiveness of Cognitive-Behavioural Therapy (CBT) for hording difficulties in 14 people with mild intellectual disabilities (ID). All participants were on the social-services case-register; residing in community properties in Birmingham. Potential participants were identified by their key-workers as having difficulty with hoarding. Prior to inclusion in the study, researchers assessed their level of hoarding and applied inclusion and exclusion criteria (e.g. people with severe learning disabilities were excluded).
Participants received 12 weekly sessions of manualised-CBT for hoarding in their homes. The researchers hypothesised that ratings of clutter, hoarding, depression and anxiety would reduce post-treatment and that improvements would be retained at follow-up.
The primary outcome measure was the Clutter Image Rating Scale (CIRS) which comprised photographs of varying degrees of clutter. Scores were derived by photographs of participants’ kitchens, bedrooms and living rooms being compared to the CIRS images by four independent raters. The measure was adapted to include more communal areas (CIRS+) and used as a separate outcome. Additional outcomes included changes in the Savings Inventory-Revised (SI-R); a self-report measure of hoarding-severity, and changes in depression and anxiety (using measures validated for people with ID).
The acceptability and safety of the intervention was assessed by monitoring treatment refusal, drop out and the change in hoarding-severity. Changes in outcomes across time (baseline; post-treatment, six month follow-up) were analysed using non-parametric tests. Effect sizes were also calculated.
In terms of acceptability, no participants refused treatment, dropped out or showed adverse responses to treatment. Regarding effectiveness, CIRS+ scores improved significantly (from baseline to post-intervention), with a large effect size. CIRS scores did not change significantly. Global SI-R scores improved significantly, with a moderate effect size. There were not significant reductions in anxiety or depression (except for the informant-reported measure of depression). In general, the results suggested that improvements were maintained at follow-up.
An identified limitation was that the CIRS was not validated in this population. However, there were additional limitations with this measure. The results were potentially confounded because 21% of participants lived in a communal property (and an unspecified additional number lived with a partner). Therefore, the amount of clutter could have been impacted by coinhabitants. Perhaps this explains why pre-post changes were only significant when the measure was extended to include more communal areas. It would have been useful to have based ratings exclusively on participants’ private bedrooms. Furthermore, it was not specified when the photographs were taken, or by whom. The photographs could have been taken at especially tidy or untidy times (by chance or by intent). To measure reliability, a range of participant photographs for each phase of the study could have been taken (to assess the daily variability of the clutter).
Steps were taken to improve reliability (e.g. applying Bonferroni corrections and inter-rater reliability checks) and safety (e.g. monitoring adverse events) and this preliminary study potentially paves the way for more tightly-controlled research. However, qualitative feedback about the acceptability of the intervention would have been useful to this end.