Gitlin LN, Corcoran M, Winter L, Boyce A & Hauck W,

2001,

Reviewed by Helen Buri, School of Health and Social Care, Sheffield Hallam University

Methodology

A randomised controlled trial was undertaken to determine the effects of an occupational therapy home environmental intervention on the function of dementia patients and self-efficacy and upset in their informal carers. One hundred and seventy-one carers were randomised to either a treatment or control group. Those carers in the treatment group received 5 visits lasting 90 minutes each from an occupational therapist, in which they received education and modification to the environment. The control group received usual care. The sample was identified mainly through medical centres and social services in Philadelphia, USA. Several inclusion and exclusion criteria were applied to ensure uniformity amongst participants. A trained interviewer screened potential participants for inclusion, gained consent and conducted the baseline interview prior to randomisation. Randomisation was stratified by gender and race. A series of outcome measures were used, some of which related to the person with dementia and others relating to the carer. Outcomes relating to the person with dementia included: recording the frequency of behavioural problems, the level of activities of daily living (ADL) dependence and the level of dependence in instrumental ADL (for example making a meal or managing money). Outcomes relating to the carer included self-efficacy and upset in managing dementia behaviours, ADL dependence and instrumental ADL dependence. Standardised measures or numerical scales reported in previous research were used to record data. Outcomes were repeated at 3 months post-baseline. The effects of the intervention were analysed using ANCOVA (analysis of covariance) and regression analysis was used to determine if certain subgroups of carers benefited more than others from the intervention.

Critical appraisal

The strengths of this trial include measures taken by the researchers to reduce bias. By stratifying the randomisation process by gender and race, they were able to ensure the intervention and control group had an equal chance of including different racial and gender groups. Although there was a 15% dropout rate by the time of the 3-month follow-up, the participants who dropped out were compared with those who stayed in, and no significant difference was found. The characteristics of the intervention and control groups were compared at baseline and both groups were significantly similar. Although the sample consisted of mainly female carers, this would seem to be representative of the carer population. The outcome measures selected had been previously validated or used in other similar studies, enhancing the validity of the results.

There are several aspects of the way the study was conducted which may have affected the validity and reliability of the conclusions drawn by the researchers. A trained interviewer carried out the outcome measurement and will probably have guessed which group participants were in when the 3-month follow-up was carried out, potentially affecting how results were recorded. Also, the measures involved carers rating perceptions about their ability to cope with caring and those in the intervention group may have perceived themselves to be coping better just by being visited by an occupational therapist, regardless of the efficacy of the treatment. As with many randomised trials in rehabilitation, therapy intervention is difficult to standardise. The authors acknowledge the variability of interventions and the difficulty for therapists in adhering to a study protocol.

Key findings

The main findings of this study were as follows:

• No significant difference was found between the intervention and control groups for the outcome measures relating to carers and all but 1 of the outcomes relating to the person with dementia.
• Carers in the intervention group reported less decline in instrumental ADL in the person with dementia at 3-month follow-up than carers in the control group, which was statistically significant (p = 0.03).
• Sub-group analysis revealed that female and minority carers in the intervention group showed better scores for self-efficacy than male and white carers, and spouse carers benefited from the intervention for upset more than non-spouse carers, to significant levels.

Appropriate statistical tests have been used in the analysis of results. The main problem is that although standard numerical measures have been used, recording data about carers' perceptions and emotions has a subjective element that may affect reliability. The significance of the statistical results should also be viewed with some caution, as the sub-group analysis tested the benefits of intervention within groups, for example comparing male against female results or spouse against non-spouse, which is different from testing the intervention group against the control group.

Social work practice/ policy implications

Despite the lack of statistically significant findings in favour of an occupational therapy home environmental modification programme, professionals should not assume that similar interventions will not benefit carers of persons with dementia. The subjective, emotional and individualised nature of informal caring mean that randomised controlled trials may not be the most suitable means of determining effectiveness. A qualitative study may help uncover what carers perceive to be useful in terms of occupational therapy intervention. Research in this field is important, as recent government legislation, including the National Strategy for Carers (Department of Health 1999) has highlighted the needs of the 5.7 million informal carers in the UK.