Barry Hounsome

Lifestyle factors playing a role in the development of late-life disability may be modifiable. There is a need for robust evidence about the potential for prevention of disability through behavior change interventions.

Advisory bodies, such as the National Institute for Health and Clinical Excellence (NICE) in the UK, advocate using preference based instruments to measure the quality of life (QoL) component of the quality-adjusted life year (QALY). Cost per QALY is used to determine cost-effectiveness, and hence funding, of interventions. QALYs allow policy makers to compare the effects of different interventions across different patient groups. Generic measures may not be sensitive enough to fully capture the QoL effects for certain populations, such as carers, so there is a need to consider additional outcome measures, which are preference based where possible to enable cost-effectiveness analysis to be undertaken. This paper reviews outcome measures commonly used in health services research and health economics research involving carers of people with dementia. An electronic database search was conducted in PubMed, Medline, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), PsycINFO, the National Health Service Economic Evaluation Database (NHS EED), Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment database. Studies were eligible for inclusion if they included an outcome measure for carers of people with dementia. 2262 articles were identified. 455 articles describing 361 studies remained after exclusion criteria were applied. 228 outcome measures were extracted from the studies. Measures were categorised into 44 burden measures, 43 mastery measures, 61 mood measures, 32 QoL measures, 27 social support and relationships measures and 21 staff competency and morale measures. The choice of instrument has implications on funding decisions; therefore, researchers need to choose appropriate instruments for the population being measured and the type of intervention undertaken. If an instrument is not sensitive enough to detect changes in certain populations, the effect of an intervention may be underestimated, and hence interventions which may appear to be beneficial to participants are not deemed cost-effective and are not funded. If this is the case, it is essential that additional outcome measures which detect changes in broader QoL are included, whilst still retaining preference based utility measures such as EQ-5D to allow QALY calculation for comparability with other interventions.

The relatively high rate of suicide among UK farmers suggests that they may suffer greater mental health problems than the general population. This paper provides a comparison of the psychological morbidity of farmers and their partners/spouses with non-farmers. The General Health Questionnaire (GHQ-12) was administered using face-to-face interviews with 784 attendees of agricultural shows in the UK. Results show that GHQ-12 scores for farmers and their partners/spouses were significantly higher (P < 0.001) than those for the non-farming population, indicating higher psychological morbidity among farmer families. Approximately 35% of farmers had scores 12 and higher (recommended cut-off for psychiatric disorders), compared to 27% of non-farmers. Within the farmers group, male respondents, those aged from 45 to 64, self-employed or not in paid employment, having a non-supervisory position and living in a rural area were characterized by higher mean GHQ-12 scores compared to correspondent subgroups from the non-farming population.

The growing number of people with dementia, and the increasing cost of care, provides a major incentive to develop and test methods of supporting them in the community for longer. Most attention has been given to pharmacological interventions, but there is increasing recognition that psychosocial interventions may be equally effective, even preferable where medication has negative side-effects. Reminiscence groups, run by professionals and volunteers, which use photographs, recordings and other objects to trigger personal memories are probably the most popular therapeutic approach to working with people with dementia, but there is little evidence for their effectiveness and cost-effectiveness. The recent inclusion of family carers in groups with people with dementia, notably in our own pilot studies, has generated informal evidence that this joint approach improves relationships between people with dementia and their carers, and benefits both.

There is wide recognition that pragmatic randomised trials are the best vehicle for economic evaluation. This is because trials provide the best chance of ensuring internal validity, not least through the rigorous prospective collection of patient-specific data. Furthermore the marginal cost of collecting economic data alongside clinical data is typically modest. UK Clinical Research Collaboration (UKCRC) does not require a standard operating procedure (SOP) for economic evaluation as a prerequisite for trial unit registration. We judge that such a SOP facilitates the integration of health economics into trials.