Implications for practice and research

• Currently knowledge and care rely upon evidence from hospital and nursing home settings and this review suggests that this knowledge is not transferable.

• There is a strong requirement for comparative studies to evaluate the effectiveness of different approaches to pain assessment and treatment among people with dementia in the community.

Context

There is an ageing population with a related increase in dementia many of whom are likely to experience pain.1 Many more people with dementia are likely to be cared for in the community than in a formal care setting. We know that people with dementia cared for in hospital and nursing home settings experience pain and are treated differently than those who do not have dementia.2 Most research related to people with dementia in pain occurs in these settings and thus provides the evidence base for care. This review seeks to understand what the situation is for those living in the community.

Findings

Bullock et al 4 report that most people with moderate to severe dementia were unable to complete standard self-report pain instruments. Although adaptations such as language and font size may improve this, the current evidence suggests the need for using other measures in addition to self-report in this group. Carers are more likely to report that the person with dementia is in pain than the people themselves. Nurses and nursing assistants are less likely to identify pain in this group of people. This discrepancy suggests that the care environment is an important factor in the accuracy of pain assessment. No good quality studies have evaluated informant-reporting tools in a community setting.

A lack of homogeneity restricted the reviewers’ ability to report on pharmacological interventions, with variable designs used and contradictory results evident. This compares with strong evidence in nursing homes that analgesia is less likely to prescribed. There was limited evidence that people with dementia are more likely to use paracetamol and less likely to be prescribed Non-Steroidal Anti-Inflammatory Drugs than those without. For opioids, the evidence is unclear; overall use is less but this increases the longer a person has dementia. Where opioid prescription occurs, they tend to be strong rather than weak opioids and use non-oral methods. It is unclear whether this relates to disease progression. The evidence on effectiveness in this group was very limited and only indicated that scheduled paracetamol use reduced pain scores. It is likely that analgesia is underutilised in this group.

The evidence for non-pharmacological treatment was very weak, with only poor quality papers evident so that no conclusions were drawn.

Commentary

Although this is not a systematic review of the literature but a qualitative synthesis, a thorough approach was adopted. However, while reputable quality appraisal tools were used to evaluate the literature, low-quality papers were not rejected from the subsequent review. Excluding these would have improved the ensuing discussion and enhance its value as an account of the evidence base.

Despite this, the review is important. It establishes that the environment where care occurs may have a role in pain management. It identifies the need to repeat robust nursing home or hospital-based studies in a community setting. In particular, it singles out the need to perform good quality studies into the effectiveness of non-pharmacological treatments.