As the dementia progresses, the ability to self-report decreases (Pesonen et al. 2009), and other methods of pain assessment become necessary. The importance of adequate pain assessment is widely acknowledged and research predominantly focused on the development of behavioral pain assessment tools. Numerous behavioral pain assessment scales have been developed. Over ten reviews have been conducted including over 30 behavioral pain assessment tools for people with dementia (e.g., Herr et al. 2006; Husebo and Corbett 2014; Lichtner et al. 2014; Lobbezoo et al. 2011; Lord 2009; Park et al. 2010; Stolee et al. 2005; van Herk et al. 2007; Zwakhalen et al. 2006). An updated overview of assessment tools to measure pain with older people can be assessed via the City of Hope Pain and Palliative Care Resource Center (http://​prc.​coh.​org/​PAIN-NOA.​htm). In general all reviews conclude that no single instrument can be recommended for broad use in clinical practice. Several instruments (e.g., PAINAD, MOBID, PACSLAC) show acceptable psychometric qualities. Almost all tools are easy to use but much harder to interpret limiting its clinical usefulness. Therefore it is often a matter of preference and taking the context of the care setting into account when making a choice for a specific scale. Most of the tools have a cutoff score for the presence of pain available, like an onset of pain score. However authors were often unable to determine pain severity scores for the behavioral tools. In itself this is not surprising since people may vary enormously in their expressive behavior. Furthermore it’s worth mentioning that most of the tools were developed and tested in nursing home residents who often experience chronic pain. Consequently reliability and validity of the tools in other settings and other pain conditions (e.g., acute pain) are limited. Table 7.2 presents an overview of frequently used observational pain assessment scales for people with dementia. Given the enormous amount of scales available, a selection of tools is presented. Only scales that have been translated in more than one language are included in Table 7.2.

As a result of the research on tool development, it became clear that (1) patients with dementia display heterogeneous and atypical pain behavior (AGS Panel on Pharmacological Management of Persistent Pain in Older Persons 2009), (2) patients with verbal capacity displayed other behavior compared to nonverbal severe dementia patients (Kaasalainen et al. 2013), (3) behaviors associated with pain in this patient population are not unique to pain (e.g., guarding sore area or facial expressions), and (4) less obvious pain cues (e.g., behavioral problems like aggression and agitation) are frequently overlooked (Kaasalainen et al. 2013; Zwakhalen et al. 2007).

All available pain observation tools make use of facial expressions of pain. These facial pain cues like frowning and grimacing seem very useful for assessing pain in patients with dementia. Utility and reliability of facial expressions to measure pain in dementia have been frequently debated. However more and more evidence shows that these facial expressions are indeed one of the strongest and key cues to determine pain in dementia (Oosterman et al. 2016; Sheu et al. 2011).

Pain assessment is a must for adequate treatment; however, it does not guarantee successful treatment. Pain management includes both pharmacological and non-pharmacological approaches to reduce the amount of pain and improve functioning and the quality of life. Often a combination of pharmacological and non-pharmacological management is combined and individualized to the patient’s needs. A variety of non-pharmacological interventions that may be helpful to reduce pain in older people are available. These are often divided into physical and psychosocial intervention. Physical interventions include massage, exercise, positioning, TENS, etc. Psychosocial interventions include comforting approaches like relaxation, music, and distraction. Many of these non-pharmacological management approaches are easy to use by care workers and family. Non-pharmacological interventions may be effective; however, it must be mentioned that evidence about the effectiveness of non-pharmacological management is often limited. Furthermore the effectiveness may depend on the person’s abilities and characteristics (age, health condition, etc.).

Evidence about treatment and side effects of pain in dementia is still limited mainly due to the fact that these patients are often excluded from medical trials (McLachlan et al. 2011). Despite this, pharmacological management of pain is very common in daily practice. However, when people with dementia are prescribed with pain medication, usually the dosage is low, and weak analgesics are often prescribed (Corbett et al. 2012). For a long time, studies consistently reported a lower use of analgesics in patients with cognitive impairment compared to patients without cognitive impairment (Morrison and Siu 2000). However lately the opposite, namely, overuse of analgesics, is also reported in a number of Scandinavian studies on pain in patients with dementia (Haasum et al. 2011; Lovheim et al. 2008). This illustrates the difficulty of tailored and adequate treatment of pain. One could state that either underuse or overuse is inappropriate and undesirable.

Age-related changes are likely to influence how drugs are metabolized and absorbed in older people. When planning pharmacological interventions, the impact of these age-related changes such as comorbidities and use of multiple medications must be considered carefully in order to avoid complications and optimize pain treatment. The overall principle therefore in pain management in older people is “Start low, go slow!” This means that, for example, in the case of opioid use in older adults, an initial dose reduction of up to 50% of the recommended dose is warranted (AGS Panel on Pharmacological Management of Persistent Pain in Older Persons 2009). Most guidelines on pain in older adults provide a clear insight in pharmacological pain treatment options.