Psychosocial Interventions in Dementia Care

It is probably not difficult for anyone to understand that the term psy­chosocial is a combination of the words psychological and social. Despite the simple explanation of the word itself, defining what psychosocial interventions exactly are is not so simple. In dementia literature several definitions could be found for psychosocial interventions and no abso­lute criteria exist to determine whether a certain intervention for people with dementia should be classified as psychosocial or not. Although psychosocial interventions are always nonpharmacological, nonphar- macological interventions are not necessarily psychosocial. They should not simply be seen as an alternative to drugs. Whereas pharmacologi­cal treatments are developed to fit symptoms of a disease, psychosocial interventions should fit the person and caregiver who suffer from dis­ease symptoms.

The aim of psychosocial interventions is to optimize the quality of life of people with dementia and their caregivers and could be directed to the patient, the caregiver, or both. Psychosocial interventions could be defined as

Interventions usually involving interaction between people, to sup­port cognition, emotion, personal relationships and a sense of con­trol in people with dementia and their family caregivers, through valued, meaningful activity and social integration.1

Psychosocial interventions became part of dementia treatment in the 1970s. At first interventions focused on the functional symptoms of the disease aiming at rehabilitation and compensation of functioning. Later psychosocial interventions became more emotion-oriented, focusing on the person with dementia and his or her own experiences and coping strategies. These days psychosocial interventions focus on the person with dementia as a whole at any stage of the disease, taking into account the persons functional capacities and subjective experiences and feelings.

Theoretical models are often used to explain and explore coping with and adaptation to dementia symptoms. A popular model that is used to develop new psychosocial treatment strategies is the stress-coping model of Lazarus and Folkman (1984) . In this model feelings of stress are explained by the way a person appraises a stressful situation and the extent to which he or she feels capable of coping with it. Stress is consid­ered a product of the interplay between the environment and the person and reducing stress could be achieved by changing a person’s perception of stressors and providing strategies to cope (for further reading on psy­chosocial models see Finnema et al. 2000).

To date no treatment exists that can cure or at least stop progression of any type of dementia. Logically, the treatments available for people with dementia and their caregivers focus on postponing cognitive decline as long as possible, alleviating behavioral and psychological symptoms dur­ing the course of the disease, and alleviating stress of caregivers. Psycho­social interventions could be effective in treating symptoms and problems related to dementia.

Professionals working in dementia care often use clinical guidelines to help them decide which treatment is most appropriate in the case of specific symptoms or situations. In many countries dementia guidelines are avail­able for geriatricians, neurologists, nurses, general practitioners, and other professionals working in dementia care. Guidelines ideally summarize the scientific evidence and best practice that is currently available and most dementia guidelines emphasize the importance of psychosocial interven­tions in the treatment of people with dementia and their caregivers. Many guidelines even recommend that psychosocial interventions should be the first choice when treating behavioral and psychological disease symptoms. The difficulty with recommendations on psychosocial interventions, how­ever, is that these are often stated rather generally. Whereas a recommen­dation for a pharmaceutical could be very clear about the dose that should be given, for how long, and when the pharmaceutical treatment should be stopped, this is far more difficult for a psychosocial intervention.

Psychosocial interventions interfere with the interaction between a person’s psychological state and social environment. Although the symptom(s) to be treated could be the same, the psychological state and social environment differ for each person with dementia. It is up to the healthcare professional to match the situation of a single person with the most appropriate psychosocial intervention. Unfortunately, directions on how to individualize psychosocial interventions are not provided by most dementia guidelines. Yet, treatment success largely depends on a person­alized approach.

Another difficulty for dementia guidelines is that scientific evidence for the effectiveness of psychosocial interventions in dementia care is growing fast. It could take months to years to develop a high-quality and evidence- based guideline, and by the time it is published new evidence is already available. Dementia guidelines should therefore be updated every few years to keep up with scientific evidence but this is done not that often.

The aim of this post on psychosocial interventions in dementia care is to summarize current knowledge and evidence for the use of psycho­social interventions in dementia care. It is not meant to give a full and systematic overview of all available evidence but to show the broad range of effects that psychosocial interventions could have in the treatment of behavioral, cognitive, and functional symptoms of people with dementia and the treatment of family caregivers.

Problems in dementia care are diverse and dementia-care services and organizations should provide access to a range of psychosocial inter­ventions that can be personalized to individual patients and caregivers. Directions on how to personalize psychosocial interventions and the basic outline of a care plan are given at the end of this post. These direc­tions are useful for professional caregivers as well as family caregivers, and apply to the different settings where people with dementia reside, at home, day care, residential care, or nursing home.

Psychosocial interventions described in dementia literature comprise a great variety of strategies and techniques and are used to manage demen­tia symptoms during the whole course of the disease. Psychosocial treat­ments could start right after diagnosis and be used till the end stage of the disease. Psychosocial interventions are available that help people with dementia to cope after they have received the diagnosis and people are still aware of their memory problems, or teach them how to use memory
aids. When the disease progresses interventions could focus on meeting the needs of the person with dementia regarding social and recreational activities, self-care, and daily structure. At the more severe stages inter­ventions could help to ease behavioral disturbances and stimulate interac­tion with others.

Scientific evidence for the effectiveness of psychosocial interventions in the treatment of dementia symptoms is scarce when applying the “gold standard” of randomized controlled trials (RCTs). The Cochrane Library,2 which includes systematic reviews and meta-analyses about the efficacy and effectiveness of treatments for many diseases and conditions, is con­sidered a reliable source of evidence-based practice and used as such by many healthcare professionals. Systematic reviews for different psychoso­cial interventions in dementia care could be found in this database but for all these interventions evidence is inconclusive.

This lack of evidence is caused by insufficient high-quality research and not because psychosocial interventions are found to be ineffective. The interventions for which some evidence is available show positive results in favor of the psychosocial intervention but more high-quality studies are needed. Negative effects of interventions are seldom reported as side effects, contrary to most pharmacological interventions used in dementia care. Personal privacy and ethical aspects could be an issue when using some psychosocial interventions, like tracking devices or subjective barri­ers for people who wander or certain sensory stimulation techniques.

The effectiveness for specific psychosocial interventions is hard to prove not only because RCTs are difficult to conduct but also because it is increasingly recognized that psychosocial care should be tailor made. Finding no effects or even undesired effects from psychosocial interven­tions could indicate that the patient’s needs and preferences did not match the specific intervention. For instance, some people prefer doing things alone or being more physically active whereas others prefer group activi­ties or like to listen to music. It is often stated in study reports of psycho­social interventions that although no overall group effect was found there seemed to be a subgroup within the study sample who benefited from that specific intervention.

Evidence for the effectiveness of one standard package of psychosocial interventions that can be recommended to all people with dementia will probably never be found. The evidence for tailored psychosocial interven­tions is growing rapidly. During the last decade, promising effects have been described in systematic reviews and papers reporting high-quality randomized controlled trials. An overview of these findings is described here.

Neuropsychiatry of Dementia: Nonpharmacologic Interventions for Inappropriate Behaviors

In the past, it was not believed that persons with dementia retained high­er-level needs or functions. We have since learned, however, that although persons with dementia differ from cognitively intact persons in their ability to articulate and independently meet higher-level needs, these needs are present nonetheless (Cohen-Mansfield and Werner 1995). Evidence shows that a large proportion of dementia-related behavior problems stem from an incongruence between the needs of the person who suffers from demen­tia and the degree to which his or her environment fulfills these needs (Bar­ton, Findlay, and Blake 2005; Cohen-Mansfield and Werner 1995; Palmer et al. 1999). In fact, it is our experience as well as that of other researchers that many behavior problems constitute a response to physical pain or dis­comfort (Cohen-Mansfield et al. 1990; Douzjian, Wilson, and Shultz 1998), feelings of loneliness or isolation (Cohen-Mansfield and Werner 1997), boredom (Buettner and Kolanowski 2003; Ice 2002), or sensory deprivation (Cohen-Mansfield 2000b). Thus, many “problematic behaviors” may rep­resent a cry for help, a result of unmet needs, or an inadequate attempt to fulfill those needs. For example, Hancock, Woods, Challis and Orrell (2006) found that sensory or physical disability (including mobility problems and incontinence) needs, mental health needs, and social needs of persons with dementia in residential care were often unmet and were associated with psychological problems such as anxiety and depression. Due to such find­ings, it is critical that the evaluation and care of unmet needs become the guiding principles of good, domain-specific patient care.

Nonpharmacologic interventions aim to address what we have learned to be the most important etiologic basis of behavioral problems in demen­tia. Similar to the notion of “person-centered care” (Touhy 2004), this approach can be better described as “informed care,” a treatment approach that is based on knowledge of the needs of persons with dementia in gen­eral and the individual in particular. Care can be enhanced by an approach of rehabilitation and restorative care, yet the main focus is that of improv­ing overall well-being and addressing the needs of the older individual with dementia, even when those needs are not obvious or articulated. Admittedly, the implementation of this type of care is more complex than prescribing a psychoactive medication, and the identification of the needs to be fulfilled is more difficult than articulating the specific psychi­atric syndrome from which the patient may be suffering. However, these interventions avoid the potentially harmful side effects that result from pharmacologic treatments (Ballard et al. 2009; Folks 2003; Gill et al. 2009). Also, reducing inappropriate behaviors via sedation with psychoactive medication can potentially rob the person with dementia of the very lim­ited resources he or she has in either expressing or attending to his or her needs (Cohen-Mansfield 2000a; Fisher and Swingen 1997), thereby dimin­ishing the ability of caretakers to detect and address the true underlying need. Most important, nonpharmacologic interventions aim to improve the quality of life of the person with dementia.

In summary, the importance of utilizing a nonpharmacologic approach for inappropriate behaviors associated with dementia is threefold: (a) it aims to address the psychosocial/environmental underlying reasons for the behavior, which have been documented in prior research, thus increas­ing quality of life; (b) it avoids the limitations of pharmacologic interven­tions, namely adverse side effects, drug-drug interactions, and limited efficacy (Cohen-Mansfield et al. 1999); and (c) when medication is effica­cious, it may mask the actual need and reduce the already compromised communication by the older person, thereby limiting the caregiver’s abil­ity to properly care for that person.

Measuring and Interpreting the Efficacy of Nutraceutical Interventions for Age-Related Cognitive Decline

The world’s population is aging rapidly, with the proportion of the population over 60 growing at a rate of around 2% per annum in the developed world (United Nations, 2009). In the most developed regions, 264 million people (21% of the population) were estimated to be 60 years and older in 2009, with this figure projected to increase to around 416 million (33% of the population) by the year 2050 (United Nations, 2009). A major societal health issue for an aging population is not only the greater incidence of neurodegenerative disorders such as Alzheimer’s disease but also the impact of normal age-related cognitive decline. Up to 50% of adults aged 64 and over have reported difficulties with their memory (Reid and MacLullich, 2006). In response to the reality of an aging population, there has been increased research focus in recent years on the development of effective interventions that may ameliorate the declines in cognitive ability.

Age-related deficits in cognitive abilities have been consistently reported across a range of cognitive domains including processing speed, attention, episodic memory, spatial ability, and executive function (Craik, 1994; Hultsch et al., 2002; Park et al., 1996, 2002; Rabbitt and Lowe, 2000; Salthouse, 1996; Schaie, 1996; Verhaeghen and Cerella, 2002; Zelinski and Burnight, 1997). While an overall decline in processing speed may explain some of the age-related variance in cognitive ability (Salthouse, 1996), processing speed alone cannot explain why a range of neuropsychological measures still remain significantly related to age once processing speed is taken into account (Pipingas et al., 2010). Further, there is growing evidence documenting a more rapid decline for certain cognitive functions in comparison to others, a find-ing which suggests that factors in addition to processing speed are also involved in cognitive decline (Buckner, 2004).

Hedden and Gabrieli (2004) differentiate three categories of cognitive decline in normal aging: (1) lifelong declines, including processing speed, working memory, and encoding of information into episodic memory; (2) late-life declines, including well-practiced tasks and those that require previous knowledge such as vocabulary and semantic knowledge; and (3) life-long stability, including autobiographical memory, emotional processing, and implicit memory. Another common distinction is often made between crystallized abilities (e.g., vocabulary and general knowledge), which remain stable until later life versus fluid abilities (e.g., attention, executive function, and memory) that decline from middle adulthood until late old age (Gunstad et al., 2006).

The reason for disproportionate declines across cognitive domains is that certain brain structures are more heavily affected by these processes than others during aging (Buckner, 2004; Grieve et al., 2005; Hedden and Gabrieli, 2004). Cortical volume decreases in the frontostriatal system are most strongly correlated with age-related cognitive decline (Bugg et al., 2006; Hedden and Gabrieli, 2004; Kramer et al., 2006; Schretlen et al., 2000; West, 1996). It has been estimated that decreases in Prefrontal Cortex (PFC) volume occur at a rate of around 5% per decade after the age of 20, in contrast to relatively small volumetric declines in the hippocampus and medial temporal lobe structures, which occur at a rate of around 2%–3% per decade (Hedden and Gabrieli, 2004; Kramer et al., 2006). In the absence of neurodegenerative diseases such as Alzheimer’s (AD), the medial temporal lobes are relatively spared by the aging process (Albert, 1997).

Age-related reductions in the brain’s gray matter are due to a number of factors including neuron apoptosis, neuron shrinkage, and lowered numbers of synapses; whereas reductions in white matter may be attributed in part to large age-related decreases in the length of myelinated axons (Fjell and Walhovd, 2010). During aging the brain suffers accumulative damage due to a number of cellular processes includ-ing reactive oxygen species formation (Halliwell, 1992), chronic inflammation (Sarkar and Fisher, 2006), redox metal accumulation (Connor et al., 1995), and homocysteine accumulation (Kruman et al., 2000). In addition to direct cellular damage, the brain is also indirectly impaired by insults to the cardiovascular system (Pase et al., 2010).

While the unfortunate decline in cognitive ability is ubiquitous, it is also evident that a great deal of variability exists in both the rate and the extent of cognitive decline experienced by individuals as they age (Shammi et al., 1998; Wilson et al., 2002). While some of the variance may be explained by genetic factors (e.g., Hariri et al., 2003; Price and Sisodia, 1998), there is also a great deal of research highlighting the importance of diet and lifestyle during aging. Chronic nutraceutical interventions hold great promise in ameliorating age-related cognitive decline because they simultaneously target multiple cellular mechanisms of cognitive decline. Many natural sub-stances already identified through in vivo as well as clinical studies have been found to have potent anti-oxidant and anti-inflammatory properties as well as being of benefit to the cardiovascular system (Ghosh and Scheepens, 2009; Head, 2009; Kidd, 1999). In order to be able to accurately assess and interpret the clinical efficacy of these natu-ral substances, it is recommended in the following review that highly accurate and specific cognitive tests are needed, together with the creation of normative databases that may be used to interpret clinical data in terms of years of cognitive function recovered.

Traditional ways of measuring cognitive decline in the elderly have involved clinical neuropsychological tests, tests that were often designed for the diagnosis of dementia. Commonly used dementia assessment scales include the mini mental state exam (MMSE; Folstein et al., 1975), the clinical dementia rating scale (Morris, 1997) and the cognitive subtest of the Alzheimer’s disease assessment scale (ADAS-cog; Mohs et al., 1983). Such scales involve structured interviews to determine the presence of dementia, and if present then the severity of dementia symptoms. While these scales may be useful in the diagnosis of dementia, they lack the sensitivity to be able to assess cognitive decline in the normal population, and hence a strong ceiling effect would be expected. Another limitation of these clinical assessment scales is that they primarily assess global cognitive function, as opposed to specific cognitive domains that may be disproportionately affected by the aging process. Further, these tests often rely on pen and paper recording, which lacks the measurement precision associated with modern computerized testing.

For these reasons it is recommended that computerized tests that target specific cognitive abilities and have a high degree of sensitivity to fluctuations in cognitive function be used for testing the efficacy of interventions for age-related cognitive decline, rather than the more traditional dementia assessment scales. The cognitive drug research (CDR; Wesnes et al., 1999) neuropsychological assessment battery has previously been found to be a particularly sensitive measure for the detection of changes to cognitive function associated with chronic nutraceutical and dietary interventions (Ryan et al., 2008; Stough et al., 2008; Wesnes et al., 2000). The computerized mental performance assessment system (COMPASS; Scholey et al., 2010), which was developed at Northumbria University, United Kingdom, has also been found to be a sensitive measure in nutraceutical intervention trials, as has the Cambridge neuropsychological test automated battery (CANTAB; Cambridge Cognition, Cambridge, United Kingdom). Using large normative samples, the CANTAB has been found to be sensitive enough to detect declines in cognitive ability associated with normal aging as well as mild cognitive impairment preceding dementia (Égerházi et al., 2007; Robbins et al., 1994).

More recently our laboratory has developed a neuropsychological assessment battery designed specifically for the assessment of age-related cognitive decline, the Swinburne University computerized cognitive aging battery (SUCCAB; Pipingas et al., 2008, 2010). The SUCCAB is a computerized test battery consisting of nine tasks designed to capture the range of cognitive functions that decline with age: immediate/ delayed word recall, simple reaction time, choice reaction time, immediate/delayed recognition, visual vigilance, n-back working memory, Stroop color-word, spatial working memory, and contextual memory (Pipingas et al., 2010). In preliminary studies from our laboratory, the tasks contained in the SUCCAB have been found to be highly sensitive to age-related cognitive decline (Pipingas et al., 2008).

When reporting the results of timed cognitive tasks times, it is informative to not only state the results of parametric statistical tests and p-values but also report the average millisecond improvement that is observed in the treatment group. Further, if the mean difference in reaction times has also been found to change in the placebo group, then the mean change in the treatment group above and beyond the change observed in the placebo group is the most informative metric.

An example of reported millisecond improvements in SUCCAB tasks that were found to be associated with a chronic intervention is provided by Pipingas et al. (2008). In a randomized, placebo-controlled trial, Pipingas et al. (2008) investigated the cognitive effects associated with 5 weeks supplementation with the Pinus radiata bark extract Enzogenol® in 42 males aged 50–65 years. Significant differences between the treat-ment and placebo groups were found for SUCCAB spatial working memory (SWM) and immediate recognition memory. The average reduction in reaction time for the SWM task in the Enzogenol group was found to be 65 ms, while reaction times for the control group were unchanged. Similarly, for SUCCAB immediate recognition memory, the average reduction in reaction time for the Enzogenol group was found to be 60 ms, while the reaction time in the control group increased by 7 ms (Pipingas et al., 2008).

The same approach can also be used for computerized cognitive tests that are scored according to accuracy (percent correct). An example of improvements to SWM accu-racy associated with a nutraceutical intervention is provided by Stough et al. (2008). In a 90 day randomized placebo-controlled trial, Stough et al. (2008) investigated the cognitive effects of Bacopa monniera in 62 participants aged 18–60 years. Significant differences between the treatment and placebo groups were found for change in CDR SWM accuracy over the 3 month period. The average improvement in accuracy for the SWM task in the Bacopa monniera group was found to be 5.44%, while the aver-age improvement in the control group was found to be 2.3%. If we make the assumption that the average improvement in accuracy for the control group is a measure of improvement due to practice effects, then we can see that there is still a 3.14% improvement observed in the treatment group above and beyond this value.

When baseline normative data are collected in regard to the results of computerized cognitive tests across a wide age range, the regression coefficient of age can be compared with the treatment effect when a nutraceutical intervention is applied.

This approach was recently used in our laboratory, whereby the SUCCAB cognitive battery was administered to 120 participants between the ages of 21 and 86 years (Pipingas et al., 2010). Significant correlations between both accuracy and reaction time measures were found across a wide range of cognitive domains including word recall, recognition memory, SWM, contextual memory, simple and choice reaction time, visual vigilance, n-back working memory, and Stroop reaction times (Pipingas et al., 2010). Regression analysis was used to predict reaction time and accuracy measures as a function of age for each cognitive domain from the SUCCAB. SWM ability was found to display the greatest degree of age-related decline amongst all cognitive measures, followed by contextual memory and immediate recognition tasks.

Here it can be seen that there is a steady linear increase in reaction time and a decrease in task accuracy from the age of 20 years onward. While further normative data for the SUCCAB is required in order to predict this relationship more accurately, this preliminary study nevertheless illustrates the strong relationship between age and cognitive decline.

An illustrative example is the previously mentioned intervention study by Pipingas et al. (2008) using P. radiata bark extract in elderly adults. Average SWM reaction time for the treatment group decreased from 1018 ms at baseline to 953 ms post-treatment. By using the regression equation for SWM reaction time as a function of age established in the SUCCAB normative study (Pipingas et al., 2010), the 65 ms improvement in reaction time can be interpreted as a SWM cognitive age-recovery of approximately 6.5 years.

Another example of this approach was a study by Durga et al. (2007), which investigated the effects of 3 year folic acid supplementation on cognition. By com-paring the regression coefficient of age with the treatment effect, the authors were able to calculate that 3 year supplementation conferred an individual the performance of someone 4.7 years younger for memory, 1.7 years younger for sensorimotor speed, 2.1 years younger for information processing speed, and 1.5 years younger for global cognitive function (Durga et al., 2007). Such interpretations of the data place cognitive change in a more meaningful context, changes that can be directly interpreted in terms of the amelioration of cognitive decline.