Abstract
Background:
Agitation and aggression are common in patients with Alzheimer’s disease and related dementias and pose a significant burden on patients, caregivers, and the healthcare systems. Guidelines recommend personalized behavioral interventions as the first-line treatment; however, these interventions are often underutilized. The Standardizing Care for Neuropsychiatric Symptoms and Quality of Life in Dementia (StaN) study (ClinicalTrials.gov Identifier # NCT0367220) is a multisite randomized controlled trial comparing an Integrated Care Pathway, that includes a sequential pharmacological algorithm and structured behavioral interventions, with treatment-as-usual to treat agitation in dementia in long-term care and inpatient settings.
Objective:
To describe the rationale and design of structured behavioral interventions in the StaN study.
Methods:
Structured behavioral interventions are designed and implemented based on the following considerations: 1) personalization, 2) evidence base, 3) dose and duration, 4) measurement-based care, and 5) environmental factors and feasibility.
Results:
The process to design behavioral interventions for each individual starts with a comprehensive assessment, followed by personalized, evidence-based interventions delivered in a standardized manner with ongoing monitoring of global clinical status. Measurement-based care is used to tailor the interventions and integrate them with pharmacotherapy.
Conclusion:
Individualized behavioral interventions in patients with dementia may be challenging to design and implement. Here we describe a process to design and implement individualized and structured behavioral interventions in the context of a multisite trial in long-term care and inpatient settings. This process can inform the design of behavioral interventions in future trials and in clinical settings for the treatment of agitation in dementia.
INTRODUCTION
Major neurocognitive disorder, commonly called dementia, is a widespread health problem of global concern for the older population [1]. Alzheimer’s disease and related disorders (ADRD) are the most common cause of dementia and account for 60% to 80% of all cases [2]. Behavioral and psychological symptoms of dementia (BPSD) are highly prevalent in ADRD and significantly contribute to illness burden [3]. Epidemiological data show that up to 80% of patients with dementia experience one or more symptoms of BPSD over the course of their illness [4]. BPSD compromise of a group of heterogeneous symptoms that include apathy, depression, wandering, inappropriate social behavior, psychosis, sleep and appetite disturbances, and agitation and aggression [5]. Among these, agitation and aggression cause significant distress to the individuals with dementia and their caregivers and impair their quality of life [3, 6]. They are associated with accelerated disease progression, emergency room visits, hospital admissions, increased use of restraints, and a higher likelihood of institutionalization [3, 8]. Agitation in dementia is estimated to result in a significant additional financial burden ($4.3 billion in the US and £2 billion in the UK per year) [9].
Management of agitation and aggression in dementia commonly involves pharmacotherapy [10]. Psychotropic medications have shown consistent but modest efficacy for treating agitation and aggression in randomized controlled trials, and they are associated with adverse effects [11]. Further, polypharmacy is common among elderly living in the community and in long term care settings leading to an increased incidence of adverse effects such as falls and confusion [12, 13]. Therefore, guidelines recommend using medications to treat behavioral symptoms of dementia only when benefits clearly outweigh the risks [14].
Behavioral interventions are free from the adverse effects associated with pharmacotherapy. Thus, they can be used in milder cases when the benefits of pharmacotherapy may not be worth the associated risks. Guidelines universally recommend behavioral interventions as the first-line treatment for agitation associated with dementia [10]. In practice, however, they are often underutilized, for several reasons. First, these interventions must be individualized to patients’ backgrounds and preferences [15, 16]. Second, the interventions require a well-structured implementation plan based on personal and environmental factors [17]. The right “dose” and duration of the intervention need to be identified [18]. Without a systematic means of monitoring for effects, a delay in onset of action might discourage care providers from utilizing these interventions or lead to early abandonment of trials. Third, patients, families, and formal care providers may have negative perceptions about the efficacy of behavioral interventions [19]. Finally, institutional factors, such as inadequate staff time, lack of training resources, and insufficient funds to support required personnel or materials can also hamper the use of behavioral interventions [18].
We designed and published a sequential medication algorithm as part of an Integrated Care Pathway (ICP) to manage agitation and aggression in ADRD [20]. The Standardizing Care for Neuropsychiatric Symptoms and Quality of Life in Dementia (StaN) study is a multisite randomized controlled trial (ClinicalTrials.gov Identifier: NCT03672201), that is assessing the effectiveness of a structured algorithm of individualized behavioral and pharmacological interventions (i.e., ICP) to treat agitation and aggression in ADRD. In this paper, we describe the design and rationale for behavioral interventions in the StaN study (sequential pharmacological algorithm is published elsewhere [20]). We discuss individualizing structured behavioral interventions; the evidence supporting the use of specific interventions; the determination of the ‘dose’ and duration, and an interdisciplinary implementation plan based on measurement-based care to address existing barriers to their use.
METHODS
StaN study overview
The StaN study enrolled participants with agitation or aggression associated with ADRD from four geriatric inpatient units and five long-term care homes (LTCH) across Canada and randomized them to ICP versus Treatment-as-usual (TAU). Participants in the ICP arm underwent a stepwise intervention comprising: a cleanup phase, during which unnecessary or ineffective medications were discontinued; a structured behavioral interventions phase including a standalone behavioral interventions for the first 3 weeks (the subject of the current article); and a sequential pharmacotherapy phase in combination with behavioral interventions [20]. The TAU consisted of the typical care that the interdisciplinary team provided at each site for agitation related to dementia. No predetermined cleanup phase, non-pharmacological interventions, algorithmic pharmacological interventions were part of TAU. Both groups were assessed by research analysts (blinded to the assignment of ICP vs. TAU) at regular intervals to measure clinical symptoms, medication use, rate of falls, and caregiver burden. Specifically, the effects of behavioral interventions are planned to be analyzed at 3 weeks, that is the end of standalone behavioral intervention phase.
Methods and rationale employed in the design of structured behavioral interventions
The interdisciplinary study team created a structured algorithm of behavioral interventions based on the following five considerations: 1) personalized selection of interventions, 2) evidence base to guide the choice of interventions, 3) dose and duration of interventions, 4) adoption of measurement-based care, and 5) environmental factors and feasibility. These considerations and their impact on the design of behavioral interventions are described in more detail below.
Personalized selection of behavioral interventions
Behavioral interventions work best when they are personalized [16]. There is strong evidence from several Randomized Controlled Trials (RCT) that personalized behavioral interventions can reduce BPSD and their impact on patients and caregivers [21, 22]. Interventions can be personalized based on persons’ previous interests and current levels of cognition and function. In several RCTs, individualization of interventions based on a person’s cognition, function, and interests was shown to reduce agitation [17, 23]. By contrast, non-personalized interventions or those selected without consideration for patients current level of function or previous interests might have less positive effects [17].
The factors that must be addressed in personalization need to be identified systematically. There are several existing approaches or tools for personalization. In the Cohen-Mansfield model, individualization of interventions is based on a person’s sense of identity. This identity is based on biological, psychosocial, and environmental factors that change throughout life [24]. The interventions developed using Montessori’s method are based on a person’s capabilities. Key components of this approach are matching demands to an individual’s capabilities through guided repetition, breaking tasks into smaller components, and introducing them by order of difficulty from easier to more difficult. This approach is continued until competence required for the task develops [17]. The Need-Driven Dementia-Compromised Behavior (NDB) model conceptualizes unmet needs as causing agitation in dementia, and thus interventions are targeted toward unmet needs [25]. According to this model, behavioral symptoms are the result of an interplay between risk factors such as cognitive deficits, physical function, premorbid personality, and precipitating factors such as physical and social environment, physiological and psychological needs. Understanding behavioral symptoms as reflecting a patient’s needs can also guide interventions tailored to address these needs [26]. The Describe, Investigate, Create, Evaluate (DICE) approach is a step-wise method to manage behavioral symptoms in patients with dementia [27]. In step one, the caregiver describes the troublesome behavior, followed by step two that is healthcare provider’s investigation of possible causes of the behavior. In step three, the caregiver and healthcare provider create a treatment plan for addressing the behavior. In the last step, healthcare provider evaluates the implementation of the plan by the caregiver and its efficacy and safety.
The P.I.E.C.E.S framework is designed to optimize care for elderly individuals with complex mental problems using a comprehensive interdisciplinary approach [28]. This approach characterizes the drivers of problematic behaviors across six domains: physical causes (P), intellectual causes (I), emotional causes (E), individual’s capabilities (C), factors related to emotional and physical environment (E), and person’s social life (S) [29]. This framework is used widely to develop individualized behavioral interventions for patients experiencing agitation or aggression with dementia [30, 31]. The PIECES framework is a comprehensive approach and systematically assesses factors in different domains that commonly contribute to agitation and aggression.
Evidence base for behavioral interventions
Several types of behavioral interventions have been shown to be beneficial for BPSD, including agitation, [32] and here we briefly summarize the relevant evidence.
Music therapy: There are two types of music therapy: receptive and active. In receptive music therapy, the person listens to music, and in active music therapy, the person engages in making music [33]. Music therapy has strong evidence for treatment of agitation in dementia from several RCTs [34, 35], large systematic reviews [36, 37], and meta-analyses [37, 38]. Results of these studies have shown that music therapy is effective in reducing agitation and aggression, and improving depression, anxiety, and socialization skills. Like other recreational activities, music therapy can reduce polypharmacy.
Exercise and motor interventions: These interventions involve individualized or group physical activity including walking, ball games, stretching, and dancing. Exercise and motor interventions (walking in particular) have promising evidence for agitation and other BPSD consisting of several RCTs [39, 40], systematic reviews [41, 42], and one meta-analysis [43]. Results of these studies showed that exercise interventions improve cognitive function, mood, functional performance, sleep, and behaviors such as agitation. Further, exercise combined with social interaction, may also improve sleep, and reduce psychotropic medication use.
Sensory stimulation: The focus of these interventions is stimulating primary senses via environmental media (such as lights, tactile surfaces, relaxing sounds, or aroma) using multisensory environments often referred to as a Snoezelen room. Snoezelen rooms are equipped with tactile, visual, olfactory, auditory, and proprioceptive sensory equipment that allows for a multisensory experience. The evidence supporting the use of a Snoezelen room for treating agitation in dementia is mixed. Some reviews concluded that sensory stimulation is effective in improving agitation, while others concluded that there is a lack of sufficient evidence for the efficacy of Snoezelen or multisensory stimulation programs [44, 45]. In several RCTs, the effect of a Snoezelen room is comparable to the effect of recreational activities and music [46, 47].
Social contact: Social contact, i.e., in-person contact with people, animals, or toys, is an integral part of many behavioral interventions for BPSD. However, only a few studies have assessed social contact as the main intervention [48]. In several RCTs, social contact has shown promise for managing BPSD [49–51]. To our knowledge, there are no systematic reviews or meta-analyses focusing solely on social contact as an intervention for BPSD.
Animal-assisted therapy (AAT): Traditional AAT (“pet therapy”) involves either a short visit from a pet or having a pet live with the patient. AAT has been used widely for individuals with dementia and has been shown to reduce agitation and aggression [52]. Recently, robotic animals have been developed as an alternative to live pets [48, 53]. Evidence from several studies [53, 54] supports the positive effects of live or robotic AAT for BPSD in general and agitation specifically. AAT also improves communication and provides pleasure for staff and residents.
Structured purposeful activities: These are individualized activities based on a person’s interests and their level of function. These interventions are wide-ranging, including activities such as simple chores (e.g., folding laundry), reading, games, physical activity, and arts. In several RCTs, structured purposeful activities tailored to a person’s interests were found effective in reducing agitation and other BPSD and improving attention or social interaction [21, 55].
Dose and duration of interventions
Despite numerous trials investigating behavioral interventions for patients with dementia, there are no clear guidelines for selecting dose and duration of these interventions. In a meta-analysis of music therapy, the dose varied from 20 minutes once a week to 30 minutes daily and the duration ranged from 5 weeks to a year [37]. In a meta-analysis for exercise interventions, the average duration of interventions was 23 weeks (ranging from 2 to 112 weeks), 3.6 sessions per week (ranging from 1 to 6 sessions), and 45 minutes per session (ranging from 20 to 150 minutes) [43]. In a systematic review using sensory stimulation, the duration of the multisensory intervention varied from 9 sessions to 18 months with 16–40 minutes per session [44]. Similarly, for purposeful activities, the dose and duration of purposeful activities were variable across studies.
Measurement-based care
Measurement-based care includes a routine assessment of symptoms and side effects using standardized tools, informing a specific action plan towards a clear outcome [56]. This approach enhances communication between care providers and leads to adequate dosing and duration of the treatment [56, 57]. Measurement-based care has been applied to pharmacotherapy in patients with BPSD; however, its use for behavioral intervention is uncommon [20]. Standardized tools such as Cohen Mansfield Agitation Inventory (CMAI) are routinely used for measurement-based care in patients with agitation in clinical trials [58]. Alzheimer Disease Cooperative Study-Clinical Global Impression of Change in agitation (ADCS-CGIC) scale [59] has been used in several trials [60, 61] targeting agitation in dementia. Originally developed to assess clinically significant changes in AD clinical trials, this scale has been adapted and validated [62] to assess change in agitation. ADCS-CGIC is easy to use and captures the global change in a patient’s clinical condition, making it appropriate for use in routine clinical settings.
Environmental factors and feasibility
Environmental factors present unique opportunities and challenges that must be taken into consideration while designing behavioral interventions for patients with BPSD. Efficiency and safety-related considerations may be different in an acute inpatient unit as compared to a LTCH and influence the choice of the intervention [14, 63–65]. Settings may also differ in terms of the number of available personnel and equipment for implementing behavioral interventions [65]. In addition, staff characteristics may influence the choice and feasibility of personalized interventions [18]. Thus, involvement of clinical leadership needs to be considered early on to facilitate staff education and to address barriers to implementation of behavioral interventions [66]. A well-organized interdisciplinary team focused on providing personalized care, and a system for ongoing identification and management of any potential barriers must be part of the process of designing and actualizing personalized behavioral interventions.
The study team deliberated considerations described above via in-person meetings and teleconferences. Based on these deliberations, the process described below was adopted to design and implement individualized structured behavioral interventions based on consensus.
RESULTS
Based on the five considerations mentioned above we created a structured algorithm to design and implement behavioral interventions for each study participant randomized to the ICP arm. This structured algorithm starts with a comprehensive assessment that leads to the development of an individualized behavioral plan. The individualized behavioral plan is carefully implemented using ongoing reviews of symptoms and critical incidents. Measurement-based care with clear decision points is used to guide the interventions. The personalized behavioral interventions are fully integrated with the pharmacological algorithm, as needed. Participants complete the individualized behavior plan based on the algorithm and the behavioral plan is fully transitioned to the clinical team. Key components of this structured algorithm are described below.
Comprehensive assessment and creation of an individualized behavioral plan
Target behaviors are identified through interviewing the participant, their family members, staff caregivers, and the clinical team. The P.I.E.C.E.S. framework [28] is then used to complete a multi-domain comprehensive assessment to detect possible underlying triggers or causes of behaviors. We chose the P.I.E.C.E.S. framework as it covers individual factors (physical, intellectual, emotional, capabilities), as well as environmental, social, and cultural factors that may be contributing towards target behaviors. Moreover, this framework is suitable for both inpatient and long-term care settings and can be applied across the full spectrum of the cognitive illness [29–31]. This assessment is used to 1) develop an individualized behavioral plan, 2) provide the multidisciplinary team with a shared understanding of the individual and the target behaviors, and 3) promote effective interprofessional communication and collaboration.
We created a detailed checklist (Supplementary Material 1) that prompts assessment of multiple factors that may be driving the agitation or aggression. First, physical causes such as pain, constipation, difficulty voiding, excessive heat or cold, thirst or hunger, ill-fitting clothing, possibility of drug or alcohol withdrawal, physical disabilities such as decreased vision, impaired hearing, and difficulties with mobility/ambulation are assessed. Second, intellectual causes such as memory deficits, language impairment, difficulty recognizing objects and people, inability to perform routine tasks such as dressing, and poor insight are assessed. Third, emotional causes such as major losses throughout life, recent losses, fears or phobias, significant trauma, loneliness, and boredom are assessed. Fourth, individual’s capabilities such as their ability to perform basic and instrumental activities of daily living, their likes and dislikes, and best and worst times of the day are assessed. Fifth, environmental factors such as recent changes in place of living, presence of noise, crowding and clutter, and lack of meaningful contact/stimuli in the environment are assessed. Sixth, social and cultural factors such as individual’s life story, their accomplishments and interests (education, work, travel, and hobbies), their support system (family, friends, and neighbors) and cultural heritage, spirituality or religion are assessed.
Further, significant behavioral incidents are recorded using an ABC checklist that has three components: Antecedents (A), which refers to patient’s status prior to the incident; Behaviors (B), which records the type of behavior and its severity; and Consequence (C), which refers to the immediate response of the clinical team (Supplementary Material 2) [67].
The PIECES checklist and the ABC checklists are discussed in an interdisciplinary team meeting to develop a comprehensive formulation of the patient’s clinical status and the target behaviors. All evidenced based behavioral interventions described above (music therapy, exercise and motor interventions, sensory stimulation, social contact, animal-assisted therapy, and purposeful activities) are considered as potential candidates for each participant. The suitability of particular behavioral interventions in a participant is discussed based on factors such as individual’s interests, life history, severity of cognitive decline, mobility, current environmental and social factors, unmet needs, and a behavioral plan consisting of interventions appropriate for the individual is formulated. The selected interventions are qualitatively graded according to their likelihood of success based on personal and environmental factors and evidence for their efficacy (Table 1). If needed, selected interventions are modified to match the participant’s interests and capabilities, and environmental factors to maximize feasibility.
Key aspects of behavioral interventions based on the review of current evidence
RCT, randomized controlled trial; SMD, standardized mean difference; CI, confidence interval; CMAI, Cohen–Mansfield Agitation Inventory; BEHAVE-AD, Behavioral Pathology in Alzheimer’s Disease Rating Scale; NPI, Neuropsychiatric Inventory; BGP, Behavioral Rating Scale in Geriatric Patients; MOSES, Multidimensional Observation Scale for Elderly Subjects; ES, effect size.
As an example, an 83-year-old female participant diagnosed with mild to moderate AD living in LTCH presented with behaviors of exit-seeking, yelling, swearing, and crying. PIECES assessment was done. No physical (P) trigger for the behaviors was found. In the intellectual domain (I), she had short-term memory loss that was contributing to behaviors. Regarding emotions (E), she felt bored and lonely most of the time, which was a trigger for agitation. She felt worse in the afternoon when she would pack her stuff as if she was returning to her condominium. From capabilities (C) perspective, she needed supervision with bathing, and regular reminders to change outfits. She also needed company and support while going out as she was known to get lost which had been the main reason for LTCH placement. She loved socializing, talking about news and books, listening to music, and physical activity such as walks. Regarding environmental factors (E), recent admission to LTCH resulted in loss of sense of independence and staying indoors for most of the time was identified as a trigger for agitation. From social perspective (S), she was an accomplished person having lived most of her life independently. Before her move to LTCH, she used to visit her friends regularly and go on walks every day. Based on the above PIECES assessment, the primary triggers for agitation were identified as recent move to LTCH, and feelings of loss of independence, being trapped in the LTCH, boredom, and loneliness. The individual was previously used to independent living, had several interests (such as music, reading books, and going for walks) and was striving for more activity and social contact. Her behavioral plan included social contact and physical activity (in the form of scheduled accompanied outdoor walks) and music therapy. Initial intervention was strolls in LTCH grounds with a companion 3–4 times per week. Subsequently listening to music while walking or sitting in the LTCH garden was included as well. The research team had a meeting with the LTCH recreational team and advised them to actively encourage participant’s involvement in social activities at the LTCH such as game nights and live concerts.
Implementation of the individualized behavioral plan
The interventions included in the behavioral plan are introduced in a phased manner, starting with one or two preferred interventions depending on personal and environmental factors described above. The decisions about whether to continue the current interventions or to switch to different interventions are made based on acceptance or tolerability of the intervention, and the standardized assessment of clinical response. In general, the selected interventions are delivered daily, and the frequency is modified based on factors such as severity and frequency of behaviors or feasibility depending on the setting. The research team collaborates with the clinical team for the ongoing implementation of the behavioral plan and to collect information about acceptance, tolerability, and success of the particular intervention. The interventions are continued if they are accepted or tolerated well and are partially or fully successful in treating the target behaviors. If the intervention is not accepted or tolerated or does not result in at least partial response, they are switched to alternative interventions during the 3-week period. So, each participant can try one or more interventions, as needed, depending on their unique situation. The Clinical Global Impression of Change scale (CGIC) is completed each week to assess the participant’s response to the current interventions. The CGIC rating (1 to 7) is determined by interviewing the patient, caregiver, and the clinical team, and by reviewing the clinical charts. If the CGIC rating is <3 (i.e., moderate or marked improvement), the behavioral intervention is considered successful and is continued throughout the study. If the CGIC rating is ≥3 (i.e., minimal improvement or worsening) after one week of starting the intervention or if the intervention is poorly tolerated or accepted, then the interventions are modified or switched.
Integration with the pharmacological algorithm
Behavioral interventions are integrated with the pharmacological algorithm throughout the study. Standalone Structured behavioral interventions are started as soon as the participant is assigned to the ICP arm of the study and are continued as standalone or main interventions for three weeks. At the end of three weeks if there is at least a moderate response (CGIC rating <3), then the behavioral interventions continue without medications; otherwise, the pharmacotherapy phase starts while the behavioral interventions continue as described above. During the pharmacotherapy phase, again the participant’s response to the integrated treatment is assessed using the CGIC at predetermined time points.
Communication and collaboration with the clinical team
Recommendations of the research team regarding the behavioral interventions for each participant are communicated to the clinical team every week. The research team members deliver the interventions in collaboration with frontline clinical staff and receive information from the clinical team about the participants and their response to the behavioral interventions. Ongoing modifications to the design and delivery of the behavioral interventions are done through continuous collaboration between the research and clinical team members. In response to the COVID-19 pandemic, the study procedures were changed to virtual, and the design and implementation of the behavioral plans continued through virtual communication with the clinical team using videoconferencing, phone, or email. Similarly, assessments were completed remotely in collaboration with the clinical team, and delivery of the behavioral interventions was delegated to the clinical teams with ongoing monitoring by the research team.
DISCUSSION
To our knowledge, this paper is the first to systematically describe the key elements in designing individualized behavioral interventions for patients with BPSD in the context of a multisite trial in long-term care and psychiatric inpatient settings. We describe a process for the selection of individualized evidence-based behavioral interventions for each participant informed by personal and environmental factors. We also describe a measurement-based care model guiding the ‘dose’ and duration of behavioral interventions and their integration with pharmacotherapy. Finally, we describe the implementation of these individualized behavioral plans through collaboration between the research and clinical teams. This approach may be useful for other studies of behavioral interventions and pharmacotherapy in long-term care and inpatient settings; it may also guide the design and implementation of behavioral interventions in clinical settings.
Some previous RCTs of pharmacotherapy for BPSD have included behavioral interventions for all participants [62, 68]. Citalopram for Agitation in Alzheimer’s Disease (CitAD) [69] is an example of a large RCT that investigated the effect of a psychosocial intervention combined with medication (citalopram or placebo) in patients with BPSD. In the CitAD trial, all participants received three weeks of psychosocial intervention. In the placebo group, there was a significant improvement in participants’ agitation after week 3 that remained stable for the rest of the study [70]. Escitalopram for agitation in Alzheimer’s disease (S-CitAD) is an ongoing trial that has incorporated a 3-week psychosocial run-in period, informed by CitAD, and psychosocial intervention is provided for the duration of the study, in conjunction with the study drug (escitalopram or placebo) [71].
The model described in the current paper can help with the integration of behavioral and pharmacological interventions in treatment trials for agitation and other BPSD in long-term care and inpatient settings. The use of an initial behavioral intervention phase has several potential benefits. First, it may facilitate optimization of behavioral interventions and help to select participants who are appropriate for pharmacotherapy. Second, a structured behavioral intervention phase helps with the medication wash-out required before randomizing participants to investigational agents. Finally, behavioral intervention for all participants helps the study to meet its ethical obligation to provide the standard of care to all participants regardless of their assignment to active or placebo pharmacotherapy in the study.
Our process of designing behavioral interventions includes three key components: comprehensive assessment, individualization of interventions, and measurement-based care. We use tools (PIECES and ABC checklist) with a variety of foci (e.g., cognition, function, capabilities, interests) to collect comprehensive information about the participant and their target behaviors. This systematic approach provides an in-depth understanding of the potential causes and contributors to agitation in the context of a person’s unique identity and environment [72]. The comprehensive assessment is also essential for the individualization of the interventions. Individualization has been shown to increase participation and engagement and prevent early abandonment of the interventions [23]. Also, individualized interventions may have a greater effect on reducing agitation and behavioral symptoms than non-individualized ones [23, 74]. Measurement-based care determines key decisions regarding the continuation or discontinuation of specific interventions. These decisions lead to appropriate discontinuation of ineffective interventions and preservation of effective interventions. Despite strong evidence supporting the use of behavioral interventions for BPSD, they can be difficult to implement and sustain [15]. While a baseline comprehensive assessment and individualization of behavioral interventions may improve BPSD [23, 76], systematic tailoring of interventions based on a measurement-based approach was not tested in previous studies. We believe that this unique model will enhance the impact of behavioral interventions and will improve their uptake in clinical settings. Partial or non-responders to the behavioral interventions receive modified individualized behavioral interventions combined with pharmacotherapy guided by an algorithm that specifies medications, with their dosages and duration based on clear decision time points [20]. Thus, results of the StaN study will provide information on factors that impact response to behavioral interventions, including the effects of the combination of behavioral and pharmacological intervention.
Implementation of the structured process involves several challenges. It may be more challenging to implement behavioral interventions than pharmacotherapy in a research study [18]. These challenges may be related to inherent nature of behavioral interventions, for example it may be harder to standardize the dose and delivery of behavioral interventions due to factors such as variations related to availability of resources across different facilities, and the preferences and training of the clinical staff [18]. Competing priorities for the research and clinical teams may also impact the delivery of behavioral interventions and measurement-based care. The model described here helps to standardize the process of design and delivery of behavioral interventions and may address some of these challenges. Further, clinical staff may be disinclined to implement new processes or change routines due to reasons such as need for extra time commitment to carry out assessments required for implementation of measurement-based care. Workload, beliefs and attitudes regarding behavioral interventions, and lack of training may be other challenges [18, 19]. Some of these challenges can be addressed by using implementation science techniques [77] identifying dedicated champions who will take on the implementation process and by providing necessary training and education to the frontline clinical staff. Some of these resources may be available through Alzheimer’s Association or other professional organizations depending on a particular jurisdiction [78, 79]. Another challenge is making decisions about duration of specific interventions. We chose three-week interval for medication clean up and standalone behavioral intervention phase and chose to start pharmacotherapy if the behavioral interventions were not successful during this interval, based on practical considerations. Patients may be at risk of worsening behaviors and other adverse outcomes due to untreated agitation or aggression if pharmacotherapy is not started in time. We implemented this standardized approach while maintaining flexibility for individual cases, as in certain situations pharmacotherapy may have to be started earlier than three weeks due to clinical or safety concerns, and in others, patients or their substitute decision makers may defer starting pharmacotherapy in favor of continuation of behavioral interventions. Future studies will have to make their own decisions regarding the duration of standalone behavioral interventions based on the nature of the study population and the nature of interventions. Finally, sustaining the behavioral interventions during transitions of care can be challenging [80]. Examples of transition include changes in setting (e.g., discharge from the inpatient unit back to home) or personnel (e.g., transfer of responsibility from the research team to the clinical team). It is important to build transition of care into the clinical workflow from the very beginning, such as a formal process of communication that facilitates the handover of information between the research or clinical team members during transitions.
Conclusion
Individualized behavioral interventions are recommended as the first-line treatment for patients with agitation or aggression associated with dementia; however, they are underutilized and are challenging to design and implement. In the current paper, we describe the process used in the StaN study to design and implement structured behavioral interventions in the context of a multisite trial in long-term care and inpatient settings. This process takes into consideration personal factors, evidence for a given intervention and environmental factors through a comprehensive assessment of the person and their environment. Further, the interventions are implemented in a structured manner using a measurement-based approach in collaboration with the clinical team. Finally, the behavioral interventions are integrated with pharmacotherapy and standardized communication processes are followed to facilitate the transition of care. This standardized process can inform the design and incorporation of behavioral interventions in future studies and in clinical settings for the treatment of agitation in dementia.
Footnotes
ACKNOWLEDGMENTS
The Standardizing Care for Neuropsychiatric Symptoms and Quality of Life in Dementia (StaN) study is funded by Brain Canada Foundation through the Canada Brain Research Fund, with the financial support of Health Canada and the Centre for Addiction and Mental Health, Additional support to StaN is provided by the Canadian Centre for Aging and Brain Health Innovation.
