Strategies to Engage Older Adults with Behavioral and Psychological Symptoms of Dementia in Exercise: A Multiple Case Study

Abstract

Alzheimer’s disease (AD) affects 5.2 million Americans in 2016 and this number will increase to 14.7 million by 2050. Unfortunately, AD cannot yet be prevented, slowed, or cured. Aerobic exercise is potentially therapeutic for Alzheimer’s disease (AD) as it favorably affects brain structure and function. A critical determinant in testing its effectiveness is optimal adherence to the exercise prescription. AD symptoms, particularly behavioral and psychological symptoms of dementia (BPSD), however, make exercise delivery and adherence challenging. The purpose of this paper is to discuss the strategies used for managing BPSD during exercise using a multiple case study design. Three cases with BPSD that met the eligibility criteria were selected from the FIT-AD Trial. The FIT-AD Trial is an ongoing randomized controlled trial that tests the effects of a 6-month moderate intensity aerobic exercise on cognition and hippocampal volume in community-dwelling persons with AD. The cases included: a 77-year-old woman with mild AD, a 79-year-old woman with mild AD, and a 69-year-old man with moderate AD. The participants exhibited anxiety, repetitive questioning, irritability, hostility, extreme distractedness, resistance to instruction, and hoarding during exercise. The symptoms were effectively addressed using person-centered strategies such as routine schedules, reassurance, timely communications, distraction, small rewards, rephrasing, immediate answers, selective withdrawal from conversations that were hostile, step by step instructions, building on rapport and relationship with staff, and affirmation. They achieved 83.3%–100% attendance and 52.8%–81.7% session adherence, respectively. In conclusion, individualized strategies for BPSD are necessary to ensure optimal exercise adherence.

Introduction

Alzheimer’s disease (AD) affects 5.2 million Americans in 2016, and this number will increase to 14.7 million by 2050 unless effective treatments are established¹. Currently, AD cannot be slowed, prevented, or cured; however, aerobic exercise interventions have shown increasing promise for improving or maintaining cognition and functional status in persons with AD^2,3. Recent advances in mechanistic studies of aerobic exercise in AD suggest that the therapeutic role that aerobic exercise plays in stabilizing or improving cognition in AD may be grounded in its ability to induce favorable changes in cerebral structure and function^4–7.

Studies further postulate that mere participation in aerobic exercise is probably insufficient to generate cognitive benefits^8,9. One theory suggests that the delivered dose of aerobic exercise has to be sufficient to induce enhanced aerobic fitness in order to produce cognitive benefits^6,9,10. Testing of such a theory calls for accurate measurement of exercise adherence to determine the actual-delivered aerobic exercise dose. In addition, measuring exercise adherence is paramount for determining the effectiveness of aerobic exercise interventions on altering AD progression. Last, exercise adherence is essential for future studies to establish the minimally effective dose of aerobic exercise to improve aerobic fitness and cognition in persons with AD, and to elucidate the dose-response relationships between aerobic exercise and cognition in AD. Unfortunately, exercise adherence has not been routinely assessed in existing exercise studies, and low adherence has been common among studies that did measure exercise adherence^8,11.

Exercise adherence for persons with AD can be very challenging given their age, medical comorbidities, and unique AD symptom combinations, particularly behavioral and psychological symptoms of dementia (BPSD). BPSD refers to a range of non-cognitive symptoms exhibited by persons with AD that are unrelated to their cognitive impairment^12,13. Behavioral symptoms are commonly divided into four subtypes: physically aggressive behaviors, such as hitting, kicking or biting; physically nonaggressive behaviors, such as pacing or inappropriately handling objects; verbal aggression, such as cursing; and verbally nonaggressive agitation, such as repetition of words or sentences. Psychological symptoms include delusions, hallucinations, depressive symptoms, apathy, anxiety, irritability, and disinhibition^13,14. Up to 90% of persons with AD show some BPSD during the course of their disease^12,15. If not properly and effectively managed, BPSD can be a significant source of distress for persons with AD and their caregivers. Currently, non-pharmacological interventions are established as the first-line treatment for BPSD.

Despite the prevalence and treatments for BPSD, little is known about the common BPSD manifested by persons with AD during aerobic exercise interventions and strategies to address them to ensure exercise adherence. Hence, the purpose of this multiple case study analysis was to discuss the strategies used for managing BPSD that occurred during exercise delivery.

Methods

Design

A multiple case study approach was used for this paper¹⁶. Three cases with BPSD that met the eligibility criteria were selected from the FIT-AD Trial. The FIT-AD Trial is an ongoing randomized controlled trial that tests the effects of a 6-month moderate intensity aerobic exercise intervention on cognition and hippocampal volume in community-dwelling persons with mild-to-moderate AD (National Institute on Aging 1R01AG043392–01A1). Participants are randomized to either moderate intensity cycling or low intensity stretching, 3 times a week for 6 months on a 2:1 allocation ratio, and followed for another 6 months. Cognition is assessed at baseline, 3, 6, 9, and 12 months, and hippocampal volume is measured at baseline, 6, and 12 months. Data collections occur at the university’s Clinical and Translational Science Institute¹⁷. The FIT-AD Trial was approved by the university’s Institutional Review Board.

This paper addresses the following research questions from each case:

1. What were the predominant BPSD that a participant manifested and were challenging to exercise adherence?

2. What strategies did the exercise interventionist find helpful to engage the participant in exercise?

3. What was the adherence rate to the prescribed exercise dose?

Case Selection

The cases were selected using the following inclusion criteria: 1) completed the 6-month exercise program; and 2) exhibited BPSD that made exercise adherence challenging. Three participants met the eligibility criteria and were included: a 77-year-old woman, a 79-year-old woman, and a 69-year-old man (see Table 1).

Table 1.

Demographics, Adherence, and Behavioral and Psychological Symptoms of Dementia

	Case 1	Case 2	Case 3
Demographics
Age	77	79	69
Gender	female	female	male
Race	Caucasian	Caucasian	Caucasian
Education	16	20	14
Living status	Single-family home, with spouse	Single-family home, with daughter and daughter’s family	Single-family home, with spouse
Baseline MMSE	21	19	15
Exercise adherence
Attendance	83.3%	100%	98.6%
Session adherence	58.3%	52.8%	81.7%
BPSD severity
Baseline	7	3	5
3 months	5	2	7
6 months	14	3	5
BPSD distress
Baseline	4	3	9
3 months	8	2	12
6 months	11	4	11

Exercise Program

The exercise sessions occurred at a local YMCA gym or in the lounge of a retirement community, whichever was closer to the participant’s residence. Transportation was provided to and from the exercise site for participants by the study staff.

Both the aerobic exercise intervention and stretching attention control are 3 times a week for 6 months. The aerobic exercise intervention was moderate intensity cycling on a recumbent stationary cycle. Moderate intensity was individualized as 50–75% of heart rate reserve or a Rating of Perceived Exertion (RPE) score of 12–14. The exercise dose was gradually increased over time from a low intensity of 50% of heart rate reserve or RPE 9 for 20 minutes in the first session to 65–75% of heart rate reserve and/or RPE 12–14 for 50 minutes a session by alternatively increasing intensity by 5% of heart rate reserve (or 1 on RPE) and 5-minute in duration. It took about 6 weeks for participants to reach the target prescribed dose. The attention control intervention was primarily seated range-of-motion exercises and stretching at low intensity. The session duration was increased over time to match to the session duration of cycling.

The exercise interventionist supervised 1–3 participants in a session. The exercise interventionist was trained using the FIT-AD Trial staff training plan and study operations manual developed by the investigators using a train-the-trainer approach¹⁷. The interventionist received the Initial On-Board Training from the principal investigator and exercise physiologist co-investigator. The Initial On-Board Training includes: 1) 2–3-hour self-study of topics on AD, person-centered dementia care, and communication with persons with AD and those with BPSD; 2) 2-hour in-person workshop on the self-study topics; 3) 2–3-hour self-study of the study protocol and operations manual; 4) 4-hour in-person workshop of the study protocol and operations manual; and 5) 1–4-week on-site field training until competence was demonstrated. Once the exercise interventionist completed the Initial On-Board Training, she provides ongoing training to other staff under the supervision of the investigators and participates in ongoing study quality improvements, which provide them the opportunity to identify their own gaps in knowledge and skills. Subsequent training of the interventionist are provided on an ongoing basis via emails, phone calls, weekly staff meetings, bi-monthly investigator meetings, treatment fidelity checks by the investigators, development of study presentations and publications, and semi-annual booster training.

Person-centered care was used as the guiding principle to exercise participation where we consider the participants’ and their families’ preferences in setting up the exercise sessions and make decisions during exercise delivery. Under this principle, three common strategies for preventing the occurrence of BPSD and reducing the impact of BPSD were used across all participants. First, the interventionist interviewed the family caregivers for BPSD triggers. BPSD triggers identified by family caregivers and staff were minimized by maintaining a consistent schedule, calling ahead for anticipated pickup delays, trying to maintain a calm demeanor, and answering questions immediately. Second, participant’s background information, including personality, family, social, and medical histories, was shared among the staff involved in exercise delivery. We spent considerable effort in getting to know the participants and educating each staff about the participants. Last, we tried to minimize staff stress by adjusting schedules to accommodate slow movers (giving them more time) and anxious participants (dropping them off first and picking them up last to reduce their commute time). This allowed staff to remain calm when interacting with participants and prevent aggravating BPSD.

Data Collection and Analysis

Exercise adherence was calculated from the exercise case report of each exercise session. The case reports were used by the exercise interventionist to document the exercise prescription for a given session and the actual-delivered exercise (e.g., assessing and recording a participant’s heart rate and RPE every 5 minutes as well as blood pressure every 10–15 minutes). Data on AD stage, BPSD severity, and caregiver distress were obtained by trained data collecting research assistants who were undergraduate students and were blinded to participant’s group assignment. The data collecting research assistants only interacted with participants for data collection¹⁷.

Adherence was operationalized in two ways: 1) attendance which was calculated as the percent of exercise sessions attended divided by the 72 prescribed exercise sessions; and 2) session adherence which defined the percent of attended exercise sessions where the participants met the prescribed session intensity and duration. Strategies for engaging participants in exercise were identified qualitatively through semi-structured questionnaires with the exercise interventionist.

AD stage was determined using the Mini-Mental State Examination (MMSE)¹⁸. MMSE assesses orientation, memory, recall, language, and visuospatial function, and is scored 0–30. A MMSE score of 19–26 indicates mild stage, and a score of 15–18 indicates moderate stage. The MMSE was administered at baseline by data collecting research assistants.

BPSD was measured using the Neuropsychiatric Inventory Questionnaire - Caregiver (NPI-Q)¹⁹ which was administered to the caregiver by the data collecting research assistants at baseline, 3 months, and 6 months. NPI-Q assesses the presence and severity of 12 symptoms and their impact on the caregiver distress. Each symptom is given a rating of 0–3 for severity and 0–5 for caregiver distress, resulting in a total score of 0–36 for BPSD severity and 0–60 for caregiver distress (higher scores implying more severe BPSD and more caregiver distress). The NPI-Q has an interscale correlation of 0.91 and 0.92 for severity and distress, respectively¹⁹.

Data on the strategies used to address BPSD were collected using two methods. First, the exercise interventionist documented BPSD and symptom manifestations, along with strategies used to address them after each exercise session on the session case report form. Second, the interventionist completed a semi-structured questionnaire at 6 months when a participant completed the assigned exercise program. The questionnaires asked the interventionist to reflect on the participant’s overall exercise participation, exhibition of BPSD, and strategies used to address BPSD over the course of the training program.

Results

Case 1. Participant 1 was a 77-year-old Caucasian woman with 16 years of education, living in a single-family home with her husband, her primary caregiver. Her past medical history includes osteoporosis, hypothyroidism, and gastroesophageal regurgitation disorder. She has been diagnosed with AD for 3 years. Her medications included memantine, 10 mg, twice a day, galantamine hydrobromide, 16mg, once a day, and levothyroxine, .05mg, once a day. Her baseline MMSE score was 21, indicating mild stage AD. The participant’s NPI-Q severity scores were 7, 5, and 14 at baseline, 3 months, and 6 months, respectively. Her NPI-Q caregiver distress scores were 4, 8, and 11 at baseline, 3 months, and 6 months, respectively.

Her primary symptoms during exercise were anxiety and asking repetitive questions, stemming from separation from her husband. She routinely informed staff that she was done and ready to go home. This cyclical pattern of anxiety leading to repetitive questioning continued unless intervened upon. The common strategies of consistent schedules and advanced communications about scheduling changes were only effective for minutes because she would forget the answers to her questions and feel anxious again. The interventionist would then try to initiate conversations (e.g., talking about her grandson) with the participant to divert her from the repetitive questioning. This was successful at times, but did require the interventionist to be fairly knowledgeable about the participant’s background (e.g., knowing her family history, past experiences, and interests). The final strategy was to address each repetitive question as if it was the first time asked, and answer with patience (Table 2).

Table 2.

Successful Strategies for Addressing Behavioral and Psychological Symptoms of Dementia during Exercise

Symptom	Strategies
Anxiety	Routine schedules, reassurance, call in advance for session changes or delays, distraction
Repetitive questioning	Distraction, small rewards, answer immediately, answer as if new
Irritability	Distraction, small rewards
Hostility	Selective withdraw from conversations that were hostile
Extreme distractedness	Step by step instructions, supervision during instructions
Resistance to instruction	Change wording (“it’s part of the rules”), drop the topic of conversation (if not imperative)
Hoarding	Ask for the item back, build on rapport and relationship with staff

The participant attended 60 of the 72 assigned exercise sessions (83.3% attendance). Her session adherence was 58.3%.

Case 2. Participant 2 was a 79-year old-Caucasian woman with 20 years of education, living at home with daughter, her primary caregiver, and her daughter’s family. Her past medical history included hypercholesterolemia and possible diabetes. She has been diagnosed with AD for four years. Her medications included donepezil, 10 mg, once a day and ibandronate, 150mg, once a day. Her baseline MMSE score was 19, indicating moderate stage AD. The participant’s NPI-Q severity scores were 3, 2, and 3 at baseline, 3 months, and 6 months, respectively. Her NPI-Q caregiver distress scores were 3, 2, and 4 at baseline, 3 months, and 6 months, respectively.

The participant exhibited repetitive questioning, irritability, hostility, and rudeness which were often linked with each other. Her repetitive questions typically centered on her dislike of the exercise program and the interventionist, although she gave verbal consent to participate in each exercise session. For this symptom, previously-described strategies did not work. Providing explanations to her questions sometimes appeared to irritate the participant further and spurred more questioning. Distraction and disengaging from conversation were effective on these occasions. Distraction was achieved by either changing the topic of conversation to one the participant enjoyed or offering a small reward such as chocolate (Table 2).

Despite her symptoms and verbalized dislike of the exercise program, the participant attended all 72 exercise sessions (100% attendance). Her session adherence was 52.8%.

Case 3. Participant 3 was a 69-year-old Caucasian man with 14 years of education, living in a single-family home with his wife, his primary caregiver. His past medical history included hypercholesterolemia, hypertension, sleep apnea, and cardiac bypass surgery. He has been diagnosed with AD for 4 years. His medications included donepezil, 10 mg, once a day, metoprolol, 25 mg, once a day, celexa, 20mg, once a day, irbesartan, 150 mg, once a day, atorvastatin, 80 mg, once a day, and mirtazapine, 7.5mg, once a day. His baseline MMSE score was 15, indicating moderate stage AD. The participant’s NPI-Q severity scores were 5, 7, and 5 at baseline, 3 months, and 6 months, respectively. His NPI-Q caregiver distress scores were 9, 12, and 11 at baseline, 3 months, and 6 months, respectively.

The participant’s BPSD included extreme distractedness, resistance to instruction, and hoarding. His extreme distractedness was exemplified by forgetting what to do in the middle of an action, such as whether to take his shoes off or put them on while holding one shoe. To address this issue, tasks were broken down into single steps, with directions given for each step to ensure he did not get “stuck” and distract other participants from their exercises. While this strategy worked well for distractedness, it conflicted with his lack of interest in following instructions. Multiple reminders had to be given to move him through an exercise session, and on occasion, he’d verbalize his dislike for “being instructed around.” However, he respected authority per family caregiver, and did not want anyone to get in trouble for his behavior. Staff capitalized on this to diffuse situations by saying “I know you do not like to be told, but it is part of the rules I need to follow when we work together.” This strategy worked well, as the participant complied almost immediately. The participant’s hoarding behavior occurred following exercise, while walking around to stretch his legs. He often collected things from other people, typically water bottles, and held them in his lap. We suspected that this behavior might actually be an extension of his profession as a hockey coach. He might just be picking up equipment (in this case, the water bottles) after practice. He also consistently attempted to take the staff’s stethoscope “for repairs.” A very effective strategy was to say, “Oops! I think you accidentally took my item by mistake!” and he would return it almost immediately (Table 2).

The participant attended 71 of the 72 assigned exercise sessions (98.6% attendance). His session adherence was 81.7%.

Discussion

This multiple case study analysis examined the BPSD that occurred during exercise sessions and the strategies used for managing the BPSD to engage participants with mild to moderate AD in exercise. Based on these three cases, we found that BPSD is common during the delivery of exercise, and can be effectively managed to improve exercise adherence. The symptoms exhibited by our participants included anxiety, repetitive questioning, irritability, hostility, extreme distractedness, and hoarding, and were unique to each individual. While some common strategies must be in place for all persons with AD, such as identifying and removing BPSD triggers, collecting background information and getting to know the participants, and educating staff about taking care of self to maintain calm and professional, the strategies need to be adapted and modified for each participant with AD. What works for one person might not have the same effect on another person and what worked yesterday might not work today; thus, a person-centered approach is needed.

Our findings are consistent with the current literature. BPSD is increasingly recognized as a result of neurochemical disturbances caused by AD pathological changes. Depending on the exact brain regions and neurotransmitters affected by AD pathology, persons with AD will show variable presence and severity of BPSD. In addition, BPSD likely results from the unmet needs that persons with AD can no longer communicate. Hence, an individualized, person-centered, non-pharmacological approach to care is paramount to successfully engage persons with AD in aerobic exercise interventions. Two of our participants demonstrated repetitive questioning behavior, yet addressing the questions only worked for one participant who was anxious, not the other one who was irritable and hostile. Although the latter participant verbalized her dislikes of the program and the staff, she continued to attend her exercise sessions. Distractions, small rewards, intentional conversations, and selective nonresponse to her negative comments worked well in these circumstances. In the third case, we combined detailed instructions with an explanation that these are rules which we must follow to balance the participant’s resistance and distractedness behaviors.

Overall, our participants achieved 83%−100% attendance and 52.8%−81.7% session adherence. The attendance rate is considerably higher than many reported in exercise trials for both AD and non-AD older adult populations⁸. Few studies have reported session adherence rates as we reported. In the past two decades, substantial advances have been made to suggest the importance of adequately delivered aerobic exercise dose in the study of aerobic exercise’s effects and mechanisms in AD^6,9,10. The ability to engage persons with AD in aerobic exercise in the context of challenging BPSD is the cornerstone for ensuring the delivery of an adequate aerobic exercise dose. Our paper demonstrates that persons with AD can successfully participate in aerobic exercise interventions even when BPSD is present with high attendance and moderate to high session adherence. These findings are foundational to future studies that will evaluate the effects of aerobic exercise interventions on symptom progressions, the minimally effective aerobic exercise dose, and the dose-response relationship in AD.

Our paper used a multiple case study design and has the inherence limitations of such a design. We only described three cases that met our eligibility criteria. It is likely that other unique symptom combinations will be manifested by other participants that will require different strategies to ensure exercise adherence. The strength of our study was to use a person-centered approach and specific strategies in the context of each participant’s family, social, and health history to manage BPSD to engage participants in exercise. The findings from this study inform the design of future exercise studies in AD. Our established strategies for engaging persons with AD in exercise can be easily incorporated into daily clinical practice.

Conclusions

Persons with mild to moderate AD exhibit various BPSD during participation in a moderate intensity aerobic exercise program. A person-centered approach using individualized strategies, removing triggers for BPSD, getting to know the person, and educating staff are essential for interacting with persons with AD to ensure optimal exercise adherence.