Study Design
The study was designed as a multicenter quasi-experimental non-randomized study. Nine nursing homes accepted to participate in this study. The nursing homes had similar physical conditions, daily routines, number of staff and patients. The allocation of the institutions (to EG or CG) was made after an initial screening of the stage of dementia of all potential participants. The homogeneity of participants in the initial and moderate stage of dementia, for both CG and EG was the criteria used for the allocation of the institutions. Blinded assessors did the assessment and data collection. Four nursing homes implemented an exercise intervention for six months (Exercise Group- EG); while five did not participate in structured physical activity program and maintained their normal routine and usual care, during the same period (Control Group - CG). Usual care includes attendance at day care facilities, visits to health professionals, receipt of medication, respite care etc.
Participants
Nine formal caregivers, in a supervising position, from nine different nursing homes, aged 28-47 years, accepted to participate in the study, proxy reporting about the effects of an exercise intervention in the IwD and reporting about the distress related with the BPSD of their care receivers. These caregivers had daily interaction with the IwD.
The eligible subject pool was restricted to older adults with the following characteristics: age ≥ 65 years, not engaged in any regular exercise training in the last year, institutionalized for more than 6 months, all diagnosed by a physician with an age-related neurocognitive disorder (dementia) at mild or moderate stage according to clinical dementia rating (CDR) (14) and lack of any diagnosed or self-reported musculoskeletal or cardiovascular disorders that contraindicate participation in moderate exercise and testing. After initial screening, formal caregivers, care receivers and institutions received a complete explanation of the purpose, risks and procedures of the study. Written informed consent was provided. The investigation was in full compliance with the Helsinki declaration(15) and the nine institutions where the intervention took place approved all methods and procedures. The Ethical Commission of the Faculty of Sports of the University of Porto also approved this study (reference: CEFADE 02.2014).
Seventy-seven older adults from both genders, aged 65–94 years, accepted to participate in this study. A total of 12 participants dropped out during the trial (15.6%). In the EG, 2 participants didn't reach 70% of attendance rated and in the CG, 4 dropouts were registered due to being bedridden, 3 due to death and 4 unwilling to participate in the last assessment. Sixty-four individuals from both genders, aged 65–93, concluded the study.
A 70% minimum attendance rate to the exercise sessions was required for participants in the EG. The attendance rate for the EG was calculated by dividing the number of exercise sessions completed by participants by the full amount of sessions they were expected to perform throughout the study. The attendance levels were 78.3% or over. The reasons for missing exercise sessions were acute diseases, behavioral disorders, unwillingness to participate in a particular exercise session and other reasons.
The participant's flow diagram is represented in Figure 1.
Exercise Intervention
The EG completed a 6-month exercise program following the recommendations of the American College of Sports Medicine (16) including aerobic, muscle strengthening, flexibility, balance and postural exercises with 2 sessions per week on non-consecutive days. Sessions included 5-15 care receivers and took place in specific rooms with peaceful and pleasant background music. Sessions lasted for 45-55 minutes and were conducted by the same exercise trainer in all settings. The trainer is specialized in exercise for older adults. The sessions were divided into 3 main parts: warm-up (5-10 min including joint mobilization, postural and stretching exercises for general activation), specific training (30-35 min, including 15 min of moderate aerobic exercises + 15-20 min motor and muscular tasks for strength and coordination/balance training) and cool down (5 min with respiratory and stretching exercises) (see Table 1). To ensure homogeneity in the comprehension of the exercises, participation was limited to individuals in mild and moderate stages of dementia. To make the exercise program more efficient and attractive, we established regular similar routines that prioritized enjoyable and familiar exercises (such as simulating walking, running, rowing). Due to the frailty condition of the participants and for safety reasons, sessions were mainly chair-based and routines of functional exercises with low coordination requirements were emphasized so care receivers could achieve the session's goals.
Sociodemographic and clinical characteristics
Formal caregivers’ supervisors and care receivers’ characteristics are given in Table 2.
Outcome Measures
All following measures were assessed at baseline and 6 months of exercise intervention. CDR test, Senior Fitness Test (SFT) and Mini-Mental State Examination (MMSE) were performed by the care receivers. The care receivers Katz index, quality of life - Alzheimer disease scale (QoL-AD) and Neuropsychiatric Inventory (NPI) were proxy-reported by the caregivers.
Dementia Stage
The CDR test (14) was used only at baseline to allocate the care receivers according to their cognitive stage. CDR is an instrument that assesses the existence and prevalence of the various stages of dementia. It comprises 6 cognitive-behavioral items covering memory, orientation, judgment and problem solving, community activities, home and hobbies, and personal care. The cut-off points were CDR=1 (mild dementia stage) and CDR=2 (moderate dementia stage).
Physical Fitness
The SFT (17) battery is considered a reliable instrument for assessing physical fitness in older adults (≥ 60 years old) including older people with cognitive impairment (18). The test items included: chair-stand test - to assess lower-body strength; arm curl test - to measure upper-body strength; 2-min step test - to assess aerobic endurance; chair sit-and-reach test - to assess lower-body flexibility; back scratch test - to assess upper-body flexibility; and 8-foot up-and-go test - to assess agility and dynamic balance.
Cognitive function
The MMSE (19) was used for a global cognitive evaluation. This instrument is clinically used to assess cognitive mental status, detect and follow the course of mental illness and can also be used as a research tool to screen for cognitive disorders and follow cognitive changes in epidemiological studies. It assesses orientation, attention, immediate and short-term recall, language and the ability to follow simple verbal and written instructions. Furthermore, it provides a total score that categorizes the individual on a scale of cognitive function ranging from 0 to 30 (19). MMSE normative values consider the subject’s educational level. Operational cut-off values for the Portuguese population are 22 (for 0 to 2 years of literacy), 24 (for 3 to 6 years of literacy) and 27 (for more than 6 years of literacy) (20).
Functional Capacity
Katz index (21), one of the most used instruments for measuring the ability to perform ADL, was used to proxy-report the physical functioning of the IwD. Katz index includes 6 items: bathing, dressing, transferring, feeding, incontinence, toileting and the sum of all items to calculate the Katz total. Independence levels for the ADL questions are recorded on a scale of 0 to 4, where 0 represents dependence and 5 represents complete independence (22).
Quality of Life
The QoL-AD(23) was used to measure the IwD QoL. The questionnaire included 13 items: physical health, energy, mood, living situation, memory, family, marriage, friends, self as a whole, ability to do chores, ability to do things for fun, financial situation, and QoL as a whole. The QoL-AD provides the participant and caregiver reports of the participant’s QoL and is scored on a 4-point Likert scale ranging from 1 to 4 (excellent), with total scores ranging between 13 and 52 points.
Behavioral and Psychological Symptoms of Dementia
The Neuropsychiatric Inventory (NPI) (24) is an instrument that proxy-reports IwD changes in BPSD over time. The NPI originally assessed 10 behavioral domains (delusions, hallucinations, agitation, dysphoria, anxiety, apathy, irritability, euphoria, disinhibition, and aberrant motor behavior). Two more domains have been added since its development: night-time behavioral disturbances and appetite and eating abnormalities (24). For each of the 12 behavioral symptoms on the NPI, caregivers rated the level of distress they experienced, due to IwD behavioral symptoms, on a scale from 1 (low) to 5 (extreme). The NPI Distress score (NPI-D) (25) is the sum of these 12 ratings (range 0–60). In the present study, we used the Portuguese version published in the Book of Scales of the Study Group on Brain Ageing and Dementia (26).
Statistical Analysis
Results were expressed as either means (standard deviations) or proportions (Table 2). Differences between groups at baseline were tested using unpaired sample t-tests, mann-whitney, and chi-square tests. The intervention effects results were expressed as a percentage of the baseline values of the CG and were examined by repeated-measures analysis of variance ANOVA (2 times [initial and final] × 2 groups [EG and CG]). When ANOVA revealed significant (time, group or interaction), Bonferroni post hoc tests were performed to evaluate pairwise differences. Between-group differences for all outcomes were adjusted to age, as a confounder. Partial eta squared values (η2p) were reported to quantify the effect sizes.
Significance level was set at 0.05 throughout the analyses. Statistical analyses were performed using SPSS 26.0.
Sample Size
The required sample size was calculated with G*Power (Version 3.1.9.2, Heinrich Heine University of Duesseldorf) (27). A power analysis based on a formulation of 80% power, a moderate effect size of 0.25 for NPI, and a significance level of 0.05 for a two-tailed test deemed that a sample of 34 subjects (17 per group) was sufficient to address the research questions.