Behavioral symptoms, a core clinical feature of dementia,

Transcription

1 Targeting Behavioral Symptoms and Functional Decline in Dementia: A Randomized Clinical Trial Laura N. Gitlin, PhD,* Paul Arthur, PhD, OTR/L, Catherine Piersol, PhD, OTR/L, Virginia Hessels, BA, Samuel S. Wu, PhD, Yunfeng Dai, MS, and William C. Mann, PhD, OTR ** BACKGROUND/OBJECTIVES: Dementia-related behavioral symptoms and functional dependence result in poor quality of life for persons with dementia and their caregivers. The goal was to determine whether a home-based activity program (Tailored Activity Program; TAP-VA) would reduce behavioral symptoms and functional dependence of veterans with dementia and caregiver burden. DESIGN: Single-blind (interviewer), parallel, randomized, controlled trial (Clinicaltrials.gov: NCT ). SETTING: Veteran s homes. PARTICIPANTS: Veterans with dementia and their family caregivers (N = 160 dyads). INTERVENTION: Dyads in TAP-VA underwent 8 sessions with occupational therapists to customize activities to the interests and abilities of the veterans and educate their caregivers about dementia and use of customized activity. Caregivers assigned to attention control received up to 8 telephone-based dementia education sessions with a research team member. MEASUREMENTS: Primary outcomes included number of behaviors and frequency of their occurrence multiplied by severity of occurrence; secondary outcomes were functional dependence, pain, emotional well-being, caregiver burden (time spent caregiving, upset with behaviors) and affect at 4 (primary endpoint) and 8 months. RESULTS: Of 160 dyads (n = 76 TAP-VA; n = 84 control), 111 completed 4-month interviews (n = 51 TAP-VA; n = 60 control), and 103 completed 8-month interviews (n = 50 TAP-VA; n = 53 control). At 4 months, compared From the *Center for Innovative Care in Aging, School of Nursing, Johns Hopkins University, Baltimore, Maryland; Department of Occupational Therapy, St. Catherine University, Minneapolis, Minnesota; Center of Innovation on Disability and Rehabilitation Research, North Florida/ South Georgia Veterans Health System, Department of Veterans Affairs, Gainesville, Florida; Department of Occupational Therapy and Jefferson Elder Care, Thomas Jefferson University, Philadelphia, Pennsylvania; Department of Biostatistics; and **Department of Occupational Therapy, University of Florida, Gainesville, Florida. Address correspondence to Laura N. Gitlin, Center for Innovative Care in Aging, School of Nursing, Johns Hopkins University, 525 Wolfe Street, Suite 316, Baltimore, MD lgitlin1@jhu.edu DOI: /jgs to controls, the TAP-VA group showed reductions in number (difference in mean change from baseline = 0.68, 95% CI = 1.23 to 0.13) and frequency by severity ( 24.3, 95% CI = 45.6 to 3.1) of behavioral symptoms, number of activities needing assistance with ( 0.80, 95% CI = 1.41 to 0.20), functional dependence level (4.09, 95% CI = 1.06, 7.13), and pain ( 1.18, 95% CI = 2.10 to 0.26). Caregivers of veterans in TAP-VA reported less behavior-related distress. Benefits did not extend to 8 months. CONCLUSION: TAP-VA had positive immediate effects and no adverse events. Because TAP-VA reduces behavioral symptoms, slows functional dependence, and alleviates pain and caregiver distress, it is a viable treatment option for families. J Am Geriatr Soc 66: , Key words: neuropsychiatric symptoms; functional decline; quality of life; caregiving; home care Behavioral symptoms, a core clinical feature of dementia, occur across disease etiologies and stages. 1 Symptoms result in significant disease burden and negative sequela including increased healthcare costs, reduced quality of life and functional dependence, rapid disease progression and more time spent caregiving, nursing home placement, and caregiver distress. 2 6 The Food and Drug Administration has not approved any pharmacological treatments. Off-label antipsychotic medications that are prescribed off label have risks, including stroke and mortality, that often outweigh their modest benefits. 7,8 Pharmacological treatments also do not address behaviors distressful to families (rejection of care). Although nonpharmacological strategies are recommended as first-line treatment, it is unclear which approaches are effective. Previously, we reported results of a pilot randomized trial testing the Tailored Activity Program (TAP), which matches activities to interests and preserved abilities of persons with dementia and instructs caregivers in their JAGS 66: , , Copyright the Authors Journal compilation 2017, The American Geriatrics Society /18/$15.00

2 340 GITLIN ET AL. FEBRUARY 2018 VOL. 66, NO. 2 JAGS use. 4 Findings showed large effect sizes, statistically significant reductions in frequencies of behaviors and time spent caregiving, and cost effectiveness The trial involved a small sample (N = 60) and a wait-list control group, suggesting need for replication controlling for attention. Other activity studies with positive outcomes were conducted in nursing homes, with small samples, and with nonrandomized designs. 12 We report on a randomized trial testing whether TAP involving veterans with dementia (TAP-VA) reduced the number and frequency by severity of behavioral symptoms (primary outcomes) more than in an attention control group. We also evaluated whether TAP-VA slowed functional decline and reduced pain, time spent caregiving, and caregiver distress with behaviors. Finally, we examined whether 4-month benefits (main endpoint) remained at 8 months. Dementia prevalence among veterans ( %) is comparable with that in the population at large, and dementia is associated with higher inpatient and outpatient service use. 13 Substantiating and expanding the evidence base for activity as a treatment option for addressing behavioral symptoms and daily function would improve dementia care. METHODS We conducted a single-blind, parallel, randomized, controlled trial comparing TAP-VA with an attention control using a 1:1 allocation ratio of participants to study groups. A trained interviewer masked to group allocation assessed participants at baseline and 4 and 8 months. Participants were aware of group assignments and asked not to discuss their assignment with the interviewer. Data safety monitoring boards and institutional review boards (IRBs) approved study and consent procedures and brochures. 14 Recruitment ( ) targeted caregivers of veterans using geriatric Veteran Affairs (VA) services using mailings that included an IRB-approved study flyer and invitational letters from medical directors. Procedures were explained to caregivers contacting the research team, who were administered a telephone screen to determine eligibility. Eligible and willing caregivers underwent a home interview within 2 weeks, followed by randomization to TAP-VA or telephone-based education (attention control). Eligible veterans were English speaking, had a Mini- Mental State Examination (MMSE) score of 23 or less, or a physician diagnosis of dementia were able to participate in 2 or more self-care activities; and were not involved in another study. Veterans taking medications from any of 4 classes of psychotropic medications (antidepressant, benzodiazepine, antipsychotic, anticonvulsant) or an antidementia medication (memantine, cholinesterase inhibitor) who were on a stable dose 60 days before enrollment were eligible. Eligible caregivers were English-speaking primary caregivers aged 21 and older and living with the veteran, were accessible by telephone, planned to live in area for 8 months, were willing to learn activities, had managed 1 or more behavioral symptoms in past month, and were not participating in another study. Caregivers taking psychotropic medications who were on a stable dose 60 days before enrollment were eligible. We stratified participants according to caregiver relationship to the veteran (spouse vs nonspouse) and then randomized within stratum. Randomization lists and two sets of randomization forms were prepared using opaque envelopes. Participants were randomized within 48 hours from baseline. Interventions TAP-VA, described previously, 14 involved up to 8 in-home sessions delivered by occupational therapists (interventionists). Interventionists conducted assessments (2 sessions) of veterans (capabilities, executive and physical functioning, fall risk, daily routines, interests), caregivers (routines, employment, readiness), and environments (lighting, seating, clutter, noise). An assessment report and activity prescriptions detailing activity goals, how to set up the environment to support activities and specific strategies for their implementation were provided (four sessions). Graded activities drew upon preserved abilities and reduced task demands, enabling engagement by minimizing distress and sensory overload and compensating for executive dysfunctions. Through demonstration with veterans, caregivers learned to use activities, manage situational distress, and understand behavioral symptoms. In the last two sessions, caregivers learned to simplify activities for future declines and how to use strategies to other care challenges (bathing). The attention control group participated in 8 telephone sessions with a masters-level team member with experience educating caregivers. Information was provided on relevant topics (home safety, dementia), with no discussion of activity or behavioral symptoms. Outcomes Primary endpoints included number and frequency by severity of behavioral symptoms. Using the Neuropsychiatric Inventory Clinician (NPI-C), the presence of behaviors from 14 domains in the past month was assessed. A score reflecting total number of endorsed behaviors (range 0 14) was derived. For each endorsed behavior, caregivers reported frequency (0 = never to 4 = very frequently ( 1/ d), and severity (0 = none to 3 = major source of behavioral abnormality). A total score was derived by multiplying frequency by severity scores for each item and then summing across 142 items from 14 domains (alpha = 0.82, range = 0 1,704); higher scores indicated greater frequency by severity. 15 Functional dependence was measured using the Caregiver Assessment of Function and Upset Scale (CAFU). 16 Items included 8 instrumental activities of daily living (IADLs; telephone, shopping, meal preparation, housework, laundry, travel, medicine, managing finances) and 7 activities of daily living (ADLs; bathing, dressing upper and lower body, toileting, grooming, eating, getting in and out of bed). For activities, caregivers indicated whether veterans were independent (score = 7), had a safety concern (excessive time required, assistive devices used; score = 6), or needed supervision (set-up, cueing but no physical help; score = 5) or physical help (4 = a little, 25% assistance, 3 = moderate, 50% assistance, 2 = a lot,

3 JAGS FEBRUARY 2018 VOL. 66, NO. 2 RANDOMIZED CLINICAL TRIAL FOR DEMENTIA CARE % assistance, or 1 = complete help, >75% assistance). A total functional dependence score was derived by summing across items and dividing by number of items (range 1 7). Lower scores represented greater dependence (alpha = 0.90). Subscale scores for IADL (alpha = 0.73) and ADL (alpha = 0.93) dependence were similarly derived. We also calculated the number of ADLs and IADLs that veterans require assistance with (score<7), deriving a total mean number of activities requiring assistance (0 15; IADL range 0 8; ADL range 0 7). Caregivers evaluated veterans pain using four Pain Intensity Scale items (How much has pain or physical discomfort interfered with day-to-day activities?) on a 5-point scale (1 = not at all, 5 = extremely). 17 A total score was derived by summing across items. Scores range from 4 to 20; higher scores indicate more pain (alpha = 0.88). Caregivers rated their perceptions of veteran s affect using 6 quality-of-life items on a 5-point scale (1 = never, 5 = several times per day). A total score was derived by summing across items. Scores range from 6 to 30; higher scores indicate greater frequency of positive emotion (alpha = 0.29). 18 We used the 10-item Centers for Epidemiologic Study Depression Scale (CES-D) for caregiver depression. Caregivers rated symptoms on a 4-point scale (0 = rarely or none of time, 3 = all of the time). 19 A total mean score was derived by summing across items. Scores range from 10 to 40; higher scores indicate greater symptomatology (alpha = 0.79). Burden was measured using the 12-item Zarit Burden Short Form scored as never (0) to nearly always (4) occurring. A total mean score was derived by summing across items. Scores range from 0 to 48; higher scores indicate greater burden (alpha = 0.87). 20 Time spent caregiving was measured according to number of hours providing ADL and IADL assistance, hours on duty, and and hours doing things for the veteran, caregivers estimated time in minutes and hours using a 24- hour day. 21 Finally, caregivers rated each occurring behavior on the NPI-C using on a 6-point scale (0 = not distressing, 5 = extremely distressing). A distress score was calculated as the mean of subscale averages for 14 behavioral domains. 15 Procedures Adverse events were tracked, interviews were checked for coding accuracy by three independent reviewers, and missing data and coding errors were resolved. Treatment fidelity was monitored using checklists applied to randomly selected (10%) case presentations at supervisory meetings for both groups. Completed intervention documentation was reviewed for protocol adherence. Interventionists did not have contact with the interviewer. Statistical Analysis Sample size calculation was based on the main study hypothesis tested at 80% power for a 2-sided alternative hypothesis using a t-test comparing the treatment groups on 4-month values. Based on previous trials, we sought a medium effect size of 0.50 using a type I error rate of.05. Although this required 64 dyads per group, because of expected attrition (20%), we enrolled an additional 16 per group or 32 dyads (N = 160 dyads, 80 dyads per group). We conducted descriptive analyses and univariate comparisons between the treatment and control groups using chi-square and Wilcoxon rank-sum tests, as appropriate. Descriptive statistics were used to examine potential differences between eligible individuals willing and not willing to participate and between completers and non-completers by 4 months. Analyses were used to characterize sample demographics, evaluate comparability between treatment arms or success of randomization in balancing groups, and determined the effect of noncompleters on that balance due to potentially important prognostic factors. We performed a primary intention-to-treat analysis to examine the effect of treatment on NPI-C (number of behaviors, total frequency-by-severity score), using analysis of covariance (ANCOVA), adjusting for prespecified covariates (caregiver relationship, baseline NPI-C score, propensity score of dropping out). The propensity score was calculated using logistic regression, with dropping out as the dependent variable and baseline characteristics found to be different between groups (Supplementary Table S2) as predictors. For ANCOVA analysis, we calculated adjusted mean differences between groups on the outcome change (baseline to 4 months). All 160 randomized study dyads were used in the analyses, with missing outcomes at 4 months imputed using predicted values from a regression model of significant baseline characteristics. Specifically, for imputing NPI-C total scores at 4 months, predictors included ADL and IADL dependence, caregiver number of medicines, and baseline frequency-byseverity behavior score. ADL dependence, caregiver strategy use score, and baseline number of behavioral symptoms were used to predict number of behavioral symptoms at 4 months. For each primary outcome, sensitivity analysis was conducted comparing the above results with those obtained from complete-case-only analysis. For secondary aims, we followed a similar approach, imputing missing data and then estimating treatment effects based on ANCOVA or logistic regression. Analyses were performed using SAS version 9.2 (SAS Institute, Inc., Cary NC). Statistical tests were 2-sided, with a significance level of.05. RESULTS Of 248 inquiries, 231 (93.1%) dyads were screened over the telephone, of which 164 (71.0%) were eligible. At baseline, 4 dyads were ineligible, resulting in 160 dyads eligible and willing to participate (n = 76 TAP-VA; n = 84 controls) (Figure 1). By 4 months, 111 (69.4%) dyads were available (51 = TAP; 60 = controls), and 49 (31.2%) were unavailable (25 = TAP; 24 = controls). There were statistically significant differences at baseline between completer and noncompleter caregivers at 4 months. Noncompleters cared for veterans with more functional dependence, behavioral symptoms, financial strain, caregiver burden, and caregiving hours than completers (all P s<.05). Noncompleters were more likely to care for older (P =.09) and non-hispanic (P =.10) veterans. (Supplementary Table S1)

4 342 GITLIN ET AL. FEBRUARY 2018 VOL. 66, NO. 2 JAGS Inquiries (n=248) Telephone Screened for Eligibility (n=231) Screened Eligible/Willing (n=164) Completed Baseline (n=160) & Randomized (n=160) Not Screened (n =17) No Response (n = 16) Not Interested (n =1) Ineligible (n = 67) Not diagnosed with dementia (n =1) Psychiatric Diagnosis (n=1) Geography (n = 3) Medical problem (n = 3) Complete assistance with eating (n = 5) Nursing Home/Assisted Living (n = 12) Veteran deceased (n = 4) Caregiver deceased (n=1) Wanted information only (n = 18) Caregiver too far away (n=4) Not enrolled in VA (n=1) Not English speaker (n=1) Other (n = 13) Ineligible (n = 4) Extreme aggression (n = 2) Refused to sign Consent (n = 2) Allocated to TAP-VA (n=76) Allocated to Attention Control (n=84) 4- month follow up (n=51) Lost to Follow Up (n=25) Refused randomization (1) Veteran with dementia died (5) Veteran with dementia moved to nursing home (5) Veteran with dementia went to hospice (1) Caregiver health problems (1) Changed mind (3) Unable to contact (9) 4- month follow up (n=60) Lost to Follow Up (n=24) Geography (4) Extreme Agitation (1) Veteran with dementia moved to nursing home (2) Veteran with dementia died (7) Caregiver died (2) Changed Mind (5) Unknown/Lost Contact (3) 8-month follow up N=50 Lost to Follow-Up (n=1) Veteran with dementia died (1) 8-month follow up N= 53 Lost to Follow-Up (n =7) Moved (1) Unable to reach (1) Nursing home placement (1) Veteran with dementia died (3) Changed mind (1) Figure 1. Study flow chart. By 8 months, 103 (64.4%) dyads were available for follow-up (n = 50 TAP; n = 53 controls). Total attrition was 57 (35.6%) caregivers (n = 26 TAP; n = 31 controls). Caregivers had an average age of , and 156 (97.5%) were female, 129 (81.1%) were white, 139 (86.9%) were the veteran s spouse, 143 (89.4%) were unemployed, 61 (38.1%) had a high school education or less, and 66 (41.3%) had financial difficulties. Veterans had an average age of , and 155 (96.9%) were male, 127 (79.4%) were white, 149 (93.1%) were married, and 65 (40.6%) had a high school education or less. Veterans had an average MMSE score of (range 0 29), and 66 (41.3%) had fair to poor health (Supplementary Table S2). Of 76 TAP-VA caregivers, 11 (15%) refused treatment: 6 for unknown reasons, 1 because of randomization, 2 because of nursing home placement, 1 because of caregiver health problems, and 1 changed his or her mind. Of the remaining caregivers (n = 65, 85%) with treatment data, an average of sessions and

5 JAGS FEBRUARY 2018 VOL. 66, NO. 2 RANDOMIZED CLINICAL TRIAL FOR DEMENTIA CARE minutes per session (range minutes per session) were completed. Of TAP-VA group, 58.5% (n = 38) completed 8 sessions, 24.6% (n = 16) 7 sessions, 0.7% (n = 1) 6 sessions, 3.1% (n = 2) 5 sessions, 7.7% (n = 5) 4 sessions, and 4.6% (n = 3) 3 or fewer sessions; 62 dyads completed more than 3 sessions, considered minimal for treatment effect. Of 84 attention control caregivers, 1 refused treatment because of time commitment (1.0%). Average call time was minutes (range = 8 57). The remaining 83 control caregivers completed an average of sessions and minutes per session. Seventy-eight of the initial 84 (92.9%) completed 4 or more sessions and 6 (7.1%) completed fewer than 4. Veteran Outcomes At 4 months, greater improvement occurred in number of behavioral symptoms (between-group difference, change from baseline to 4 months = 0.68, 95% CI = 1.23 to 0.13, P =.02) and frequency by severity scores (difference =.24.3, 95% CI = 45.6 to 3.1, P =.02) for TAP-VA participants than the control group. After adjusting for baseline frequency by severity score, veteran age, and caregiver relationship (spouse vs nonspouse), we estimated that 69.7% of TAP-VA participants showed improvement (elimination of behaviors or reduction in frequency by severity scores), compared with 46.4% of controls. The number needed to treat to prevent one additional bad outcome was 4.7. We observed greater improvements in TAP-VA veterans than controls for all secondary outcomes except affect. Reductions were found in number of functional activities requiring assistance (difference = 0.80, 95% CI = 1.41 to 0.20, P =.009), functional dependence (difference = 4.09, 95% CI = 1.06, 7.13, P =.009), and pain (difference = 1.18, 95% CI = 2.10 to 0.26, P =.01) (Table 1). Caregiver Outcomes At 4 months, TAP-VA caregivers had less distress with behavioral symptoms than controls (difference = 0.07, 95% CI = 0.14 to 0.01, P =.03). TAP-VA caregivers did not improve statistically significantly more than controls in depression, burden, or time caregiving, although all change scores favored intervention (Table 2). There were no differences between TAP-VA and the attention control group at 8 months. DISCUSSION Activities customized to interests and abilities immediately improved clinical features of dementia (behavioral symptoms, functional dependence), with no adverse events. We found a preventive effect (fewer behavioral symptoms at 4 months for TAP-VA participants than controls), a maintenance effect (TAP-VA caregivers provided assistance with a similar number of ADLs at 4 months whereas controls reported increases in assistance required), reductions in frequency by severity of behaviors and pain, and slowing of functional dependence. TAP-VA also benefited caregivers by reducing distress with behavior. Table 1. Comparison Between Tailored Activity Program (TAP-VA) (n = 76) and Attention Control (n = 84) Group Outcomes for Veterans TAP-VA Attention Control D TAP-VA D Control Baseline 4 Months Change Baseline 4 Months Change Difference Mean (95% Confidence Interval) Least Squares Mean Standard Deviation a P-Value Outcome Number of behavioral symptoms b b 0.68 ( 1.23 to 0.13).02 Neuropsychiatric symptom frequency x severity b ( 45.6 to 3.1).02 Number of ADLs and IADLs needing assistance with b 0.80 ( 1.41 to 0.20).009 Level of ADL and IADL dependence (total) b 4.09 ( ).009 Pain b 1.18 ( 2.10 to 0.26).01 Emotion, affect b b 0.47 ( ).30 Number of ADLs needing assistance with b 0.61 ( 1.08 to 0.14).01 Number of IADLs needing assistance with b b 0.25 ( ).09 Level of ADL dependence b 2.37 ( ).02 Level of IADL dependence b 1.57 ( ).04 a From mixed-model analysis using all 160 randomized dyads. Covariates included baseline values of dependent variable, veteran age, relationship (spouse vs not spouse), and propensity score. b Significant within-group change.

6 344 GITLIN ET AL. FEBRUARY 2018 VOL. 66, NO. 2 JAGS Table 2. Comparison Between Tailored Activity Program (TAP-VA) (n = 76) and Attention Control (n = 84) Group Outcomes for Family Caregivers Outcome TAP-VA Attention Control Baseline 4 Months Change Baseline 4 Months Change Least Squares Mean Standard Deviation a D TAP-VA D Control Mean Difference (95% CI) P-Value Psychosocial Depressive symptoms b 0.59 ( ).31 Burden ( ).65 Distress b ( 0.14 to 0.01).03 Time spent caregiving Hours on activities of daily b ( ).12 living Hours on instrumental ( ).08 activities of daily living Hours on duty b 0.28 ( ).78 Hours doing things ( ).13 a From mixed-model analysis using all 160 randomized dyads. Covariates included baseline values of dependent variable, veteran age, and relationship (spouse vs not spouse). b Significant within-group change. Results are consistent with those of other TAP trials (initial pilot, TAP-hospital, 22 TAP Brazil, 23 TAP-Australia), 24 extending reported benefits to pain and functional dependence. It is disappointing that 4-month benefits did not endure to 8 months, yet expectations for long-term effects may be unrealistic without repeated exposure to the treatment (activity engagement). Unlike pharmacotherapies, activity use depends upon caregivers. We could not determine whether caregivers discontinued activity use or activities needed modifications. Why do customized activities reduce behavioral symptoms, improve function, and decrease pain? The etiology of symptoms is unclear. Conceptual models suggest that symptoms are consequences of heightened vulnerabilities to physical and social environments due to neurodegenerative processes. 25 Customized activities may help veterans remain physically active, reducing pain, and meaningfully engaged with purpose, an enduring need. 26 Instructing caregivers in ways to support function may reduce dependency. 27 Future research should identify the underlying mechanisms by which tailored activities have their effects, including neurobiological links. Activities may affect gene expression networks and cellular health, 28 physiological stress, 29 and circadian rhythms, 30 which are potential pathways as to how activity engagement may have its effects. Study limitations include use of a single veteran setting, small sample size, and that medication change effects are unclear. We relied on caregiver self-report, although state-of-the-science and psychometrically sound measures were used. Furthermore, noncompleters were more distressed and financially strained than completers and reported more behavioral symptoms and functional dependencies. The differential between noncompleters and completers brings into question the generalizability of results and also if exposure to TAP-VA eariler on in the disease progress is warranted. A different dropout rate between study arms and higher-than-expected attrition resulted in loss of power, with some findings not reaching statistical significance. Because TAP-VA caregivers received more attention than control group caregivers, contributions of attention to outcomes are unclear. Regardless of limitations, because poor treatment options for behavioral symptoms and functional dependence are limited, TAP-VA offers a viable approach for addressing behavioral symptoms and functional declines, affording important immediate benefits to veterans and caregivers. ACKNOWLEDGMENTS We would like to acknowledge the contributions of our research staff, including interventionists, and our study participants. Financial Disclosure: Primary funding source was the Veterans Administration Health Services Research and Development Service (VA-IIR ). Conflict of Interest: Dr. Gitlin is an inventor of a training program for the intervention reported in this study, for which the university is entitled to fees. This arrangement has been reviewed and approved by the Johns Hopkins University in accordance with its conflict-of-interest policies. Author Contributions: Gitlin: Study concept and design, interventionist training and attentional control group materials, interpretation of data, primary responsibility for manuscript preparation. Arthur: Participant screening and assessment, analysis and significant process results, critical edits, manuscript review. Piersol: Refine intervention and train interventionists, critical review of manuscript. Hessels: Coordination of study-related activities, randomization of study participants, assurance of interview and intervention integrity, critical review of manuscript. Wu: Analytical design and statistical analyses, developed tables, writing methodological sections of manuscript. Dai: Analyses, critical review of manuscript for statistical accuracy. Mann: Adaptation of study concept to Veterans

7 JAGS FEBRUARY 2018 VOL. 66, NO. 2 RANDOMIZED CLINICAL TRIAL FOR DEMENTIA CARE 345 Administration guidelines, overseeing scientific integrity and interpretation of data, critical review of manuscript. Sponsor s Role: The funding agency did not have a role in the design and conduct of the study including data collection, management, analysis and interpretation of data; and preparation, review or approval of the manuscript. REFERENCES 1. Lyketsos CG, Carrillo MC, Ryan JM et al. Neuropsychiatric symptoms in Alzheimer s disease. Alzheimers Dement 2011;7: Gitlin LN, Kales HC, Lyketsos CG. Nonpharmacologic management of behavioral symptoms in dementia. JAMA 2012;308: Rabins PV, Schwartz S, Black BS et al. Predictors of progression to severe Alzheimer s disease in an incidence sample. Alzheimer s Dement 2013;9: Chan D-C, Kasper JD, Black BS et al. Presence of behavioral and psychological symptoms predicts nursing home placement in community-dwelling elders with cognitive impairment in univariate but not multivariate analysis. J Gerontol A Biol Sci Med Sci 2003;58A:M548 M Jutkowitz E, MacLehose RF, Gaugler JE et al. Risk factors associated with cognitive, functional, and behavioral trajectories of newly diagnosed dementia patients. J Gerontol A Biol Sci Med Sci 2017;72A: Jutkowitz E, Kuntz KM, Dowd B et al. Effects of cognition, function, and behavioral and psychological symptoms on out-of-pocket medical and nursing home expenditures and time spent caregiving for persons with dementia. Alzheimers Dement 2017;13: Maust DT, Kim HM, Seyfried LS et al. Antipsychotics, other psychotropics, and the risk of death in patients with dementia. JAMA Psychiatry 2015;72: Dennis M, Shine L, John A et al. Risk of adverse outcomes for older people with dementia prescribed antipsychotic medication: A population based e-cohort study. Neurol Ther 2017;6: Gitlin LN, Winter L, Burke J et al. Tailored activities to manage neuropsychiatric behaviors in persons with dementia and reduce caregiver burden: A randomized pilot study. Am J Geriatr Psychiatry 2008;16: Gitlin LN, Winter L, Vause Earland T et al. The Tailored Activity Program to reduce behavioral symptoms in individuals with dementia: Feasibility, acceptability, and replication potential. Gerontologist 2009;49: Gitlin LN, Hodgson N, Jutkowitz E et al. The cost-effectiveness of a nonpharmacologic intervention for individuals with dementia and family caregivers: The Tailored Activity Program. Am J Geriatr Psychiatry 2010;18: Cohen-Mansfield J, Marx MS, Dakheel-Ali M et al. Can agitated behavior of nursing home residents with dementia be prevented with the use of standardized stimuli? J Am Geriatr Soc 2010;58: Krishnan LL, Petersen NJ, Snow AL et al. Prevalence of dementia among Veterans Affairs medical care system users. Dement Geriatr Cogn Disord 2005;20: Gitlin LN, Mann WC, Vogel WB et al. A non-pharmacologic approach to address challenging behaviors of veterans with dementia: Description of the Tailored Activity Program-VA randomized trial. BMC Geriatr 2013;13: de Medeiros K, Robert P, Gauthier S et al. The Neuropsychiatric Inventory-Clinician rating scale (NPI-C): Reliability and validity of a revised assessment of neuropsychiatric symptoms in dementia. Int Psychogeriatr 2010;22: Gitlin LN, Roth D, Burgio LD et al. Caregiver appraisals of functional dependence in individuals with dementia and associated caregiver upset: Psychometric properties of a new scale and response patterns by caregiver and care recipient characteristics. J Aging Health 2005;17: Parmelee P. Assessment of pain in the elderly. In: Lawton MP, Teresi J, eds. Annual Review of Gerontology. New York: Springer, 1994, pp Albert SM, Del Castillo-Castaneda C, Sano M et al. Quality of life in patients with Alzheimer s disease as reported by patient proxies. J Am Geriatr Soc 1996;44: Santor DA, Coyne JC. Shortening the CES-D to improve its ability to detect cases of depression. Psychol Assess 1997;9: Bedard M, Molloy DW, Squire L et al. The Zarit Burden Interview: A new short version and screening version. Gerontologist 2001;41: Mahoney DF. The Caregiver Vigilance Scale: Application and validation in the Resources for Enhancing Alzheimer s Caregiver Health (REACH) project. Am J Alzheimers Dis Other Demen 2003;18: Gitlin LN, Marx KA, Alonzi D et al. Feasibility of the Tailored Activity Program for Hospitalized (TAP-H) patients with behavioral symptoms. The Gerontologist 2016;57: Pires Camargo Novelli MM, Machado SC, Balestra de Lima G et al. The Brazilian version of the Tailored Activity Program (TAP-BR) to manage neuropsychiatric behaviors in persons with dementia and reduce caregiver burden in Brazil: A randomized pilot. Alzheimers Dement 2016;12:P O Connor CM, Clemson L, Brodaty H et al. The Tailored Activity Program (TAP) to address behavioural disturbances in frontotemporal dementia: A feasibility and pilot study. Disability and Rehabilitation 2017: Gitlin LN, Piersol CV, Hodgson N et al. Reducing neuropsychiatric symptoms in persons with dementia and associated burden in family caregivers using tailored activities: Design and methods of a randomized clinical trial. Contemp Clin Trials 2016;49: Irving J, Davis S, Collier A. Aging with purpose. Int J Aging Hum Dev. 2017: Gitlin LN, Hodgson N, Choi S. et al. Interventions to address functional decline in persons with dementia: Closing the gap between what a person does do and what they can do: In: Park R, ed. Neuropsychology of Alzheimer s Disease and Other Dementias, 2nd Ed. Oxford, UK: Oxford University Press. 28. Epel ES, Puterman E, Lin J et al. Meditation and vacation effects have an impact on disease-associated molecular phenotypes. Transl Psychiatry 2016;6:e Potier F, Degryse J-M, de Saint-Hubert M. Impact of caregiving for older people and pro-inflammatory biomarkers among caregivers: A systematic review. Aging Clin Exp Res 2017;29: Hodgson N, Safi A. Timing of activities and their effects on circadian rhythm in the elderly with dementia: A literature review. J Sleep Disord Ther 2014;3. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Table S1. Comparison of Non-completers to Completers Table S2. Baseline Characteristics by Treatment Group Please note: Wiley-Blackwell is not responsible for the content, accuracy, errors, or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.