Developing and Testing the Enhancing Active Caregiver Training (EnACT) Intervention for Dementia Family Caregivers (EnACT)

Developing and Testing the Enhancing Active Caregiver Training (EnACT) Intervention for Dementia Family Caregivers

Developing and Testing the Enhancing Active Caregiver Training (EnACT) Intervention for Dementia Family Caregivers

Persons with Alzheimer's disease and related dementias (ADRD) experience behavioral symptoms such as agitation, combativeness, depression, and apathy. These behaviors increase caregiver stress, which leads to negative outcomes, such as poor health, depression, and increased caregiver burden. This project seeks to develop and test an intervention to enhance active caregiver skill training to prepare caregivers to better manage behavioral symptoms in order to improve ADRD caregiver wellbeing. This intervention is called Enhancing Active Caregiver Training (EnACT). The study team will develop and conduct initial testing of EnACT techniques in a small group intervention with family caregivers of persons living with dementia. This intervention will first develop and refine EnACT through small group meetings with ADRD caregivers. Next, EnACT will be tested by two groups of ADRD caregivers in order to assess how achievable and useful caregivers find it to be. In addition, EnACT will be evaluated for its impact on caregiver preparation, stress process, and well-being.

Specific Aim 1: Develop and iteratively refine the EnACT intervention for ADRD caregivers. A series of three sequential focus groups will be used to gather ADRD caregiver feedback about EnACT-intervention techniques, including frequency, duration, delivery, feasibility, acceptability, and relevance of the intervention. Between focus groups, the investigators will refine and edit the EnACT intervention in an iterative process. Intervention scenarios and activities will be chosen and refined as part of Aim 1 in preparation for testing feasibility and acceptability during Aim 2 and mechanisms of action in Aim 3. Setting and Participants: The investigators will partner with community-based services through the Utah Caregiver Support Program and the Utah Alzheimer's Association to recruit 10 ADRD caregivers to participate in a series of three focus groups. Formal caregivers, persons under the age of 18, and individuals who have English fluency ratings of none or poor will be excluded. Data Source: During Focus Group 1, participants will view previously developed video segments in order to identify those that caregivers prefer for intervention practice. Participants will also explore intervention activities to assess which are most useful and acceptable. Revisions to the process will be made based on reactions and feedback. During Focus Group 2, participants will work through a revision of the steps based on Focus Group 1 feedback using a second video segment focused on behavioral symptoms in ADRD. A prototype intervention manual will be developed based on feedback from the first two focus groups. Focus Group 3 will involve reviewing developed prototype materials (i.e., manuals, videos) and a discussion on acceptability, the logistics for incorporating this intervention into caregiver training, challenges, and benefits of activities. Feedback from this focus group will be incorporated into the EnACT intervention in preparation for Aims 2 and 3. Measures: Each focus group will be audio-recorded and transcribed. The PI will also document observations and thoughts using field notes and journaling in order to account for bias. Focus group questions will focus on participant reactions to intervention steps (view, practice, reflect) to identify which activities are meaningful, facilitate or impede participation, and are of most help in ADRD caregiving. Participants will complete a demographic questionnaire to document age, caregiving experience, gender identity/sexual orientation, race/ethnicity, and education. All materials developed during the process of intervention development, including edits and revisions, will be retained as an audit trail. Analytic Plan: All qualitative data will be analyzed in an iterative process to inform intervention development. NVivo software will be used to organize and analyze data and to document an audit trail. Coding will occur in two phases: 1) a priori codes to organize data based on the theoretical framework and intervention development, and 2) pattern coding within each a priori code. Team members (PI, RA, and co-primary mentors) will meet to resolve conflicts in coding and to help group patterns into major themes. Descriptive statistics will be used to describe participant demographics. Sample Size/Power Calculation: Guidelines for acceptable focus group size range from 4 to 12 participants, with a minimum of three focus groups. To account for potential attrition, a sample of 10 caregivers will be recruited to participate in three focus groups. This will provide enough time to elicit feedback regarding all elements of the intervention. Specific Aim 2: Evaluate the feasibility and acceptability of the EnACT intervention. For Aim 2, the investigators will conduct an NIH Stage 1b pilot study to test the feasibility and acceptability of the EnACT intervention. A randomized waitlist control design will be used. Participants will be randomly assigned to one of two groups. Group A will first participate in the intervention (three meetings offered every 2 weeks at Weeks 2, 4, and 6), followed by 8 weeks of follow-up. T1 will be a pretest gathered at enrollment. Group B will wait 8 weeks and then start the intervention at Weeks 8, 10, and 12, with a posttest at Week 14. Data will be gathered at six time points no less than 2 weeks apart over the course of the control, intervention, and follow-up. Setting and Participants: Thirty ADRD caregivers from the Utah Caregiver Support Program and Utah Alzheimer's Association will be recruited and randomly assign them to the group-based EnACT intervention (n = 15) and a waitlist control group (n = 15). Eligibility criteria will include 1) being a primary, informal ADRD caregiver, 2) the ability to read and speak English, and 3) being age 18 or older. Caregivers will test the intervention developed and refined in Aim 1. A postintervention debriefing focus group will occur at Week 8 for Group A and Week 14 for Group B in order to gather feedback on the process and activities and any recommendations participants have for improving the intervention. Location: The intervention will take place at locations associated with our community partners' caregiver support groups, including senior centers, the Division of Aging & Adult Services, and offices of the Utah Chapter of the Alzheimer's Association. Frequency and duration of the intervention: The focus groups will meet every 2 weeks for a total of three 60-minute sessions. It is anticipated that participants will work through two scenarios per session. The frequency and duration have been chosen in consideration of caregiver burden. Delivery of intervention: The EnACT intervention will be led by an interventionist who will be recruited, trained, and supervised by the PI. The interventionist will also have access to the intervention manual, which will be developed during Aim 1 of the study. Measures and Data Source: Feasibility will be assessed by gathering screening data, recruitment rates, documentation of the randomization process, intervention adherence, completion rates, and fidelity. Intervention procedures will be assessed by examining participation rejection rates for each activity. Acceptability will be measured using observational data documented in video recordings, questions about participant satisfaction, and a postintervention focus group. Questions will focus on ease of participation, helpfulness, relevance, benefits/challenges, and comfort with each activity. The final survey will include open-ended questions targeting reflection on caregiving practices, assumptions, the caregiving environment, and influences on practice change. In addition, participants will be asked for suggestions for improvement, preference compared to other programs they have experienced, and barriers to or benefits of use as part of the postintervention focus group debriefing. Surveys will be used to capture demographics, mechanisms of the EnACT intervention, and outcomes. Video will be used to document audio and visual activities at each of the intervention meetings. These data will allow us to assess participant reactions to the intervention activities, which will influence our understanding of the acceptability of the intervention. The post-intervention focus group will be recorded using digital audio in order to assess for overall acceptability of the intervention, effectiveness, participant satisfaction, and facilitators of and barriers to participation. Analytic Plan: NVivo software will be used to organize and analyze qualitative data and document an audit trail. Digital audio recordings will be professionally transcribed. Video and written documentation will be directly imported and analyzed in the software. Coding will occur in two phases: first, a priori codes will be used to identify elements of feasibility and acceptability, and second, pattern coding within each a priori code. The PI and RA will meet weekly to analyze data, and monthly with the larger team (PI, RA, and mentors) to resolve conflicts in coding and to group patterns into major themes. Descriptive statistics will be used to summarize levels of feasibility and acceptability. Power Analyses: The primary goal is to assess the feasibility and acceptability of the EnACT intervention within the context of a randomized study, as well as the study protocol. Based on research best-practices guidance from NIH and noted experts, pilot studies-due to their smaller sample sizes and the frequent design adjustments necessary to maximize recruitment, retention, and quality assessment of outcomes-can not definitively test hypotheses, nor can they provide reliable effect-size estimates. Nevertheless, the proposed pilot will assess whether a subsequent full-scale RCT (NIH Stage III) modeled after this pilot is logistically feasible and acceptable. Specific Aim 3: Examine potential mechanisms of change over time and their subsequent impact on proximal and distal outcomes. Setting and Participants: Using the same sample from Aim 2, the investigators will assess the influence of the intervention on constructs from the guiding framework using multilevel growth modeling. Outcome measures will be administered a total of six times, beginning at enrollment and no less than every 2 weeks throughout waitlist control, intervention, and follow-up (see Figure 2). To decrease burden, participants will complete surveys in a manner of their choosing (i.e. phone, Internet, or in-person). Procedures and Measures: Instruments will include those measuring demographics, attributes of imagined interactions, capacity to adapt, appraisal of caregiving demands, perceived stress, and caregiver well-being. Analytic Approach: Data will be stored in a secure database built using Research Electronic Data Capture (REDCap). Descriptive statistics will be used to assess demographic data and caregiver practice. Multilevel growth modeling, using a linear mixed-effects approach, will be used to examine individual change, population average, and interindividual differences in initial status; and rate of change regarding caregiver perceived stress, well-being, capacity to appraise demands and adapt, as well as attributes of imagined interactions. Constrained longitudinal data analysis, which works well with mixed-effects growth modeling, will be used to assess the ability of the intervention to produce a lower regression coefficient between perceived stress and uncontrolled objective stress. This will also give greater power to the analysis. Sample Size/Power Analysis: As an NIH Stage 1b pilot study, the goal is to assess the plausibility of this work in order to guide the next stage of research. Thus, it should not be expected that the sample size will have outstanding power. Constrained longitudinal data analysis will be used on the data set (N = 30) to exploit randomization and integrated with the full longitudinal impact data in order to give greater power than more limited analyses. Simulations conducted in SAS and analyzed under the integrated model (see Statistical Design and Power) yield a respectable power of .70 against the null hypothesis when the hypothesized slope difference corresponds to a medium effect size of .5. While there are limitations in a small sample, one strength of this study is the analysis of both whether the intervention works and how it works. Analysis done on the growth curve will be accompanied by statistics that show 1) how well it works, 2) for whom it works, and 3) confidence level in these results.

We will use a randomized waitlist control design. Participants will be randomly assigned to one of two groups. Group A will first participate in the intervention (three meetings offered every 2 weeks at Weeks 2, 4, and 6), followed by 8 weeks of follow-up. T1 will be a pretest gathered at enrollment. Group B will wait 8 weeks and then start the intervention at Weeks 8, 10, and 12, with a posttest at Week 14. Data will be gathered at six time points no less than 2 weeks apart over the course of the control, intervention, and follow-up.

A series of three sequential focus groups will be used to gather ADRD caregiver feedback about EnACT intervention techniques, including frequency, duration, delivery, feasibility, acceptability, and relevance of the intervention. Between focus groups, we will refine and edit the EnACT intervention in an iterative process. Intervention scenarios and activities will be chosen and refined as part of Aim 1 in preparation for testing feasibility and acceptability.

A Stage 1b pilot study will be conducted to test the feasibility and acceptability (Aim 2) of the EnACT intervention and potential mechanisms of change over time and their subsequent impact on proximal and distal outcomes (Aim 3). This arm will use a randomized waitlist control design.

The intervention has three components within each session. Step 1, View, involves group viewing of a caregiving scenario, developed during previous research, which portrays a situation in which a caregiver is dealing with a challenging behavioral symptom. Step 2, Practice, involves small-group rehearsal activities that facilitate understanding, making choices, and examining possible responses. Step 3, Reflect, involves self-reflection and group processing surrounding choices and observations.

The Survey of Imagined Interactions is a 7-point Likert scale, with higher scores representing more of a quality of imagined interactions. Four subscales will be used to assess specific characteristics and functions of imagined interactions. These subscales include: the rehearsal function with a minimum value (worse outcome) of 4 and a maximum value (better outcome) of 28, the characteristic of Specificity with a minimum value (worse outcome) of 4 and a maximum value (better outcome) of 28, Proactivity with a minimum value (worse outcome) of 4 and a maximum value (better outcome) of 28, and Discrepancy with a minimum value (better outcome) of 6 and a maximum value (worse outcome) of 42.

Capacity to Adapt will be measured using the Resilience Scale. This is a 7-point Likert scale, with a minimum value 25 (worse outcome) and a maximum value of 175 (better outcome). This scale measures one's ability to adapt, specific to constructs of meaning, self-confidence, perseverance, serenity, and loneliness.

This outcome will be measured using the Revised Caregiving Appraisal Scale (5-point Likert scale) which asks ADRD caregivers to assess the following five subscales: Burden with a minimum value (better outcome) of 9 and a maximum value (worse outcome) of 45 , Satisfaction with a minimum value (worse outcome) of 6 and a maximum value (better outcome) of 30, Mastery with a minimum value (worse outcome) of 6 and a maximum value (better outcome) of 30, Demand with a minimum value (better outcome) of 3 and a maximum value (worse outcome) of 15, and Impact with a minimum value (better outcome) of 3 and a maximum value (worse outcome) of 15.

The Perceived Stress Scale (5-point Likert scale) measures self-reported stress. This scale has a minimum value (better outcome) of 0 and a maximum value (worse outcome) of 40.

The Modified Caregiver Burden Inventory (5-point Likert scale) comprises five subscales: Time, Physical, Social, Emotional, and Developmental. This scale has a minimum value (better outcome) of 0 and a maximum value (worse outcome) of 96. We will also use the Positive Aspects of Caregiving Scale (5-point Likert scale) to measure the perceptions of the benefits of caregiving. This scale has a minimum value (worse outcome) of 9 and a maximum value (better outcome) of 45.

Demographics will be measured, including the following: age, education, gender identity/sexual orientation, race/ethnicity, employment, salary, caregiving experience, frequency of theatre attendance, care-recipient relationship, caregiver support, time as caregiver, type of caregiver.

Inclusion Criteria: providing primary, informal, support to family members with Alzheimer's disease or related dementia (ADRD). 18 and older ability to read and speak English. Exclusion Criteria: Formal caregivers Persons under the age of 18 individuals who have English fluency ratings of none or poor

Data collected as part of this study will be shared in compliance with the NIH data-sharing policy. This data will be de-identified to ensure that the identities of research subjects cannot be readily ascertained. Audio and video recordings will not be included in data sharing due to the identifying nature of these data. Transcripts from audio recordings will be de-identified and available. All data sharing (either with individual investigators or public databases) will be done in compliance with our human research participants' wishes, who indicate on their consent form whether their data may be shared with other investigators.

When will the data be shared: Prior to the on-line publication date of each paper, the PI will work with relevant investigators to assemble a complete dataset with full documentation (including metadata, protocols, etc.).

Investigators working under an institution with a Federal Wide Assurance (FWA) may request access to data. The request for data will require, at a minimum, a short proposal describing the research question, the aims of the study, and the data requested. This dataset will be carefully screened to make certain that all identifying information has been removed (to the standards of the IRB). We will make the data and associated documentation available to other users under a data-sharing agreement that provides for at a minimum the following information: (1) a commitment to use the data only for stated research purposes and not to identify any individual participant; (2) a commitment to secure the data using appropriate computer technology; (3) sound scientific research questions; (4) a commitment to acknowledge the project and funding sources in all publications; and (5) a commitment to submit accepted manuscripts to PubMed Central per NIH policy.

Walter E, Pinquart M. How Effective Are Dementia Caregiver Interventions? An Updated Comprehensive Meta-Analysis. Gerontologist. 2020 Nov 23;60(8):609-619. doi: 10.1093/geront/gnz118.

Gitlin LN, Marx K, Stanley IH, Hodgson N. Translating Evidence-Based Dementia Caregiving Interventions into Practice: State-of-the-Science and Next Steps. Gerontologist. 2015 Apr;55(2):210-26. doi: 10.1093/geront/gnu123. Epub 2015 Feb 17.

Laver K, Milte R, Dyer S, Crotty M. A Systematic Review and Meta-Analysis Comparing Carer Focused and Dyadic Multicomponent Interventions for Carers of People With Dementia. J Aging Health. 2017 Dec;29(8):1308-1349. doi: 10.1177/0898264316660414. Epub 2016 Jul 25.

Cheng ST. Dementia Caregiver Burden: a Research Update and Critical Analysis. Curr Psychiatry Rep. 2017 Aug 10;19(9):64. doi: 10.1007/s11920-017-0818-2.

Onken LS, Carroll KM, Shoham V, Cuthbert BN, Riddle M. Reenvisioning Clinical Science: Unifying the Discipline to Improve the Public Health. Clin Psychol Sci. 2014 Jan 1;2(1):22-34. doi: 10.1177/2167702613497932.