A 5-month qualitative and quantitative methods pilot study was conducted. For objective one, descriptive statistics and qualitative design were used to collect data from content experts and 5 consecutively enrolled informal caregivers of beneficiaries with AD in the HCBS program to modify the intervention. Consecutive enrollment is known to be effective in practice-based research. [16-18] For objective two, descriptive statistics and open-ended question design were used to examine clinician knowledge (pre-/post-test) and satisfaction with training. For objective three, descriptive and pre-/post-intervention design were used to examine feasibility, acceptability, usability, and satisfaction of caregivers with the intervention. For objective four, pre-/post-intervention design was used to examine beneficiary outcomes. Internal Review Board (IRB) approval of the study protocol were obtained.
Study setting and participants
The setting for our study were three HCBS sites in Michigan from the parent trial. A convenience sample of clinicians, beneficiaries who had an AD diagnosis, and their informal caregivers were included.
Three groups were eligible to participate: clinicians, informal caregivers, and beneficiaries. First, clinicians (RNs, SWs, and OTs) were eligible if they were employed at the waiver sites. Second, informal caregivers of a beneficiary with an AD diagnosis were eligible if they were 18 years of age or older. Finally, beneficiaries were eligible if they were 65 years of age or older with an AD or Dementia diagnosis.
Three caregiver engagement strategies that are effective are use of short and succinct evidence-based interventions, use of high-quality clinicians, and use of supporting materials that are tailored to the needs of the person being cared for. The parent trial intervention was evidence-based and used high-quality clinicians, however, the supporting materials (toolkit) were not tailored for those with AD or dementia. [6-7, 11-15] Another problem is that older adults’ needs (44.3% of ADLs [e.g., bathing and grooming] and IADLs [e.g., shopping and banking]) are often not met by an informal caregiver. This may be occurring due to an incongruence between caregiver training and competencies needed to perform care. Caregiver competency domains include medical and nursing skills, assessment, collaborating, and communication; however, most training programs focus on problem solving, use of community resources, and communication. This indicates caregivers may need training on assessment, nursing skills, home exercises, and planning. Powell and colleagues  recommend use of a toolkit when implementing an intervention; and use of a toolkit is known to increase adherence when using an evidence-based intervention. To address gaps in the toolkit, content tailored to care of those with AD or other dementias and additional nursing skills (e.g., assessment) were needed prior to engaging caregivers to assist with providing the intervention.
For objective one, study staff added content to the toolkit from the Alzheimer’s Disease Education and Referral (ADEAR) Center website and the literature. The toolkit was emailed to content experts and mailed to informal caregivers (after consent) for review. Study staff collected data (e.g., edits, additions, deletions, suggestions, and items needing clarity) in an Excel sheet. Data were shared with content experts and an iterative process was used to finalize the toolkit. The revised toolkit was reviewed and approved by the IRB and printed (color, wire bound, with tabs denoting each section) at a cost of $25.32 per toolkit (see Additional file 1).
For objective two, study staff prepared the training module (1-hour, online, see Additional file 2) and surveys in Qualtrics (baseline and pre-test and post-test and satisfaction) for clinicians. An email that included links to the consent, baseline, and pre-test survey; training module; toolkit; and post-test and satisfaction survey was sent to clinicians asking them to participate. If interested, the clinician clicked on the link and completed the informed consent prior to data collection, then completed the training and surveys.
For objectives three and four, beneficiaries with AD and their informal caregivers were identified by managers at the sites, then the manager called the caregiver and beneficiary to explain the study. If both were interested, the manager completed a screening form and notified study staff. After receipt of the screening form, study staff mailed the consent forms and a toolkit to the caregiver, waited a week, then called the caregiver and/or beneficiary to conduct informed consent. If interested, the caregiver and/or beneficiary signed the forms and mailed them to the study staff. For beneficiaries who did not have the capacity to understand informed consent, the caregiver acted as the surrogate, which is common practice in AD trials. [27-29]After receipt of the signed consent forms, study staff informed the site that the caregiver and beneficiary were enrolled and the clinician should provide the intervention. Study staff then completed the baseline, monthly (2-4), and exit surveys via phone and obtained pre-intervention Minimum Data Set-Home Care (MDS-HC) data from the electronic health record.
Measures and data collection
Data on toolkit modifications from content experts and caregivers were collected in field notes. Feasibility data were collected on the screening form (caregivers/beneficiaries) and in online surveys (clinicians). Clinician age, sex, race, ethnicity, discipline, knowledge (pre-/post-test), and satisfaction were collected via online surveys. Caregiver age, sex, race, ethnicity, education level, and relationship to beneficiary (baseline); self-efficacy (baseline/exit); and acceptability, usability, and satisfaction with the toolkit (months 2-4/exit), measured on a scale of 1 to 10, with 10 being the highest, were collected via phone. Beneficiary age, sex, race, ethnicity, and health status (ADLs, IADLs, pain, depression, and falls), ED visits, and hospitalizations were collected from the MDS-HC assessment completed just prior to consent (pre-intervention) and via caregiver phone calls at exit (post-intervention). The General Self-efficacy (GSE)  and MDS-HC  are described in the parent trial protocol paper. Briefly, the GSE is a 10-item tool (Cronbach’s alpha .79-.90)  and the MDS-HC is a self-reported, person-centered assessment for the collection of beneficiary minimum essential nursing data, with reliability and validity, and used in the HCBS program since 1993.  Cronbach’s alpha at baseline was .91 for the caregiver self-efficacy scale.
Data were summarized using qualitative and quantitative methods. Thematic analysis was used to examine field notes. Descriptive statistics were used to examine socio-demographic characteristics and satisfaction. Feasibility was reflected by the proportion of consenting caregivers and beneficiaries out of those who were eligible and approached. A proportion of caregivers who completed the exit assessment out of those who consented was also used. Counts of toolkit use reported by caregivers (months 2-4 and exit) were summarized to reflect acceptability. Self-efficacy of caregivers at baseline and exit were compared and was summarized using paired t-tests. Preliminary efficacy of the intervention delivered via engagement of informal caregivers to use the toolkit for AD or dementia beneficiaries were summarized from matched pairs t-tests for continuous variables and McNemar’s tests for binary variables.