Training dementia care professionals to help caregivers improve the management of behavioral and psychological symptoms of dementia using the DICE Approach: a pilot study

Abstract

Most persons living with dementia (PLWD) exhibit behavioral or psychological symptoms of dementia (BPSD) over the course of the illness. The DICE Approach (DICE) is a framework that enables caregivers to identify, evaluate, and manage BPSD. This pilot pre-post test study examined the effects of DICE training on dementia care professionals’ self-efficacy, knowledge, and attitudes regarding care of patients with BPSD. Participants underwent either in-person DICE training or, during the pandemic, online training. Case consultations were offered as additional learning opportunities in challenging situations. Of 134 participants in the trainings, 122 (91.0%) provided survey data for one or more instruments before and after training. Participants experienced significant improvement in knowledge and attitudes with respect to BPSD and improvement in self-efficacy with respect to helping caregivers respond to BPSD. Training dementia care professionals in DICE can improve their capacity to support caregivers in the management of BPSD.

Introduction

The growing number of persons living with dementia (PLWD) presents a major challenge to society, especially to affected individuals and their families. Currently, there are more than 6 million individuals living with dementia in the United States and 11 million family caregivers providing unpaid care for individuals with dementia.¹ As the number of older Americans grows rapidly, so too will the number of individuals living with dementia and family caregivers.

Most persons living with dementia will exhibit at least one behavioral or psychological symptom of dementia (BPSD), such as wandering, aggression, and hallucinations, over the course of their illness.² Adverse consequences of BPSD include unplanned hospitalizations and nursing home placement.³ Caregivers are more likely to experience caregiver burden, including stress and depression, when the individual they are caring for has BPSD.³

Determining and addressing the underlying causes of BPSD can be difficult because patients themselves may have difficulty providing a history and because many factors may contribute. For example, behaviors may result from environmental factors such as distressing noises and lights, routine changes, or caregiver factors like critical tone. Patient factors such as medications and medical problems (e.g., pain, infection) may also predispose to or precipitate BPSD.²

There is no simple solution to addressing BPSD in all patients. A patient-centered approach is required to address specific behavioral concerns and underlying factors. Non-pharmacological approaches are the recommended first-line treatment for managing BPSD.^2,4 Non-pharmacological approaches offer significant benefit for individuals with BPSD and their caregivers; however, they are used infrequently or inconsistently.⁵ Instead, psychotropic medications are commonly used, despite minimal evidence of benefit and substantial accompanying risks.³

According to Kales and colleagues⁶, a major reason that non-pharmacological approaches are not used to manage BPSD is the absence of training for front-line providers, resulting in discomfort addressing BPSD themselves, and in turn, being ill-equipped to educate caregivers. To address this, Kales and colleagues organized an interdisciplinary expert panel to improve the assessment and management of BPSD.⁷ This resulted in the creation of the DICE Approach (“Describe, Investigate, Create, Evaluate,” referred to as DICE going forward), an algorithmic method of assessing and managing BPSD by examining the three main interacting contributors related to BPSD – the person with dementia, the caregiver, and the environment. DICE is designed to help caregivers systematically describe the behavior in detail and who is affected, investigate all potential causes of the behavior including environmental and caregiver factors, create an intervention to address those causes, and evaluate whether the intervention plan was successful. Kales and colleagues subsequently demonstrated that a one-day, multi-modal training on DICE could significantly improve family and formal caregivers’ confidence and knowledge of individualized dementia care.⁶

There is no literature available on the effectiveness of training dementia care professionals on using DICE to in turn help family caregivers assess and manage BPSD. The aim of this pilot study, therefore, was to assess the effects of a comprehensive, one-day training program on DICE for dementia care professionals working with caregivers of PLWD affected by BPSD. We hypothesized that dementia care specialists participating in the training would report increased knowledge, attitudes, and self-efficacy regarding working with PLWD with BPSD and increased self-efficacy in assisting caregivers to manage BPSD.

Methods

Sample and Setting

Training on DICE was offered to dementia care professionals who work directly with family caregivers of PLWD in Wisconsin. Dementia care professionals may work as part of county, community, or health care organizations. Invitations were sent via email to dementia care specialists at Aging and Disability Resource Centers (ADRCs) and Tribal Agencies, dementia care leads from managed care organizations, county and community social workers at senior centers and senior coalitions, outreach specialists from the Alzheimer’s and Dementia Alliance of Wisconsin and the Alzheimer’s Association, healthcare professionals in geriatrics and memory care practices, and supervisors in these settings. No individuals were excluded from participating in the training.

All procedures were reviewed and determined to be exempt by the Institutional Review Board at the University of Wisconsin-Madison. There was a waiver of informed consent due to the exempt status of the study. Consent was implied by participation in attending the training and completing the surveys.

Intervention

Participants attended a comprehensive, one-day in-person training that was designed by the research team and Dr. Kales based on Adult Learning Theory.⁸ The training included didactics, problem solving with case examples, and questions and answers from the participants, to provide an opportunity for participants to converse and share their experiences working with clients with BPSD. Didactic topics included factors that contribute to BPSD and how to use DICE with family caregivers. All participants received the DICE manual (https://diceapproach.com/).

Three annual trainings (2017, 2018, 2019) were held in person. Due to the COVID-19 pandemic, in-person training was transitioned to online training in 2020. Participants completed an online modular training program created by the developers of DICE.⁶ The online version included pre-recorded modules on the same topics covered during the in-person training didactic sessions. The amount of time to complete the online training was approximately three hours. Participants had the flexibility to start, stop, and resume the modules as needed and to participate from a location that was convenient to them. While online-modules were not as interactive as in-person trainings, they included a case-based format as well as “e-simulations” at the end of training to test participants’ ability to apply the approach in two cases (one home-based and another facility-based).

Once participants completed an in-person or online program, they were asked to implement DICE with family caregivers who were caring for individuals with BPSD and seeking assistance. Content experts (social worker, nurse practitioner and geriatric psychiatrist) offered case consultations to DICE participants for additional case-based learning opportunities, particularly with challenging care situations.

The training materials are freely available on the Wisconsin Alzheimer’s Institute website (https://wai.wisc.edu/dementia-capable-wisconsin/dice/), and online DICE training is available for a nominal annual fee.

Measures

The impact of the training was evaluated using both quantitative and qualitative approaches. Pre- and post-training data were collected in paper format for the in-person trainings and electronically for the online training. Data from the pre- and post-training surveys included primary outcome measures of trainees’ self-efficacy, knowledge, and attitudes for working with persons with dementia and BPSD, and self-efficacy of assisting caregivers to manage BPSD. Data on participant demographics and professional characteristics and setting were collected on the pre-training survey. A follow-up electronic survey of participants was conducted in January 2021 to capture experiences with implementing DICE with PLWD and their caregivers.

Self-efficacy

Surveys completed pre- and post-training included questions written with the DICE developers on self-efficacy of assisting caregivers with managing BPSD (9 items).⁶ Self-efficacy was scored as the sum of rating responses to questions that asked “How well can you train caregivers to …” manage various aspects of dementia and BPSD, rated on a 4-point Likert scale from 0 (not at all) to 3 (very much).

Knowledge

Surveys completed pre- and post-training included 9 questions from the Knowledge About Memory Loss and Care test (KAML-C).¹⁰ The KAML-C questions were scored as the total number of correct answers (maximum score of 9). Surveys also included questions developed with the DICE developers on knowledge of principles related to DICE (6 items).⁶ Knowledge questions were scored by either having a correct answer (3 questions) or by self-rated level of understanding (3 questions) on a 4-point scale of 0 to 3. Professionals who were not providing direct support to family caregivers, such as those in supervisory roles and physicians, were excluded from completing the KAML-C items on the surveys.

Attitude

Surveys completed pre- and post-training also included the 20-item Dementia Attitudes Scale (DAS).¹¹ The DAS was scored by responses that used a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree); 6 attitude questions with ideal responses in a negative direction were reverse scored for a total maximum score of 140. DAS comfort and knowledge related questions were each grouped and scored as a subscale (10 items, maximum score of 70, for each). Professionals who were not providing direct support to family caregivers, such as those in supervisory roles and physicians, were excluded from completing the DAS items on the surveys.

Sociodemographic information

The sociodemographic information collected included gender, race and/or ethnicity, educational background, professional role and experience, and professional organization.

Training satisfaction

The post-training survey included the following question: “Any comments about this program that you would like to share with the program team?” This question generated comments about training satisfaction.

Experiences using DICE

The follow-up survey included both quantitative and qualitative response questions on implementation of DICE with caregivers, including utilization, facilitators, and barriers. The following questions were included on the survey:

• Are you currently using the DICE Approach?

• Estimate the total number of caregivers you’ve used DICE with since you first started using DICE.

• What factors have contributed to the success of using the DICE approach? (Why was the DICE approach successful?)

• What has been the biggest barrier to using the DICE approach, or to using it successfully with caregivers? (Or why are you not using DICE at all?)

Data Analysis

Program data were managed using the REDCap electronic data capture tools hosted at the University of Wisconsin School of Medicine and Public Health.⁹ Differences in instrument scores collected before and immediately after trainings were calculated to detect change in knowledge, attitudes, and self-efficacy with training. Differences in overall scores (sum of all items, for DAS, DICE self-efficacy and BPSD knowledge ratings), subscale scores (for DAS Comfort and DAS Knowledge subscales) and individual items from the instruments were determined. The total number of correct answers were scored for the three fixed-response DICE knowledge questions and the KAML-C instrument. Differences between results collected before and after training for respondents who submitted complete data for the instrument for both time points were tested for significance using the Wilcoxon signed rank test. Wilcoxon rank tests were used to compare differences in results between individuals trained in-person versus those completing training online. Statistical significance was set at p<0.05.

Means, standard deviations, frequencies and percentages were used to summarize quantitative survey data using SAS version 9.4 for Windows (Copyright© 2013 SAS Institute Inc., Cary, NC, USA). Responses to qualitative questions from the end of program experiences survey were summarized using a thematic analysis approach completed by 2–3 team members per question among the training team and program evaluators.¹²

Results

Demographics of training participants

DICE training was completed by 134 professionals across the state, among whom 122 (91.0%) provided survey data for one or more instruments before (at baseline) and immediately after (post-) training (Table 1). Training participants who provided any survey data were similar to the overall group of participants. Nearly 60% worked for a county-based organization (e.g., ADRC, senior coalition), a multi-county ADRC, or a Tribal Agency. Thirty-two percent were Dementia Specialists or Leads and 26% were Social Workers and/or Case Managers. Most of the participants were trained in-person (83%). Ninety three percent were women; 3% self-identified as Asian or Asian American, 2% as Black or African American and one individual self-identified as Latina or Latino. Participants were well educated and experienced in their professional roles.

Table 1.

Baseline Characteristics for Training Participants

	Training Participants (n=122)
Characteristics	n	%
Organization
County*, ADRC, or Tribe	70	57.4%
Community or Advocacy Services**	10	8.2%
Healthcare	8	6.6%
Homecare	7	5.7%
Long-term or Managed Family Care	12	9.8%
State Government	6	4.9%
Unknown	9	7.4%
Professional Role
Dementia Care Specialists or Dementia Lead***	38	31.1%
Social Worker/Case Manager	38	31.1%
Nurse/Nurse Practitioner	8	6.6%
Physician	3	2.5%
Information, Resources, Caregiver Service Specialists†	16	13.1%
Other	6	4.9%
Director/Manager/Supervisor to above professionals	13	10.7%
Training Format
In-person, full-day	91	74.6%
In-person, half-day	10	8.2%
Online	21	17.2%
Gender
Female	112	91.8%
Male	9	7.4%
Race and Ethnicity ¶
American Indian or Alaska Native	0	-
Asian or Asian American	4	3.3%
Black or African American	2	1.6%
Hawaiian Native or Pacific Islander	0	-
White, Latina or Latino	1	0.8%
White	110	90.2%
White, unknown ethnicity	2	1.6%
Other than above	1	0.8%
Other than above, unknown ethnicity	1	0.8%
Education and Experience
Technical or 4-year College (13–16 years)	65	53.3%
Graduate School (17–23+ years)	55	45.1%
More than 6 years in role	43	34.4%
Years in professional role, mean (SD), median [range]	6.7 (8.3) 3.25 [0.0–40.0]

Key: ADRC, Aging and Disability Resource Center.

^*County based organization such as Area Agencies on Aging and Senior Coalitions.

^**Non-profit organization offering social services, education, advocacy or access to resources.

^***These Dementia Care Professionals are employed by ADRCs to support persons with dementia living at home in the community; Dementia Care Specialists also support building dementia-friendly communities including training other staff at ADRCs.

^†Specialists employed by ADRCs to provide information and access to resources to persons with dementia or family members.

^¶Ethnicity is not Latina or Latino unless otherwise noted.

Knowledge, self-efficacy, and attitudes of dementia and BPSD

Complete knowledge and self-efficacy instrument data were available for most training participants (Table 2). The KAML-C and DAS instruments were administered to the subset (60%) of in-person participants who were in direct-service positions and thus more likely to use DICE in their practice. These individuals were very similar to the larger group of training attendees, although were more likely to be employed by county-based organizations, ADRCs or Tribal organizations, were less likely to have attended graduate school (35%) and had on average one year less experience in their profession (data not shown).

Table 2.

Dementia Knowledge, Self-Efficacy, and Attitudes Scores at Baseline and Post-Training

Scale and Sub-scale Totals	N (%)**	Baseline Mean (SD) Median [Range]	Post-Training Mean (SD) Median [Range]	p-value¶
Total KAML-C* scores† (9 items, max score=9)	72 (59.0%)	7.9 (1.1) 8.0 [4–9]	8.0 (1.0) 8.0 [5–9]	0.040
Self-efficacy for managing BPSD scores† (9-items, max score=27)	92 (75.4%)	19.0 (5.3) [7–27]	22.6 (3.8) 24.0 [9–27]	<0.0001
Dementia Knowledge scores† (3-items, max score=3)	92 (75.4%)	2.7 (0.5) 3.0 [1–3]	2.8 (0.5) 3.0 [1–3]	0.549
BPSD* Knowledge scores (3-items, max score=9)	122 (100%)	5.9 (1.8) 6.0 [2–9]	7.4 (1.5) 7.0 [2–9]	<0.0001
Dementia Attitudes Scale† Total Score (max score=140)	82 (67.2%)	125.0 (9.6) 126.5 [101–140]	127.2 (8.6) 129.0 [105–140]	0.002
Comfort Score† (max score=70)	84 (67.2%)	58.8 (7.5) 60.0 [36–70]	60.0 (6.8) 61.0 [36–70]	0.024
Knowledge Score† (max score=70)	83 (67.2%)	66.0 (4.3) 67.0 [51–70]	66.8 (4.1) 68.0 [48–70]	0.012

^*KAML-C, Knowledge about Memory Loss and Care⁷; BPSD, Behavioral and Psychological Symptoms of Depression;

^**% of N=122;

^†Instrument not administered to all participants;

^¶p-value by Wilcoxon signed rank test.

Knowledge of memory loss in early-stage dementia, as measured by the KAML-C, was high before training (mean of 7.9 correct answers out of maximum score of 9) and increased slightly following training (mean 8.0, p=0.04) (Table 2). Self-efficacy on ability to assist caregivers to manage BPSD using principles of DICE also increased with training (p<0.0001). The greatest improvements in individual self-efficacy ratings included improved ability to assist caregivers on dealing with challenging behaviors (increase from 24% to 40% of participants rating their ability to help as “very much”, the highest rating), deciding what to do about risk (22% to 40%), offering choice to persons with dementia in everyday care (42% to 65%) and engaging persons with dementia in activities (34% to 59%). Knowledge questions regarding dementia and BPSD showed significant improvement after training, with the largest increase in understanding that patient, caregiver and environmental factors all contribute to behaviors (increase from 51% to 72% rating their understanding as “very much”).

The DAS Total Score and the DAS Knowledge and Comfort subscale scores showed increases in dementia knowledge and attitudes (feelings and behaviors towards persons living with dementia) after training (Table 2). In addition to overall score and subscale score increases, specific questions on knowledge or attitudes also showed statistically significant improvement after training. The percent of individuals “strongly agreeing” increased by 7 to 15% on each of the following items: feeling confidence and reward in working with persons with ADRD; understanding importance of knowing history and familiarity with items for persons with ADRD; and admiration of persons’ coping skills.

When examining change in scores for the online versus in-person participants, baseline to post-training gains for the BPSD Knowledge questions were similar regardless of training mode (Table 3). Self-efficacy on ability to assist caregivers to manage BPSD using principles of DICE improved significantly for the in-person participants. One of the self-efficacy questions was missed for those trained online, however improvement across the remaining 8 questions showed a gain in self-efficacy for the online participants. In-person participants who received the DAS instrument (n=61/96, 64%) showed improvements in dementia attitudes (DAS) scores, most notably in comfort, from baseline to post-training. However, improvements following training were not statistically significant for online participants in the DAS Total Score or Comfort and Knowledge Scores (n=21). DAS Knowledge Scores at both baseline and post-training were lower for the online versus in-person participants. Online participants also had less experience and less education before training (data not shown).

Table 3.

Dementia Knowledge, Self-Efficacy, and Attitudes Scores at Baseline and Post-Training by Training Mode

Scale and Subscale Totals	Training Mode	N	Baseline Mean (SD) Median [Range]	Post-Training Mean (SD) Median [Range]	p-value*
Total KAML-C scores (9 items, max score=9)	In-person†	51	7.8 (1.0) 8.0 [6–9]	7.9 (1.0)¶ 8.0 [5–9]	0.328
	Online	21	8.0 (1.3) 8.0 [4–9]	8.3 (0.9)¶ 9.0 [6–9]	0.094
Self-efficacy for managing BPSD (8-items, max score=24)§	In-person	96	17.0 (4.7) 17.0 [6–27]	20.7 (3.4) 21 [9–24]	<0.0001
	Online†	20	17.2 (4.4) 16 [10–24]	19.1 (3.6) 19 [14–24]	0.011
Dementia Knowledge scores (3-items, max score=3)	In-person	92	2.7 (0.5) 3.0 [1–3]	2.8 (0.5) 3.0 [1–3]	0.549
	Online†	21	N/A	N/A	N/A
BPSD Knowledge (3-items, max score=9)	In-person	101	5.9 (1.8) 6.0 [2–9]	7.4 (1.5) 8.0 [2–9]	<0.0001
	Online	21	5.7 (1.7) 6.0 [3–9]	7.2 (1.4) 7.0 [5–9]	0.045
Dementia Attitudes Scale Total Score (max score=140)	In-person†	61	125.8 (9.0) 127 [101–140]	128.2 (8.3)¶ 130 [106–140]	0.002
	Online	21	122.8 (11.0) 123 [104–140]	124.0 (9.0) 125 [105–140]	0.397
Comfort Score (max score=70)	In-person†	61	58.9 (7.6) 61 [37–70]	60.8 (6.5) 62 [42–70]	0.004
	Online	21	59.5 (5.9) 58 [48–70]	59.0 (5.4) 58 [49–70]	0.992
Knowledge Score (max score=70)	In-person†	61	66.9 (3.1)** 68 [54–70]	67.4 (3.9)** 69 [48–70]	0.021
	Online	21	63.3 (5.9)** 65 [51–70]	65.0 (4.5)** 66 [56–70]	0.264

^*Test for significant differences from baseline to post-training within training group by Wilcoxon signed rank test.

^**Test for significant differences in baseline, post-training and difference from baseline to post-training across training groups by Wilcoxon rank test, significant at p<0.05;

^¶borderline significant at p≤0.10.

^†Instrument not administered to all participants.

^§8 of the 9 items used due to missing data for online learners. N/A indicates data not available.

Approximately 50 case consultations were completed with 12 participants following training. Consultation themes included specific discussions on application of the “Describe,” “Investigate” and “Create” steps of DICE, as well as discussion of caregiver health issues and caregiving burden.

Training satisfaction

There was high satisfaction with training and with DICE, as evidenced by comments received immediately after training and following consultations. Illustrative comments included:

• “This training was helpful and easy to follow. I feel able to use these steps and techniques in daily practice.”

• “Excellent approach to assisting caregivers with loved ones. I will be using the information and tools provided by your program.”

• “Excellent feedback and ideas from the DICE team. The available expertise to families dealing with dementia in a rural area has been invaluable!”

• “Consultations with [project manager] have been very helpful and useful experiences.”

Experiences using DICE

The follow-up survey was sent to 104 participants with valid email addresses. Thirty-six (35% of those invited) completed the survey. Compared to the overall group of participants, survey respondents were more likely to be working with the county-based organizations, ADRCs, or Tribal Agencies (72% versus 57% among original training attendees).. The survey respondents were otherwise similar to the larger group of training attendees.

Among the respondents, 69% had used one or more steps of DICE since training and 56% were currently using DICE in their practice. Of the 69% who had used DICE, 84% agreed with finding the approach easy to use and 76% agreed they were comfortable using the approach. When asked to identify factors that contributed to the success of using DICE, themes among 22 responses included having an effective training approach that included learning with peers in a group class, receiving consultations after the training, and having staff mentor other staff (Table 4). Attendees also found the multifactorial approach to be helpful, as exemplified by this participant statement: “The DICE Approach is a whole person/caregiver approach that focuses on a multifactorial approach, which makes it successful.”

Table 4.

Successes with and barriers to using DICE

Feature	What contributed to the successful use of DICE?	What are the barriers to using DICE?
Effective training	Learning with peers in group, support with consultations after training; trained staff mentoring others
Format	Logical stepwise approach that promotes knowledge and self-efficacy	Time-consuming when used formally; time needed for follow-up with caregivers lacking confidence
Caregiver engagement	Caregiver willingness to use approach and follow-through on steps	Caregiver feeling overwhelmed or in crisis; lack of understanding of dementia; wanting “easy fix” of medication; having no caregivers
Healthcare team	Healthcare provider buy-in; use in clinic when educating family caregivers	Healthcare team being unaware; need for better referral channels to dementia care specialists in community

Participants reported that the following caregiver factors were barriers to using DICE: 1) feeling overwhelmed or being in crisis; 2) lack of knowledge about dementia; or 3) wanting an “easy fix” through medications. As noted by one DICE user: “Caregivers are burnt out and do not want to try anything more; they are at the point of looking for outside assistance.” Thirty-one percent of the survey respondents had a family caregiver refuse the approach. Lack of DICE awareness by the healthcare team caring for the PLWD was an additional barrier to successful implementation. Healthcare provider awareness and engagement through training in DICE, as well as referrals to community-based dementia care professionals working directly with caregivers of PLWD was a noted solution for reducing medication use. Only about half of the respondents reported using all four steps of DICE and 53% indicated they had modified the approach for implementation. Modifications included less formal use of the approach with only certain steps applied and delivery through group sessions.

Discussion

As the number of PLWD continues to increase, more family caregivers will need support in managing behavioral and psychological symptoms of dementia. This study provides preliminary evidence that training dementia care professionals in the DICE Approach may help them better serve family caregivers of PLWD. Specifically, training dementia care professionals in DICE was feasible and acceptable and it resulted in improved knowledge, self-efficacy and attitudes regarding the care of people of BPSD. Many participants reported using DICE with their clients and had favorable comments about the usefulness of the approach. Participants reported caregiver-related factors that reduced use of DICE. Lack of healthcare provider knowledge about DICE was a barrier to implementation of medication changes. About half of participants who used DICE used all four steps of the approach, with the other half modifying DICE to use only parts of the algorithm with caregivers of PLWD. Furthermore, the positive outcome of the training led to system-level changes through statewide adoption of DICE.

There are several limitations to this study. First, the study included a homogeneous sample of dementia care professionals (mostly White, non-Latinx women), which may limit applicability to other settings. Improved access to training for individuals who are historically underrepresented based on race or ethnicity is needed in the state to ensure better representation among the dementia care workforce, to in turn better reach underserved persons who are at increased risk for developing dementia. Second, the outcomes presented are for the trained professionals only. It is unclear what, if any impact, the training had on ultimate beneficiaries of this intervention, PLWD and their caregivers. Third, the high levels of education and experience of the dementia care professionals in the state resulted in ceiling effects that may have impacted the ability to detect changes in outcome measures. Finally, due to the low response on the follow-up survey, findings from the subgroup of participants cannot be generalized to all participants with certainty.

The COVID-19 pandemic posed a challenge and an opportunity. The intervention had to change from in-person to online. Although both in-person and online trainees experienced gains in self-efficacy to assist caregivers to manage BPSD, in-person training with peers may have had additional benefits for improving self-efficacy and attitudes, perhaps because participants found the in-person sessions more engaging and collegial. This would be consistent with Bandura’s argument that self-efficacy may be strengthened by observing others performing tasks, in this case, seeing colleagues present cases and receive feedfback.¹¹

As far as we know, this is the first study evaluating training of dementia care professionals from county, community, and health care organizations in DICE. The findings add to the limited literature on educating dementia care professionals in the management of BPSD. The findings support developing a train-the-trainer program to promote sustainability of DICE training to individuals in social service organizations and governmental agencies providing community-based dementia care. The results of this study will help refine the training to ensure sustainability and future expansion to other settings, including a clinical setting (project underway). Furthermore, future studies should assess the effect of the training on longer-term outcomes including care processes and patient outcomes.

Conclusions

BPSD are common, distressing and potentially dangerous to PLWD and their caregivers. The DICE Approach is a well-accepted framework for educating caregivers about how to assess and address BPSD. Training community-based dementia care professionals in DICE may improve their knowledge, attitude, and self-efficacy for understanding and working with PLWD experiencing BPSD and their caregivers. Governmental agencies and healthcare organizations may wish to consider adopting DICE as a framework for training their employees in addressing BPSD.

HIGHLIGHTS.

• Behavioral and psychological symptoms of dementia (BPSD) are common and problematic.

• The DICE Approach helps dementia caregivers assess and address BPSD.

• We developed training for dementia professionals to implement DICE with caregivers. During the pandemic, the training was converted from in-person to virtual.

• In this pilot study, we found that training was satisfactory and effective at increasing knowledge and self-efficacy.