Reducing Inappropriate Medication Use for BPSD

Investigators previously developed a low-cost, practical, patient- and care partner-centric, evidence-informed systematic approach (the "DICE Approach" or DICE), to assess and manage behavioral and psychological symptoms of dementia (BPSD).

The goals of this proposal are to refine and test the application of DICE in primary care clinics by: (a) using existing clinic staff to deliver DICE; and (b) using the electronic medical record to identify and recruit PLWD (persons living with dementia) and their care partners (n=100) based on criteria that are clinically meaningful and inclusive of the maximum number of participants in the most equitable way.

Clinic-based social workers in four primary care practices at University of California Davis (UCD) will coordinate behavioral management using DICE with care partners, PLWD and other clinic providers. Outcomes will include: 1) feasibility/ability to carry out the approach; 2) acceptability to PLWD and their care partners; and 3) the ability to measure psychiatric medication use and health care use in the electronic medical record. Findings from this study will guide the design of a much larger future study using the DICE Approach to improve outcomes for PLWD and their care partners.

PRÉCIS

Intervention Structure, Implementation Protocol, Fidelity/Adherence Monitoring Plan:

In this study, Investigators will refine and pilot the application of The DICE Approach (DICE) in 4 primary care clinics by: (a) leveraging existing clinic staff to deliver DICE to care partner-PLWD dyads; (b) using electronic resources to identify and recruit PLWD-care partner dyads based on clinically relevant and minimum inclusion/exclusion criteria; and c) evaluating clinically relevant outcomes using the electronic medical record (EMR). DICE will be implemented in four large primary care clinics within the UCD Health System over a 6-month period (n=100 PLWD-care partner dyads; 25 dyads per site, allowing comparison of implementation and outcomes by site characteristics). Key "study champions" at each of the four primary care clinics will be the licensed practical nurse assigned to each clinic who will serve as the Onsite DICE Coordinator (ODC).

As a minimal risk study embedded in primary care practices, Investigators plan to obtain a waiver of informed consent for PLWD, care partners and clinic staff in order to reduce introduction of artificial care resources or biases wherever possible. Additionally, a Health Insurance Portability Act (HIPAA) waiver will be obtained for subject identification (see a. below) and outcome ascertainment from the EMR. There will be minimal personal health information (PHI) data collected. At the time the agreement is finalized, the Internal Review Board (IRB) of Record to make the HIPAA determinations, or if necessary UC Davis IRB will make the HIPAA determinations before releasing the acknowledgment of a reliance on an external IRB.

Investigators have arranged with University of California Davis LVN supervisors (Ms. Elder and Ms. Skillsky) to do a 6-hour training for all of the LVNs on a single day. This will be a hybrid of in person (kick off session and ending brainstorming session) and watching the online modules together in a campus auditorium. Clinicians will then use the approach in a "test" case in study months 2-3 during which they will engage in as needed coaching with a DICE trainer (a geriatric psychiatrist and education specialist) to ensure they are using the approach with fidelity. The DICE trainer will also be available as needed for coaching during the remainder of the study for approach related questions. Manualization and training with follow-up coaching have been shown to be effective implementation strategies that can enhance fidelity of implementation of an evidence-based program. Other clinic personnel (social workers and PCPs {primary care providers}) will also receive the manual, an email overview of the study, and be invited to train using the website or at the on-site training. Following training and case consultation, the ODCs will then use the DICE Approach for any PLWD they encounter during the 6-month study period (months 4-9). Primary outcome metrics will be collected using the electronic medical record and by estimating time spent in the approach and strategies extracted from clinic notes.

The clinicians will meet with the PLWD-Care partner dyad by phone to determine if any BPSD are present. If no BPSD are currently present, the ODC will educate the care partner about DICE, including minimizing risk factors to prevent behaviors, and plan to continue monitoring (26) for BPSD during the 6-month study period. If BPSD are occurring, then the "Describe" and "Investigate" Steps of DICE will be triggered.

Each step of DICE will be documented by the clinicians in their clinic note within the EMR. In the "Describe" step, the ODC will obtain an accurate characterization of the behavior and the context in which it occurs as well as quantify the frequency and severity of the most troublesome behavior (as defined by the care partner) using the one-item question Investigators have used in prior trainings (16, 17): "Please rate from 0 to 4 the severity and frequency of the behavior (0=none or never; 1=mild and/or occasionally; 2=moderate and/or sometimes; 3=severe and/or frequently; and 4=very severe and/or daily). If the severity and frequency are different, pick the higher score (e.g. a behavior that is daily but mild should be scored as a 4)". In the "Investigate" step, the ODC will identify possible underlying and modifiable causes of the behavior.

With information obtained in the "Describe" and "Investigate" steps, the ODC will work with the care partner to create and implement a treatment plan to manage BPSD. The ODC will brainstorm with the care partner (e.g. what are activities the PLWD enjoys) and instruct the care partner in behavioral and environmental strategies. These approaches could include enhancing effective communication with the PLWD, creating meaningful activities, dealing with environmental challenges including ensuring safety, and simplifying tasks and creating structured routines. The ODC will also consult with other relevant clinic personnel on the Create treatment plan as needed; for example, a plan may involve having the provider assess and manage infection, constipation or pain. From prior experience with DICE, these first three steps (Describe, Investigate, Create) can be accomplished in 30 minutes.

Following the "Create" step, and in a 2-week follow-up contact by telephone, the ODC will engage in the "Evaluate" step with the care partner, during which the ODC will determine if recommended strategies were used and their effectiveness. Effectiveness will be measured by repeating the frequency/severity question asked of the care partner in Describe so that the impact of the strategy can be assessed quantitatively comparing the pre- and post-strategy scores at 2 weeks' time; any downward movement of the score will be categorized as effective. , If the DICE generated strategies were effective, the ODC will monitor and surveil for BPSD with monthly contact. If the DICE generated strategies were not effective, the ODC will problem-solve with the care partner and other team members to determine if a) strategies were used and if so correctly; or b) if strategies are not effective and additional "Create" recommendations are needed. Additionally, the DICE trainer (MB) will be available to ODC for coaching regarding the DICE Approach as needed during the study period.

Study Title Training Dementia Care Professionals to Help Care Partners Improve the Management of BPSD Using the DICE Approach

Objectives Primary: To pilot a clinical implementation of The DICE Approach to not only provide clinicians tools to address behavioral issues with PLWD patients via their caregivers but also to determine the feasibility of ascertaining and analyzing the primary (psychotropic medication use) and secondary clinical outcomes (hospitalizations, ED Emergency Department) visits and nursing home placement of PLWDs), including whether they differ for participants of color.

Secondary: To inform the design of a future large-scale ePCT (Engineering, Procurement, Construction, Turnkey), including its relevance to people from diverse racial, ethnic, socio-economic, and educational backgrounds who are living with dementia and their care partners. DICE is centered around reducing the "knee-jerk" use of psychiatric medications for sedation among PLWD, particularly when the underlying causes of BPSD (e.g. infection; pain; stressful environment) have not been explored.

Design and Outcomes In this study, Investigators will refine and pilot the application of DICE in 4 primary care clinics by: (a) leveraging existing clinic staff to deliver DICE to care partner-PLWD dyads; (b) using electronic resources to identify and recruit PLWD-care partner dyads based on clinically relevant and minimum inclusion/exclusion criteria; and c) evaluating clinically relevant outcomes using the electronic medical record (EMR). DICE will be implemented in four large primary care clinics within the UCD Health System over a 6-month period (n=100 PLWD-care partner dyads; 25 dyads per site, allowing comparison of implementation and outcomes by site characteristics). Key "study champions" at each of the four primary care clinics will be the licensed practical nurse (LVN) assigned to each clinic who will serve as the Onsite DICE Coordinator (ODC).

Feasibility measures will include: 1) rate of enrollment of PLWD-care partner dyads; 2) number of contacts between ODCs and dyads; and 3) time requirements for ODCs implementing DICE (as estimated by the ODCs and corroborated by clinic notes).

Acceptability will be measured by the acceptance rates of the ODC's recommendations of strategies for BPSD (generated by use of DICE) among dyads, as well as by other providers within the primary care practice. Exit interviews of participants (ODCs, other providers and care partners) will be conducted to further deepen the knowledge of intervention acceptability and implementation challenges.

Clinical outcomes: all UCD clinical sites (inpatient and outpatient) use the same EMR (EPIC), so ascertaining clinical outcomes will be uniform across clinics. The primary clinical outcome is the feasibility of measuring psychotropic medication use. Mean PLWD participant psychotropic medication use during the intervention period will be compared to historical controls (mean rate of medication use among PLWD for the four participating clinics in the 6 months prior to the intervention). Secondary clinical outcomes will be the feasibility of measuring rates of hospitalizations, ED visits and nursing home placement ascertained from the EMR. Mean PLWD participant health care utilization during the intervention period will be compared to historical controls (mean rate in the 6 months prior to the intervention). The feasibility of obtaining the nursing home placement variable from the EMR will be established during this pilot study by ascertaining nursing home placement from care partners and then determining if it subsequently is recorded in the EMR.

Interventions and Duration The DICE Approach intervention will be piloted for 6 months during the study period. As above, mean PLWD participant psychotropic medication use during the intervention period will be compared to historical controls (mean rate of medication use among PLWD for the four participating clinics in the 6 months prior to the intervention). Secondary clinical outcomes will be the feasibility of measuring rates of hospitalizations, ED visits and nursing home placement ascertained from the EMR. Mean PLWD participant health care utilization during the intervention period will be compared to historical controls (mean rate in the 6 months prior to the intervention).

Sample Size and Population Investigators will examine data across 4 clinics to assess the feasibility of a multi-site study. Investigators selected 25 dyads per clinic as our sample size (total n=100) to have the ability to examine feasibility measures within clinic as well as across clinic with some accuracy. For example, the degree of accuracy around an enrollment rate of 75% overall is 18% (width of the 95% confidence interval), allowing a tighter margin of error overall. Within clinic, accuracy is larger (width of 95% confidence interval 36%), suggesting more uncertainty, but allowing us to examine clinic-specific enrollment to ensure that there is consistency across clinics.