EVALUATING THE SOCIAL LIFE TEMPLATE FOR RESIDENTS WITH DEMENTIA

Transcription

1 EVALUATING THE SOCIAL LIFE TEMPLATE FOR RESIDENTS WITH DEMENTIA FINAL REPORT January 30, 2013 Farida K. Ejaz, Ph.D., LISW-S 1 Margaret Calkins, Ph.D. 2 Ashley M. Bukach, B.S. 1 Special Assistance: Mahum Abbas, B.S. 1 & Alycia Conway, B.A. 1 Margaret Blenkner Research Institute, Benjamin Rose Institute on Aging (MBRI) 1 IDEAS Consulting, Inc. 2

2

3 Acknowledgments This work is made possible through a grant from the Mount Sinai Healthcare Foundation. The project team would like to thank their interviewers: Sarah Schwartz Branka Primetica We would also like to thank the following staff members at Stone Gardens for their assistance with this project: Ross Wilkoff Marina Sanchez Bianca Williams Kelly Henderson

4

5 TABLE OF CONTENTS EXECUTIVE SUMMARY... E - 1 OVERVIEW OF THE REPORT... 1 SECTION 1: INTRODUCTION... 1 Background... 1 Evaluation of the SLT conducted by BRIA and IDEAS... 2 SECTION 2: EVALUATION CONDUCTED BY BRIA... 5 Research Methods... 5 Data Analyses... 7 Findings... 8 SECTION 3: EVALUATION CONDUCTED BY IDEAS INSTITUTE Behavior Observations Findings SECTION 4: DISCUSSION AND RECOMMENDATIONS Fidelity of the Training on the SLT Staff Outcomes Resident Outcomes REFERENCES... R - 1 APPENDIX... A - 1 Supplemental Scales... A - 1 Behavior Observation Form... A - 3

6

7 EXECUTIVE SUMMARY The evaluation of the Social Life Template (SLT) project was conducted by two research institutes: the Margaret Blenkner Research Institute of the Benjamin Rose Institute on Aging (BRIA) and IDEAS Institute. BRIA staff was largely focused on conducting a quantitative analysis while IDEAS staff conducted behavioral observations. BRIA Evaluation Overview BRIA staff used a longitudinal, quasi-experimental design to conduct the study. The experimental site was Helen s Place, the newly opened unit for persons living with dementia (PLWD) in which the SLT was implemented. The comparison unit was Stone Gardens and PLWD from there were selected to participate in the study. The SLT was not implemented at Stone Gardens. Data were collected at two points in time: Time 1 (T1) approximately 3-4 months after residents had moved into Helen s Place and at Time 2 (T2) approximately 3 months after T1. A true baseline could not be collected due to reasons such as Institutional Review Board delays and obtaining family consent prior to approaching PLWD. Data were collected from the following sources: 1) In-person interviews with staff at Helen s Place 2) In-person interviews with residents at Helen s Place and at Stone Gardens 3) Resident and staff records Certain groups were not included in the evaluation such as staff in the comparison unit and family members because of budgetary constraints. The major areas that the BRIA evaluation focused on were: Results 1) The fidelity with which the staff at Helen s Place were trained on all modules of the SLT 2) Staff perceptions of and experiences with the SLT 3) The extent to which the SLT resulted in positive resident outcomes The fidelity with which the staff at Helen s Place were trained on all modules of the SLT A train-the-trainer model was used to conduct the training. Data from the staff training records indicate that although the trainers self-trained themselves on all of the 12 modules of the SLT, they did not train the care staff on all of the modules. The care staff was trained on anywhere between 2-6 modules. Perhaps, the trainers did not have E - 1

8 enough lead time to train the care staff before residents began moving into Helen s Place. It is also possible that although care staff was not trained on all of the modules, the trainers felt that the care staff understood enough about the general principles of the SLT to implement these with confidence. In fact, the majority of the care staff was very satisfied with their training. However, our data also demonstrated that modules that staff was not trained on, were not implemented at Helen s Place such as PLWD being able to lead committees, organize activities and maintain the library. Our recommendation is that the initial training of the trainers should not be conducted as a self-training, but in collaboration with Dr. Camp and his staff who developed the SLT. This would provide an interactive method of the trainers providing feedback to the developers on whether the training needs to include all 12 modules; or whether the modules become repetitive after a while; or whether some of the modules need to be dropped that are most unlikely to be implemented with PLWD. Based on our findings from this project (see section on staff and resident findings), it is evident that some principles and components were implemented more than others. Thus, we suggest that Dr. Camp and his team set up a meeting to discuss refinement of the SLT with the staff at Helen s Place prior to implementation at another site. Staff perceptions of and experiences with the SLT A total of 13 staff completed both the T1 and T2 interviews. Staff overall, were satisfied with the training that they had received on the SLT, although a few would have preferred to have had more training. They were also confident about implementing it with PLWD, although in certain areas they believed that the level of dementia of the person would dictate what they could implement. They provided positive examples of how the SLT helped them deal with the challenges faced by PLWD, such as calming residents who became agitated and re-directing negative behaviors. Overall, staff believed that the SLT was easy and simple to implement, created a sense of community and helped with communicating with PLWD. However, they believed that the SLT was most practical to implement with residents with mild dementia, and during the course of the study this opinion was re-affirmed by them. They also felt that PLWD could make choices regarding the clothes to wear and food to eat but certain higher level cognitive tasks such as leading or directing an activity or managing the library were less likely to be successfully implemented with PLWD. Our recommendation is to include on-going communication and feedback between staff and the original developers in order to refine pieces of the SLT. For example, certain components of the SLT that are not practical to implement with PLWD could be removed or refined. The intervention could also be targeted to persons with mild dementia. Another strategy could be to further refine and test the components of the E - 2

9 intervention that are likely to work with varying levels of dementia (mild, moderate and severe). The extent to which the SLT resulted in positive resident outcomes A total of 10 residents (5 from Helen s Place and 5 from Stone Gardens) completed both the T1 and T2 interviews. Residents at both units were similar on the key outcomes at the start of the study. During the study period, findings demonstrated that residents in both the experimental and comparison units had a good Quality of Life, were satisfied with leisure activities and had positive comments to make about living on their units. Residents participated more often in activities such as getting together with friends, listening to music and watching movies and sports activities but rarely went shopping or worked on hobbies or attended a religious activity. With respect to certain principles or components of the SLT, residents reported that they often had enough activities to choose from and participated in group activities with other residents. However, they rarely reported leading or organizing a group activity or being part of a committee to welcome new residents or participating in intergenerational activities. During the course of the study, residents at Helen s Place became more satisfied with the quality of the care staff that provided direct services to them. On the other hand, residents at Stone Gardens became less satisfied with their meals and dining experience and exhibited more depressive symptoms. Perhaps the success of the SLT lies in the impact it has on staff that take care of PLWD. However, since our sample sizes were very small in the experimental and comparison units and some of the staff who worked at Helen s Place also worked at Stone Gardens, we cannot make definitive conclusions that the results were due to implementing the SLT. We recommend that the SLT is implemented in a larger, more generalizable and controlled clinical trial to demonstrate its effectiveness on resident outcomes. However, prior to conducting this larger study we recommend that the SLT is further refined and targeted to an appropriate group of PLWD. IDEAS Evaluation Overview In addition to interviews and quality of life measures, direct observations were made of residents and staff over the course of two days. A behavior mapping recording sheet was created that focused on both routine activities and components of the SLT that were featured in the training modules. Location, activity, level of engagement, quality of interactions, cues and mood was recorded every 20 minutes. Field notes were made providing more context and detailed descriptions of what was taking place. Occasionally, actual bits of conversation were written down if it related specifically to E - 3

10 aspects of the SLT. Due to budgetary constraints, it was not feasible to conduct observations at Stone Gardens for comparison purposes. Results Some elements of the SLT were clearly incorporated into the daily life of Helen s Place, such as offering residents choice, particularly at meals and occasionally for activities, and use of daily calendars. Other elements such as leading activities, use of activity stations, intergenerational visits and field trips/outings, were not observed. While this may be partly due to the limited time frame of the observation period, it supports the data from the staff and resident interviews about certain components of the SLT not being incorporated into Helen s Place. We recommend that the developers of SLT meet with staff and either conduct additional training on how to implement some components of the SLT, or adjust the SLT program to better fit with the abilities of residents, possibly having different models for individuals at different levels of cognitive acuity. E - 4

11 OVERVIEW OF THE REPORT This report is divided into four sections. The first section deals with the introduction and the core common components of the evaluation conducted by the Benjamin Rose Institute on Aging (BRIA) and IDEAS Institute. Section 2 focuses on the evaluation conducted by BRIA. Section 3 focuses on the evaluation conducted by IDEAS. Section 4 summarizes the major findings from sections 2 and 3 and provides recommendations for the further testing and implementation of the Social Life Template (SLT). SECTION 1: INTRODUCTION Background There are approximately 35.6 million people internationally with a diagnosis of moderate to severe dementia (Alzheimer s Disease International and World Health Organization, 2012). For years, it was assumed that once an individual is diagnosed with dementia, they no longer have meaningful decision-making capacity. At the time, researchers stated that the loss of medical decision making capacity (competency) is an inevitable consequence of Alzheimer's disease (Marson & Harrell, 1999). However, there is increasing evidence that, despite the cognitive, behavioral and mental impairments resulting from the disease, many persons living with dementia (PLWD) can communicate choices and preferences regarding issues that are meaningful to them (Gillies, 2000; Harris, 2002). In fact, current research on Quality of Life (QoL) for PLWD points to the importance of autonomy, choice, decision-making, and dignity as critical components of person-centered care (Barnes, 2006; Boyle, 2008; Dwyer, Andershed, Nordenfeldt, & Ternestedt, 2009; Koppelman, 2002; Nay, 2002; Venturato, 2010). Research also suggests that involving PLWD in shared decision making is likely to reduce agitation and negative behaviors (Chenoweth et al., 2009; Fossey et al., 2006, Whitlatch et al., 2006). Further, there is evidence that the sense of self continues even through the late stages of dementia (Sabat, 2001), suggesting that an understanding of personal needs and values is critical, even when PLWD begin to lose their communication abilities (Moyle et al., 2011). Respecting the wishes of PLWD becomes an important part of helping such individuals maintain or enhance their QoL. For individuals living in shared residential settings, QoL is influenced by factors such as meaningful use of time through activities of interest, social engagement, positive staff attitudes and an environment that promotes social connectedness (Moyle & O Dwyer, 2012). Many non-pharmacological interventions are being developed for PLWD and some of these are being tested in residential care facilities. One such intervention called the Social Life Template (SLT) was developed by Dr. Cameron Camp. The purpose of the SLT is to ensure that PLWD maintain or enhance 1

12 their QoL, are engaged in meaningful activities, have a sense of belonging to a community and a sense of control and independence. Key features include involving PLWD to organize and participate in a wide range of personally meaningful activities. The SLT is grounded in Montessori principles and builds on years of research conducted by Dr. Camp that have proven successful with PLWD in nursing homes and adult day care centers (Camp, 2006; Camp, Orsulic-Jeras, Lee & Judge, 2004; Camp & Nasser, 2003; Camp, Zeisel & Antenucci, 2011; Judge, Camp, & Orsulic-Jeras, 2000). Evaluation of the SLT conducted by BRIA and IDEAS The SLT was implemented at Helen s Place, a newly created dementia care unit at Stone Gardens, an assisted living facility that is part of Menorah Park Center for Senior Living. Residents began moving into Helen s Place at the end of March of An evaluation of the SLT was conducted collaboratively by two research institutes: the Margaret Blenkner Research Institute of the Benjamin Rose Institute on Aging (BRIA) and IDEAS Institute. The BRIA evaluation of the SLT was focused on: Establishing the fidelity with which staff at Helen s Place received training on all modules of the SLT/intervention: o Whether supervisors were trained on all 12 modules of the SLT o Whether supervisors trained the rest of the staff at Helen s Place on all of the 12 modules of the SLT Staff perceptions of and experiences with the SLT: o Whether the staff found the training to be satisfactory o Practicality of the SLT Staff perceptions of the practicality of implementing the SLT with residents suffering from varying levels of dementia, i.e., mild, moderate or severe dementia Staff perceptions of whether the key principles of the SLT could be implemented with residents with dementia Level of confidence in implementing the SLT at Helen s Place o Maintained or increased staff job satisfaction The extent to which the SLT resulted in positive resident outcomes o Maintained or enhanced residents QoL and other outcomes at Helen s Place compared to Stone Gardens 2

13 The evaluation conducted by IDEAS Institute involved observations of residents and staff and was focused on: The extent to which core elements of the SLT were evidenced in the interactions and daily routine of residents and staff at Helen s Place. Assessing the extent to which residents at Helen s Place had the opportunity to participate in the core principles of the SLT o Had choice and control o Opportunities for meaningful engagement o Took on leadership roles Obtaining Institutional Review Board (IRB) Approval Since residents were expected to begin moving into Helen s Place in April of 2013, the IRB approval process was initiated in March of 2013 by researchers at Benjamin Rose Institute on Aging (BRIA). Dr. Calkins subsequently began obtaining approval from the IRB of IDEAS Institute. Informed consent forms were designed for residents and staff. However, because of discrepancies in the two institute s approval processes, modifications requests were made and resolved. Other modification requests were made to the IRB because of processes stipulated by the administration at Stone Gardens. The administrative staff requested that research staff obtain consent to participate from all family members prior to approaching PLWD to get their consent. This was different from the procedure initially approved by the IRB of obtaining prior consent only from family members of residents who had a legal guardian or were too impaired to sign a consent form on their own. Thus, another modification request needed to be made to the IRB for changing the process initially approved. In addition, staff was to provide researchers with residents Mini Mental State Examination (MMSE) scores prior to starting the research. However, research staff realized that most of the PLWD in the comparison group at Stone Gardens did not have a current MMSE score in their records. In fact, the MMSE had not been completed for some residents since 2009 and Since staff at Stone Gardens did not have time to complete the MMSE, research staff decided to take on this task and complete an MMSE for all residents who did not have one completed in Thus, we had to send in another modification request to the IRB to add this component to the study and could not begin data collection until the IRB approved this additional modification. Such modifications resulted in substantial delays and therefore prevented collection of a true baseline measure of residents when they first moved in to Helen s Place. Data collection began in June 2013 when final approval was obtained from the IRB. 3

14 Eligibility to Participate in the Study Three criteria were used to include residents in the study: Obtaining family or legal guardian s consent for resident/relative to participate Obtaining resident consent to participate (for a resident to be eligible to participate in the study, both family members and the resident had to consent to be included in the study). Resident having a Mini Mental State Examination score of 11 or higher (MMSE) (Folstein & McHugh, 1975). Three criteria were also used to include staff in the study: Be employed at Helen s Place Received training on the SLT Provide consent Procedures Introductory Letters to Explain the Study An introductory letter was sent to families/legal guardians to obtain their consent for their relative to participate in the study. Once family consent was received, residents were approached to provide their consent. With respect to the staff, an introductory letter explaining the study was distributed to staff before they were asked to participate in the study. Project staff then met with each staff member to review the consent form and answer any questions about the study. Only staff members who consented to participate were interviewed. Staff Training on the SLT Prior to residents moving in, staff at Helen s Place began training on the SLT. A trainthe-trainer approach was used to train staff on the SLT. Initially one manager selftrained herself on the training materials on the 12 modules of the SLT provided by Dr. Camp. She then trained a second manager on the 12 modules of the SLT. They then used the materials to train care staff employed at Helen s Place. 4

15 SECTION 2: EVALUATION CONDUCTED BY BRIA Research Methods Design The evaluation utilized a quasi-experimental, longitudinal study design over a four month period (June October 2013). Data were collected at two points in time: Time 1 (T1): within 3-4 months of residents moving in, i.e. June and July of 2013; and Time 2 (T2): three months after T1 (in October). At both T1 and T2, residents and staff had the option to refuse participation or reschedule the interview for a different date and time. Based on the researchers prior experience, it was determined that the most effective way to collect data from PLWD would be through in-person interviews. For consistency, data were also collected via in-person interviews from the staff. Data Collection Data were obtained from the following sources: Staff at Helen s Place Residents/PLWD who were admitted to Helen s Place and began participating in the SLT (intervention group) and a comparison group of residents/plwd residing at Stone Gardens. Resident records at both Helen s Place and Stone Gardens obtained from the Nursing Office Staff records regarding the number of trainings on the SLT conducted at Helen s Place. Initially, research staff was also going to collect data from family members; and from staff in the comparison group; however, due to budget limitations, these components were eliminated from the evaluation. Sample Staff A total of 23 staff members were responsible for implementing the SLT at Helen s Place and were considered eligible to participate in the T1 staff interview. Five staff members were ineligible because they were new hires and had not yet been trained on the SLT and two staff members refused participation. The remaining 16 staff members completed a T1 interview with research staff. Two of those staff respondents left employment prior to the T2 interview and one additional staff member was not 5

16 scheduled during the month of the T2 interviews. Therefore, a total of 13 staff members completed both the T1 and T2 interviews including one Administrator, one manager, and 11 care staff, yielding a 57% response rate. The findings in this report are focused on the 13 staff that completed both T1 and T2 interviews. Residents All of the 14 PLWD residing at Helen s Place at the time of the study were considered for participation in this project. Of the 14 residents, a total of six residents were not included as either they or their family member refused to participate. One additional resident was ineligible due to a low score on the MMSE indicating that he/she was not appropriate for interviewing. The remaining 7 PLWD participated in a T1 interview. One of the residents/respondents passed away prior to the T2 interview and another refused participation in the T2 interview, resulting in a total of 5 residents respondents from Helen s Place that completed both the T1 and T2 interviews. In an effort to recruit a comparison group, 14 residents who had cognitive impairment but resided in a non-dementia specific unit at Stone Gardens were selected by staff. Of the 14 who were initially recruited, 7 refusals were obtained either from the family or the PLWD. One additional PLWD was ineligible due to hospitalization. The remaining 6 PLWD participated in a T1 interview with project staff. One resident/respondent refused participation in the T2 interview, resulting in a total of 5 PLWD that had complete data in both T1 and T2. Overall, 10 out of a possible 28 PLWD/respondents completed the T1 and T2 interviews, yielding a 36% response rate. The responses reported here are for the 10 residents who completed both T1 and T2 interviews. Questionnaires Two separate questionnaires were developed for residents and staff. The questionnaires were developed based on a review of the existing literature on the SLT, dementia care, and residential care settings and included measures on QoL and satisfaction. Drafts of the BRIA questionnaires were reviewed by staff at IDEAS, Dr. Cameron Camp, developer of the SLT and Mr. Ross Wilkoff, Administrator at Stone Gardens. The resident and staff questionnaires were pretested with at least three residents and three staff members. Final versions were developed that incorporated the changes based on the pretesting. The T2 questionnaire included the same questions that were asked in T1. 6

17 The final staff questionnaire included the following: a) Background Information b) Employment Information c) Training on the SLT and perceptions about its usefulness and practicality for residents with dementia d) Job Satisfaction The final resident questionnaire included the following domains: a) Leisure Activities b) Principles of the SLT (although the SLT was not implemented at Stone Gardens we examined whether similar principles were being implemented for residents at Stone Gardens) c) Personal Values d) Choice and Autonomy e) Meals & Dining Experience f) Administrative Staff g) Care Staff (staff providing direct care services) h) Overall Quality of Staff i) Depression j) Quality of Life k) Overall Satisfaction with Care and Services Background information on residents was also obtained from their chart records such as their demographic characteristics and prior living arrangements. Data Analyses Data from the in-person interviews were collected on paper and were entered into SurveyMonkey. Data were then exported from SurveyMonkey into the Statistical Package for Social Sciences (SPSS) software for analysis. The first phase of the analysis involved running descriptive statistics on all items in the questionnaires. Subsequently researchers constructed scales on the domains listed in the resident and staff questionnaires (see above). The scales were constructed using a summative process of all items in a particular domain unless an item lacked variance (determined at the 90:10 ratio of variability in at-least two of the response categories). Subsequently, Cronbach s alpha for each scale was examined to determine the scale s internal reliability. Individual items were dropped if doing so led to an improvement in the reliability of the scale. 7

18 The following scales were created from the staff survey: 1) Compendium of Resident Tasks, 2) Compendium of Staff Tasks, and 3) Job Satisfaction. The following scales were created from the resident survey: 1) Leisure Activities, 2) Principles of the SLT, 3) Personal Values, 4) Choice & Autonomy, 5) Meals & Dining Experience, 6) Administrative Staff, 7) Care Staff, 8) Overall Quality of Staff, 9) Depression, 10) Quality of Life, and 11) Overall Satisfaction with Care and Services. In the second phase of the analyses, non-parametric statistical tests were run due to the small sample size. First, the Mann-Whitney test was used to examine if the two independent samples of residents at Helen s Place (Intervention Group) and Stone Gardens (Comparison Group) were different at T1 - the starting point of the evaluation. Then, the Wilcoxon test was run to examine if each group changed over time, i.e., from T1 to T2. The significance level for all tests was set at.10. Due to the small sample size, however, significant differences should be viewed with caution. Findings Fidelity with which staff received training on the SLT Manager Training There were two program managers who conducted the SLT training for the care staff. The head manager self-trained on all 12 modules of the SLT, and the associate manager was trained by the head manger on all 12 modules. The Administrator selftrained himself. Care Staff Training The majority of the care staff reported being trained on the SLT by the head manager (85%). Fewer staff reported being trained by the associate manager (39%). These percentages do not add up to 100% because some of the staff was trained on different modules by both supervisors. As of October 2013, the maximum number of modules on which the 11 care staff was trained was 6 (Range: 2 6). Despite only completing half of 12 SLT modules at most, when asked how many modules they completed, many of the care staff thought they had completed all of the modules. Table 1 on the next page shows the number of care staff out of 11 that completed each module. 8

19 Table 1. Number of Care Staff Trained on Specific SLT Modules Number of Care Staff Trained 10 Staff Interview Findings 8 0 Module 1 Resident-Driven Community 6 Personal Care 7 Dining 10 Engaging the Passive Resident 8 Problem Behaviors #1 9 Problem Behaviors #2 2 Committees 3 Visitors Center and Families 4 Intergenerational Programming 5 Activity Stations 11 Incorporating New Residents 12 Creating and Maintaining a Sense of Community Findings presented in this section focus on the 13 staff members that participated in both T1 and T2 interviews. The 13 respondents included one Administrator, one manger, and 11 care workers. Background Characteristics Table 2. Characteristics of Staff Respondents Characteristic Age (Mean) Female 9 % of Staff (Unless Otherwise Indicated) 27 (Range: 20 48) Race: African American 77% Caucasian (Non-Hispanic) 15% Hispanic/Latino 8% Marital Status: Single (Never Married) 77% Married (Living with Spouse) 15% Divorced 8% Highest Level of Education: High School Diploma 62% Associate s Degree 23% Master s Degree 15% Licensure or Certification:* State Tested Nursing Assistant (STNA) 69% Licensed Nursing Home Administrator (LNHA) 8% 92%

20 Table 2. Characteristics of Staff Respondents % of Staff Characteristic (Unless Otherwise Indicated) Employment Type: PRN 46% Full-Time 39% Part-Time 15% Number of Years Worked in Nursing Home or Assisted Living Facility (Mean) 8 (Range: 1 23) Typical Daily Caseload (Mean) 7 (Range: 7 14) *Note: Percentages do not add up to 100% because not all staff was licensed. Satisfaction with Training At T1 and T2, the majority of the staff members reported being very satisfied with their training on the SLT (67%) and the remaining respondents were somewhat satisfied (33%). Some of the qualitative or open-ended comments that staff made regarding the training included: I understand the SLT, but there s still more I can learn, but I can implement it with the residents with what I know. I was trained on it for one day so how am I supposed to train the residents to do the absolute perfect thing when I m not sure about it. I do not remember everything that is in the SLT but I ve been working 4.5 years with dementia residents so it s is fairly easy to know what to do with them. The SLT makes a lot of sense because it s a routine. It s better than medicating them. Training was good taught me a lot about the SLT. I m very comfortable using it because I know what to do and how to do it. Practicality of the SLT Staff members were asked how practical it is to implement the SLT with residents with three levels of dementia: mild, moderate, and severe. Staff believed that it was most practical to implement the SLT with residents who had mild dementia, followed by those with moderate and then those with severe dementia. After implementing the SLT over time, staff ratings significantly improved for the practicality with which the SLT can be implemented for residents with mild dementia 10

21 (T1 mean = 2.33, SD =.778; T2 Mean = 2.67, SD =.651) but not for those with moderate or severe dementia. Compendium of Resident Tasks Staff were asked to rate how frequently they believed that residents with dementia could complete certain tasks central to the SLT. Answers to those questions were combined to create the Compendium of Resident Tasks Scale. There was no significant change in staff perceptions on this scale from T1 to T2. With respect to the individual items in this scale, tasks that received the high scores at T1 and T2 indicated that staff believed residents with dementia could choose the clothes they want to wear, choose the activities they enjoy, participate in group activities with other residents, and help themselves to a snack or beverage whenever they want to. Tasks that staff felt residents with dementia could not do as easily included running a committee, maintaining the library, and leading or organizing a group activity. See Table 3 on the next page for more information. When staff members felt residents with dementia could not complete a key task most of the time, they were asked to provide a reason why. The most common reason for all of the tasks was that residents were too cognitively impaired followed by physical impairment and a lack of resident desire to complete the particular task. For example, for leading or organizing a group activity at T2, 9 staff members (70%) felt that residents could not complete this task because they are too cognitively impaired, and one staff member felt that often, residents do not have the desire to complete this task. One staff member complained to the interviewer that questions were very hard to answer because answers depended on the level of dementia that the residents had. Another staff person believed that a lot of the residents were very withdrawn and did not want to participate in the activities. 11

22 Table 3. Compendium of Resident Tasks Scale Individual items had the following response categories: 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time Can residents with dementia T1 Mean T2 Participate in religious activities Participate in group activities with other residents Help themselves to a snack or beverage whenever they want to Choose the activities they enjoy Understand the signs for where things are located Choose the clothes that they want to wear Set the table for a meal Take a shower or bath whenever they want to Participate in intergenerational activities Understand a calendar listing daily events and activities Lead or organize a group activity Maintain the library Run a committee Scale Reliability =.85; Range = 0 (Least often) to 39 (Most often) Compendium of Staff Tasks Staff were asked to rate how frequently they believed that staff could perform certain tasks related to the SLT for residents with dementia. Answers to those questions were combined to create the Compendium of Staff Task Scale. There was no significant difference in Staff Task Scale ratings from T1 to T2. With respect to individual items all of them received high mean scores indicating that staff felt they could do these tasks for residents with dementia. Highest scores at T1 and T2 related to staff giving extra time to residents who act out, but other tasks were also rated highly. See Table 4 on the next page for more information. 12

23 Table 4. Compendium of Staff Tasks Scale Individual scales items 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time How often can staff T1 Mean T2 Give extra time to residents who act out Redirect aggressive behavior Try to teach residents with memory problems how to remember things Try to motivate passive residents to get involved in activities Try to find reasons why some residents act out Understand the signs and symptoms of depression Scale Reliability =.43; Range= 0 (Least often) to 18 (Most often) Staff comments related to this section included: A resident got upset about a lady coming into his room, so I redirected him by taking him for a walk and he was calm when we returned for dinner. This is an example of how an aspect of the SLT seems to work. Even when residents are having a difficult time/day it [SLT] allows me to see the possibilities rather than loss. They are still capable of doing things. it [SLT] may not click in for residents the first time, but repeating helps. Confidence Implementing the SLT The majority of staff reported being very confident about implementing the SLT (67% at T1 & 85% at T2), and the confidence level remained stable from T1 to T2. No one reported being less than fairly confident. Qualitative or open-ended questions to this question included comments from staff such as: I feel like I ve been working with them/residents with dementia for a long time, so I try to put it in a way that they can understand. The SLT is a good way to communicate with residents. 13

24 I ve seen how the approach has been successful. It s simple/straightforward. To help with memory they have to be engaged in activities. Before it gets too severe, the mind needs to be stimulated. I see the residents in activities and I see the results. SLT develops a sense of community. For example, there is an outing at Helen s Place today and every resident is going. When do you ever see 100% participation in an activity? I believe that s a result of the SLT. Dr. Camp came up with a very good idea. They [residents] did not know multiplication at the beginning and now they do. They feel like movies, activities, happy They have life and I m living with them. I have been doing this for 5 years. I feel confident because of my experience working with people with dementia. Job Satisfaction Overall Job Satisfaction Respondents were asked a series of questions on different aspects of job satisfaction. Answers to these questions were combined to form the Job Satisfaction Scale. Staff job satisfaction remained consistently high and did not change significantly from T1 to T2. With respect to individual items on the scale, the job aspects with the highest mean scores at T1 and T2 related to the amount of responsibility they had, the amount of time they had to get the job done, comfort with approaching management with concerns, and their opportunity to grow. Only one item, teamwork among staff, at T2 had a mean score of below 3. Table 5. Job Satisfaction Scale Individual scales items 1 = Very Dissatisfied, 2 = Somewhat Dissatisfied, 3 = Somewhat Satisfied, 4= Very Satisfied How satisfied are you with T1 Mean T2 Amount of time you have to get your job done Opportunity to grow Comfort with approaching management with concerns Job security

25 Table 5. Job Satisfaction Scale Individual scales items 1 = Very Dissatisfied, 2 = Somewhat Dissatisfied, 3 = Somewhat Satisfied, 4= Very Satisfied How satisfied are you with T1 Mean T2 Recognition you get for your work Amount of control you have over your job Attention paid to your observations or opinions Supplies you use on the job Way employee complaints are handled Feedback about how well you do your job Teamwork among staff Opportunities for promotion Way management and staff work together Way this facility is managed Fringe benefits Attention paid to suggestions you make Scale Reliability =.95; Range= 0 (Least satisfied) to 68 (Most satisfied) Staff comments related to their job satisfaction included: Would like the residents to have [more] opportunities to do smaller groups, activities meaningful to them. Lack of this happening leads to job dissatisfaction. Would like better communication between staff and HP upper management would help ability to implement SLT as a team. We ve come a long way. We re working on getting better as far as employee/management communication. Staff seems to be more open to trying new things and learning. We still have a little ways to go, but it s getting better. 15

26 Respondents were asked overall, how satisfied they are with their job. All but one staff member reported being very satisfied at T1, and all staff members reported being very satisfied at T2. Recommendation of Job at Facility All staff members indicated that they would recommend taking a job at Helen s Place at T1 and T2. One staff member commented: I would recommend this job because it is high in pay and it is rewarding to work with people with dementia. This facility gives them life again. They don t just eat, sleep, get drugged up. They are able to participate and live. Resident Interview Findings All findings reported in this section are for residents that completed both the T1 and T2 in-person interviews. Background Characteristics of Residents As previously mentioned, all of the 10 residents that completed the T1 and T2 interviews were PLWD. Five of these resided at Helen s Place and the other 5 resided at Stone Gardens. Of the residents for whom we had demographic information, all of them were white, widowed, and had at least one child. See Table 6 for more information on resident demographics. Table 6. Demographics of Residents by Facility Characteristic (% unless otherwise indicated) Age (Mean) White, non-hispanic Widowed Female Mean MMSE Score College Graduate and beyond Child/Child-in-Law is Primary Care Giver Helen s Place (n=5) Stone Gardens (n=5) % 100% 100% 100% 20% 80% % 20% 100% 100% 16

27 Helen s Place and Stone Gardens Residents Similar at T1 on Key Measures/Domains Researchers found that residents on both units at T1 did not score significantly differently on any of the key domains: leisure activities, principles of The SLT, personal values, choice, meals and dining experience, administrative staff, care staff, overall quality of employees, depression, quality of life, and satisfaction with care and services. Thus, it is likely that their domain scores were similar at the start of the project when Helen s Place opened and that differences over time are more likely to be the result of where they resided (experimental unit or comparison unit). Changes from T1 to T2 on Key Measures/Domains Researchers examined significant differences between residents in the intervention group from T1 to T2. The same procedure was used for residents in the comparison group. Higher scores on each of the domains reflect more positive results except for depression. Care Staff Scale There appeared to be an improvement in perception of residents at Helen s Place with respect to their care staff (z = -1.60, p =.11; i.e., close to the.10 level of significance). The mean scale score improved at Helen s Place from 8.60 (SD = 2.79) at T1 to 9.80 (SD = 2.49) at T2. With respect to the individual scale items, residents at both Helen s Place and Stone Gardens very highly rated the patience of the care staff and the timeliness with which they are given their medications. Table 7. Care Staff Scale Individual scales items 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time HP Mean SG Mean T1 T2 T1 T2 Are the staff patient with you Do you get your medications on time Do the employees who take care of you know what you like and don t like Do the employees explain your care and services to you Scale Reliability =.85; Range= 0 (Least often) to 12 (Most 8.60* 9.80* often) *Change in T1 & T2 scores approached significance at the.10 level. 17

28 Overall Quality of Employees While Care Staff Scale scores approached having a significant improvement for Helen s Place residents, scores for the Overall Quality of Employees scale significantly decreased over time for both residents of Stone Gardens (z = -1.63, p =.10) and Helen s Place (z = -1.84, p =.07). At Stone Gardens, ratings decreased a full scale point (T1: M = 8.20, SD = 1.30; T2: M = 7.20, SD = 1.79). At Helen s Place, ratings decreased more than a full scale point (T1: M = 8.40, SD =.89; T2: M = 7.00, SD = 1.00). With respect to individual items, residents at Helen s Place and Stone Gardens indicated that staff are friendly and treat them with respect Most of the time. Residents indicated that staff only teach residents how to remember things sometimes. During the course of the study there were some staff turnover and we are unsure if the changes in staff led to an overall dissatisfaction with staff over time. Table 8. Overall Quality of Staff Scale Individual scales items 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time HP Mean SG Mean T1 T2 T1 T2 Are the people who work here friendly Do the employees treat you with respect Do the staff teach or help you how to remember things Scale Reliability =.60; Range = 0 (Least often) to 9 (Most often) 8.40** 7.00** 8.20** 7.20** **Change in T1 & T2 scores significant at the.10 level. Qualitative comments on the staff included the following positive and negative statements: They [staff] are very good to me. Very nice to me but I am not a jerk. Others are jerks and it makes it hard for the staff. Resident of Helen s Place Staff are all sweet but most don t have any life experience. If you give them authority they run with it. Resident of Helen s Place the people are nice. Resident of Stone Gardens 18

29 Depression While the Depression Scale scores for Helen s Place residents remained consistent over time, residents at Stone Gardens saw an increase at T2 (M = 9.20, SD = 4.82) from T1 (M = 7.20, SD = 4.32). This increase approached significance (z = -1.60, p =.11). Table 9. Depression Scale Individual scales items 0 = Never, 1 = Sometimes, 2 = Often During the past week, how often did you HP Mean SG Mean T1 T2 T1 T2 Enjoy life Feel happy Feel lonely Feel sad Have trouble keeping your mind on what you were doing Not seem to be able to get going Feel depressed Feel like eating Feel that everything you did was an effort Feel that people disliked you Sleep restlessly Scale Reliability =.83; Range = 0 (Least often) to * 9.20* (Most often) *Change in T1 & T2 scores approached significance at the.10 level. Meals & Dining Experience Although this scale did not significantly change for Helen s Place residents, scores significantly decreased over time for resident of Stone Gardens (z = -2.04, p =.04) from a mean of (SD = 3.27) at T1 to (SD = 2.68) at T2. With respect to individual items, both residents at Helen s Place and Stone Gardens gave the lowest ratings to whether they could help set the table if they wanted to. See Table 10 on the next page.

30 Table 10. Meals & Dining Experience Scale Individual scales items 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time HP Mean SG Mean T1 T2 T1 T2 Is your food served at the right temperature Are the dining room staff courteous and friendly Can you choose to sit at a table with your own friends Do you get a variety of foods that you can choose from Can you get snacks and drinks whenever you want to Can you help set the table if you want to Scale Reliability =.73; Range = 0 (Least often) to 18 (Most often) *Change in T1 & T2 scores significant at the.05 level. No Changes from T1 to T2 on Key Measures/Domains Quality of Life (QoL) * 10.80* Both residents at Helen s Place and Stone Gardens experienced a fairly high QoL and their QoL was maintained or remained the same from T1 to T2. In terms of the individual items in this scale, residents reported good to excellent responses on their relationship with the person they felt closest to, with friends, and with their quality of life as a whole. They believed that they enjoyed fair to excellent physical health, energy level, mood, living situation, relationship with family members, ability to do chores, things for fun, and their financial situation. It appears that more residents at Helen s Place reported their memory as being fair to good compared to tenants at Stone Gardens who were more likely to report their memory as good to excellent. Table 11. Quality of Life Scale Individual scales items 1 = Poor, 2 = Fair, 3 = Good, 4= Excellent How would you rate your HP Mean SG Mean T1 T2 T1 T2 Physical health Relationship with family members Relationship with friends

31 Table 11. Quality of Life Scale Individual scales items 1 = Poor, 2 = Fair, 3 = Good, 4= Excellent How would you rate your HP Mean SG Mean T1 T2 T1 T2 Relationship with person you feel closest to Ability to do chores Life as a whole Energy level Financial situation Living situation Memory Mood Ability to do things for fun Scale Reliability =.83; Range = 0 (Highest quality) to 48 (Lowest quality) Qualitative comments by residents on their quality of life at the facility included the following types of statements: It s the best place after home. Resident of Stone Gardens I think it is a great place to live, and I am fortunate to be here. -- Resident of Stone Gardens Living here is nice. Everyone is pleasant. Resident of Helen s Place My cousin lives here, but I don t see him often. -- Resident of Helen s Place Leisure Activities Overall, all residents in Helen s Place and Stone Gardens reported being somewhat or very satisfied with the activities at T1 and T2. No one reported that they were dissatisfied or very dissatisfied with the activities. Overall scores for the intervention and comparison groups did not change significantly from T1 to T2. See Table 12 below. With respect to the individual questions that were asked on leisure activities, residents at both Helens Place and Stone Gardens participated more often in activities such as 21

32 getting together with friends, listening to music, doing exercises and meeting with family but rarely went out shopping, worked on hobbies (some reported that they did not have hobbies) and almost two-thirds didn t attend a religious activity in the past two weeks. Table 12. Leisure Activities Scale Individual scales items 0 = Not at all, 1 = A few times, 2 = Often In the past two weeks, how often did you HP Mean SG Mean T1 T2 T1 T2 Go to the movies, theater, museum, or sporting event Exercise (stretch, walk, dance) Get together with family Go out to eat Listen to music Get together with friends Go to the beauty shop or barber shop Play games/puzzles Read a book or listen to a story Attend a religious activity Work on a hobby Go shopping Scale Reliability =.83; Range = 0 (Least often) to 24 (Most often) Principles of the SLT Residents reported that sometimes to most of the time they had enough activities to choose from and participated in group activities with other residents; however, residents rarely reported leading or organizing a group activity, being part of a committee to welcome new residents, or participating in intergenerational activities. Overall scores for the intervention and comparison groups did not change significantly from T1 to T2. See Table 13 on the next page. 22

33 Table 13. Principles of the SLT Scale Individual scales items 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time How often HP Mean SG Mean T1 T2 T1 T2 Are there enough activities to choose from Do the staff tell you about the activities on a daily basis Do you get a calendar listing the activities on a daily basis Have you participated in group activities with other residents Have you lead or organized a group activity Have you been part of a committee to welcome new residents Have you participated in intergenerational activities Scale Reliability =.62; Range = 0 (Least often) to 21 (Most often) Supplemental Information Additional information was collected from residents on the domains listed below. For each of the domains, there was no significant difference in scores comparing T1 to T2. For more information on the domain and their individual items, please refer to the Appendix. Personal Values page A - 1 Choice & Autonomy page A - 1 Administrative Staff page A - 2 Overall Satisfaction with Care and Services page A - 2 SECTION 3: EVALUATION CONDUCTED BY IDEAS INSTITUTE Behavior Observations Observations were conducted from 8 am to 2:30 pm and from 11 am to 7 pm on two weekdays in October (different weeks). Staff remarked that these were fairly typical days, with the exception of the fire alarm that went off once on one of the two days. The floor plan of Helen s Place was categorized into 8 different spaces, so resident s locations could be identified. The behavior coding sheet focused on elements of the SLT and the training staff received prior to the opening of Helen s Place. Specific items that were coded include: 23

34 Level of engagement in either group or individual activities was coded on a 5 point Likert type scale (leading activity, active participation, semi-participation, passive participation, un-engaged) Quality of resident interactions with others (residents, staff and families) was coded on a five point scale (strong positive/friendly, mild positive, neutral, mild negative, strong negative/unfriendly) Residents mood was coded on a 5 point scale (happy/laughing, mildly positive, neutral, mildly negative, agitated/distressed) Support for resident choice (3 point scale- staff offer limited choices, staff tell/suggest resident what to do, staff do for resident) Use of the daily calendars (verbal or physical) Use of other cues (physical environment, verbal/gestural cues) Observations were recorded every 20 minutes for six study participants who were in public areas of Helen s Place (exclusive of bedrooms). The coding sheet also included space for qualitative notes and a list of program elements from the SLT (use of activity stations, intergenerational visits, family visits, field trips, task force/committee meetings, and time between scheduled activities) that were of particular interest (these being the elements that might distinguish the SLT from other dementia care programs). The timing of the observations was structured to capture as many of the normal daily activities as possible. If a resident moved between areas of the unit, or engaged in different behaviors during a 20 minute period, the primary location and behavior were recorded. Findings Participants Observations were made of the six residents who had agreed to participate at T1 who were still at Helen s Place in October (note that observations took place before T2 interview data was collected, when one additional resident declined to be interviewed). This included five males and one female. Residents were identified by ID code on the behavior mapping forms. Five staff was also observed, although staff was not identified individually. Quantitative Frequency data from each specifically coded item will be summarized first, followed by a summary of the qualitative notes. There were 44 data observation points (each 20 minute period of observations being one data point), and with six residents included in the observations, there was a total of 264 possible observations. The actual number of data points was 115, meaning that 56% of the time residents were not out in the shared 24

35 areas of the unit (they were either off unit or in their bedrooms). The results presented below are based on 115 observations, or data points. The majority of time (91%) when residents were in public areas of the unit, they were with other people residents or staff. This was coded as group interaction even if they were not actively doing something together (see below for more details on level of activity). A small percentage of time (9% or 10 data points) residents were alone, either doing something (usually eating) or doing nothing. In terms of the type of interactions observed, in only one instance was a resident actually leading an activity they were making floral centerpieces and one resident was showing another how to select a flower and put it in her vase. Residents were actively engaged 44% of the time, which included eating during meals, participating in making the centerpieces, or participating verbally in one of the word games (chicktionary or complete the familiar phrase). Semi participation, which included sitting in on one of these activities and following what was going on, but not actively participating by answering questions or making a centerpiece, occurred in 31% of the group observations. Sixteen percent of observations were passive (awake and in the group, but not really following what was going on) while eight percent was unengaged (eyes closed, possibly sleeping during the activity or focused on a TV program/watching it). When residents were not with others but were out on the shared areas of the unit, they were passively engaged 66% of the time (usually sitting and doing nothing, but not apparently asleep) and actively participating 33% of the time (usually eating a meal after the other residents had finished theirs). When considering the quality of interactions, separate notations were made for resident resident, resident staff, resident family, (all resident initiated communications) and staff resident (staff initiated communication). Overall resident initiated communications were primarily mildly positive (30%) or neutral (20%), with seven percent being strongly positive and only 2 percent being mildly negative. No interactions were observed that were strongly negative. Staff initiated interactions were all on the positive end of the spectrum, with the majority (64%) being mildly positive, 25% being strongly positive/friendly, and 11% being neutral. Residents mood was primarily coded as neutral (79%) with 16% of observations being mildly positive, 3% being happy/laughing, and less than 1% being mildly negative (1 observation). There were several programmatic elements that we were particularly interesting as being innovative features of the SLT. The first was support for resident choice. This was coded as staff offering residents several options (as in a choice of food or beverages, and a choice of activities), staff telling/suggesting to the residents what to do 25

36 (such as it s time for this activity without offering a choice of alternatives), or staff essentially doing for the resident (effectively eliminating choice). The vast majority of the time (88%) staff offered residents choices and waited for the resident to make a choice. In 12% of observations staff basically told residents what to do. In no case was staff observed doing something for the residents. For example, even if staff knew a resident preferred to drink her coffee with milk, it was always offered as an option for the resident to decide ( would you like milk in your coffee, as usual, or not? ). Similarly, the use of personal calendars was of interest. In the 115 data points, calendars were physically or verbally referred to only nine times (0.07% of the time). Seven times the discussion was with the actual calendar and twice it was just referred to. Finally, the use of other cues, either physical, environmental or verbal was not observed. Qualitative On each data collection sheet notes were made about details of what was going on during each 20 minute period. These are not in-depth notes, but mostly short notations. The most common notation (in 16 of 40 data points or 40% of the time) relates to there being no conversation. This was found during meals (the only conversation often being a response to staff asking someone what they wanted), during times in between activities when residents might be sitting in the dining room or the family room, or even during some activities, such as the happy hour. An exception to this lack of conversation was when family was present during meals. They were offered a place at the table and the same food as the residents. There was definitely more conviviality (at least at the table when the family sat) when there was a non-resident to get the conversation started and help keep it going. It is also clear from the notes that residents had a great deal of latitude as to when they eat their meals. Although there are clearly meal times, there were several instances of residents coming out late and eating breakfast at 11 am, or lunch at 4:00. It was clear this was the preferred pattern of one individual, and was routinely accommodated. But it also happened when a resident had been out with a friend and missed lunch. He mentioned that he was hungry, and the staff offered him a number of alternatives. Another repeated notation related to the friendly relationships and conversations that some residents had with staff. There were conversations about going to a club to find a man (for the staff), and several instances were noted of residents (one in particular) and staff who were really joking around and having fun. Notes were also made about caregiving conversations, and the extent to which they reflected the SLT. For example, when staff said to a resident "You've got to take a shower today. It doesn't have to be now, but we could just get it over with. This reflects the sometimes difficult balance between being task focused ( you ve got to take a shower today ) and always trying to offer choices ( it doesn t have to be now ). Other notes relate to residents making 26

37 decisions as when different residents were asked to pick out the music to be played during a meal. SECTION 4: DISCUSSION AND RECOMMENDATIONS Fidelity of the Training on the SLT Although Dr. Camp had provided a training manual in terms of a PowerPoint presentation for each module, it appears that the training on the SLT was not conducted with fidelity. Since the SLT is in its inception stages, we recommend that in the initial phases, those who are trained to become trainers are educated by Dr. Camp and/or his staff rather than train themselves on the materials provided to them. The recommended interactive training between the developers of the SLT and the staff being trained to become trainers is likely to provide invaluable feedback to the developers, particularly with regard to the length and scope of the training and areas that are likely to work with PLWD. As pointed out by researchers, educating staff about new programs before implementation and soliciting staff input into strategies for implementation are examples of positive pre-implementation activities to develop an evidence based program (Bass & Judge, 2010). These researchers have also suggested that the training include role playing, pilot cases and practice sessions with a structured review and feedback; and that training should be on-going during the initial implementation phase because some information can be learned after the basics of the implementation are experienced (Bass & Judge, 2010). We suggest that prior to another round of testing the SLT, Dr. Camp meets with the staff at Helen s Place to review strategies that worked; and to use the information to refine the SLT. Further, it was evident that trainers did not train the care staff on all of the 12 modules of the SLT. None of the care staff was trained on more than half of the modules, and the modules that staff was not trained on were not implemented at Helen s Place. It is possible that the trainers did not have time to conduct the training on all of the modules for the care staff, or that the care staff did not have the time to participate in the training because residents began moving in to the experimental site. It is also possible that staff were not trained on modules that trainers felt would be more difficult to implement with PLWD, such as forming a committee, conducting intergenerational programming and having a visitors center for families. In fact, the observations conducted by IDEAS corroborated that intergenerational activities were not being routinely conducted, that activity stations were not used and that residents were seldom engaged in leading activities. Another speculation for the lack of training on all of the 12 modules is that the training was too repetitive, and/or too long, and that staff believed that by getting trained on a few modules, the core principles of the SLT were covered. In fact, many of the care staff reported that they had been trained on all of the modules, but that was not the case after we reviewed the information provided by the trainers. Thus, a further exploration of 27

38 why staff was not trained on all of the modules is critical to refine the training on the SLT. This is one of the first steps in developing an evidence-based program. Without this step, it is difficult to establish if the intervention (the SLT program) resulted in positive outcomes for PLWD at Helen s Place. Thus, it is reiterated that a meeting between Dr. Camp and the supervisors and staff at Helen s Place is important to refine the SLT. Another limitation to evaluating the fidelity of the training was that four staff members who worked in the experimental unit also worked on the comparison unit. It is therefore likely that bias was introduced and that the comparison site was contaminated with the effects of the training that these four staff had received on the SLT. Further, Dr. Camp is likely to have discussed some of the principles of his Montessori-based training with staff in the comparison unit. Perhaps, the care provided at Helen s Place was too similar to that in the comparison unit at Stone Gardens. It would be ideal to have a strictly controlled clinical trial to examine the efficacy of the intervention (Prohaska & Etkin, 2010). Staff Outcomes Despite these drawbacks and the fact that some staff wanted further training, the staff overall had high levels of job satisfaction and would recommend others to take a job at Helen s Place. They also had positive comments to make about the SLT. They felt confident about implementing it with PLWD, and believed that it was simple, easy, created a sense of community and helped with communicating with PLWD. They also reported that the SLT provided ideas on how to deal with the challenges faced by PLWD. Some believed that their prior experience with PLWD or earlier training on dementia helped them apply the principles of the SLT more broadly to residents at Helen s Place. They provided examples of how the SLT helped them to calm residents who became agitated and helped to re-direct negative behaviors. The findings from the behavior observations corroborated these findings. There were no observations of resident agitation or distress, even when the fire alarm went off (at a very loud decibel level) and it took almost 20 minutes for the fire department to arrive and turn it off. Staff were able to explain to residents and maintain a calm atmosphere (this, evidently, not being the first time a resident had pulled the alarm). A core element of the SLT relates to offering residents meaningful choices. Observations showed that staff did offer choices to residents. This is particularly true during meals, where the table-side service easily allows residents to see what is available and make a choice. Choices for activities were more limited, with generally only one activity taking place at a time (which, by definition, eliminates or minimizes choice). There were a few instances of staff going over the calendars and making sure the resident knew about activity options both on and off Helen s Place. Some of this is 28

39 tailored to the individual resident (some residents being more interested in particular activities or doing hobbies, or able to leave Helen s Place than others), but most of the time a single activity was announced and residents invited to participate, or not. However, as PLWD on Helen s Place were not very skilled at initiating any kind of activity, if staff were not directing some activity, a number of the residents would sit and do nothing. Sometimes staff was off doing other care activities, but sometimes they simply didn t work at getting an activity going. There was also very little evidence of residents leading activities. There was one instance of one resident helping another during the floral arrangement activity. Observations also demonstrated that residents liked having a personal daily calendar, as several pulled it out multiple times a day to check it. However, the calendar of activities was not always followed (one day they weren t handed out until about 3pm), so there was more in-between time when nothing much was happening. There were several other components of the SLT that were not evidenced. The SLT called for a series of Task Force/Committee meetings that would be resident led, but this, apparently, was not happening (staff questioned didn t know about any resident committees). The activity stations were not used or referred to during the period of observations. There were no intergenerational activities, either on the unit or in the building on the two days that observations were taking place. The Visitor s Center, although present, was not used by anyone, and the medication cart was parked in front of it when it was on the unit. Perhaps, the findings from the behavior observations may be related to staff perceptions that the SLT was most practical to implement with residents with mild dementia and that PLWD could do certain tasks more easily than others. For example, PLWD could choose the clothes they wanted to wear and the activities they enjoyed, but were less able to complete tasks that required higher-level cognitive skills, such as running a committee, maintaining the library or leading/organizing an activity. However, these latter elements of the SLT are the ones that staff did not receive training in, so we are unsure if staff perceptions are related to their lack of training on these modules, or because they actually believe that PLWD cannot perform these higher level cognitive tasks. Another dimension to add to these findings was that staff, overall, had complaints about the lack of communication between them and upper management. We are unsure if this dissatisfaction was related to not being trained on all of the modules of the SLT, or if it was related to other factors such as the challenges of setting up a new dementia care unit, and the staff turnover that occurred during the course of this study. Thus, our recommendations include re-enforcing the issue of having on-going communication and feedback on various levels: between care staff, supervisors and upper management to resolve issues of concern in general; and between trainers, care 29

40 and the original developers to refine the SLT for wider applicability with PLWD. Another recommendation is to test whether some components of the SLT are impractical to implement with PLWD, or to refine the intervention to target it specifically for persons with mild dementia. After further evaluation of the SLT, another strategy could be to develop and test various components of the intervention to better fit with the abilities of PLWD who have differing levels of cognitive acuity (mild, moderate and severe). Resident Outcomes Residents on both units were not significantly different on key outcomes at T1 demonstrating that they were likely on a level plane at the start of the project. During the course of the study, residents in both the experimental and comparison sites demonstrated good QoL, were satisfied with leisure activities and had positive comments to make about living on their units. However, during the study, it appeared that residents at Helen s Place became more satisfied with the quality of the care staff while residents at Stone Gardens became more dissatisfied with their meals and dining experience and had higher levels of depression. This evaluation suggests that residents at Helen s Place positively experienced those components of the SLT that staff was trained on, specifically managing problem behaviors, engaging the passive resident, dining, personal care and choice. Residents appreciated what the care staff did for them at Helen s Place. On the other hand, components of the SLT that staff were not trained in, such as PLWD running a committee, maintaining the library, and leading or organizing a group activity were not evident at Helen s Place, and were not observed or reported by residents. Thus, based on our evaluation of how the SLT was implemented, it appears that the intervention was successful with respect to helping the care staff effectively interact and manage PLWD, and that PLWD appreciated how they were taken care of. Perhaps, if all staff at Stone Gardens had also been trained on some components of the SLT, the residents there might not have had an increase in their depression scores or experienced greater dissatisfaction with their meals and dining experience. However, because our sample sizes were very small in the experimental and comparison units, we cannot make definitive conclusions about the success of the SLT. We recommend that the SLT is implemented in a larger, more generalizable and controlled clinical trial to demonstrate its effectiveness on resident outcomes. In fact, as Prohaska and Etkins (2010) state multiple replications of the intervention with diverse target populations and in environments such as community settings will help establish its application and generalizability. 30

41 REFERENCES Alzheimer s Disease International and World Health Organization. (2012). Dementia: A public health priority in UK: World Health Organization. Barnes, S. (2006). Space, choice and control, and quality of life in care settings for older people. Environment and Behaviour, 38, Bass, D. M., & Judge, K. S. (2010). Challenges implementing evidence-based programs. Generations, 34(1), Boyle, G. (2008). Autonomy in long-term care: A need, a right or a luxury? Disability and Spirituality, 23, Camp, C. J. (2006). Montessori-Based Dementia Programming TM in long-term care: A case study of disseminating an intervention for persons with dementia. In R. C. Intrieri & L. Hyer (Eds.). Clinical applied gerontological interventions in long-term care (pp ). New York: Springer. Camp. C. J., & Nasser, E. H. (2003). Nonpharmacological aspects of agitation and behavioral disorders in dementia: Assessment, intervention, and challenges to providing care. In P. A. Lichtenberg, D. L. Murman, & A. M. Mellow (Eds.). Handbook of dementia: Psychological, neurological, and psychiatric perspectives (pp ). New York: John Wiley & Sons. Camp, C. J., Orsulic-Jeras, S., Lee, M. M., & Judge, K. S. (2004). Effects of a Montessori-based intergenerational program on engagement and affect for adult day care clients with dementia. In M. L. Wykle, P. J. Whithouse, & D. L. Morris (Eds.). Successful aging through the life span: Intergenerational issues in health (pp ). New York: Springer. Camp, C. J., Zeisel, J., & Antenucci, V. (2011). Implementing the I m Still Here Approach : Montessori methods for engaging persons with dementia. In P. E. Hartman-Stein & A. La Rue (Eds.). Enhancing cognitive fitness in adults (pp ). New York: Springer. Chenoweth, L., King, M. T., Jeon, Y. H., Brodaty, H., Stein-Parbury, J., Norman, R., et al. (2009). Caring for Aged Dementia Care Resident Study (CADRES) of personcentered care, dementia-care mapping, and usual care in dementia: A clusterrandomised trial, The Lancet Neurology, 8(4), Dwyer, L., Andershed, B., Nordenfeldt, L., & Ternestedt, B. (2009). Dignity as experienced by nursing home staff. International Journal of Older People Nursing, 4, Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, Fossey, J., Ballard, C., Juszczak, E., James, I., Alder, N., Jacoby, R., & Howard, R. (2006). Effect of enhanced psychosocial care on antipsychotic use in nursing home residents with severe dementia: Cluster randomized trial, BMJ, 332(7544), Gillies, B. A. (2000). A memory like clockwork: Accounts of living through dementia. Aging and Mental Health, 6(2), Harris, P.B. (Ed.). (2002). The person with Alzheimer s Disease: Pathways to understanding the experience. Baltimore, MD: Johns Hopkins University Press. R - 1

42 Judge, K. S., Camp, C. J., & Orsulic-Jeras, S. (2000). Use of Montessori-based activities for clients with dementia in adult day care: Effects on engagement. American Journal of Alzheimer s Disease, 15(1), Koppelman, E. R. (2002). Dementia and dignity: Towards a new method of surrogate decision making. Journal of Medicine and Philosophy, 27, Marson, D., & Harrell, L. (1999). Executive dysfunction and loss of capacity to consent to medical treatment in patients with Alzheimer's disease. Seminars in Clinical Neuropsychiatry, 4(1), Moyle, W., & O Dwyer, S. (2012). Quality of life in people living with dementia in nursing homes. Current Opinion in Psychiatry, 25(6), Moyle, W., Venturto, L., Griffiths, L., Grimbeek, P., McAllister, M., Oxlade, D., & Murfield, J. (2011). Factors influencing quality of life for people with dementia: A qualitative perspective. Aging & Mental Health, 15(8), Nay, R. (2002). The dignity of risk. Australian Nursing Journal, 9, 33. Prohaska, T. R., & Etkin, C. D. (2010). External validity and translation from research to implementation. Generations, 34(1), Sabat, S. R. (2001). The experience of Alzheimer s Disease: Life through a tangled veil. Oxford: Blackwell. Venturato, L. (2010). Dignity, dining and dialogue: Reviewing the literature on quality of life for people with dementia. International Journal of Older People Nursing, 5, Whitlatch, C.J., Judge K. S., Zarit, S. H., & Femia, E. (2006). A dyadic intervention for family caregivers and care receivers in early stage dementia. The Gerontologist, 46, R - 2

43 APPENDIX Supplemental Scales & Behavior Observation Form

44

45 Table 14. Personal Values Scale Individual scales items 0 = Not so important, 1 = Somewhat important, 2 = Very important How important is it for you to HP Mean SG Mean T1 T2 T1 T2 Do things with other people Come and go as you please Be with family Be with friends Do things for yourself Have time to yourself Spend your money how you want Do hobbies and activities that you like Organize your daily routines in your own way Be with children Scale Reliability =.84; Range = 0 (Least important) to 20 (Most important) Table 15. Choice and Autonomy Scale Individual scales items 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time HP Mean SG Mean T1 T2 T1 T2 Can you go to bed when you like Can you take a bath or a shower whenever you want to Are the rules here reasonable Do the people who work here let you do the things you are able to do yourself Are you free to come and go as you are able Scale Reliability =.79; Range = 0 (Least often) to 15 (Most often) A - 1

46 Table 16. Administrative Staff Scale Individual scales items 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time Would you feel comfortable speaking up to the people in charge when you have a problem HP Mean SG Mean T1 T2 T1 T Do your problems get taken care of Are the people in charge available to talk with you Do you know who to go to here when you have a problem Scale Reliability =.96; Range = 0 (Least often) to 12 (Most often) Table 17. Overall Satisfaction with Care and Services Scale Individual scales items 0 = Never, 1 = Hardly Ever, 2 = Sometimes, 3 = Most of the Time HP Mean SG Mean T1 T2 T1 T2 Do you feel like you are getting your money s worth here Would you recommend this place to a family member or friend Overall, do you like living here Scale Reliability =.89; Range = 0 (Least often) to 9 (Most often) A - 2

47 Date Engagement Level Group Engagement Level Individual Quality of Interactions R--> R interaction S-->R interactions S--> R support for choice 1-3 R--> S interactions R & F interactions Calendar Use - Phys verb Cues Mood Time Behavior Observation Form R1 R2 R3 R4 R5 R6 S1 S2 S3 S4 S5 Events to Observe other locations Meals 10 ADL care 11 Use of activity stations 12 Intergenerational 13 Family visits 14 Time between scheduled activities 15 Field trips Task Force/Committee meeting Archival Data Review Notes from task Force Meetings Visitor Center- use, frequency, upkeep Evidence of using w/w/w/w/w to problem solve Communications with families A - 3