Dementia Care Networks Study

Transcription

1 Dementia Care Networks Study August 2002 Louise Lemieux-Charles, PhD Larry W. Chambers, PhD, FACE, FFPHM(UK) Kevin Brazil Rhonda Cockerill Susan Jaglal J. Kenneth Le Clair Carole Cohen Bill Dalziel Barbara Schulman Gertrude Cetinski Marge Dempsey Maureen Montemuro Funding provided by: Canadian Health Services Research Foundation Ontario Ministry of Health and Long-Term Care

2 Principal Investigator: Dr. Louise Lemieux-Charles Department of Health Policy, Management and Evaluation Faculty of Medicine University of Toronto 12 Queen's Park Crescent W. Toronto, ON Canada M5S 1A8 Telephone: (416) Fax: (416) Dr. Larry W. Chambers President and Chief Scientist University of Ottawa Institute on Health of the Elderly SCO Health Service Élisabeth-Bruyère Health Centre 43 Bruyère St. N. Ottawa, ON Canada K1N 5C8 Telephone: (613) Fax.: (613) This document is available on the Canadian Health Services Research Foundation web site ( For more information on the Canadian Health Services Research Foundation, contact the foundation at: 11 Holland Avenue, Suite 301 Ottawa, Ontario K1Y 4S1 Telephone: (613) Fax: (613) Ce document est disponible sur le site Web de la Fondation canadienne de la recherche sur les services de santé ( Pour de plus amples renseignements sur la Fondation canadienne de la recherche sur les services de santé, communiquez avec la Fondation à l adresse suivante : 11, avenue Holland, bureau 301 Ottawa (Ontario) K1Y 4S1 Courriel : communications@fcrss.ca Téléphone : (613) Télécopieur : (613)

3 Dementia Care Networks Study Principal Investigators: Louise Lemieux-Charles, PhD 1 Larry W. Chambers, PhD, FACE, FFPHM(UK) 2 Co-Investigators: Kevin Brazil 3 Rhonda Cockerill 4 Susan Jaglal 5 J. Kenneth Le Clair 6 Carole Cohen 7 Bill Dalziel 8 Barbara Schulman 9 Gertrude Cetinski 10 Marge Dempsey 11 Maureen Montemuro 12 1 Department of Health Policy, Management and Evaluation, University of Toronto 2 University of Ottawa Institute on Health of the Elderly 3 Department of Clinical Epidemiology and Biostatistics, McMaster University; St. Joseph s Health System Research Network, Hamilton 4 Department of Health Policy, Management and Evaluation, University of Toronto 5 Department of Physical Therapy, University of Toronto 6 Division of Geriatric Psychiatry, Queen s University; Regional Geriatric Psychiatry Program PCCC, Kingston 7 Department of Psychiatry, University of Toronto; Community Psychiatric Services for the Elderly, Sunnybrook and Women s College Health Sciences Centre, Toronto 8 Geriatric Medicine, University of Ottawa; Ottawa Regional Geriatric Program; Care of the Elderly Program, SCO Health Service, Ottawa 9 SCO Health Service, Ottawa 10 Alzheimer Society of Hamilton 11 Alzheimer Society of Niagara Region 12 St. Peter s Hospital, Hamilton Acknowledgments: This project was funded by the Canadian Health Services Research Foundation and the Ontario Ministry of Health and Long-Term Care. The investigators would like to acknowledge the genuine support of each network and its members in all phases of the study. Network members assisted in the pilot testing of instruments, the recruitment of caregivers and care recipients, and the completion of survey questionnaires; they also provided insights into the meaning of the results. We would also like to acknowledge the assistance of the research project staff members: Debbie Jones, Julie Bodden, Wendy McGuire, You-ta Chuang, and Carrie McAiney. Gina Lockwood assisted with the data analysis. Professor Dean Behrens gave generously of his time in assisting with the interpretation of the network data

4 Table of Contents Key implications for Decision Makers... i Executive Summary... ii Context...1 Financial and Social Burdens of Care...1 Coordination of Care Processes...3 Networks of Care A Possible Solution...4 Implications...4 Approach...7 Results and Discussion...11 Four Networks...11 Network Effectiveness and Patterns of Service Delivery...12 Perceptions of and Satisfaction with Processes of Care...16 Policy Development...19 Further Research...20 Appendix References...23

5 Key Implications for Decision Makers Planning and provision of dementia services must respond to the nature of the illness and to the unique needs of and relationship between a care recipient and caregiver. Individuals with dementia and their caregivers are a high-risk population. Over 54% of care recipients and 25% of caregivers reported poor health days, and entry into the formal health care system is generally precipitated by a crisis. Dementia care networks enable the exchange of information about care recipients and caregivers. They also facilitate the development of new initiatives. Care recipients and caregivers perceptions of the effectiveness of dementia care networks are influenced by the availability of specialized services and the scope of services in the local community. In communities where specialized services are developed and agencies work together, networks are perceived to have changed the delivery of services through information sharing. The three most significant factors determining care recipients and caregivers satisfaction with care are coordination of care, awareness of services, and family physician care. The planning and provision of dementia care must: Deal with the difficulty that care recipients and caregivers sometimes have accepting that a problem exists; Provide care that addresses the physical, social, emotional, and financial status of caregivers and care recipients; Help identify the most appropriate assistance required in consultation with different care providers, family members, and friends; and Provide services in ways that allow for easy entry into and exit from the service system as care recipients and caregivers needs change. The greater the extent to which care recipients received emotional and informational support from members of their social support network, the more likely they were to report that they were satisfied with the processes of care. The more a caregiver is supported by in-home services, the more likely they are to report satisfaction with awareness and coordination of services. Lack of support for caregivers at risk may lead to the premature institutionalization of care recipients, thereby entailing higher costs to the health care system. At Community Care Access Centres in Ontario, the shift in policy away from providing homemaking and personal care will lower care recipients and caregivers satisfaction with care and increase costs for them. Where network members share multiple ties with members of different groups within the network they are more likely to perceive that the network has been effective in sharing facilities, joint training, shared administrative and client information, and the offering of joint clinical programs. i

6 Executive Summary Networks of care are seen as a way to overcome the fragmentation of services that can result from involving many different providers, types of care, and varieties of specialization. The Dementia Care Networks Study investigated four community-based dementia care networks located in Ottawa, Toronto, Hamilton, and the Niagara region. Our research focused on the evolution, structure, and processes of the networks and on how these networks served the needs of care recipients and caregivers who were using community-based or ambulatory care services provided by acute-care agencies. A model of network effectiveness guided the study. The findings have implications for care recipients, caregivers, health and social service professionals, and government policy makers. Though we studied each network as a whole, we also examined the relationships that existed among groups of agencies within them. The types of relationships we examined were based on activities related to administrative functions (e.g., shared programs) and service delivery functions (e.g., referrals received and sent). General findings indicate that members perceived higher administrative and service delivery effectiveness when network members shared multiple ties with members of different groups within the network (clique overlap) as opposed to the sharing of ties across the network. The centralization of network structure (i.e., the extent to which a network is dominated by a few agencies in terms of service delivery activities) was also found to be related to the perception of service delivery effectiveness. Three major outcome variables were identified as they pertained to the care recipient/caregiver dyad: satisfaction with coordination of services; awareness of services; and satisfaction with family physician care. Care recipients and caregivers assessment of network effectiveness as measured by perception of and satisfaction with processes of care indicated that when care recipients receive emotional and informational support from members of their informal network, they are more likely to be satisfied with the care they are receiving. Caregivers, meanwhile, are more satisfied when the health care system increases the in-home services provided to the people for whom they are caring. ii

7 Specific findings related to network processes and outcomes included: Service delivery indices of case coordination and shared client information were positively correlated with service delivery effectiveness. This suggests that care providers perceive service delivery to be most effective when case coordination and sharing of client information is concentrated within a few agencies. The number of cliques in the referrals received activity was negatively correlated with both administrative effectiveness and service delivery effectiveness. This suggests that overall service delivery effectiveness decreases when agencies receive referrals only from the cliques in which they are involved. Administrative effectiveness was positively correlated with clique overlap in the shared administrative information activity. This finding suggests that a high number of agencies involved in a high number of cliques facilitates the exchange of administrative information, which, in turn, enhances administrative effectiveness at the network level. Clique overlap indices of shared client information and joint clinical programs were positively and significantly correlated with both administrative and service delivery effectiveness. This finding suggests that the degree of influence within cliques advances administrative and service delivery effectiveness. Specific findings related to care recipients and caregivers perceptions of processes of care included: The frequency of in-home services was positively related to satisfaction with coordination and awareness of services the more in-home services a dyad received, the higher the level of satisfaction. The more hours a caregiver spent meeting the personal care needs of a care recipient, though, the less satisfied a dyad was with coordination. If a dyad was receiving help with personal care, both parties were also satisfied with the family physician s awareness of services available for dementia patients. iii

8 Emotional support available to care recipients was positively related to both satisfaction with coordination and awareness of services; it did not influence assessment of physician services. Physical companionship available to care recipients was correlated with assessment of coordination; it did not influence opinions of awareness or physicians. Satisfaction with coordination decreased with the hours of care provided by a caregiver, and increased with the emotional support received by a care recipient and the frequency of in-home services received by a caregiver. Satisfaction with awareness increased with the age of a care recipient, assistance provided to a caregiver with personal care, emotional support given to a care recipient, and frequency of in-home services offered to a caregiver. Satisfaction with physicians care increased with a care recipient s age, when a caregiver and a care recipient were female, and when a caregiver received in-home help. Method We conducted a Care Provider Survey to evaluate service providers perceptions of administrative and service delivery effectiveness, and a Steering Committee Survey to determine steering committee members perceptions of the processes and outcomes of network management. A Care Recipient and Caregiver Survey gauged perceptions of and satisfaction with care processes. We supplemented this research with literature reviews as well as interviews and focus groups held with key stakeholders. iv

9 Context Financial and Social Burdens of Care The impact of Alzheimer s disease and other dementias on society cannot be ignored. The Canadian Study of Health and Aging (CSHA) estimates the prevalence of dementia in Canada to be 252,600 cases and that there are 60,150 new cases per year. CSHA data suggest that the average duration of dementia is 5.8 years (6.7 years for women and 4.6 years for men). 1 Because dementia creates major life changes for families (which are often secondary victims of the disease) and places significant demands on their time and resources, the number of people directly and/or indirectly affected by these diseases is much greater than previously thought. 2 Most communities in Ontario possess an array of home care services and ambulatory clinics designed to care for people with Alzheimer s disease and other dementias. In the CSHA, community-dwelling subjects with dementia received an average of 25 hours assistance per week; those without dementia received help for an average of six hours per week. The reality is, however, that many people with dementia receive no help from the formal health care system, even when their impairment is severe. 3 Family members and friends the informal care system provide 75% to 85% of the care needed by frail Canadian seniors. 4 The caregiving role includes practical assistance with daily activities and personal care, emotional support, mediation with formal organizations and service providers, financial help, and possibly sharing a household. 5 1

10 Using CSHA data, Ostbye and Crosse estimated that the annual health service cost for dementia care in the formal system is six billion dollars. 6 Hux et al., also using CSHA data, reported that the annual cost of care increases with the severity of Alzheimer s disease, from $9,451 for mild cases to $36,794 for severe ones. Institutionalization comprised the main portion of this financial burden for those with severe disease. 7 Financial strain has also been reported by 25% of Canadian informal caregivers due to the cost of supplies, medications, renovations, and equipment. 8 Other financial effects include the economic impact of taking time off from work, and postponed or lost career opportunities. 9 The social consequences of caregiving include restrictions on activities and time spent with family members. 10 Further, in the CSHA one third of employed caregivers reported work disturbances due to caregiving; these disruptions were reported twice as frequently among caregivers of people with dementia. 11 As many as half the people caring for a frail elderly person report physical or emotional strain associated with caregiving. 12 Likewise, in the CSHA s longitudinal analyses, caregivers of people who were newly diagnosed with dementia showed significant increases in depressive symptoms when the care recipients remained in the community, but not when they were admitted to an institution. 13 Caring for an elderly person with dementia over a long term was also associated with chronic health problems among caregivers. 14 Sir Martin Roth was the first clinician/researcher to include in a standardized diagnostic process the involvement of family members and friends who provide care and support for people with 2

11 Alzheimer s disease and other dementias. 15 Today, there is widespread acceptance that, in addition to diagnosis, care must be provided to what Roth identified as the dyad i.e., the care recipient and their informal caregiver(s). Given the wide repercussions of Alzheimer s disease and other dementias, it is widely regarded as crucial that education, research, care, and public policy surrounding these conditions address the concerns and requirements of both care recipients and caregivers. Coordination of Care Processes In its 1994 vision paper, the Alzheimer Society of Metropolitan Toronto emphasized the importance of coordinated care processes for care recipient/caregiver dyads. The Society also noted, however, that ensuring continuity and coordination of care is difficult, not only as a result of clients multiple needs, but because of 1. a lack of knowledge among family members and sometimes even service providers of what can be done at early stages of dementia; 2. a lack of coordination between institutional and community services; and 3. the disorienting nature of the disease itself. The Alzheimer Society of Ontario, through its Blueprint for Action, supports further research on this topic. 16 Caring for individuals who have Alzheimer s disease and other dementias requires that health and social care be linked in some manner to avoid fragmentation of care processes. Such linkage is, however, difficult to achieve. Compounding the problem is the fact that, despite the recognition that processes of care must be better integrated, little research has been done on how 3

12 to better connect institutional and community care, including the relationship between primary care and specialized services. 17 Networks of Care A Possible Solution Networks of care appear to be a viable solution when the specialization and types of care offered to a client group increase and when no single provider offers all the services those clients require. They are also seen as a way to overcome the fragmentation of services that can result from involving many different providers, types of care, and varieties of specialization. Networks appear to be especially important when care recipients and their caregivers have multiple problems and needs, often stretching over long periods. In light of these criteria, people who have Alzheimer s disease and other dementias as well as their caregivers are an ideal target group for network care. 18 In Ontario the formalization of networks labelled as dementia care networks began in The provincial government s Strategy for Alzheimer Disease and Related Dementias includes an initiative (# 9) to establish dementia care networks. A Work Group is currently developing an implementation framework for dementia networks. In January 2002 it launched a resource handbook A Guide to Developing a Dementia Network designed to assist local communities with their formation. 19 Implications At the commencement of the Dementia Care Networks Study in September 1999, the Ontario government released its Strategy for Alzheimer Disease and Related Dementias. Many of these 4

13 initiatives began during our study s data collection phase ( ). Of particular relevance to our study, initiative # 9 (Co-ordinated Specialized Diagnosis and Support) was set in motion while we were nearing completion of our data collection. Initiative # 9 includes the design of local, specialized support networks to assist people with dementia, their families, and caregivers. To that end, communities across the province in 2002 are being offered a small amount of funding to establish dementia care networks. During the course of the Dementia Care Networks Study, the Ministry of Health and Long- Term Care also established the Dementia Networks Work Group. One of the study s coprincipal investigators (Dr. Louise Lemieux-Charles) is a member of the Work Group, as are representatives from the Alzheimer Society of Ontario and other governmental and nongovernmental organizations. As part of its activities, the Work Group wrote and published A Guide to Developing a Dementia Network. Network case studies undertaken by the Dementia Care Networks Study helped to identify and define core elements of network implementation and operation that were subsequently incorporated in this handbook. These included initiation activities, development, evolution, evaluation, boundaries, interfaces, and leadership/champions. Input from members of the Dementia Care Networks Study group is acknowledged in the document. In addition to its online presence at the Ontario Ministry of Citizenship s Web site, the Guide is being made available to local communities that want to set up dementia care networks. Because the Guide was released during the final stage of the Dementia Care Networks Study, it did not have any impact on our data collection activities. 5

14 When considering the development of networks, our findings suggest that the availability of specialized services and the scope of services in the local community influence care recipients and caregivers perceptions of the effectiveness of dementia care networks in relation to service delivery. We found that in communities where specialized services are developed and agencies work together, networks are perceived to have changed the delivery of services through processes of information sharing. Findings also have implications for the delivery of services to individuals with dementia and their caregivers. In September 1999 the Ontario Ministry of Health and Long-Term Care announced that Community Care Access Centres would contract out home care services through a bidding process. Under the new system, both not-for-profit as well as for-profit agencies would be invited to submit proposals. Thus, for the first time requests for proposals for longterm care services were introduced in Ontario. Previously, the ministry had not funded for-profit agencies to provide these services. Because this change was introduced at the start of the Dementia Care Networks Study, its full impact on service delivery, including the operation of dementia care networks, was likely small. However, as of the completion of our study this new system is operating throughout the province. The Dementia Care Networks Study did not attempt to assess the effects of for-profit agencies on the delivery of services or on the operation of dementia care networks. Given these changes, our findings suggest that any system needs to acknowledge and have the ability to respond to the nature of dementia and to the unique needs of and relationship between a care recipient and caregiver (a dyad ). Because of the nature of dementia, services need to be 6

15 provided in ways that allow for easy entry into and exit from the service system as care recipients and caregivers needs change. Also, in the area of service planning, the formal care system must take into account dyads informal care networks. Where caregivers lack support from their informal networks, they may need help in learning how to mobilize their networks or may need additional help from the formal system to compensate for this lack of support. In 2002 the Ministry of Health and Long-Term Care announced it would not increase the amount of resources available to Community Care Access Centres for home care services. This policy affects family members and friends who provide care to people with Alzheimer s disease and other dementias, as well as acute and long-term care facilities. The shift in Community Care Access Centre policies away from homemaking and personal care will affect care recipients and caregivers satisfaction with care and increase the costs of care for them. We found that the more a caregiver is supported by in-home services, the more likely they will be to report satisfaction with awareness and coordination of services. Lack of support for caregivers who are at risk may lead to the premature institutionalization of care recipients, thereby entailing higher costs to the health care system Approach Networks are complex entities and there is a lack of empirical evidence revealing why some are effective while others are not. 20 The study of networks as a unit of analysis focuses on the structure of relationships among organizations, where the network consists of nodes linked by a set of relationships. In our study, we investigated network management by asking: What conceptual frameworks can be used for evaluating networks? 7

16 With regard to measuring network effectiveness, we asked: How can the components of care networks be measured? What care processes result in care recipient/caregiver satisfaction with care received? How do care providers assess the effectiveness of networks? Monitoring of networks, meanwhile, raised these questions: What indicators reveal that networks of care have started? What indicators reveal that care networks are being sustained/maintained? We studied four community-based dementia care networks and targeted care recipients and caregivers who were using community-based services or ambulatory care services provided by acute-care agencies. Membership ranged in size from 17 to 24 agencies/programs. We examined the networks evolution over a two-year period, their patterns of relationships, and perceived effectiveness. Case studies of the four networks were developed to describe their origins and evolution, key events in their development, defining characteristics (including membership, structure, resources, accountability, and leadership), and network activities, processes, and achievements. A cross-sectional survey using social network analysis was used to describe patterns of service delivery and the mix and distribution of services among providers in the four networks. 21 Within each network we analyzed the degree to which agencies are connected ( density ), the extent to which a network is dominated by one agency ( centrality ), and the presence of cliques (agency groupings) and clique overlap (the degree to which members share multiple ties with members of more than one grouping). 8

17 In health service organizations, two components strongly influence the effectiveness of care delivery: administrative activities and service delivery activities. Using Provan s Measuring Inter-Organizational Relationships questionnaire, we asked key informants from each agency to indicate how their organizations were linked with other groups in their respective networks. 22 We classified these network relationships as administrative links and service links. Our analysis of network effectiveness focused on service providers perceptions of administrative and service delivery effectiveness, and on steering committee members perceptions of the processes and outcomes of network management. Figure 1 describes the conceptual framework that guided our study. Figure 1. The conceptual framework of the relationship between network structure and network effectiveness and processes in the Dementia Care Networks Study Administrative links - Shared staffs/volunteers - Shared facilities - Joint training programs/workshops - Shared administrative information Service links - Referrals sent - Referrals receive - Case coordination - Joint clinical programs - Service contacts - Shared client-related information Network structure - Density - Centralization (degree of influence) - Clique overlap Network Effectiveness Network level - Meeting network s goals - Promoting collaboration among members Agency level - Meeting agency s goals - Promoting collaboration with other members Caregiver/Care recipient (Dyad) level - Perception of and satisfaction with processes of care 9

18 A cross-sectional survey of care recipients and caregivers was also conducted to gauge dyads perceptions of and satisfaction with care processes. Across the four network sites 267 dyads were surveyed (n=574). Community Care Access Centres, family physicians, and local societies of the Alzheimer s Association selected the dyads. Based on a literature review, focus groups, and expert opinion, we hypothesized that there are discernable relationships between specific characteristics of care recipients and caregivers and their experiences with dementia care networks. Our conceptual model (see figure 2) suggests that care recipient and caregiver characteristics demographic variables (age, gender, living arrangements, and general health status), personal support networks (number of friends and group affiliations), emotional supports, and service utilization (type and frequency of services received) influence perceptions of and satisfaction with processes of care. Figure 2. Conceptual model of processes of care (care recipient/caregiver level of analysis) Demographics - Age - Gender - Marital status - Education - Time spent caregiving - General health status - Employment # of hours Perceptions of Processes of Care - Satisfaction with process - Continuity - Coordination of care - Want/expressed and felt need - Appropriateness of services - Timeliness - Awareness of services - Availability - Experience with physicians Personal Support Networks - Number of friends/family living nearby - Frequency of contact - Group affiliation (religious, self-help, social) Emotional Supports - Affection - Positive social interaction - Emotional and informational support Service Utilization - Services received - Use of services - Services paid for living arrangement 10

19 A main contribution of the Dementia Care Networks Study was the development of operational measures of network effectiveness. We reviewed the literature on caregiving for people with Alzheimer s disease and other dementias, as well as other relevant literature and studies, in order to identify instruments that could be used or adapted for use in evaluating dementia care networks. In many instances, we developed new instruments, and our work forms the foundation for future use of these new instruments in different settings and with different actors. All instruments were pilot tested and the written case studies were validated with members of each of the networks steering committees. There is now a need to apply the new instruments beyond the four dementia care networks that participated in our study. Appendix 1 describes the activities and measures related to each area of investigation Results and Discussion Four Networks The four networks we studied were representative of collaborative entities in which agencies voluntarily agree to work together to improve service delivery by realigning some aspects of their services and contributing resources to fund joint initiatives. They were bounded networks because they had defined their membership through interest and representation rather than according to the representativeness of all existing relationships. At the time our study ended, none had formalized as a network entity with a system mandate or an overall, centralized infrastructure funded by network members. 11

20 Network Effectiveness and Patterns of Service Delivery Table 1 summarizes and compares network density and centralization across networks. Table 1. Summary of network density and centralization Network Density A B C D Administrative links Shared staff Shared facilities Joint training Shared admin. info Mean SD Service delivery links Referrals sent Referrals received Case coordination Shared client info Service contracts Joint clinical programs Mean SD Centralization Administrative links Shared staff Shared facilities Joint training Shared admin. info Mean SD Service Delivery Links Referrals sent Referrals received Case coordination Shared client info Service contracts Joint clinical programs Mean SD

21 We measured the importance of each agency within their respective network (degree of centrality), and then calculated the degree of centralization of each network. Degree of centralization provides information about how information flow is organized around core agencies within a network. Network density measures provide information about networks overall structure and cohesion. We found that across all four networks, network density, as measured by number of links among agencies, was low in relation to the number of administrative links and higher in relation to the number of service delivery links. Centralization, as measured by the degree to which specific organizations are central within a network, was low to moderate across the networks (the higher the index, the more centralized the network). When examining the number of cliques, the largest number of agencies involved in a large number of cliques occurred in case coordination and sharing of client-related information. Table 2 compares care providers perceptions of network effectiveness by network. Table 2. Summary of network effectiveness Network effectiveness A B C D Administrative effectiveness Mean SD Service delivery effectiveness Mean SD When examining the relationship between network effectiveness and network structure, we found that network density was not significantly correlated with either administrative or service delivery effectiveness. These results ran counter to our theoretical assumption that a high level of integration among members as measured by network density is related to perceived network effectiveness. Network centralization was not significantly correlated with administrative 13

22 effectiveness, suggesting that a network s administrative structure of power and influence is unrelated to administrative effectiveness. However, both the network centralization indices of case coordination and shared client information were positively correlated with service delivery effectiveness. This suggests that the perception of service delivery effectiveness is greatest when case coordination and sharing of client information is concentrated within a few agencies. We also found that the number of cliques in the referrals received activity was negatively correlated with both administrative effectiveness and service delivery effectiveness. This result suggests that overall service delivery effectiveness was decreased when agencies only received referrals from the cliques in which they were involved. Clique overlap in shared administrative information activity was, however, positively and significantly correlated with administrative effectiveness. This correlation suggests that a high number of agencies involved in a high number of cliques facilitated the exchange of administrative information, which, in turn, increased administrative effectiveness at the network level. In addition, clique overlap indices of shared client information and joint clinical programs were positively and significantly correlated with both administrative and service delivery effectiveness. This finding suggests that service delivery effectiveness could be increased when a few agencies have great influence across cliques. Table 3 illustrates the relationship between clique overlap and administrative and service delivery effectiveness. 14

23 Table 3. Relationship of clique overlap to administrative effectiveness and service delivery effectiveness Clique overlap Administrative effectiveness Service delivery effectiveness Administrative links Shared staff Shared facilities.997**.894+ Joint training * Shared admin info.983**.916* Service delivery links Referrals sent Referrals received Case coordination Shared client info.921*.988* Service contracts Joint clinical programs.821+*.844+ N=4, + p<.10;* p<.05;**p<.01 (one-tailed test) In sum, we found there were few significant relationships between administrative network structure and administrative effectiveness. More importantly, the clique overlap of shared administrative information activity played a key role in promoting both administrative and service delivery effectiveness. Our results suggest that the degree of influence within cliques, instead of within an entire network, strongly promotes administrative and service delivery effectiveness. Our results also show that among the four aspects of network structure examined in our study (density, centrality, cliques, and clique overlap), clique overlap plays a principal role in promoting administrative and service delivery effectiveness. The centralization of network structure in service delivery activities also promoted service delivery effectiveness in our sample. 15

24 These results are interesting in light of the assumption that network structures increase integration, whereas our results demonstrate that a high level of integration does not necessarily lead to high network effectiveness in human service organizations. Instead, clique overlap and centralization play principal roles in promoting network effectiveness. These results are consistent with previous finding. 23 Perceptions of and Satisfaction with Processes of Care Three major outcome variables were identified as they pertained to care recipients: satisfaction with coordination of services; awareness of services; and satisfaction with family physician care. Overall, respondents reported satisfaction with the care they were receiving. Care recipients were primarily female (64%), had moderate levels of dementia (62%), and had an average age of 81 years. Caregivers were also primarily female (72%) either a daughter or wife and had an average age of 63 years. In Network A, care recipients were more likely to live alone (27%) than in the other sites; this was statistically significant (p= 0.006). Similarly, except in Network A, over 70% of caregivers lived with care recipients. Overall, 54% of care recipients reported their health as fair or poor and experienced an average of 18.4 unhealthy and 13.0 disability days per month, compared with caregivers (25% fair or poor health and 12.1 unhealthy and 3.8 disability days). On average, 20%, 49%, and 60% of care recipients received in-home services such as nursing, personal care, and cleaning respectively. These percentages did not vary by site and the majority of care recipients (60%) paid for services. There were, though, large variations across the sites 16

25 regarding whether or not caregivers received in-home services. In Network A, only 31% of caregivers received in-home services, compared with 43% in C, 54% in B, and 66% in D (p =0.001). The frequency of in-home services for caregivers was highest in D and C, and lowest in A. The Wenger scale was used to measure personal support networks. In Wenger s study, all sites had high proportions of family dependent (37%) and locally integrated networks (29%). 24 According to Wenger, these networks make more demands on community services than does residential care. In addition, they provide high levels of practical help; however, with increasing dependency, demands on the caregiver (who is usually a daughter or wife) increase. There were, however, differences among the sites we studied. In Network A, where a much smaller proportion of caregivers live with care recipients, caregivers were significantly less likely to assist directly with personal care or household activities. The following points emerged when we examined the relationships between processes of care and other variables (note that we did not discover differences among the sites surveyed). 1. The frequency of in-home services was positively related to satisfaction with coordination and awareness of services, such that the more in-home services received by a dyad, the higher the level of satisfaction. As well, the more hours a caregiver spent meeting the personal care needs of a care recipient, the less satisfied a dyad was with coordination. However, if a dyad was receiving help with personal care, then both parties were also satisfied with the family physician s knowledge and awareness of services available for dementia patients. 17

26 2. Emotional support available to care recipients was positively related to both satisfaction with coordination and awareness of services; it did not influence assessment of physician services. Physical companionship available to care recipients was correlated with assessment of coordination; it did not influence opinions of awareness or physician services. 3. Satisfaction with coordination decreases with the hours of care provided by a caregiver, and increases with the emotional support received by a care recipient and the frequency of in-home services received by a caregiver. 4. Satisfaction with awareness increases with the age of a care recipient, assistance provided to a caregiver with personal care, emotional support given to a care recipient, and frequency of in-home services offered to a caregiver. 5. Satisfaction with the care provided by family physicians increases with the age of a care recipient, when a caregiver and a care recipient are female, and when a caregiver receives in-home personnel help. Our findings led us to conclude that the more a caregiver was involved in the direct care of a care recipient, the less satisfied they were with coordination of services; whereas, if more inhome services were provided by the health care system, caregivers satisfaction with coordination and awareness of services increased. Meanwhile, if a care recipient received emotional and informational support from members of their social support network, they were more likely to be satisfied with processes of care. 18

27 Our findings highlight the fact that the relationship between the informal care system and the formal care system is dynamic, and that attention needs to be given to increasing caregivers capacity to provide care and respond to risks associated with caregiving. In order to optimize satisfaction with coordination and awareness of services, the health care system needs to provide appropriate levels of personal support services so that caregivers can provide emotional support. In addition, increasing the knowledge and awareness of family physicians about dementia care services will positively influence the perception of the processes of care. Another major finding establishes the importance of the location of the network or site, which was independent of the other variables. Policy Development By working with policy makers throughout the study to identify key elements in the formation of networks throughout the province, the research team was able to inform concurrent policy development. The team accomplished this by 1. providing pre-existing evidence and research to inform the development of the policy implementation at its outset; 2. providing a vehicle for bilateral exchange of information to inform research directions as well as policy development; and 3. by collaboratively identifying future directions in both research and policy implementation. The Dementia Care Networks Study affirms the critical importance of understanding the significant contributions of specific communities in terms of effecting network outcomes. This result underscores the critical role played by over-arching policy directions, while allowing 19

28 implementations that are sensitive to the foundation and nature of community activities and developments. Further Research The Dementia Care Networks Study examined networks of care that developed as a result of a felt need within particular communities to coordinate services. Their formation varied based on the kind of specialized services involved and on the type and scope of services that were available. By and large network members perceived membership in a formal network as facilitating the sharing of information about clients at the agency level; however, our study found that members did not believe that information sharing led to changes in service delivery. Meanwhile, in communities in which network members developed and operated specialized services together, members perceived the network as having changed the delivery of services through information sharing processes. Further research is needed to determine the conditions under which improving relationships among agencies will lead to enhanced service delivery. Networks developed initiatives that members believed would not have been possible without the network structure. Having a formalized set of relationships among members facilitated the development of these initiatives. Further research is needed to determine the extent to which a formal network structure facilitates the diffusion of these initiatives. Networks were also found to strengthen existing links and to develop new, effective coordination at the system level. Mapping network relationships must occur between organizations at two levels: administrative and clinical. Mapping must address the critical components of each level. 20

29 The research design component of the Dementia Care Networks Study identified and developed an evaluation framework, tools, and methods that are key to evaluating policy implementation at the level of consumers, networks, and the health care system. Monitoring the development and operation of networks is required in order to better inform provincial policies. The tools that were developed during the study will be adapted for use in the formal evaluation of the networks being developed as a result of the Ontario government s policy initiatives. Results of these evaluations will be compared with our present findings. For example, the study of sub-groups within networks is more likely to explain effectiveness than is integration across networks. We found that clique overlap played a key role in promoting administrative and service delivery effectiveness. To gain a better understanding of patterns of service delivery, networks will therefore be encouraged to examine overlapping groups within their communities whose members share multiple ties with the members of one or more other organizations. 21

30 Appendix 1 Dementia Care Networks' Study activities October 1999 to April 2002 Perspective Network members Type of activity and time period Preliminary interviews - December 1999 to April 2000 Instrument/ data sources Network Dimensions and Processes outline Key issues addressed - Agency information - Network dimensions and processes Care recipients Care Recipient and Caregiver Survey - October 2000 to June 2001 Care Recipient Questionnaire - Characteristics - Personal support network - Emotional support - Perceptions of processes of care Caregivers Network service providers Care Recipient and Caregiver Survey - October 2000 to June 2001 Care Provider Survey - May to September 2001 Caregiver Questionnaire Care Provider Questionnaire - Characteristics - Personal support network - Emotional support - Agency information - Administrative links - Service links - Network effectiveness Network steering committee members Survey of Steering Committee Members - March to April 2002 Steering Committee Questionnaire - Importance of network to member organizations - Network leadership - Network decision making - Value of network to service delivery Dementia care networks Network descriptions and evolution - December 1999 to March 2002 Agency and network documents, literature, and interviews Network - Development and evolution - Characteristics - Participation - Champions - Activities - Future directions Community Care Access Centre managers Chart-assisted interviews - April to November 2001 Chart-assisted interview with service provider Care recipient and caregiver - Use of services - Referrals made to other service providers - Contact with other service providers Policy-makers and other key stakeholders Collaboration throughout the study 22

31 References 1 Canadian Study of Health and Aging Working Group. (1994b). Canadian Study of Health and Aging: Study methods and prevalence of dementia. Can Med Assoc J, Canadian Study of Health and Aging Working Group. (in press). Patterns and health effects of caring for people with dementia: The impact of changing cognitive and residential status. Gerontologist. 2 Maslow, K. & Whitehouse, P. (1997). Defining and measuring outcomes in Alzheimer disease research: Conference findings. Alzheimer Disease and Associated Disorders, 11 (Supp. 6), 189. Schulz, R., O Brien, A. T., Bookwala, J. & Fleissner, K. (1995). Psychiatric and physical morbidity effects of dementia caregiving: Prevalence, correlates, and causes. Gerontologist, 35, Canadian Study of Health and Aging Working Group. (in press). Canadian Study of Health and Aging Working Group. (1994a) Patterns of caring for people with dementia in Canada: The Canadian Study of Health and Aging. Can J Aging, 13, Grunfeld, E., Glossop, R., McDowell, I. & Danbrook, C. (1997). Caring for elderly people at home: The consequences to caregivers. Can Med Assoc J, 157, Chappell, N. L. (1985). Social support and the receipt of home care services. Gerontologist, 25, Peek, C. W., Zsembik, B. A. & Coward, R. T. (1997). The changing caregiving networks of older adults. Res Aging, 19, Novak, M. (1997). Aging and society: A Canadian perspective (3 rd ed.). Toronto: ITP Nelson. Thompson, E. H., Futterman, A. M., Gallagher-Thompson, D., Rose, J. M. & Lovett, S. B. (1993). Social support and caregiving burden in family caregivers of frail elders. J Gerontol, 48, S245 S Keating, N., Fast, J., Frederick, J., Cranswick, K. & Perrier, C. (1999). Eldercare in Canada: Context, content and consequences. Ottawa: Statistics Canada. Horowitz, A. (1985). Family caregiving to the frail elderly. Annu Rev Gerontol Geriatr, 5, Ostbye, T. & Crosse, E. (1994). Net economic costs of dementia in Canada. Can Med Assoc J, 151, Hux, M. J., O Brien, B. J., Iskedjian, M., Goeree, R., Gagnon, M., Gauthier, S. (1998). Relation between severity of Alzheimer s disease and costs of caring. Can Med Assoc J, 159, Keating et al. Alcock, D., Danbrook, C., Walker, D., Hunt, C. (1998). Home care clients, providers and costs. Can J Public Health, 89, Keating et al. Wagner, D. L. & Neal, M. B. (1994). Caregiving and work: Consequences, correlates, and workplace responses. Educational Gerontology, 20, Keating, et al. 11 Canadian Study of Health and Aging Working Group (1994a). 12 Schulz, R., Mittelmark, M. B., Burton, L., Hirsch, C. & Jackson, S. (1997). Health effects of caregiving: The caregiver health effects study an ancillary study of the cardiovascular health study. Ann Behav Med, 19, Canadian Study of Health and Aging Working Group. (in press). 14 Canadian Study of Health and Aging Working Group. (1994a; in press). 15 Roth, M., Huppert, F. A., Tym, E. & Mountjoy, C. Q. (1988). CAMDEX: The Cambridge Examination for Mental Disorders of the Elderly. Cambridge: Cambridge University Press. 23