Longitudinal multiple case study on effectiveness of network-based dementia care towards more integration, quality of care, and collaboration in primary care.

Introduction: The rising incidence and policies to keep dementia patients in their own homes are increasingly putting pressure on primary care systems and budgets. The DementiaNet program stimulates development of primary care networks of medical, nursing and welfare professionals for community-dwelling dementia patients through practice facilitation. This study aimed to provide insight into the merits and drawbacks of this program, mechanisms and which contextual factors influenced them. Methods: In this longitudinal mixed methods multiple case study, primary care professionals with shared caseloads of dementia patients, were enrolled to form networks in the DementiaNet program. Data collection consisted of continuously kept logs, yearly structured interviews to rate the network maturity score (range 0-24), yearly quality of care assessment through a sum score of quality indicators (range 0-100), and in-depth interviews regarding experiences and perceived effects. Quantitative data were analyzed through mixed models; qualitative data with thematic analysis. Results were integrated by combined interpretation. Results: Thirteen networks were successfully initiated in the program, consisting of a median of 9 professionals. Overall, the networks showed an average yearly increase of 2.03 (95% CI 1.20-2.96) on network maturity and 8.45 (95% CI 2.80-14.69) on quality indicator sum scores. Mixed methods interpretation revealed patterns regarding network and contextual factors enabling the transition towards more mature networks and better quality of care. Discussion: Time trends in network maturity and quality of care indicators showed overall improvements. Several enabling factors for the transition to network-based care were identified including strong and adequate leadership (preferably with leaders from primary care practice), high involvement of motivated primary care physicians, high acquaintanceship with other network members, and network seize with a compact network that operates in a relatively small geographical area. Conclusions: Participation in the DementiaNet program was associated with increased network maturity and subsequent beneficial effects on quality of care. Adaptation towards a more mature network seemed to favour quality of care improvements. Lesson learned: The multiple case study design demonstrated its value in the evaluation of DementiaNet as example of a complex health care innovation by incorporating interactions and contextual dependency. Limitations: The main limitation of the study was the limited follow-up. The DementiaNet approach demands considerable changes in behaviour and practice from large numbers of actors; such adaptations require time and will be different per network. Indeed, networks work different in speed of change and improvement goals. Nonetheless, these initial results show improvements even over one and two year timeframes. Suggestions for future research: A succeeding study will be initiated to ensure longer followup of existing networks, while simultaneously expanding the number of networks. [ABSTRACT FROM AUTHOR]