Search

Introduction: Implementing integrated care in primary care settings is essential to establish a sustainable health care system for older adults living at home. Measuring integrated care performance facilitates implementation and improvement of integrated care services in daily practice, yet valid instruments are still lacking. Based on the Rainbow Model of Integrated Care (RMIC) a measurement tool (MT) has been validated for renal care patients and care providers. In the present study we aimed to validate the RMIC-MT for healthcare professionals working in an integrated primary elderly care setting. Theory/Methods: The RMIC-MT is a 36-item questionnaire, covering all aspects of integrated care according to the RMIC. Per integration domain professionals are asked to rate a set of questions. In a cross-sectional study, the RMIC-MT was sent to local networks of primary elderly care professionals in the Netherlands by email between February 2020 and February 2021. Confirmatory factor analysis (CFA) was used to build upon the previous results of the RMIC-MT for renal care by using the standard fit indices: root-mean-square error of approximation (RMSEA) (=0.06), standardized root-mean-square residual (SRMR) (=0.08), comparative fit index (CFI) (=0.90) and Tucker-Lewis index (TLI) (=0.90). Maximum likelihood was used to estimate the model parameters and goodness-of-fit indices. Results: 323 professionals filled out the RMIC-MT, the majority were general practitioners, community nurses, practice nurses and case managers. Median time needed to complete the RMIC-MT was 14 minutes. Some professionals did not complete the questionnaire because the questionnaire was too long or perceived as irrelevant. The model passed the goodness-to-fit test by confirmatory factor analysis: RMSEA=0.046, SRMR=0.055, CFI=0.895, TLI=0.882. The factor structure of 9 categories with a total of 36 items was thereby confirmed. Conclusions: We successfully validated the RMIC-MT for the primary elderly care setting. We adjusted the name of one category from 'triple aim' to 'outcome measurements'. Application of the RMIC-MT enables professional networks in this setting to evaluate and improve their care integration and contribute to better quality and more sustainable primary elderly care. Implications for applicability/transferability, sustainability, and limitations: The sample size was small but sufficient for the CFA. Recruitment was difficult which may reflect difficulties in future use in daily practice. The RMIC-MT was considered too long by some professionals. Developing a shorter version may be a promising strategy to increase completion in daily practice. RMIC-MT application can enable professionals to improve people-centered integrated primary care by using the feedback reports for improvement plans. Feasibility and added value of the tool should be studied after this small-scale implementation. Broader implementation of the RMIC-MT for elderly care should be investigated, including other contexts and countries. [ABSTRACT FROM AUTHOR]

Introduction: Rearrangements in primary care are needed to cope with aging populations, resulting in an increased prevalence of dementia and increasing health care costs. An essential strategy to establish a sustainable health care system for people with dementia is implementing integrated care in primary care settings. With DementiaNet, an integrated network-based approach, we facilitated collaboration between healthcare professionals from medical, care and social domains, and helped increasing their dementia expertise. Our aim is to evaluate the effects of the DementiaNet approach on quality of care, network collaboration and number of crisis admissions. Theory/Methods: In a longitudinal mixed methods multiple case study, 34 primary care networks participated; they adopted the DementiaNet approach in which a two-year tailor-made program was offered. Networks started between 2015 and 2020 and follow-up ended in 2021. Yearly, networks participated in data collection on their self-reported quality of care and crisis situations. Network collaboration was assessed yearly with scores on the eight integration domains of the Rainbow Model of Integrated Care based on semi-structured interviews with network leader(s). Descriptive statistics were used to identify changes in mean scores over time. Network characteristics and interview data was used to gain deeper understanding of the quantitative outcomes. Linear mixed models will be used to analyze the differences over time for quality of care and network collaboration. Results: Networks consisted on average of nine professionals (mainly a general practitioner, community nurse, practice nurse and case manager). Follow-up was 2-6 years. On average, quality of care of the networks increased for all measured indicators, of which the percentage of people discussed in a multidisciplinary meeting increased the most from 55% (range 0-100%) to 80% (range 48-100%). Advanced networks had less crisis situations compared to starting networks, 25% versus 30% (p=0.038) of their caseload of persons with dementia had a crisis situation. The quality of care was not associated with the number of crisis situations. Network collaboration increased as well over time and these results sustained or further improved after the two-year program ended. Conclusions: The DementiaNet approach improved quality of care and network collaboration and decreased the number of crisis situations of local primary dementia care networks. These effects sustained after the two-year program ended. Because of the improved collaboration, more advanced networks may be able to better communicate and act sooner in case of an oncoming crisis. Implications for applicability/transferability, sustainability, and limitations: Evaluation of a complex program such as DementiaNet requires an innovative design, flexibility and time. Previous studies did not find improvements after implementing integrated care, we showed that these improvements take at least two years to establish. Nevertheless, these effects were successfully sustained by the networks after the two years of support. Data is partly self-reported by network members. Despite the detailed data-collection protocols and definitions that were provided, the quality of the data may have suffered because of this strategy. Future research should focus on identifying if the DementiaNet approach is able to decrease healthcare costs. [ABSTRACT FROM AUTHOR]