Tese de Doutoramento Saúde Pública Global 2022 Faculdade de Ciências Médicas, Universidade NOVA de Lisboa

Introduction: Relationship quality between a person with dementia and a family carer potentially promotes health and quality of life outcomes for both. However, evidence supporting practical advice on how to tailor psychosocial interventions to relationship aspects is still limited. This may be due to overreliance on methods that miss the complexity of individual and dyadic characteristics and changing circumstances throughout the course of dementia. Little is also known regarding risk profiling i.e., who could benefit the most from interventions. This thesis aimed to provide a better understanding on the role of relationship quality in dementia, using both quantitative and qualitative methods. Methods: We used data from the multicentre and multimethod Actifcare (ACcess to TImely Formal Care) EU-JPND project (involving Germany, Ireland, Italy, Norway, Portugal Sweden, The Netherlands and United Kingdom). Evaluations included qualitative explorations and a one-year prospective cohort study (with baseline assessments and follow-ups at 6 and 12-months) of 451 dyads (n = 902) of community dwelling-people with mild-to-moderate stages of dementia and their informal (family) carers in the Actifcare European countries. We complemented part of the analyses with extended longitudinal data from the second Actifcare wave (36-months follow up). A varied set of methods were employed, namely regression models, latent growth model (LGM), latent profile analysis (LPA) and thematic analysis. Results: Our findings suggest that relying only on a cross-sectional snapshot or a short follow-up period to study relationship quality may be misleading. Whilst with a 1-year period relationship quality scores for the people with dementia did not change, the Portuguese data from the 3-year longitudinal study suggested that both carers and people with dementia perceived a deterioration in relationship quality overall. Factors influencing the perceived relationship quality also differed over time. At baseline, a more positive perspective of the relationship quality was associated with carers’ lower stress and higher sense of coherence, and being a spouse. Over one year, higher levels of neuropsychiatric symptoms were linked to decline in carers’ relationship quality, whereas social support was associated with more positive relationship quality trajectories. Overall quantitative and qualitative findings showed that changes in the relationship quality may be perceived differently by the person with dementia and their family carer, and that divergent perspectives often lead to tension within the relationship. The results also highlighted differences between spouse and non-spouse carers’ perceived relationship quality. Our analyses revealed the individual trajectories and risk profiles in terms of relationship quality. The more vulnerable profiles among people with dementia and carers were determined by higher levels of carers’ stress and lower sense of coherence, among other factors, and showed a decline in quality of life over time. Finally, the mixed-methods approach demonstrated that not all aspects of relationship quality are affected in the same way, and that changes occur at different stages. Closeness feelings played an important protective role in the likelihood of institutionalisation of the person with dementia. Qualitative data revealed that ‘mixed-feelings’ (ambivalence) can be a complex experience, arguably more difficult to manage than negative feelings alone. Furthermore, our qualitative findings pinpointed perceived social networks, particularly families, and receiving formal services’ support as facilitators of positive appraisals of relationship quality. Conclusion: This research contributed to a better understanding of the role of relationship quality in dementia and underlined its close link with quality of life. Practitioners may offer responsive and meaningful interventions based on a straightforward risk appraisal of person-carer dyads. Given societal limited resources, it can also inform the strategy of developing psychosocial interventions by targeting potentially modifiable aspects of the relationships. Optimising aspects of relationship quality, monitoring psychosocial needs and cherishing the resources of different dyads and families, will potentially improve quality of life and health outcomes in dementia