Depression and chronic diseases in old age : understanding their interplay for better health

Sammanfattning: Late-life depression is intricately linked with somatic diseases. This thesis aimed to systematically explore this complex interplay. Specifically, we investigated: 1) the symptom-level interconnectedness between depression and somatic diseases, 2) the association of depression with somatic multimorbidity accumulation, 3) the role of somatic disease burden in depression development, and 4) the association of somatic burden with transitions across depressive states in older adults. Data were gathered from the Swedish National Study on Aging and Care in Kungsholmen (SNAC-K), a populationbased study comprising 3,363 individuals aged 60+ years who underwent clinical assessments over a 15-year follow-up. Study I. Using a network approach, we aimed to describe the interconnectedness between depressive symptoms and somatic disease burden in older people. We found that sadness, pessimism, anxiety, and suicidal thoughts were central to the network, whereas somatic symptoms of depression appeared peripherally with fewer interconnections. When examining the association between depressive symptoms and measures of somatic disease burden, we found that suicidal thoughts, reduced appetite, and cognitive difficulties were bridge symptoms, linking late-life depression with somatic health. Study II. We investigated the impact of depression severity and phenotypes (i.e., affective, anxiety, cognitive, and psychomotor) on the progression of somatic multimorbidity over 15 years. Compared to those without depression, individuals with major (β'year: 0.33, 95%CI: 0.06-0.61) and subsyndromal depression (β'year: 0.21, 95%CI: 0.12-0.30) presented an accelerated accumulation of somatic multimorbidity. An increase in the cognitive phenotype burden (and not in the other three) was associated with faster accumulation of somatic diseases in old age (β'year: 0.07, 95%CI: 0.03-0.10). Study III. We aimed to examine the association between quantitative and qualitative measures of somatic disease burden and the incidence of depression in older adults. Each additional somatic disease was associated with an increased hazard of depression over a 15-year follow-up (HR 1.16, 95%CI: 1.08-1.24). Individuals presenting with disease patterns of sensory/anemia (HR 1.91, 95%CI: 1.03-3.53), thyroid/musculoskeletal (HR 1.90, 95%CI: 1.06-3.39), and cardiometabolic patterns (HR 2.77, 95%CI: 1.40-5.46) had higher depression hazards compared to those without multimorbidity. In the subsample of multimorbid participants, the cardiometabolic pattern remained associated with a higher depression risk (HR 1.71, 95%CI: 1.02-2.84) compared to the unspecific pattern. Study IV. We examined the course of old-age depression by investigating 15-year transitions along the depressive continuum and exploring time-varying factors associated with specific transition patterns. Over the follow-up, 19.1% had ≥1 transitions across depressive states (no depression, subsyndromal depression [SSD], depression), while 6.5% had ≥2 transitions. A higher number of somatic diseases was associated with progression from no depression to both SSD (HR 1.09, 95%CI: 1.07-1.10) and depression (HR 1.06, 95%CI: 1.04-1.08), and with lower recovery rates from SSD (HR 0.95, 95%CI: 0.93- 0.97) and depression (HR 0.96, 95%CI: 0.93-0.99). A richer social network was linked to lower transition rates to depressive states (HRNoDep-SSD 0.81, 95%CI: 0.70-0.94; HRNoDep-Dep 0.58, 95%CI: 0.46-0.73; HRSSD-Dep 0.66, 95%CI: 0.44-0.98), and higher recovery rates (HRSSD-NoDep 1.44, 95%CI: 1.26-1.66; HRDep-NoDep 1.51, 95%CI: 1.34-1.71). Being physically active was associated with higher recovery rates (HRSSD-NoDep 1.49, 95%CI: 1.28-1.73; HRDep-NoDep 1.20, 95%CI: 1.00-1.44). Conclusions. Our findings suggest that several dimensions of complexity characterize the interconnection of depression and somatic disease burden in old age. A symptomlevel characterisation of depression, along with a consideration of subsyndromal severity, may help clarify the comorbidity of depression and somatic diseases, as well as predict health decline in people with depressive symptoms. Similarly, recognizing disease patterns may help improve risk stratification for depression development in clinically complex older adults. Last, the natural course of depression in late life is dynamic and involves complex patterns of transitions through symptom severities, which can be influenced by the time-varying burden of somatic diseases. Developing person-centered care that integrates these complexities could enhance resilience and contribute to better health in old age.