Trial Lecture – time and place
See Trial Lecture.
Adjudication committee
- First opponent: Professor Kjeld Andersen, Syddansk Universitet
- Second opponent: Seniorforsker Eva Biringer, Helse Fonna, Haugesund
- Third member and chair of the evaluation committee: Professor Emeritus Stein E. Opjordsmoen Ilner, University of Oslo
Chair of the Defence
Associate Professor Elfrida Hartveit Kvarstein, University of Oslo
Principal Supervisor
Senior Consultant Cecilie Bhandari Hartberg, Oslo University Hospital
Summary
The overall aim of the study was to gain knowledge of anxiety symptoms among older adults in specialist mental health services. We explored severity of anxiety symptoms as measured with the Geriatric Anxiety Inventory (GAI) and how often anxiety was described at admittance, performed a principal component analysis of the GAI, further explored the relationship between anxiety and depression and investigated if anxiety symptoms predicted cognitive decline, mortality, or persistence of anxiety.
We used data from a regional quality register (KVALAP) in a cross-sectional design for the first three sub-studies. In the first article, data from 473 admissions (414 patients) was included. In the second, 428 patients were included, of which 218 with a diagnosis of depression were also included in the third. In the fourth, longitudinal part (mean follow-up time of 32.6 (SD 10.5) months), there were 201 patients, 150 were re-examined, while 51 patients died before follow-up.
GAI was used to assess anxiety symptoms (range 0-20). Using a cut-off of 8/9, which indicates having a possible diagnosis of anxiety, the prevalence of anxiety symptoms was: depression 65.3%, psychosis 28%, dementia 38.8%, mania 33.3% and surprisingly anxiety 66.7%. Of 236 patients with a GAI score >8, only 22 (9.3%) were reported to have a comorbid anxiety disorder by the treating physician. The PCA of the GAI with oblimin rotation resulted in a two-component solution, labeled 1) worries (explained variance 46.3%, Cronbach’s alpha 0.92) and 2) physical symptoms (explained variance 7.1%, Cronbach’s alpha 0.85). The patients with depression had higher scores than the other diagnostic groups. Among patients with depression, higher GAI scores were significantly associated with the MADRS (beta = 0.233, p 0.002) and suicidality (beta = 0.206, p 0.006). No associations between GAI score at baseline and cognitive decline, death, or persistence of anxiety symptoms were found at follow-up.
Additional information
Contact the research support staff.