The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here:

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

Life Satisfaction and the Oldest-Old. Results from the longitudinal population study "Good Aging in Skåne".


  • Åsa Enkvist

Summary, in English

The overall aim of this thesis was to study life satisfaction (LS) and factors associated to LS in the group aged 78 and above. As an aid in interpreting the findings, Baltes’ and Baltes’ theory of selective optimisation with compensation (SOC) was used. All four studies included the same studypopulation who were followed for three years in the study Good Aging in Skåne, a national, randomized, population-based study of the elderly. Neugartens’ life satisfaction index A (LSI-A) was the outcome variable in all four studies.

In study I the aim was to examine what factors affect LS in the oldest-old in a multi disciplinal perspective. Health-related as well as psychological and socio-economical factors were included. In a logistic multiple regression model the number of symptoms, marital status, health locus of control (HLoC) internal and powerful others, depressive mood and age could predict LS three years later. Independence in functional ability was related to unchanged LS, stratified for age and gender during the 3-year follow-up. The clinical implementations were that recognizing and treating factors that affect LS and are reachable for intervention in this age-group should be highlighted in clinical practice. Relieving symptoms and paying attention to personality factors that modify LS seem to be key-factors in the care of elderly.

Study II set out to describe the prevalence and experienced severity of symptoms and its’ relation to LS in the oldest-old. Scores on the LSI-A were related to scores on a modified version of the Göteborg QoL instrument, covering 32 common symptoms.

The prevalence of symptoms was high: musculosceletal symptoms like pain were reported by 74%; 80% had depressive symptoms and 68 % general fatigue. Less than six percent of men and women reported no metabolic symptoms or symptoms related to the head. In a linear multiple regression model four groups of symptoms could predict LS three years later: depressive-, tensive-, GI-symptoms and musculoskeletal symptoms.

The clinical implication of this study is that careful attention should be paid to the elderly patients’ description of his/her symptoms in the above areas since this could have the potential to increase the patients’ LS.

In study III, we wanted to describe the change in functional ability (FA) in the oldest-old during three years and examine its relation to LS. FA was measured according to Hulter Åsbergs’ ADL-scale. The greatest decline in ADL was seen in the eldest group (87-93 years) in which 51% reported decline during the three years. The group that reported decline in I-ADL had a mean LSI-value of 23.0 compared to 26.4 in the group that was unchanged. A decline in ADL had a stronger negative effect on LS in the younger group. In a linear multiple regression model impairment of FA, depressive mood and the number of symptoms predicted lower LS after controlling for age. Being married and scoring higher on power and internal HLoC was associated with higher LS. The findings point to the importance of rehabilitation and training of this group not only for the purpose of physical restoration but also due to its potential to increase the LS of this population. If further research confirms that the association is not as


strong for the very oldest, then perhaps this needs consideration when planning rehabilitation and treatment for this group.

Study IV set out to explore if cognitive abilities could predict LS three years later in the oldest-old. Correlations between 13 cognitive tests related to six different cognitive domains, and Neugartens’ LSI-A three years later were calculated for the whole sample as well as stratified for factors known to affect cognitive abilities; age, depressive mood, dementia, stroke, sex, education and functional capacity.

A linear multiple regression model was constructed for each cognitive domain separately with LSI-A at re-examination as dependent variable. The models were adjusted stepwise for age, sex, education, functional ability and depressive mood. Processing speed (B=0.118, p=0.020) and spatial ability (B=0.453, p=0.014) remained significantly associated to LS after adjustment. Possible clinical implications were discussed.


  • Geriatric Medicine

Publishing year





Lund University Faculty of Medicine Doctoral Dissertation Series



Document type



Division of Geriatric Medicine


  • Gerontology, specializing in Medical and Health Sciences
  • Medical and Health Sciences


  • Life Satisfaction
  • Oldest-Old



Research group

  • Geriatric Medicine


  • ISSN: 1652-8220
  • ISBN: 978-91-87449-61-1

Defence date

24 September 2013

Defence time


Defence place

Medelhavet, Skåne University Hospital, Malmö


  • Yngve Gustafson (Professor of Geriatrics)