and changes that occur in later life (59, 60). However, a great part of this survey sample reported good mental health. One possible explanation may be that those with more severe mental conditions were among the non- respondents. A great deal of the individuals in the current study did not live alone, and in general, they reported to have good overall health, which may be parts of the explanation to the positive result regarding mental health. Still, one factor to have in mind is that mental problems in older people are claimed to be unrecognized and under- reported (61).
The assessed sense of coherence in the current study sample was stronger in the group with higher self-care ability, although it did not emerge as a predictor in the logistic regression analysis. Exploring the older persons’ sense of coherence may provide a better understanding of how some older people appear to have more strengths and abilities to compensate for age-related declines and losses than others (16, 62). Sense of coherence is a problem- solving and resource-oriented concept, focussing on the individuals’ perceptions of life as comprehensive, manage- able and meaningful (14), and is shown to be closely related to perceived health as well as quality of life and life satis- faction (15, 63). The obtained results showed that sense of coherence is an important issue in relation to self-care ability.
However, in addition to variables like marital status, type of dwelling, profession, social contact, frequency of contact, the size of the residential municipality did not emerge as predictors for self-care ability in this study.
Methodological considerations
There were 35% of the invited persons that responded the questionnaire. This is a low response rate compared to another Norwegian postal survey (48) that reported a response rate of 43%. However, it was only sent one reminder to those who did not respond according to rules for distribution of randomized addresses in Norway.
The number of male respondents in the study sample was higher compared to the invited sample, which may indicate that the males in this study were more willing, or capable, to complete the questionnaire. The mean age of the respondents was also significant lower compared with the dropouts. A reported methodological problem related to population health-surveys among older people is that the final sample tends to constitute the youngest, health- iest and least vulnerable individuals (64, 65). Recruiting older persons, especially frail, rural-dwelling women, are found to be a challenging and persistent problem, and the individuals with the most severe health conditions are often underrepresented (66). Consequently, a generaliza- tion to the population, in general, has to be performed with caution. Maybe a more balanced sample, e.g. with more women and very old persons responding, would
show some other results regarding self-care ability, because of more severe mental or physical health conditions. Nevertheless, although a considerable part of the invited persons did not complete and return the questionnaire, the remaining sample is considered to represent an important database for investigating several conditions concerning the older population living at home in this part of Norway.
However, with a cross-sectional design it is not possible to find causal connections between self-care ability and related factors. But the result has revealed that a number of aspects are closely related to self-care ability, either as the affecting or the affected factors. For instance, a poorer nutritional condition may cause, but also be the result of, declined health and lower self-care ability.
The aim of this study was to investigate several aspects related to many persons’ self-care ability and health. Although the methodological approach was appropriate, one problem with such comprehensive postal surveys is the low response rate. However, a response rate around 30% is not unusual in similar studies among older people (6, 35). In a Canadian study implemented in a corre- sponding sample (67), the postal enquiry was followed up by a telephone contact, resulting in a higher response rate. A similar procedure could have been used in our study, but on the other side, a telephone reminder also involves some ethical concerns regarding the aspect of voluntariness (42).
Another weakness with surveys is that the questionnaire can be more or less incomplete, i.e. all questions will not be answered by all respondents. Missing data in this survey were considered to be completely at random. However, some of the respondents had not filled out all pages in the questionnaire, which can indicate difficulties to turn over the pages. Missing data regarding the instruments were handled rather strictly. When more than five of the ques- tions were unanswered in the instruments with a neutral score (i.e. SASE, ASA and SOC), they were not replaced with the neutral score. In those cases where missing data not were replaced, the instrument scores were excluded in the analyses. In the other instruments, NUFFE and GHQ, missing data were not replaced. Thus, the data regarding the instruments can therefore be concerned as valid.
Correlations between some of the independent variables in the logistic regression analysis revealed that some of the variables were highly correlated with each other, for example SOC and GHQ. However, Altman (40) states that it is advantageous to use stepwise regression, which was used in present study, because misleading findings attrib- uted to high correlations cannot occur with this regression model.
Conclusion
The results of this survey study among home-dwelling older people in rural areas in Norway revealed that more than three-fourths of the respondents had higher self-careability that was closely related to perceived health, health- related issues, self-care agency, mental health, nutritional state and sense of coherence. The instruments used for mapping different aspects in this study are based on theoretical assumptions that are somewhat inter-related, e.g. the importance of health preserving and enabling resources, ability to find meaning and purpose in life, to have a positive self-conception and the ability to act appropriately. Issues that especially explained self-care ability were risk factors like perceived helplessness, receiving family help, risk for undernutrition and impaired mental health, and reinforcing factors were self-care agency, activity and preparing food. When self-care ability is reduced in older people, caregivers have to be aware about how this can be expressed, and also be aware of their responsibility for identifying and mapping needs for appropriate support and help, and preventing unnecessary and unwanted dependency.
However, further studies are necessary to investigate self-care ability and related factors both in rural and urban
elucidate lived experiences of self-care and features that may influence health and self-care among older home- dwelling individuals to be able to give support.
Author contributions
The study was designed by Bjørg Dale, Ulrika So ̈derhamn and Olle So ̈ derhamn. Bjørg Dale and Ulrika So ̈ derhamn performed the data collection. Ulrika So ̈derhamn carried out the analyses. Bjørg Dale and Ulrika So ̈ derhamn drafted the manuscript and all authors revised it critically. Olle So ̈derhamn supervised the project. All authors read and approved the final manuscript.
Funding
The survey study was carried out with financial support from the Norwegian Research Council (project number 18785). The authors declare that they have no conflicts of interest.
home-dwelling older people. Moreover, it is
important to