**4. Discussion**

The aim of this study was to investigate factors associated with health-related quality of life among people applying for HA in Sweden. For this purpose, the study used multiple regression analysis to test the influence of variables already known from previous research to have an impact on health-related quality of life, including personal factors, activities, participation, and environmental factors.

The findings conclude that three variables were significantly associated with health-related quality of life measured with the VAS in this sample: age, ADL dependence, and the social aspect of usability in the home. Variables significantly associated with the health summary index were age, ADL dependence and the self-care aspect of usability.

Age in our data had a skewed distribution, and thus, was reclassified into three categories. The younges<sup>t</sup> age category covered a wide range, 25–75 years, and included 27.95% of the participants. The regression analysis showed that this age group rated their health significantly lower than the other older groups; a fact, which was not discovered when age was used as a continuous variable. Thus, categorization helped clarify the relationship between age and self-rated health in this case. In the general population, younger people rate their health higher than older people [5], but in this study, those aged under 75 years rate their health lowest compared to other groups.

A potential explanation might be that their expectations on life in general are similar to others in the same age, but when comparing their own life with others they feel much more restricted and impacted by their activity restrictions, in this study demonstrated by their higher level of ADL dependence, and thus perceive their health as lower. When it comes to the older age group, they might perceive their activity limitations and dependence as a natural consequence of ageing, and thus it has less impact on their self-rated health compared to those that are younger [20].

In our study, higher dependency in ADL was associated with low health-related quality of life. Moreover, the self-care aspect of usability was associated with a lower index score. This is in line with earlier studies focusing on older people [31,32]. A combination of functional limitations and physical demands may have negative influence on the ability to perform different activities, and therefore increase dependency in ADL [33] seems to lead to lower self-rated health [22]. An earlier study exploring the reasons for applying for a HA [34] revealed that among other things, increasing independence in ADL and not having to rely on family, friends, and home care services to perform different activities was a major argumen<sup>t</sup> for an application.

The social aspect of usability consisted of three items: socializing with friends and family in the home, contacting others via telephone/computer, and watching TV/listening to radio. The findings showed that participants, who were more satisfied with this aspect of usability, also rated their health higher. Social contact and possibility to engage in social activities is related to higher self-rated health in several previous studies [35–37]. Some of the participants in this study are considered as frail elderly and might have limited possibility to engage in social activities outside the home, which makes the own home and its usability more important.

A limitation of the study is the low representation of people with cognitive impairment, due to a higher frequency of missing data among this population. Cognitive impairment could not be assessed for 19.2% of the sample, due to fatigue, low mood, or disability among these participants. Thus, it is likely that the prevalence of cognitive impairments among persons applying for HA in Sweden might be higher than what actually is observed in our sample. In addition, only people able to communicate in Swedish were included in this study. This could be due to disability, such as aphasia or dementia related language difficulty, but also due to cultural reasons. Earlier research has showed that older immigrants might rate their health worse than citizens born and raised in Sweden [38]. This limitation as well might partially affect the generalizability of the results to the general Swedish population but does not affect the internal validity of our results and the consistency of the associations found.

It is also important to note that ADL dependence, as measured by the ADL-staircase, was calculated into sum-scores. This differs from the original version of the instrument [36] but has been previously applied in other studies [20,39]. Most important though, ratings of difficulty in activity performance are included in this study, potentially generating detailed information on the individuals' dependence in ADL. Still, further validation of the scale is needed [40].

This study had a cross-sectional design and only investigated factors associated with health-related quality of life among people applying for HA. However, given the aim of the HA, it is possible that both health-related quality of life and the strength of these associations would change after the HA. Future research could involve a longitudinal design exploring how implementations of HA and/or a rejection of applications affect health-related quality of life and associated factors.
