Association Between Higher-Level Functional Capacity and Subjective Well-Being in Older Adults with Chronic Conditions: A Cross-Sectional Study

Abstract Aims The aim of the present study was to examine the associations between higher-level functional capacity and subjective well-being (SWB) in older adults with chronic conditions. Materials We used data from a self-administered questionnaire survey conducted in two day-service facilities and one home-visit nursing station. We performed multiple linear regression analyses with SWB measured by the revised Philadelphia Geriatric Center Morale Scale and higher-level functional capacity including instrumental self-maintenance, intellectual activity, and social role. Results 140 persons were included in the analyses. We did not find an association between instrumental self-maintenance and SWB (B = 0.30, 95% confidence interval [CI]: −0.15, 0.75). On the other hand, intellectual activity (B = 0.77, 95%CI: 0.19, 1.34) and social role (B = 0.88, 95% CI: 0.34, 1.43) were associated with SWB. Conclusions Being intellectually active and performing social role may be important to increase SWB in older adults with chronic conditions.


Introduction
An important goal in physical and occupational therapy in geriatric field is that older adults, even those with chronic conditions, continue to live well in their communities.A gerontologist, Lawton proposed a seven-level theoretical model of competence in daily living, in ascending order of complexity: life maintenance, functional health, perception and cognition, physical self-maintenance, instrumental self-maintenance, effectance (also called intellectual activity), and social role. 1 In modern society, where it is assumed important to support older adults in social participation 2,3 as well as activities of daily living, this classic model including social elements have been still useful.
Of the seven factors, three higher-level factors (instrumental self-maintenance, intellectual activity, and social role) are included in a concept "higher-level functional capacity" (also called "higher-level competence"). 4everal studies have demonstrated that higher-level functional capacity is associated with important health outcomes.Two longitudinal studies demonstrated that the higher-level functional capacity, especially intellectual activity, was associated with lower mortality. 4,5Another longitudinal study showed that trajectories of the higher-level functional capacity was associated with lower mortality, lower medical costs, and lower long-term care costs. 6oreover, a longitudinal study showed that impaired higher-level functional capacity, especially intellectual activity, was a predictor of stroke in community-dwelling older adults. 7Thus, the higher-level functional capacity has been associated with the breakdown of daily living, such as the onset of severe diseases and death.
In addition, several studies have examined the association between the higher-level functional capacity and subjective well-being (SWB).The SWB is a comprehensive concept of the positive-negative affective dimension proposed by Larson,8 including life satisfaction, morale, and contentment.Two studies examined association between the sum score of three factors of the higher-level functional capacity and the SWB, 9,10 and two other studies examined whether single similar variables, such as instrumental activities of daily living 11 or social activities 12 are associated with the SWB.
However, the significance of the higher-level functional capacity in older adults with chronic conditions remains unclear.Although one study demonstrated that the social role was associated with adverse outcomes, including death, institutionalization, and functional decline in older adults with chronic conditions, 13 no study has demonstrated association between the higher-level functional capacity and the SWB in older adults with chronic conditions.There is no need for older adults with chronic conditions to remain in activities below physical self-maintenance.Rather, it would be significant for them to engage in activities and/or occupations related to the higher-level functional capacity.
Moreover, no study has dealt with all three factors of higher-level functional capacity simultaneously and compared the associations between each of three factors and the SWB.Higher-level functional capacity has a multidimensional structure, 4 meaning that instrumental self-maintenance, intellectual activity, and social role are distinct factors; thus, practical approaches for each factor may be different.This may be crucial in physical and occupational therapy for older adults with chronic conditions because it is difficult to approach all factors equally, and they need to be prioritized.Understanding which factors in the higher-level functional capacity are associated with the SWB in older adults with chronic conditions is important for physical and occupational therapists.
The aim of the present study was to examine the associations between the higher-level functional capacity and the SWB in older adults with chronic conditions.

Procedure
We initiated a longitudinal study in May 2020 using a self-administered questionnaire conducted in two day-service facilities and one home-visit nursing station located in the central area of Setagaya ward, Tokyo, Japan (the study facilities).The first follow-up survey was conducted between late September and mid-November in 2020.We used a dataset from the first follow-up survey in other studies for other backgrounds and objectives.The present study was the second use of almost the same data for the above-mentioned distinct background and objective.This study was approved by the Institutional Review Board of Tokyo Metropolitan University Arakawa Campus (approval number: 20016).

Participants
A total of 281 individuals who registered with the study facilities during the study period were included; responses were obtained from 156 participants.The study facilities were associated with Japan's long-term care insurance system. 14The exclusion criteria were age < 65 years, no careneeds level certification, and missing data for any of the study items.
The participants provided written informed consent and then filled out the questionnaire.In general, people who are unable to answer the self-administered questionnaire are often excluded from this type of survey, but we consider this strategy is not appropriate when studying themes in older adults with chronic conditions.Therefore, for participants who had difficulty writing due to hand dysfunction, we asked them to read and answer questions, and their families or assistants to write their answers in the questionnaire.Participants with cognitive dysfunctions that could influence their responses, their families, or assistants, who knew their living conditions very well, completed the informed consent process and responded to the questionnaire partially except for questions on self-rated health and the SWB.However, it was considered that such cases might lead bias.For example, in our field observation to monitor the survey, the first author observed participants with cognitive decline and family members as proxy tended to overstate their abilities.In addition, although we asked the older adults with chronic conditions (study participants) to answer about their SWB themselves, the possibility cannot be ruled out that family members may be answering by proxy.So, we handled the possibilities of bias due to the representation and less credibility of answering in persons with dementia by sensitivity analysis, as shown in the section on statistical analysis.

Participant's attributes
We collected data related to age, gender, household status, self-rated health, 15 and Barthel index 16 by questionnaires.Care-needs level, 14 and degree of multi-morbidity data were collected by referring to medical or care records.
To define the degree of multi-morbidity, we investigated whether each participant had the following diseases or conditions: orthopedic diseases, stroke, neurodegenerative diseases (e.g.Parkinson's disease, not including dementia), spinal cord injury, peripheral nerve, heart, pulmonary, kidney, psychiatric diseases (except for dementia), dementia, diabetes mellitus, and other diseases.We defined the sum of disease domains as the degree of multi-morbidity (e.g. if a participant had knee arthritis, stroke, heart disease, and diabetes mellitus, the degree of multi-morbidity was defined as 4).

Higher-level functional capacity
We used the Tokyo Metropolitan Institute of Gerontology Index of Competence (TMIG-IC) 4 to measure the higher-level functional capacity.This index was developed based on Lawton's model of competence 1 and consists of three subscales: five items for instrumental self-maintenance, four items for intellectual activities, and four items for social role. 4For each item, positive and negative answers were scored 1 and 0, respectively, and a higher score for each subscale indicated a better status for each factor.This index has good test-retest reliability. 4We processed each subscale in the analyses to achieve the aim of the present study.The details for the TMIG-IC were shown in Supplemental material 1.

Subjective well-being (SWB)
We used the revised Philadelphia Geriatric Center Morale Scale (PGCMS) developed by Lawton as an index of the SWB. 17,18In this 17-item questionnaire, positive and negative answers were scored as 1 and 0, respectively, and a higher total score indicated better SWB.The PGCMS consists of three subscales (agitation, attitude toward own aging, lonely dissatisfaction).However, we used mainly the sum score, assuming that analyses of individual subscales would not provide more practical implications beyond analyses with the sum score.The analyses of each subscale were considered supplementally to interpret the results of the main analyses.The PGCMS was widely used in Japan and results of a validation study 19 and our additional analysis demonstrated the applicability of PGCMS in Japan (See detail in Supplemental material 1).

Statistical analysis
We performed multiple linear regression analyses using the PGCMS as a dependent variable.The independent variables were each subscale of the TMIG-IC (instrumental self-maintenance, intellectual activities, and social role).First, we performed linear regression analyses without adjusted variables.Second, we performed analyses with the following adjusted variables: age, gender, household status, self-rated health, physical self-maintenance (Barthel index), care-needs level, and degree of multi-morbidity.After analysis for adjusted models, we confirmed the residual homoscedasticities by the residuals vs. fitted plot, residual normalities by the Q-Q plot, and multicollinearities by variance inflation factor (VIF).Third, since there were possibilities of bias due to answers given by representing families or assistants and low credibility of answers given by persons with dementia, sensitivity analyses were performed, excluding the cases of representation and dementia.Finally, we performed supplemental analyses using subscale scores in the PGCMS as dependent variables.We analyzed the data using R version 4.2.0 for MacOS, the R package {Tableone}, 20 and {epiDisplay} 21 to create Tables 1 and 3.In addition, we also used the R package {car} 22 to calculate VIF.The statistical significance level was defined as α = 0.05.

Results
Of the 156 respondents, excluding 9 participants younger than 65 years and seven with missing data in the PGCMS, 140 were included in the analysis.The attributes of the participants are summarized in Table 1.The mean age was 81.6 years (standard deviation [SD]: 7.7), and the mean value of the PGCMS was 9.9 (SD: 4.2).Regarding the prevalence of diseases (overlapping distribution), the top-five highest majority in this study participants were orthopedic diseases (46.4%), stroke (30.7%), diabetes mellitus (17.9%), heart disease (15.7) and neurodegenerative diseases (13.6%.mainly Parkinson's disease).
The scores for each subscale of the TIMG-IC and the mean values of the PGCMS are presented in Table 2.The mean value of the PGCMS tended to be higher with increasing scores for the social role.
The results of the linear regression analysis are shown in Table 3.We did not find an association between the instrumental self-maintenance and the PGCMS (B = 0.30, 95% confidence interval [CI]: −0.15, 0.75).In contrast, the intellectual activity (B = 0.77, 95%CI: 0.19, 1.34) and the social role (B = 0.88, 95% CI: 0.34, 1.43) were associated with the PGCMS.We have confirmed that the residual homoscedasticity and the residual normality were maintained in models for intellectual activity and social role.Moreover, multicollinearities in each model were in acceptable level.
Sensitivity analyses excluding 15 cases (10 people with dementia and 5 cases of representation due to other reasons) demonstrated consistent  results.Supplemental analyses with each subscale of the PGCMS as dependent variables showed that the intellectual activities were associated with attitude toward own aging (B = 0.32, 95% CI: 0.11, 0.52), and the social role was associated with attitude toward own aging (B = 0.39, 95% CI: 0.19, 0.59) and lonely dissatisfaction (B = 0.34, 95% CI: 0.11, 0.56).See details in Supplementary material 2.

Discussion
The present study examined the associations between the higher-level functional capacity and the SWB in older adults with chronic conditions.In this section, we discuss interpretation to statistical value, interpretation for reasons of results, benefits of the present study for physical and occupational therapy and limitation of the present study.

Interpretation of statistical value
Results of regression analyses demonstrated regression coefficients of the intellectual activity and the social role were 0.77 and 0.88.Although they were not large, considering the importance of the SWB, these associations were meaningful.The regression coefficient of the intellectual activity in adjusted model were higher than that in crude model.We assumed careneeds level and multi-morbidity as negative confounders between intellectual activity and SWB because high care-needs levels and multi-morbidity (i.e. more care needs and more illnesses) may increase intellectual activity to search for information to overcome them but have negative impact on subjective well-being.Including these variables in the adjusted model may remove such negative confounders.Indeed, the regression coefficient of intellectual activity in the adjusted model was significant.
On the other hand, regression coefficients of instrumental self-maintenance were lower, and we cloud not find significant associations with the SWB.As referring to 95%CI, upper limits were lower than regression coefficients of the intellectual activity and the social role.
In all models, coefficients of determination were small.However, we should consider (1) the SWB is a complex outcome and large number of factors may be associated with the SWB, and (2) the range of possible scores for each subscale of the TMIG-IC is 0-5 (that means each factor of the TMIG-IC cannot have as large a variance as the PGCMS, so the coefficient of determination is smaller even if the associations between these factors are strong).Moreover, as the aim of the present study was to examine the associations between higher-level functional capacity and SWB, estimators we should focus on were regression coefficients and 95%CI.

Interpretation of reasons for the results
We found that the intellectual activity and the social role were significantly associated with the SWB.We describe our interpretation of the reasons for this result.Lawton described morale, one of the concepts included in the SWB, as "a basic sense of satisfaction with oneself, " "a feeling that there is a place in the environment for oneself, " and "a certain acceptance of what cannot be changed." 17 Our supplemental analyses showed that the intellectual activities were associated with attitudes toward one's own aging, and the social role were associated with attitudes toward one's own aging and lonely dissatisfaction.Considering the above-mentioned definition by Lawton, 17 the results of the present study showed that the intellectual activity could enable older adults with chronic conditions to see their aging and/or impairments in both positive and negative ways, and deal with them realistically.In addition, performing social role may lead to a feeling that there is a place in the environment for themselves.Social interactions in such places might change their attitude toward what cannot be changed.Thus, the intellectual activity and the social role may lead to better SWB in older adults with chronic conditions.However, we did not find an association between instrumental self-maintenance and SWB.Although a previous study in older adults aged 90 years or older demonstrated similar results, 11 we attributed the results of the present study to participants having a care-needs certification based on Japan's longterm care insurance system.In this system, people can receive professional care services within the level of care-needs certification.Thus, lower capacity in instrumental self-maintenance can be compensated by professional care services (for example, shopping services by helpers, meal preparation by helpers, food delivery, and taxis with drivers well-trained in care techniques).In addition, family members may also represent tasks related to instrumental self-maintenance in many cases in Japan (e.g.preparing meals, managing their finances, and transportation by car).In these ways, although older adults with chronic conditions have a lower capacity in instrumental self-maintenance, professional care services or support from family members can compensate for it and satisfy their demands.Conversely, demands related to the intellectual activity and the social role may be filled only when individuals engage in these activities and/or occupations by themselves, not when others compensate or represent them.This might explain why we did not find association between the instrumental self-maintenance and the SWB; however, both the intellectual activity and the social role were associated with the SWB.

Integration with previous studies
A previous study found that social activities, such as teaching skills or passing on experiences to others, participating as volunteers, and belonging to senior citizen clubs, were associated with higher score in the subjective well-being in healthy older adults. 12These activities could be regarded as the intellectual activities and performing the social role.Hence, the results of the present study were consistent with the previous study.Moreover, there were some studies demonstrated the importance of the intellectual activity and the social role for health status.Previous studies have demonstrated that intellectual activity was associated with the lower incidence of stroke 7 and the lower mortality 5 in older adults.Additionally, the social role was associated with the lower risk of adverse outcomes, including death, institutionalization, and functional decline in older adults with chronic conditions. 13Moreover, the SWB was associated with the lower mortality in older adults. 23,24he results of the present study and previous study demonstrate that being intellectually active and performing social role may be important to increase the SWB and to maintain health status in older adults even with chronic conditions.

Benefits of the present study for physical and occupational therapy
Based on these considerations, we suggested that physical and occupational therapists should appropriately assess and support the performance of the intellectual activity and the social role of older adults with chronic conditions.It is true that major areas of interest for many physical and occupational therapists are improvement of physical function or capacities for basic or instrumental dairy living.Nonetheless, if other factors are more associated with the SWB and health status, physical and occupational therapists should also focus on such factors.In particular, when the activities related to instrumental self-maintenance can be replaced by professional care services or support from family members, it will be more important to focus on the intellectual activity and the social role.In addition, as mentioned in the introduction, if clinicians need to prioritize practical approaches in order of importance, we suggest focusing on the intellectual activity and the social role rather than the instrumental self-maintenance.Moreover, further studies for supporting the intellectual activity and the social role were needed, as well as research aiming to improve physical functions and capacities for basic and instrumental activities.In these ways, the benefit of the present study was to show clinicians and researchers the importance of the intellectual activity and the social role in physical and occupational therapy for older adults with chronic conditions.

Limitation of the present study
The present study has some limitations.First, the characteristics of the community should be considered.The present study was conducted in an area with one of the highest average incomes in Japan, 25 and few people in lower income group were included.Moreover, the study is conducted in Japan, a country with a relatively well-developed care system.Therefore, the present study's findings should be carefully applied to practices in communities or countries where the proportion of poor people is high, or there is no well-developed care system.Second, this study was conducted during the coronavirus disease 2019 pandemics.Although the number of infected people was not high (approximately 200 people per day) and there were no restrictions on activities, this situation might influence both the higher-level functional capacity and the SWB and modify the associations between these variables.Further studies during non-pandemic conditions are needed.Third, we could not collect data on variables related to economic status, which is generally associated with SWB 8 and may also be associated with higher-level functional capacity.Therefore, residual confounding may occur.However, even if economic status was a confounder, the influence on the estimated regression coefficients might be small since the study contained few people in lower income group.

Conclusion
Being intellectually active and performing social role may be more important to increase the SWB in older adults with chronic conditions.

Table 3 .
results of regression analyses with PGcMs as a dependent variable.

Table 2 .
each subscale of tiMG-ic and mean values of the PGcMs.