INTRODUCTION

Functional impairments refer to limitations in basic activities of daily living (ADL; such as bathing) or instrumental activities of daily living (IADL; such as handling finances or using the telephone). Such limitations in household management behaviours are also considered as aspects of disability.1 A more pronounced discussion of the terminology is also given by Bruce.2 In accordance with the Disablement Process model (suggested by Verbrugge and Jette1) acute and chronic diseases can ultimately contribute to body impairments. When such body impairments increase, functional impairments markedly increase.1

It has been demonstrated that functional impairments are associated with outcomes such as admission to a nursing home or mortality.3 The number of individuals with functional impairments is likely to increase for reasons of demographic ageing, stressing the importance of this topic.

Various—mostly socioeconomic and health-related—determinants of functional impairment have been identified. For example, it has been shown that ageing, low social support, or increased depressive symptoms are associated with functional impairment.4-6 However, thus far, far less is known about the association between personality and functional impairment. For example, based on data from 265 primary care patients aged 60 years and over, one study showed that higher openness to experience was associated with lower functional impairment.7 A study conducted in a rural Japanese community with 676 older adults showed lower extraversion to be associated with a risk of future functional decline.8 Furthermore, a German study9 based on longitudinal data from the Berlin Ageing Study showed that neuroticism predicted future functional health. Additionally, they found that decreases in extraversion predicted subsequent decreases in functional health. Moreover, another recent study (using data from eight longitudinal samples from the United States, Japan, and England) showed that personality factors are associated with both ADL and IADL limitations in various cohorts.10 More precisely, this study10 showed that higher neuroticism was associated with a higher probability of concurrent and incident ADL and IADL limitations, whereas higher openness to experience, higher extraversion, and higher conscientiousness were associated with lower risk. Additionally, while higher agreeableness was associated with a lower probability of concurrent ADL and IADL limitations, it was not associated with incident limitations.

In sum, there is limited evidence regarding the association between personality and functional impairment among older adults. Hence, our purpose was to clarify the association between personality and functional impairment among older adults based on data from the established Survey of Health, Ageing and Retirement in Europe (SHARE).

Personality is often divided into five main traits (known as the Big Five11): agreeableness (referring to the tendency to be trusting and cooperative), conscientiousness (referring to the tendency to be organized and careful), extraversion (referring to the tendency to be social and assertive), neuroticism (referring to the tendency to experience negative emotional states such as being nervous, insecure, or anxious), and openness to experience (referring to the tendency to be imaginative, curious, and open-minded).

Previous studies have also shown that personality factors are associated with various lifestyle factors as well as health-related factors. For example, it has been demonstrated that they are associated with smoking behaviour, alcohol intake, or obesity.12-14 These factors could contribute to functional impairments.15 Moreover, personality factors are associated with depressive symptoms16 which could contribute to functional decline.17 Furthermore, personality factors are associated with other health-related factors such as self-rated health or chronic illnesses18, 19 which could have an impact on functional impairments.20, 21

METHODS Sample

For the current study, data were taken from wave 7 (year 2017) of the SHARE study.22 This wave was used for reasons of data availability (i.e., personality was only assessed in wave 7).

This study examined community-dwelling individuals living in private households aged 50 years and over (and their spouses) in various countries of Europe (plus Israel). In the SHARE study, nationally representative samples were drawn. Additionally, it is worth noting that the SHARE study includes various topics such as occupational status, social relations, and health. Börsch-Supan et al. gave further details with regard to the SHARE study.23

The Ethics Committee of the University of Mannheim reviewed and approved the SHARE study (waves 1 to 4). Most recently in 2020 the Ethics Council of the Max-Planck-Society reviewed and approved wave 4 and the following waves (waves 5 to 8) of the SHARE project. Participants gave oral consent prior to being interviewed.

Dependent variables

We used two assessments of impairments in basic activities of daily living and two measures of impairments in instrumental activities of daily living as outcome measures (in each case, assessing the report of ‘any difficulty’: 0 = absence of difficulty, 1 = presence of difficulty). In accordance with the STROBE guidelines,24 different indices were used to test the robustness of our findings (also called ‘sensitivity analysis’).

The first ADL index25 corresponds to the sum of the tasks ‘eating, cutting up food’, ‘bathing/showering’, and ‘dressing’, ranging from 0 to 3.

The second ADL index refers to the sum of the tasks ‘eating, cutting up food’, ‘bathing/showering’, ‘dressing’, ‘getting in or out of bed’, and ‘walking across a room’, ranging from 0 to 5. Both scales were adapted from Katz et al.26

The first IADL index corresponds to the sum of ‘managing money’, ‘telephone calls’, and ‘taking medications’, ranging from 0 to 3. The second IADL index refers to ‘managing money’, ‘telephone calls’, ‘taking medications’, ‘preparing a hot meal’, and ‘shopping for groceries’, ranging from 0 to 5. Both tools were adapted from Lawton and Brody.27

All scores correspond to the sum of activities the participants reported as having difficulties in performing. Thus, count scores (which do not have specific labels) were computed. For each count score, higher scores correspond to higher impairments. However, due to small frequencies, all indices were dichotomized (0 = not reporting any difficulties; 1 = reporting at least one difficulty in performing).

Independent variables

Our key independent variables were the personality factors. They were quantified using the 10-item Big Five Inventory (BFI-10)28 covering agreeableness, conscientiousness, extraversion, neuroticism, and openness to experience. This tool can be considered as an established personality inventory. Each dimension ranges from 1 to 5, with higher values corresponding to higher agreeableness, higher conscientiousness, higher extraversion, higher neuroticism, and higher openness to experience.

In the regression analysis, adjustment was made for the sociodemographic covariates of age (in years), sex, family status (married, living together with spouse/registered partnership; married, living separated from spouse; never married; divorced; widowed), and educational level (primary education; secondary education; tertiary education; according to ISCED-9729). Further, adjustment was also made in the regression analysis for these health-related covariates: self-rated health (from 1 = excellent to 5 = poor), cognitive functioning (ranging from 0 = worst to 10 = best; adapted from the Ten-Word Delay Recall Test30), and the number of physical illnesses (sum score, ranging from 0 to 11, for the following conditions: high blood pressure or hypertension; high blood cholesterol; stroke or cerebral vascular disease; diabetes or high blood sugar; chronic lung disease; arthritis, including osteoarthritis, or rheumatism; cancer or malignant tumour; stomach or duodenal ulcer, peptic ulcer; Parkinson's disease; cataracts; hip fracture or femoral fracture).

Statistical analysis

The sample characteristics are first displayed stratified by ADL index 1. Moreover, effect sizes (i.e., Cohen's d) were calculated. Subsequently, multiple logistic regressions were used to clarify the association between personality factors and functional impairment. In further analysis, explorative examination was made as to whether country moderates the association between personality factors and functional impairment (by using the interaction term: personality factor × country). With regard to the countries, Austria was used as the reference category. Interaction terms for each personality factor and each country were added (worth repeating: with Austria as the reference category).

The significance level was set at 0.05. All statistical analyses were performed using Stata 16.0 (Stata Corp., College Station, Texas).

RESULTS Bivariate analysis

Sample characteristics for the analytical sample stratified by ADL index 1 are shown in Table 1.

Table 1. Sample characteristics stratified by ADL index 1 (n = 70 028) ADL index 1: not reporting any difficulties ADL index 1: reporting at least one difficulty Total P-value N = 63 633 N = 6395 N = 70 028 Gender <0.01 Male 27 446 (91.2%) 2640 (8.8%) 30 086 (100.0%) Female 36 187 (90.6%) 3755 (9.4%) 39 942 (100.0%) Age 67.7 (9.2) 74.5 (10.4) 68.3 (9.5) <0.001 Marital status <0.001 Marital status: Married and living together with spouse; registered partnership 18 897 (86.8%) 2868 (13.2%) 21 765 (100.0%) Other: Married, living separated from spouse; never married; divorced; widowed 44 736 (92.7%) 3527 (7.3%) 48 263 (100.0%) Educational level <0.001 Primary education 21 987 (87.1%) 3263 (12.9%) 25 250 (100.0%) Secondary education 27 024 (92.4%) 2225 (7.6%) 29 249 (100.0%) Tertiary education 14 622 (94.2%) 907 (5.8%) 15 529 (100.0%) Self-rated health (from 1 = excellent to 5 = poor) 3.1 (1.0) 4.2 (0.8) 3.2 (1.0) <0.001 Chronic conditions (count score from 0 to 10, with higher values reflecting more chronic conditions) 1.2 (1.2) 2.1 (1.5) 1.2 (1.2) <0.001 Cognitive functioning (from 0 to 10, with higher values reflecting better cognitive functioning) 5.3 (1.8) 4.2 (2.0) 5.2 (1.8) <0.001 Agreeableness (from 1 to 5, higher values reflect higher agreeableness) 3.7 (0.8) 3.6 (0.9) 3.7 (0.8) 0.02 Conscientiousness (from 1 to 5, higher values reflect higher conscientiousness) 4.1 (0.8) 3.9 (0.9) 4.1 (0.8) <0.001 Extraversion (from 1 to 5, higher values reflect higher extraversion) 3.5 (0.9) 3.4 (0.9) 3.5 (0.9) <0.001 Neuroticism (from 1 to 5, higher values reflect higher neuroticism) 2.6 (1.0) 2.9 (1.1) 2.7 (1.0) <0.001 Openness to experience (from 1 to 5, higher values reflect higher openness to experience) 3.3 (0.9) 3.2 (1.0) 3.3 (0.9) <0.001 ADL, activities of daily living.

In the total sample, the average age was 68.3 years (SD = 9.5 years; ranging from 50 to 105 years) and about 57% of the individuals were female. In sum, 9.1% of the individuals reported having at least one difficulty with ADL index 1 (similarly, 10.3% reported at least one difficulty with ADL index 2, 4.2% with IADL index 1, and 8.5% with IADL index 2).

In bivariate analysis, ADL index 1 was associated with all explanatory variables. Additionally, in Table 2, the frequency of impairments is displayed for our analytical sample.

Table 2. Frequency of impairments in ADL and IADL indices (n = 70 028) N (%) ADL index 1 0 63 633 (90.9%) 1 4013 (5.7%) 2 1810 (2.6%) 3 572 (0.8%) ADL index 2 0 62 841 (89.7%) 1 3921 (5.6%) 2 1625 (2.3%) 3 756 (1.1%) 4 526 (0.8%) 5 359 (0.5%) IADL index 1 0 67 091 (95.8%) 1 2029 (2.9%) 2 553 (0.8%) 3 355 (0.5%) IADL index 2 0 64 093 (91.5%) 1 3246 (4.6%) 2 1371 (2.0%) 3 677 (1.0%) 4 342 (0.5%) 5 299 (0.4%) ADL difficulties: dressing Not impaired 65 068 (92.9%) Impaired 4960 (7.1%) ADL difficulties: walking across a room Not impaired 68 602 (98.0%) Impaired 1426 (2.0%) ADL difficulties: bathing or showering Not impaired 66 614 (95.1%) Impaired 3414 (4.9%) ADL difficulties: eating, cutting up food Not impaired 69 053 (98.6%) Impaired 975 (1.4%) ADL difficulties: getting in or out of bed Not impaired 67 465 (96.3%) Impaired 2563 (3.7%) IADL difficulties: preparing a hot meal Not impaired 67 750 (96.7%) Impaired 2278 (3.3%) IADL difficulties: shopping for groceries Not impaired 65 624 (93.7%) Impaired 4404 (6.3%) IADL difficulties: telephone calls Not impaired 69 119 (98.7%) Impaired 909 (1.3%) IADL difficulties: taking medications Not impaired 69 116 (98.7%) Impaired 912 (1.3%) IADL difficulties: managing money Not impaired 67 649 (96.6%) Impaired 2379 (3.4%) ADL, activities of daily living; IADL, instrumental activities of daily living.

We also calculated Cohen's d for the association between personality factors (agreeableness, conscientiousness, extraversion, neuroticism, and openness to experience) and basic functional impairment (ADL index 1). Cohen's d was 0.03, 0.26, 0.10, −0.29, and 0.12, respectively.

Moreover, we calculated Cohen's d for the association between the personality factors and instrumental functional impairment (IADL index 1). Cohen's d was −0.003, 0.36, 0.21, −0.34, and 0.30, respectively. Similar effect sizes are present for the association between personality factors and both ADL index 2 and IADL index 2 (results not shown, but available upon request).

Regression analysis

Findings of multiple logistic regressions are shown in Table 3 (with four outcomes: ADL index 1, ADL index 2, IADL index 1, and IADL index 2).

Table 3. Determinants of functional impairment (0 = not reporting any difficulties; 1 = reporting at least one difficulty) Independent variables (1) (2) (3) (4) ADL index 1 ADL index 2 IADL index 1 IADL index 2 Gender: Female (Reference category: Male) 0.91** 0.99 1.07 1.27*** (0.85–0.96) (0.94–1.05) (0.98–1.17) (1.19–1.36) Age 1.04*** 1.04*** 1.07*** 1.07*** (1.04–1.05) (1.04–1.04) (1.06–1.07) (1.06–1.07) Marital status: Married and living together with spouse; registered partnership (Reference category: Other) 0.78*** 0.79*** 0.74*** 0.68*** (0.73–0.82) (0.75–0.84) (0.68–0.81) (0.64–0.73) Education: - Secondary education (Reference category: Primary education) 0.95 0.98 0.82*** 0.86*** (0.89–1.02) (0.92–1.05) (0.74–0.90) (0.80–0.93) - Tertiary education 0.88** 0.89** 0.83** 0.82*** (0.81–0.96) (0.82–0.97) (0.73–0.95) (0.75–0.91) Self-rated health (from 1 = excellent to 5 = poor) 2.81*** 2.81*** 2.10*** 2.89*** (2.71–2.92) (2.71–2.91) (1.99–2.21) (2.78–3.02) Chronic conditions (count score from 0 to 10, with higher values reflecting more chronic conditions) 1.21*** 1.20*** 1.17*** 1.16*** (1.18–1.23) (1.17–1.22) (1.13–1.20) (1.14–1.19) Cognitive functioning (from 0 to 10, with higher values reflecting better cognitive functioning) 0.93*** 0.92*** 0.72*** 0.81*** (0.91–0.95) (0.91–0.94) (0.70–0.73) (0.80–0.83) Personality characteristics: Agreeableness (from 1 to 5, higher values reflect higher agreeableness) 1.05** 1.04* 1.10*** 1.04* (1.01–1.09) (1.01–1.08) (1.05–1.16) (1.01–1.09) Conscientiousness (from 1 to 5, higher values reflect higher conscientiousness) 0.82*** 0.83*** 0.76*** 0.79*** (0.79–0.85) (0.80–0.86) (0.72–0.80) (0.77–0.82) Extraversion (from 1 to 5, higher values reflect higher extraversion) 1.06*** 1.06*** 1.00 1.00 (1.03–1.09) (1.03–1.09) (0.95–1.04) (0.97–1.03) Neuroticism (from 1 to 5, higher values reflect higher neuroticism) 1.12*** 1.13*** 1.15*** 1.12*** (1.08–1.15) (1.10–1.16) (1.11–1.20) (1.08–1.16) Openness to experience (from 1 to 5, higher values reflect higher openness to experience) 1.04