Quality of Life Research

, Volume 19, Issue 3, pp 351–361

Relationship of quality of life to dispositional optimism, health locus of control and self-efficacy in older subjects living in different environments

Authors

    • Department of GeriatricsMedical University of Lodz
  • Violetta Jachimowicz
    • Department of GeriatricsMedical University of Lodz
    • The President Stanisław Wojciechowski Higher Vocational State School of Kalisz
Article

DOI: 10.1007/s11136-010-9601-0

Cite this article as:
Kostka, T. & Jachimowicz, V. Qual Life Res (2010) 19: 351. doi:10.1007/s11136-010-9601-0

Abstract

Purpose

To describe the relationship of dispositional optimism, health locus of control and self-efficacy to quality of life (QOL) in older subjects differing in level of disability and institutionalisation.

Methods

The study was conducted in the three groups of subjects aged ≥ 65: 110 relatively healthy community-dwelling elderly, 102 independent elders who voluntarily decided to live in veteran home and 112 inhabitants of a long-term care home. Life orientation test—revised (LOT-R), multidimensional health locus of control (MHLC) and generalised self-efficacy scale (GSES) together with a multidimensional assessment were performed with each subject. QOL was assessed using the Euroqol 5D questionnaire, the Nottingham health profile and the satisfaction with life scale (SWLS).

Results

QOL generally decreased with growing level of dependence and institutionalisation. LOT-R, MHLC and GSES were important and independent correlates of QOL in all three environments of older subjects. The relationship of education, smoking habit, physical activity, strength and mobility measures to psychological characteristics was different in the three groups of elders. LOT-R, MHLC Powerful Others, MHLC Chance and GSES were the most important QOL correlates in veteran home group, while MHLC Internal was most significant in long-term care home inhabitants.

Conclusions

Data of this cross-sectional study suggests that the veteran home elderly, as a group ‘in transition’ between community and institution, should be the first target of psychological preventive and health-promoting measures aimed at improving QOL in older population.

Keywords

Older peoplePhysical activityPhysical functionMuscle strengthDisabilityInstitutionalisation

Abbreviations

PA

Physical activity

MMSE

Mini-mental state examination

GDS

Geriatric depression scale

ADL

Activities of daily living

IADL

Instrumental activities of daily living

LOT-R

Life orientation test—revised

MHLC

Multidimensional health locus of control

GSES

Generalised self-efficacy scale

QOL

Quality of life

NHP

Nottingham health profile

SWLS

Satisfaction with life scale

VAS

Visual analogue scale

ANOVA

Analysis of variance

OR

Odds ratios

CI

Confidence intervals

Introduction

The elderly are the fastest growing segment of the population. Psychological factors have been shown to determine survival, use of medical services, health-promoting behaviours and quality of life in older subjects [15]. Dispositional optimism, health locus of control and self-efficacy are considered among key psychological profile measures in this age group.

Dispositional optimism is a measure of life engagement and generalised positive outcome expectancies for the future. Optimism scores significantly decrease with advancing age [6]. Dispositional optimism has been associated with a variety of physical and mental health outcomes [2, 7, 8]. A graded and independent protective relationship has been found between dispositional optimism and all-cause and cardiovascular mortality in old age [3, 9, 10].

The health locus of control is the extent to which a subject believes he or she can affect his or her health status. The multidimensional health locus of control (MHLC) scales are widely used to measure beliefs about determinants of a person’s health [11]. Health locus of control scales are usually associated with socio-economic status and self-rated health. Elderly subjects, women, and subjects with fewer years of education showed more ‘external’ belief [12]. External locus of control measures correlated negatively with dispositional optimism [8]. Health locus of control is viewed as one of the factors that predisposes individuals to use medical services [5]. On the other hand, health locus of control beliefs are also consequences of health-related behaviours and events such as utilisation [5]. An external orientation may also be advantageous in situations in which little personal control is possible [13].

Self-efficacy is a useful indicator of general adaptational outcomes [1]. A person who believes in his/her abilities to produce a desired effect is able to conduct a more active life course [14]. Individuals with an internal health locus of control and high generalised self-efficacy are more likely to benefit from health education programmes than those with an external locus of control and low self-efficacy [15]. Self-efficacy was one of the determinants for the self-reported participation in physical activity [16].

Dispositional optimism, health locus of control and self-efficacy are personal psychological beliefs and resources that, being determinants of anticipation, may also influence QOL. A number of previous studies have assessed the relationship of these psychological measures to QOL. In several cohort studies, higher dispositional optimism has been shown to have a beneficial influence on QOL [7, 17, 18]. In a recent study performed in a community sample of older subjects, self-rated health was positively related to optimism [2]. Both in general and older population studies, better QOL was usually related to stronger internal and weaker external locus of control [1922]. Self-efficacy predicted QOL in people with arthritis or in frail elderly [1, 4].

Recent data indicates that relationships of psychological measures may vary across groups differing in health status [23]. Previous studies relating QOL to dispositional optimism, health locus of control or self-efficacy in the elderly have treated their samples as homogenous groups. The strength of association between these psychosocial factors and QOL may be diverse in various populations of elderly, being influenced by sociodemographic factors, concomitant diseases, functional status and health-promoting behaviours [4]. Interestingly, studies describing the relationship of QOL to dispositional optimism, health locus of control and self-efficacy in older subjects living in different environments have not been found in available literature. Therefore, the aim of the present study was to describe the relationship of dispositional optimism, health locus of control and self-efficacy to QOL in older subjects differing in level of disability and institutionalisation. It was hypothesised that these psychological measures would be correlated with QOL. It was further hypothesised that the strength of these associations might be different across subgroups of older population.

Methods

Subjects

The study was conducted in 2007–2008 using three groups of subjects aged ≥ 65 from Kalisz and Ostrów Wielkopolski, two cities of Central-Western Poland. The first group of respondents were 110 community-dwelling elderly, participants of the University of the Third Age. The second group were 102 independent elders who voluntarily decided to live in assisted living homes—the Combatant and Veterans Houses in Kalisz and Ostrów Wielkopolski. The third group were 112 inhabitants of long-term care home in Kalisz. The criteria for taking part in the study were age, ability to participate in physical function (balance and gait) tests, verbal communication efficiency and patient’s approval of the study. Of 159 participants of the University of the Third Age in Kalisz and Ostrów Wielkopolski who were 65 or older, 110 volunteered to participate in the study. From 148 residents of veteran homes aged ≥65, 102 agreed to participate. From 180 inhabitants of long-term care home in Kalisz, 13 were under 65 years, 6 subjects refused to participate, and 49 residents with severe dementia or bed-ridden were excluded. Therefore, 112 elderly residents of long-term care homes participated in the study. The study has been approved by an ethical committee, and written informed consent was obtained from all the subjects.

Protocol and measures

A multidimensional assessment (comprehensive geriatric assessment) was performed with each subject. All questionnaires were completed with the same qualified investigator (VJ) during a personal interview that lasted about 3 h. The assessment included demographic and social variables, health status, physical activity (PA), physical function and mental status. Demographic and social variables assessed included age, gender and education (in years). Health status variables included smoking habit, resting blood pressure, concomitant and previous diseases and regularly taken medications. Disease data was collected from patients’ documentation and diagnoses made by General Practitioners.

PA was assessed using two popular PA questionnaires: the Seven Day Recall PA Questionnaire [24] and the Stanford Usual Activity Questionnaire [25]. These questionnaires have been chosen because of their high validity demonstrated in older individuals [26]. The Seven Day Recall total score (daily energy expenditure over past week) and the Stanford Moderate (six habitual moderate activities) index were calculated and used for further comparisons. These two PA indices are expressed as PA-Energy and PA-Stanford, respectively.

Physical disability was evaluated by the activities of daily living (ADL) [27] and the instrumental activities of daily living (IADL) [28]. Grip strength was measured on both sides, and mobility problems in elderly patients were assessed using Tinetti test [29]. The Tinetti balance and gait evaluation combines assessment of balance (maximum score 16 points) and gait (maximum score 12 points) components. The maximum total score is 28 points. In general, patients who score in the range of 19–24 have an increased risk of falls, while patients who score below 19 have a high risk. Mental functions were evaluated using the mini-mental state examination (MMSE) [30] and 15-item geriatric depression scale (GDS) [31].

Psychological profile

A validated Polish adaptation [32] of the Life Orientation Test—Revised (LOT-R) was used to assess dispositional optimism. This scale comprises 10 items, including only 6 diagnostic ones [33]. A validated Polish version [34] of Form B of the multidimensional health locus of control (MHLC) scales was used to evaluate the health locus of control [11]. Internal, Chance, and Powerful Others MHLC were calculated and used for further analyses. A validated Polish version [35] of the generalised self-efficacy scale (GSES) developed by Schwarzer and Jeruzalem [14] was used to assess self-efficacy. QOL was assessed using the Euroqol 5D questionnaire, the Nottingham health profile (NHP) and the satisfaction with life scale (SWLS).

Euroqol 5D questionnaire is a widely used and validated generic instrument that has five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression [36]. Each dimension has three levels: no problems (no limitations), some problems and severe problems. In addition, each subject was asked to describe his/her overall actual health state using the Euroqol visual analogue scale (VAS), where 0 denotes the worst imaginable health state and 100 indicates the best imaginable health state.

The Nottingham Health Profile (NHP) was also applied to assess quality of life [37]. Polish validated translation has been used [38]. The first part of the questionnaire contains 38 questions of yes/no response and evaluates six physical and psychosocial dimensions of life: energy, pain, emotional reactions, sleep, social isolation and physical abilities. Final score for each dimension ranged from 0 (all answers of ‘no’ in the dimension, denoting absence of distress) to 100 (all answers of ‘yes’, denoting maximal distress). The second part contains seven general yes/no questions concerning paid work, looking after home, social life, home life, sex life, interests and hobbies and vacations. This part was not used in the present study.

Well-being was measured by the satisfaction with life scale (SWLS) [39], Polish adaptation by Juczyński [40]. The scale comprises five statements indicating one’s satisfaction with life.

All the measures used in the present study have been previously used and validated in general and/or geriatric medicine for many years. Internal consistency (Cronbach’s alpha) obtained in our sample for different interviews (0.71 for MNA, 0.91 for MMSE, 0.83 for GDS, 0.78 for ADL, 0.90 for IADL, 0.78 for LOT-R, 0.78 for MHLC, 0.73 for MHLC Internal, 0.78 for MHLC Powerful Others, 0.67 for MHLC Chance, 0.93 for GSES, 0.76 for Euroqol 5D, 0.93 for NHP and 0.84 for SWLS) is comparable with previous reports.

Statistical analysis

The one-way analysis of variance (ANOVA) with Tukey post hoc testing, Kruskal–Wallis test, chi-square test (3 × 2 and 2 × 2 with Yates’ correction) and single logistic regression was used to compare the groups. In bivariate analyses, Pearson product moment correlation coefficients and one-way ANOVA were calculated for the entire studied population as well as separately for each group. General linear model and multiple logistic regression were used to select variables that independently predict LOT-R, MHLC, GSES and QOL. Comparison of regression lines and ANOVA interactions was used to corroborate the different relationship of psychological profile measures to independent variables in the three groups. The Euroqol 5D dimensions (no problem vs. any problem) were dichotomised in both bivariate and multivariate analyses. Odds ratios (OR) and confidence intervals (CI) with 95% confidence limits were calculated. Results are presented as mean ± standard deviation. The limit of significance was set at P = 0.05 for all analyses.

Results

Baseline characteristics, dispositional optimism, health locus of control and self-efficacy in the three groups

Table 1 shows the baseline participant characteristics for the three groups of the study. The three groups were virtually identical as to gender composition. Community-dwelling elders were the youngest and better educated. Community-dwelling elders had the lowest prevalence of hypertension, musculoskeletal disorders and cataract and were also taking less medication than other groups. Long-term care home inhabitants had the highest incidence of diabetes, respiratory diseases, history of myocardial infarction and stroke. Community-dwelling elderly had the highest physical activity indices, IADL, grip strength and Tinetti test scores. Long-term care home inhabitants had the lowest physical activity, ADL, IADL, strength and mobility indices. Community-dwelling elderly had the highest MMSE. Long-term care home inhabitants had the highest GDS and the lowest MMSE.
Table 1

Baseline participant characteristics

 

Community-dwelling elderly (n = 110)

Veterans home residents (n = 102)

Long-term care home inhabitants (n = 112)

P-value for factor

Age (years)

69.6 ± 4.6

77.0 ± 7.6*

78.9 ± 6.9*

P < 0.001

Men (%)

23.6

21.5

22.3

NS

Education (years)

13.5 ± 2.5

8.7 ± 3.1*

7.6 ± 3.5*†

P < 0.001

Current smokers (%)

8.2

10.8

12.5

NS

Systolic blood pressure (mmHg)

132 ± 15

135 ± 17

135 ± 16

NS

Diastolic blond pressure (mmHg)

82 ± 9

79 ± 10

78 ± 9*

P = 0.028

Hypertension (%)

45.4

59.8*

59.8*

P = 0.048

Ischaemic heart disease (%)

16.4

27.5

20.5

NS

Post myocardial infarction (%)

6.4

9.8

17.9*

P = 0.022

Chronic heart failure (%)

17.3

32.0

26.8

P = 0.06

Post stroke (%)

1.8

7.8

18.8*†

P < 0.001

Diabetes (%)

10.9

14.7

17.9*

NS

Respiratory diseases (%)

5.5

7.8

17.9*†

P = 0.006

Musculoskeletal disorders (%)

67.3

85.3*

81.3*

P = 0.004

Gastroduodenal disorders (%)

18.2

22.6

17.0

NS

Osteoporosis (%)

23.6

29.4

32.1

NS

Urinary incontinence (%)

13.6

21.6

14.3

NS

Cataract (%)

11.8

38.2*

35.7*

P < 0.001

Glaucoma (%)

7.3

14.7

6.3

NS

Medications (number)

3.2 ± 2.7

5.4 ± 3.6*

5.9 ± 3.5*

P < 0.001

PA-energy (kcal kg−1 day−1)

43.8 ± 4.8

37.5 ± 4.4*

33.7 ± 2.1*†

P < 0.001

PA-stanford

4.1 ± 1.2

1.9 ± 1.5*

1.4 ± 1.3*†

P < 0.001

Activities of daily living

5.9 ± 0.2

5.7 ± 0.6

5.4 ± 0.9*†

P < 0.001

Instrumental activities of daily living

8.0 ± 0.2

6.9 ± 1.9*

4.8 ± 2.7*†

P < 0.001

Left hand grip strength (kg)

32.3 ± 9.0

23.5 ± 10.1*

21.8 ± 9.7*

P < 0.001

Right hand grip strength (kg)

35.4 ± 9.0

27.5 ± 9.7*

23.5 ± 9.9*†

P < 0.001

Tinetti balance score

15.3 ± 1.0

12.1 ± 3.6*

11.6 ± 4.1*

P < 0.001

Tinetti gait score

12.0 ± 0.2

10.0 ± 2.9*

8.3 ± 3.4*†

P < 0.001

Tinetti total score

27.2 ± 1.1

22.0 ± 6.3*

19.9 ± 7.2*†

P < 0.001

Geriatric depression scale

3.4 ± 2.9

3.9 ± 3.7

4.8 ± 3.8*

P = 0.007

Mini-mental state examination

29.1 ± 1.1

26.2 ± 3.4*

23.2 ± 4.8*†

P < 0.001

* Significantly different (P < 0.05) from community-dwelling group

† Significantly different (P < 0.05) from Veterans home residents

LOT-R was the lowest in long-term care home inhabitants (Table 2). The highest MHLC Internal was in veteran home elders. MHLC Powerful Others was higher in veteran home elders when compared to community-dwelling elderly. MHLC Chance was lower in community-dwelling when compared to other groups. GSES was comparable in all three groups. In the entire studied population, LOT-R, MHLC Internal and GSES were significantly positively intercorrelated. MHLC Internal, MHLC Powerful Others and MHLC Chance were also significantly positively intercorrelated. LOT-R was adversely related to MHLC Chance. These associations remained similar when assessed separately in the three groups.
Table 2

Dispositional optimism, health locus of control and self-efficacy in the three groups

 

Community-dwelling elderly (n = 110)

Veterans home residents (n = 102)

Long-term care home inhabitants (n = 112)

P-value for factor

Life orientation test—revised

16.2 ± 3.9

16.0 ± 5.5

14.0 ± 4.1*†

P < 0.001

MHLC internal

26.2 ± 5.3

28.8 ± 6.3*

26.4 ± 7.4†

P = 0.004

MHLC powerful others

24.9 ± 6.8

28.6 ± 8.1*

27.1 ± 6.4

P < 0.001

MHLC chance

24.3 ± 6.5

31.2 ± 4.3*

30.7 ± 3.9*

P < 0.001

Generalised self-efficacy scale

30.9 ± 5.5

31.8 ± 7.7

30.4 ± 7.0

NS

* Significantly different (P < 0.05) from community-dwelling group

† Significantly different (P < 0.05) from Veterans home residents

Association of dispositional optimism, health locus of control and self-efficacy with other variables in the three groups

The relationship of LOT-R, MHLC and GSES to other variables was, in several aspects, substantially different in the three groups of elders. Education level was correlated with LOT-R and GSES only in veteran home residents. MHLC Powerful Others and MHLC Chance were influenced by education level in community-dwelling and veteran home elderly, not in long-term care elders. Current smoking was strongly related to increased GSES and modestly (P = 0.08) to MHLC Internal in long-term care home inhabitants, not in the two other groups (Fig. 1). In community-dwelling and veteran home groups current smoking was associated with lower MHLC Powerful Others. The impact of urinary incontinence was most evident in the veteran home group where LOT-R (Fig. 2), MHLC Internal and GSES decreased more steeply with the presence of the disease when compared to other groups.
https://static-content.springer.com/image/art%3A10.1007%2Fs11136-010-9601-0/MediaObjects/11136_2010_9601_Fig1_HTML.gif
Fig. 1

Relationship of GSES to smoking habit in the three groups of older subjects

https://static-content.springer.com/image/art%3A10.1007%2Fs11136-010-9601-0/MediaObjects/11136_2010_9601_Fig2_HTML.gif
Fig. 2

Relationship of LOT-R to urinary incontinence in the three groups of older subjects

The relationship of physical activity, strength and mobility measures to psychological characteristics was more profound in the two older groups in comparison with community-dwelling elderly.

QOL data in the three groups

Table 3 shows NHP and SWLS data in the three groups. All NHP indices were more favourable (lower) in community-dwelling elders when compared to long-term care residents, while all except for NHP sleep and NHP social isolation were more favourable when compared to veteran home residents. NHP emotional reactions and NHP social isolation were also lower in veteran home elders in comparison with long-term care elders. SWLS was comparable in all three groups.
Table 3

Quality of life (NHP and SWLS) in the three groups

 

Community-dwelling elderly (n = 110)

Veteran home residents (n = 102)

Long-term care home inhabitants (n = 112)

P-value for factor

NHP energy

19.4 ± 33.0

50.7 ± 43.2*

60.1 ± 38.2*

P < 0.001

NHP pain

15.9 ± 22.6

37.7 ± 34.1*

45.8 ± 33.7*

P < 0.001

NHP emotional reactions

14.8 ± 18.5

25.1 ± 29.2*

33.6 ± 19.9*†

P < 0.001

NHP sleep

36.4 ± 35.7

46.3 ± 37.4

48.8 ± 36.5*

P = 0.03

NHP social isolation

11.1 ± 19.3

18.2 ± 23.1

33.0 ± 31.4*†

P < 0.001

NHP mobility

11.6 ± 15.4

41.2 ± 27.4*

46.3 ± 24.4*

P < 0.001

Satisfaction with life scale

21.2 ± 5.6

21.9 ± 6.2

20.8 ± 6.6

NS

* Significantly different (P < 0.05) from community-dwelling group

† Significantly different (P < 0.05) from veteran home residents

Table 4 shows Euroqol 5D scores in the three groups. In relation to community-dwelling elderly, veteran home and long-term care home residents more often reported problems with mobility, self-care and had also lower score on the visual analogue scale. Interestingly, veteran home inhabitants had less frequent anxiety/depression problems when compared to community-dwelling elderly.
Table 4

Quality of life (Euroqol 5D) in community-dwelling elderly, veteran home residents and long-term care home inhabitants

 

Community-dwelling elderly (n = 110)

Veteran home residents (n = 102)

Long-term care home inhabitants (n = 112)

P-value for factor

Mobility no problems, problems (%)

69, 31

39, 61

38, 62

P < 0.001

Odds ratios (95% confidence intervals)

1.00

3.46* (1.96–6.12)

3.59* (2.05–6.26)

 

Self-care no problems, problems (%)

93, 7

80, 20

80, 20

P = 0.009

Odds ratios (95% confidence intervals)

1.00

3.11* (1.30–7.45)

3.12* (1.32–7.37)

 

Usual activity no problems, problems (%)

79, 21

75, 25

79, 21

NS

Odds ratios (95% confidence intervals)

1.00

1.23 (0.64–2.34)

1.02 (0.54–2.00)

 

Pain/discomfort no problems, problems (%)

28, 72

31, 69

23, 77

NS

Odds ratios (95% confidence intervals)

1.00

0.86 (0.48–1.55)

1.30 (0.71–2.38)

 

Anxiety/depression no problems, problems (%)

28, 72

53, 47

35, 65

P < 0.001

Odds ratios (95% confidence intervals)

1.00

0.35* (0.20–0.62)

0.73 (0.42–1.30)

 

Visual analogue scale

72.1 ± 14.8

54.5 ± 22.9*

52.4 ± 19.1*

P < 0.001

Bivariate relationships of Euroqol 5D dimensions to group affiliation

* Significantly different from community-dwelling group

† Significantly different from Veterans home residents

Association of QOL with dispositional optimism, health locus of control and self-efficacy for the entire studied population

Table 5 shows the relationship of QOL data to LOT-R, MHLC and GSES for the entire studied population. All QOL measures were favourably related to LOT-R, MHLC Internal and GSES. The relationship of QOL to MHLC Powerful Others and MHLC Chance, however, was entirely different; they were directly related to the majority of NHP dimensions, negatively to VAS, and MHLC Powerful Others was higher in subjects with mobility, self-care, pain/discomfort and anxiety/depression problems. All QOL measures were also significantly interrelated (not shown in the table). The relationship of Euroqol 5D data to NHP was much stronger when compared to SWLS association to Euroqol 5D or NHP.
Table 5

Relationship of QOL data to LOT-R, MHLC and GSES for the whole studied population (n = 324)

 

LOT-R

MHLC internal

MHLC powerful others

MHLC chance

GSES

NHP energy

r = −0.37***

r = −0.18***

r = 0.25***

r = 0.31***

r = −0.22***

NHP pain

r = −0.33***

r = −0.22***

r = 0.28***

r = 0.25***

r = −0.21***

NHP emotional reactions

r = −0.47***

r = −0.13*

r = 0.13*

r = 0.23***

r = −0.30***

NHP sleep

r = −0.25***

r = −0.22***

r = 0.08

r = 0.08

r = −0.16**

NHP social isolation

r = −0.42***

r = −0.12*

r = 0.05

r = 0.16**

r = −0.30***

NHP mobility

r = −0.34***

r = −0.19***

r = 0.28***

r = 0.36***

r = −0.25***

Satisfaction with life scale

r = 0.50***

r = 0.32***

r = 0.03

r = 0.04

r = 0.43***

Euroqol 5D mobility

F = 16.9***

F = 14.4***

F = 10.6**

F = 3.43

F = 16.9***

Euroqol 5D self-care

F = 12.8***

F = 19.7***

F = 5.63*

F = 1.58

F = 15.7***

Euroqol 5D usual activity

F = 10.7**

F = 34.4***

F = 3.00

F = 0.04

F = 29.1***

Euroqol 5D Pain/discomfort

F = 7.94**

F = 17.7***

F = 8.38**

F = 1.73

F = 15.0***

Euroqol 5D anxiety/depression

F = 18.8***

F = 15.2***

F = 4.51*

F = 1.34

F = 14.9***

Euroqol 5D visual analogue scale

r = 0.37***

r = 0.21***

r = −0.20***

r = −0.26***

r = 0.26***

P < 0.05; ** P < 0.01; *** P < 0.001

LOT-R, MHLC and GSES, together with other independent variables, were used in multivariate analyses to select data independently predicting QOL measures. After adjustment for other confounders, LOT-R was found to contribute to NHP emotional reactions, NHP social isolation, SWLS and Euroqol 5D VAS. MHLC Internal was seen to contribute to NHP pain, Euroqol 5D self-care, Euroqol 5D usual activity and Euroqol 5D pain/discomfort. MHLC Powerful Others independently predicted NHP pain. MHLC Chance contributed to NHP mobility and GSES to SWLS.

Association of QOL with dispositional optimism, health locus of control and self-efficacy in the three groups

The relationship of QOL data to LOT-R, MHLC and GSES was different in the three groups: usually, but not always, being strongest in veteran home elderly. For 65 associations explored, 19 were statistically significant in community-dwelling elderly, 47 in veteran home residents and 21 in long-term care inhabitants (Table 6). In multivariate analyses, LOT-R, MHLC and GSES independently predicted six QOL measures in community-dwelling elderly, eight in veteran home residents and four in long-term care inhabitants.
Table 6

Number of statistically significant relationships of 15 assessed QOL measures to LOT-R, MHLC and GSES in the three groups

 

LOT-R

MHLC internal

MHLC powerful others

MHLC chance

GSES

Community-dwelling elderly

5

5

5

0

4

Veteran home residents

13

9

8

4

13

Long-term care home inhabitants

5

10

0

0

6

The different relationships of QOL to LOT-R, MHLC and GSES in the three groups have been corroborated by several statistically significant interactions. LOT-R was related to NHP pain, NHP emotional reactions, NHP sleep, NHP mobility, Euroqol 5D pain/discomfort, Euroqol 5D anxiety/depression and Euroqol 5D VAS more strongly in veteran home residents than in the two other groups (Fig. 3). MHLC Internal influenced Euroqol 5D self-care and Euroqol 5D usual activities far more in long-term care elders than community-dwelling elderly. Finally, MHLC Powerful Others was related to Euroqol 5D pain/discomfort problems and Euroqol 5D anxiety/depression problems more visibly in veteran home residents when compared to the two other groups (Fig. 4). GSES was correlated with NHP pain, NHP sleep, NHP mobility, Euroqol 5D pain/discomfort, Euroqol 5D anxiety/depression and Euroqol 5D VAS more strongly in veteran home residents than in the two other groups (Fig. 5).
https://static-content.springer.com/image/art%3A10.1007%2Fs11136-010-9601-0/MediaObjects/11136_2010_9601_Fig3_HTML.gif
Fig. 3

Relationship of Euroqol 5D VAS to LOT-R in the three groups of older subjects

https://static-content.springer.com/image/art%3A10.1007%2Fs11136-010-9601-0/MediaObjects/11136_2010_9601_Fig4_HTML.gif
Fig. 4

Relationship of Euroqol 5D pain/discomfort problems to MHLC powerful others in the three groups of older subjects

https://static-content.springer.com/image/art%3A10.1007%2Fs11136-010-9601-0/MediaObjects/11136_2010_9601_Fig5_HTML.gif
Fig. 5

Relationship of NHP pain to GSES in the three groups of older subjects

Discussion

This is the first study to relate data on the association of important personal resources, i.e. dispositional optimism, health locus of control and self-efficacy, with valid and reliable [4143] measures of QOL in different populations of older people. As expected, dispositional optimism was the lowest in the long-term care home group. Surprisingly, both internal and external locus of control were the highest in veteran home elders. These subjects voluntarily decided to live in a veterans’ home where they were allowed to benefit from organised meals, basic daily help (e.g. laundry), medical resources and social events. Self-efficacy was comparable in the three groups. Our data indicates that dispositional optimism, internal health locus of control and self-efficacy are important and independent correlates of higher QOL in all three environments of older subjects. The relationship of MHLC Powerful Others and MHLC Chance to QOL was less pronounced and gave opposite results. Important differences were observed in the strength of associations among the three groups; LOT-R, MHLC Powerful Others, MHLC Chance and GSES were the most important correlates of QOL in the veteran home group, while MHLC Internal was the most important in long-term care home inhabitants.

In the present study, multidimensional data from comprehensive geriatric assessment has been examined as potential correlates for psychological characteristics. The overall impact of independent variables on three psychological measures indicates that younger, well-educated, physically active and fit individuals, who tend to be men, tend to have higher LOT-R, MHLC Internal and GSES, along with lower external locus of control (MHLC Powerful Others and MHLC Chance). The influence of disability and concomitant diseases on psychological measures manifested itself as decreased LOT-R, MHLC Internal and GSES and increased MHLC Powerful Others and/or MHLC Chance. However, the relative impact of different diseases seems very diverse. Common cardiovascular disorders like hypertension or ischaemic heart disease had relatively small effect. This finding is in accord with recent data that cardiovascular risk profile was not predictive of subsequent dispositional optimism and self-rated health [44]. On the other hand, disorders often debilitating towards normal function, such as urinary incontinence, cataract or post stroke status, influenced the majority of psychological measures investigated in the study. Of interest is also the considerable influence of depression (GDS).

The relationship of independent variables to LOT-R, GSES and MHLC was, in several aspects, substantially different in the three groups of elders. Education level influenced psychological measures in community-dwelling and veteran home elderly, not in long-term care elders. This group was characterised by the lowest education level. Therefore, it seems that some ‘floor effect’ may be observed here and, beneath a certain level, education has no detectable influence on the psychological characteristics of older individuals.

Current smoking was related to increased MHLC Internal and GSES in long-term care home inhabitants. In community-dwelling and veteran home groups, current smoking was associated with lower MHLC Powerful Others. Therefore, older subjects who smoke, and especially those in long-term care homes, seem to have a more advantageous psychological profile. This is probably due to the fact that long-term care home smokers were younger and more physically and mentally fit (significantly higher muscle strength and Tinetti scores, tendency towards better ADL, IADL, MMSE) when compared to their non-smoking peers. Smoking may also give them the impression of being more in control of their own lives. Of course, this should not pave the way to approbation of a smoking habit in long-term care institutions. Nevertheless, this contrast with associations previously described in the general population and should be taken into consideration while programming care for long-term care residents.

In the present study, the relationship of physical activity, strength and mobility measures to psychological characteristics was more profound in the two older groups in comparison with community-dwelling elderly. In veteran home residents, the highest correlations between the majority of independent variables and psychological profile measures were observed. This group had both the highest MHLC Internal and MHLC Powerful Others, which may be important for good health habits and adherence to medical procedures. Therefore, our data indicates that the beneficial influence of physical activity and performance measures on a desirable psychological profile may be especially important in the oldest frail elders with multiple concomitant diseases. These groups of older subjects should be first targeted with physical activity and performance-improving intervention programmes.

In our study, QOL generally decreased with growing level of dependence and institutionalisation. This supports the results of previous studies indicating declining QOL with advanced age, lower functioning level, coexisting diseases and institution placement. Interestingly, the Euroqol 5D anxiety/depression score was higher in community-dwelling elderly than both institutionalised groups. This finding is in agreement with the results of our previous study, where in older individuals with cardiovascular diseases, hospitalisation decreased overall QOL but improved the perception of well-being in relation to anxiety/depression [45]. This probably suggests higher reassurance and feeling of safety in subjects having continuous and more complex care provided for them, despite their older age and poorer health status. Differences in QOL level were clearly evident with both Euroqol 5D and NHP, supporting their validity and usefulness in older populations [4143]. In contrast, SWLS did not differ among the three populations studied, indicating its limited application in QOL research in the elderly.

Several previous studies have assessed the relationship of dispositional optimism, health locus of control or self-efficacy to QOL in the elderly. Optimism predicted better QOL in cross-sectional [2] and prospective studies [7, 17, 18]. In people with arthritis, higher generalised self-efficacy was associated with greater psychological well-being, both cross-sectionally and longitudinally [1]. Self-efficacy and outcome expectancies (dispositional optimism) were lower in older adults with generalised anxiety disorder [46]. Self-efficacy strongly predicted QOL measured with the 36-item short form (SF-36) questionnaire in frail older people [4]. Health locus of control scales were significantly associated with individual and neighbourhood socio-economic status, as well as with self-rated health in people under 75 years of age [20]. Elderly subjects with internal orientation of locus of control living in an urban community had better self-reported health status, cognitive and personal–social functioning [21]. For centenarians, an apprehensive personality and low levels of control over health were additional correlates of poor subjective health [22].

Our data indicates that while dispositional optimism, internal health locus of control and self-efficacy are important correlates of higher QOL in all environments of older subjects, significant differences were observed in the strength of associations among the three groups. Generally, dispositional optimism, health locus of control and self-efficacy predicted QOL most powerfully in veteran home residents, indicating this group to be the first target for health-promoting activities. An explanation may be that the veteran group represents older people ‘in transition’ between community and institution level. These subjects seem particularly vulnerable to the influence that psychological personal resources may have on QOL. On the other hand, MHLC Internal was the most important QOL correlate in long-term care home inhabitants. This is in accord with the results of Quinn et al. [22], indicating that low levels of control over health were additional correlates of poor QOL in centenarians. Therefore, particular attention should be paid to sustaining a high internal health locus of control in institutionalised, oldest-old and frail elderly.

Alongside examined psychological resources, QOL was related to sociodemographic variables, accompanying diseases, performance measures and physical activity. The most important determinants detracting from QOL were depression and concomitant diseases (mainly cataract, urinary incontinence and musculoskeletal disorders). Main determinants enhancing QOL were a high level of physical activity as well as high physical and cognitive functioning expressed with geriatric scores and tests (ADL, IADL, Tinetti, hand grip strength and MMSE). This is in accord with considerable available literature on QOL determinants in the elderly [4752] and few previous reports that simultaneously assessed QOL and psychological resources with other QOL determinants: mainly physical exercise [53, 54]. In one available study, 20 weeks’ exercise intervention had a positive effect on both increased internal health locus of control and perceived health status in older persons [53]. In another study, exercise + dietary weight loss resulted in improved mobility-related self-efficacy connected with improvements in pain in 316 overweight or obese older adults with symptomatic knee osteoarthritis [54]. The key research challenge is determining which psychosocial factors, concomitant diseases, aspects of physical and cognitive functioning and health-promoting behaviours contribute most to QOL. Our data adds to existing literature on this issue, indicating that the influence of sociodemographic, medical and functional factors is partially complemented by psychological resources. Psychological models should include dispositional optimism, health locus of control and self-efficacy, as this considerably improves medical model in predicting QOL in the elderly.

Several shortcomings of the present study should be acknowledged. The cross-sectional method of this study leaves the question of direction of causality open. Psychological profile measures may act both as dependent and independent variables. The results pertain to volunteers with an adequate level of physical and cognitive functioning and who are able and willing to participate in multiple tests. Especially, relatively healthy and active participants of the University of the Third Age should not be considered as a representative group of community-dwelling elderly. Other sociodemographic variables, such as occupational history or childhood and war experiences, might influence the current psychological profile of older people. Finally, patterns of associations observed in this study may be different in other populations of older subjects, especially in other cultures.

Conclusions

We conclude that dispositional optimism, internal health locus of control and self-efficacy are associated with healthy ageing, expressed by higher QOL. This association, partly mediated by health status, performance and health behaviours, is best visible in older subjects with borderline functional independence. This group of the older population should be the first target of psychosocial preventive and health-promoting measures. Educational interventions, promoting physical activity, enhancing fitness measures and treating concomitant diseases detracting from QOL might be most effective in elderly subjects living ‘in transition’ between a community and institutional environment. To confirm those findings, further prospective well-designed studies are indispensable.

Copyright information

© Springer Science+Business Media B.V. 2010