The Impact of Mobility on Quality of Life Among Older Persons

Abstract

Objective: The purpose of this cross-sectional study was to determine the extent and manner in which mobility is related to quality of life (QOL) in a representative sample of older adults. Method: Two models were assessed using structured equation modeling. The first hypothesized that mobility would have a direct effect on QOL along with other variables, including; number of health conditions and satisfaction with functional capacity, life essentials (e.g., transport, access to health services, and conditions of living space) and personal relationships. The second hypothesized that mobility would have its greatest influence on QOL through satisfaction with functional capacity. Results: While both models found that mobility and all three measures of life satisfaction were significantly associated with QOL, goodness of fit indices were higher for the second model. Discussion: The findings from this study suggest that mobility interventions designed to enhance QOL should address satisfaction with functional capacity as well.

Keywords

physical function physical activity well-being

Positive quality of life (QOL) is considered to be an essential component of successful ageing (Brett et al., 2012), and as a result, enhancing QOL has become the focus of numerous initiatives designed for implementation with older adults including both the World Health Organization’s Active Ageing (World Health Organization [WHO], 2002) and Age-Friendly Cities (WHO, 2007) strategies. A wide range of factors influence the QOL of older adults, such as age, gender, levels of mobility, mental and physical health, autonomy and independence, economic standard of living, social support, and emotional and psychosocial adjustment (Baernholdt, Hinton, Yan, Rose, & Mattos, 2012; Borders, Aday, & Xu, 2004; Fusco et al., 2012; La Grow, Alpass, Stephens, & Towers, 2011; La Grow, Yeung, Towers, Alpass, & Stephens, 2011; Levasseur, Desrosiers, & St-Cyr Tribble, 2008; Salkeld et al., 2000; Steinman & Allen, 2012). However, mobility (i.e., assisted or unassisted movement within and outside the home) appears to be the most amenable of these factors to intervention, including medical innovations (Parker, Gurusamy, & Azegamy, 2010), environmental change and adaptation (Rosenberg et al., 2009; Scanaill et al., 2006), adoption of new technologies and aids (Gao, Ebert, Chen, & Ding, 2012; McCreadie & Tinker, 2005), physical activity adjustments (Vogel et al., 2009), strength training (Liu & Latham, 2009), and use of driving assistance and public transport passes (Wang, Wang, Lin, & Chiang, 2009). Furthermore, mobility limitations have been found to have a negative impact on emotional and psychosocial adjustment, autonomy, and independence (Drewnowski & Evans, 2001; La Grow et al., 2011; Salkeld et al., 2000), and, in turn, QOL. Indeed, Salkeld et al. (2000) found that older women claimed to prefer death to the loss of independence and self-determination resulting from a significant mobility-limiting event.

In a sample of older adults who have difficulty seeing, La Grow, Alpass et al. (2011) found that satisfaction with life, self-reported perception of mobility, and total number of diagnosed health conditions (in that order) made significant and unique contributions to explaining variance in QOL. Yeung, La Grow, Towers, Alpass, & Stephens (2011) followed up this study using a more sophisticated method of analyses (confirmatory factor analysis and structural equation modeling) and a more comprehensive measure of satisfaction with life. The latter consisted of 11 items which resulted in three factors of satisfaction with life: Satisfaction with functional capacity, life essentials (e.g., transport, access to health services, and conditions of living space), and personal relationships. This more detailed analysis found that five variables (total number of diagnosed health conditions, mobility, satisfaction with functional capacity, satisfaction with life essentials, and satisfaction with personal relationships) accounted for 57% of the variance observed in QOL. Mobility was found to have the greatest total (direct and indirect) effect on QOL followed by satisfaction with functional capacity, satisfaction with personal relationships, and satisfaction with life essentials in that order. Mobility was also found to have a direct and significant effect on satisfaction with functional capacity, life essentials, and personal relationships. Satisfaction with functional capacity was found to have a direct effect on satisfaction with life essentials and personal relationships. Total number of health conditions was found to have a significant direct effect on mobility, satisfaction with functional capacity, and satisfaction with personal relationships but not on QOL.

The findings by La Grow, Alpass et al. (2011) and Yeung et al. (2011) both suggest that improving one’s mobility may be an effective way to promote QOL among older persons, while the latter suggests a pathway for doing so. However, as the samples in these studies were limited to those who have difficulty seeing, it is possible that these findings may be unique to this group as loss of mobility is a fundamental consequence of the onset of visual impairment (Brouwer, Sadlo, Winding, & Hanneman, 2008). As would be expected, La Grow, Alpass et al. (2011) found both mobility and QOL to be significantly poorer among those who had difficulty seeing than those who did not (La Grow, Aplass et al., 2011). The purpose of this study, therefore, is to replicate that conducted by Yeung et al. (2011) with a representative sample of community-dwelling older persons to determine if these findings have wider applicability.

Method

Study Sample

As with Yeung et al. (2011), the sample used for this analysis was drawn from participants in the second wave of a large longitudinal study known as the Health, Work and Retirement (HWR) study conducted between May and June of 2008 (see Alpass et al., 2007). Funded by the Health Research Council of New Zealand, the HWR study is a nationally representative longitudinal study of New Zealanders aged 55 to 70 (in 2006) in the transition from work to retirement. As one of the world’s few national-level longitudinal studies of aging, the HWR specifically oversamples older Māori (New Zealand’s indigenous population) in order to facilitate analyses of health and aging of indigenous people. However, in order to produce New Zealand-representative models that can be compared to international older populations, the current analysis and that of Yeung et al. (2011) used population stratification weights to adjust for significant differences in age, gender, and sex ratios between this sample and its comparative New Zealand population-base. Study procedures were approved by the Massey University Human Ethics Committee (MUHEC 05/90) and carried out in accordance with tenants of the Treaty of Helsinki.

Yeung et al. (2011) selected only those who reported having difficulty reading ordinary newsprint even when wearing glasses (N = 356) for their study, while this study included all 2,473 participants including those who reported having difficulty seeing. Participants in this study ranged in age from 56 to 72 years with a mean age of 63.3 (SD = 4.52), 53.0% were women and 72.0% were married or partnered. The mean number of diagnosed chronic conditions reported was 1.97 (±1.84) with an interquartile range of 2 (3 -1).

Measures

This application of Yeung et al.’s (2011) model to a wider sample of older adults (i.e., not limited to those visually impaired) utilized the same independent (total health conditions, mobility, satisfaction with functional capacity, satisfaction with life essentials, and satisfaction with personal relationships) and dependent (QOL) variables used in that study.

Independent variables. Total health conditions was a summary of the number of chronic health conditions from a list of 28 (e.g., diabetes, heart trouble, hernia, arthritis) that participants had been diagnosed with at any stage. Mobility was assessed on a single self-report item asking “How well are you able to get around?,” with responses made on a 5-point Likert-type scale ranging from 1 “Very poor” to 5 “Very well.” Life satisfaction was assessed using 11 items from the World Health Organization’s Quality of Life: BREF assessment instrument (WHOQOL-BREF; Skevington, Lofty, & O’Connell, 2004) which had been selectively included in the HWR second data collection wave. With responses on a 5-point Likert-type scale ranging from 1 “Very dissatisfied” to 5 “Very satisfied” the participants indicated how satisfied they were with their (a) ability to perform activities of daily living, (b) capacity to work, (c) health, (d) yourself, (e) sleep, (f) transport, (g) access to health services, (h) conditions of your living space, (i) personal relationships, (j) sex life, and (k) support you get from friends. A confirmatory factor analysis (CFA) by Yeung et al. (2011) found three stable factors stemming from these items; Satisfaction with functional capacity (items 1-5), satisfaction with life essentials (items 6-8), and satisfaction with personal relationships (items 9 -11). We undertook a CFA for the current study and confirmed the item loading pattern and proposed 3-factor model remains stable in the wider HWR sample (see Table 1), with resulting CFA fit indices confirming that this 3-factor model has excellent fit to the data (RMSEA [Root Mean Square Error of Approximation] = 0.04; GFI [Goodness of Fit Index] = 0.99; AGFI [Adjusted Goodness of Fit Index] = 0.99; CFI [Comparative Fit Index] = 0.99). Reliability analysis revealed adequate to good internal consistency for these subscales (see Table 2).

Table 1.

Rotated Factor Loadings From the Confirmatory Factor Analysis for the 3-Factor Solution for Life Satisfaction Scale (n = 2,473).

Item	Content	Factor 1	Factor 2	Factor 3
1	How satisfied are you with your ability to perform your daily living activities?	0.85	0.23	0.12
2	How satisfied are you with your capacity to work?	0.84	0.22	0.07
3	How satisfied are you with your health?	0.78	0.22	0.18
4	How satisfied are you with yourself?	0.64	0.22	0.44
5	How satisfied are you with your sleep?	0.52	−0.01	0.27
6	How satisfied are you with your transport?	0.22	0.79	0.07
7	How satisfied are you with your access to health services?	0.16	0.77	0.12
8	How satisfied are you with the conditions of your living space?	0.19	0.67	0.30
9	How satisfied are you with your personal relationships?	0.25	0.22	0.79
10	How satisfied are you with your sex life?	0.21	0.03	0.77
11	How satisfied are you with the support you get from friends?	0.09	0.37	0.59

Note: Bolded text indicates factor loading.

Table 2.

Mean, Standard Deviation, and Observed Range for all Variables and Cronbach’s α for Satisfaction With Functional Capacity, Life Essentials, and Personal Relationships.

Scale	Mean (SD)	Observed Range	Cronbach’s α
Quality of life	4.35 (0.70)	1-5	—
Total health conditions	1.97 (1.84)	0-19	—
Mobility	4.48 (0.80)	1-5	—
Satisfaction with functional capacity	19.84 (3.42)	5-25	.79
Satisfaction with life essentials	13.20 (2.06)	3-15	.78
Satisfaction with personal relationships	11.62 (2.19)	3-15	.70

Note: Statistics have been calculated using post-stratification weights to ensure generalizability to the normally resident 55 to 70 years old New Zealand population.

Dependent Variable

The global QOL item from the WHOQOL-BREF (The World Health Organization Quality of Life-BREF; see Skevington et al., 2004) was used to assess QOL. It asks “How would you rate your quality of life?,” with responses made on a 5-point Likert-type scale ranging from 1 “Very poor” to 5 “Very good.” This single item is the principal “benchmark” against which subdomains of the WHOQOL-BREF are compared, and thus reflects QOL at its broadest domain. This single item has been utilized widely in research on older adults as a key indicator of QOL (e.g., Murray, Brett, Starr, & Deary, 2011; Tabah et al., 2010), and it correlates highly with all subdomains of the SF-36 indicator of physical and mental health status (Bonomi, Patrick, Bushnell, & Martin, 2000).

Data Analysis

Structural equation modeling was employed to explore the relationships existing between the five independent variables and QOL using two models. These models were tested via observed variable path analysis using maximum likelihood parameter estimation (AMOS 18.0). A variety of fit indexes were used to test the adequacy of both a measurement and a structural model to the data, including the GFI (Joreskog & Sorbom, 1989), the CFI (Bentler, 1990), and the RMSEA (Steiger, 1990). These indices gauge how well the estimated covariance matrix implied by the model reproduces the observed population covariance matrix. A good fitting model is indicated by GFI, AGFI, and CFI values at or above 0.90 and RMSEA values below 0.05 (Thompson, 2000). Since extensive nonnormality may lead to the overestimation of chi-square values, the underestimation of certain fit indexes (such as CFI), and the underestimation of the existing standard errors of parameter estimates (West, Finch, & Curran, 1995), the bootstrapping method was deployed (Bollen & Stine, 1993). The two modeling approaches explored in the earlier study (Yeung et al., 2011), were assessed here as well. The first proposed that all five variables would have a direct and accumulative effect on QOL, while the second model treated the three factors of satisfaction as mediators between total health conditions, mobility, and QOL.

Results

Table 2 illustrates the descriptive statistics for the independent and dependent variables. As can be seen in Figure 1, the initial model used in this study parallels a standard multiple linear regression and predicted 52% of the variance observed in QOL. Four of the five independent variables contributed significantly to this predicted variance. Satisfaction with functional capacity (β = 0.32, p < .001) made the greatest independent contribution to QOL, followed by satisfaction with personal relationships (β = 0.19, p < .001) and mobility (β = 0.18, p < .001).

Figure 1.

Initial model.

An alternative model based on the findings of Yeung et al. (2011) was then proposed for comparison with the initial model. In this model, we hypothesized that restrictions in mobility may influence QOL directly but that its primary mode of influence would be indirect, based on its principal role in influencing each of the three factors of life satisfaction. This was based on the notion that increased mobility should reflect positively on one’s satisfaction with functional capacity, potential ability to access life essentials, and capacity to engage in personal relationships.

Both the initial model (see Figure 1) and the alternative model (see Figure 2) resulted in the same degree of prediction of QOL (i.e., 52%), however the pathways of influence apparent in the alternative model provide for a more comprehensive interpretation of the mechanisms involved. As before, mobility was found to have contributed directly to QOL (β = 0.15, p < .001). However, it can also be seen to have had a significant relationship with satisfaction with functional capacity (β = 0.68, p < .001), satisfaction with life essentials (β = −0.22, p < .001), and satisfaction with personal relationships (β = −0, 29, p < .001). As found in the initial model, total health conditions were not found to make a significant direct contribution to the prediction of QOL. However, it was found to have had a significant relationship with satisfaction with functional capacity (β = −0.19, p < .001), satisfaction with life essentials (β = 0.10, p < .001) and satisfaction with personal relationships (β = −0.29, p < .001). The three factors of satisfaction with functional capacity (β = 0.36, p < .001), personal relationships (β = 0.19, p < .001), and life essentials (β = 0.16, p < .001) were all found to have had a significant and direct effect on QOL.

Figure 2.

Alternative model.

The fit indices for the initial and alternative models (see Figures 1 and 2) suggest that both fit the data well, but the indices indicate that the goodness of fit is stronger for the second model. The results reported for this model were based on 2,000 bootstrap samples yielding a nonsignificant Bollen-Stine bootstrap p value of .06 and thus indicating that this model did not have to be rejected. Although the Bollen-Stine bootstrap p was trending toward significance we can assume the model is still a good fit because (a) Bollen-Stine bootstrap p values are sensitive to large sample sizes (such as in the current study) that can substantially inflate type II error (see Enders, 2002), and (b) the alternative fit indices were very good. The second model therefore yielded a significantly better fit and was therefore accepted as the final model for explaining the prediction of QOL in this sample.

The accepted model demonstrated a number of indirect paths to QOL. The total effects in the model therefore were decomposed to determine the size of direct and indirect effects. Table 3 shows the SEM regression estimates which highlight the direct, indirect, and total effect of each of the 5 independent variables on QOL. While direct effects are commonly used when interpreting the results of SEM analysis, Bollen (1989) states that total effects present a more comprehensive indication of the influence of one construct on another. The results, presented in Table 3, show that satisfaction with functional capacity had the greatest total effect on QOL (β = 0.66, p < .001), followed by mobility (β = 0.50, p < .001). These two indicate high effects respectively. All other variables were found to have a small effect only.

Table 3.

Direct, Indirect and Total Effects (Beta Coefficients) Illustrating the Effect Size for Each Predictor Variable on Quality of Life.

Scale quality of life	Mobility	Total health conditions	Satisfaction with functional capacity	Satisfaction with lifeessentials	Satisfaction with personal relationships
Direct effect	0.15	−0.02	0.35	0.16	0.19
Indirect effect	0.35	−0.09	0.31	0.05	—
Total effect	0.50	−0.11	0.66	0.21	0.19

Note: Absolute Beta values less than 0.10 indicate a “small” effect, value about 0.30 a “medium” effect; and those greater than 0.50 indicate a “large” effect (Kline, 1998).

Discussion

In this study, we examined whether mobility would have a similar effect on QOL, satisfaction with functional capacity, life essentials and personal relationships in a large representative sample of older adults (N = 2,473) as that found in a much smaller sample (N = 356) of those who had difficulty seeing (Yeung et al., 2011). The amount of variance in QOL explained by the five variables under investigation was reasonably similar (0.57 and 0.52 respectively) across the two studies. Like the earlier study, mobility was found to have both significant direct and indirect effects on QOL. In fact, mobility was found to have an almost identical total effect on QOL in this study (0.50) as that found by Yeung et al. (2011; 0.53), despite the difference in the size of the samples and the potential difference in cause, degree or importance of mobility limitations between the two samples. However, satisfaction with functional capacity (0.66) was found to have a substantially greater total association with QOL in this study than in the earlier study (0.52).

Our current and prior modeling of the relationship between mobility and QOL illustrates that these two factors are highly correlated, but a large portion of that relationship is mediated by satisfaction with functional capacity.

The relationship between mobility and the other two factors of life satisfaction varied from study to study, however. Yeung et al. (2011) found that mobility contributed positively and significantly to all three factors of life satisfaction, yet the current study found that while mobility and satisfaction with functional capacity were positively related, an inverse relationship was evident between mobility and the other two factors of life satisfaction. It is unclear from the current cross-sectional analysis why there is an inverse relationship between mobility and two of the life satisfaction components when these relationships were positive in our prior analysis. We are unable to glean any understanding from alternative work as this is the first study to attempt modeling of these factors in a general population of older adults. Further exploration with longitudinal analysis of this sample is required in order to understand whether this finding reflects key differences in the relationships between mobility and satisfaction with life essentials (e.g., transport, access to health services, and conditions of living space) and personal relationships between those who have difficulty seeing and those who do not, or whether it reflects a transition between mobility states in general (e.g., from unlimited to limited).

The final model accepted here demonstrated that mobility should be thought as multidimensional and more than instrumental in character. The findings from this study may prove to be important to the well-being of older people as mobility appears to be the most amenable to intervention of the factors generally found to influence QOL in this age group. However, the findings reported here and earlier by Yeung et al. (2011) suggest that mobility interventions should focus on increasing one’s satisfaction with functional capacity including such things as the capacity to carry out activities of daily living and work. As a comprehensive approach to enhancing mobility and QOL in older adults health care professionals might consider focusing on multiple facets of this model; providing aids or assistance to deal with environmental challenges while improving physical status by increasing endurance for walking. Further research is needed to examine a broad array of psychological, social, biological, and physical variables as mediators of the relationship existing between mobility and QOL. Continued prospective and experimental research to advance our understanding of methods to determine the impact of mobility interventions on QOL is also required as are controlled studies of the impact of changed states of mobility on QOL.

The limitations of SEM should be considered when interpreting the results of this study. Although SEM is a statistical method that helps explain complex and multivariate relations, it does not allow for examination of reciprocal relations. Furthermore, statements of causality based on SEM must be treated with caution given the correlational nature of the data (Kline, 1998), as well as, the cross-sectional nature of this study. It should also be noted that the number of independent variables included in this study were limited to five. Therefore, this study does not constitute an exhaustive exploration of the factors which could impact QOL. In fact, the amount of variance explained by the two models (i.e., 52%) suggests that there may be other variables that should be added for better understanding of the determinants of QOL. Finally, this study may be limited by the fact that both the important variables of mobility and QOL were measured by single self-report items. The findings of this study, therefore, should be thought of as providing preliminary evidence only.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by the Health Research Council of New Zealand [Grant number HRCo5/311.

References

Alpass

Towers

Stephens

Fitzgerald

Stevenson

Davey

(2007). Independence, well-being, and social participation in an aging population. Annals of the New York Academy of Science, 1114, 241-250.

Baernholdt

Hinton

Yan

Rose

Mattos

(2012). Factors associated with quality of life in older adults in the United States. Quality of Life Research, 21, 527-534.

Bentler

P. M.

(1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238-246.

Bollen

K. A.

(1989). Structural equations with latent variables. New York, NY: John Wiley.

Bollen

Stine

(1993). Bootstrapping goodness-of-fit measures in structural equation models. In Bollen

Long

J. S.

(Eds.), Testng structural equation models (pp. 111-135). Newbury Park, CA: SAGE.

Bonomi

A. E.

Patrick

D. L.

Bushnell

D. M.

Martin

(2000). Validation of the United States’ version of the World Health Organization Quality of Life (WHOQOL) instrument. Journal of Clinical Epidemiology, 53, 1-12.

Borders

Aday

(2004). Factors associated with health-related quality of life among an older population in a largely rural western region. Journal of Rural Health, 20, 67-75.

Brett

C. E.

Gow

A. J.

Corley

Pattie

Starr

J. M.

Deary

I. J.

(2012). Psychosocial factors and health as determinants of quality of life in community-dwelling older adults. Quality of Life Research, 21, 505-516.

Brouwer

D. M.

Sadlo

Winding

Hanneman

M. I. G.

(2008). Limitations in mobility: Experiences of visually impaired older people. British Journal of Occupational Therapy, 71, 414-421.

10.

Drewnowski

Evans

W. J.

(2001). Nutrition, physical activity, and quality of life in older adults: Summary. Journals of Gerontology: Series A, 56A, 89-94.

11.

Enders

C. K.

(2002). Applying the Bollen-Stine Bootstrap for goodness-of-fit measures to structural equation models with missing data. Multivariate Behavioral Research, 37, 359-377.

12.

Fusco

Ferrini

Santoro

Lo Monaco

M. R.

Gambassi

Cesari

(2012). Physical function and perceived quality of life in older persons. Aging Clinical and Experimental Research, 24(1), 68-73.

13.

Gao

Ebert

Chen

Ding

(2012). Design of a mobile social community platform for older Chinese people in urban areas. Human Factors and Ergonomics in Manufactoring & Service Industries, 0, 1-24. doi:10.1002/hfm.20523

14.

Joreskog

K. G.

Sorbom

(1989). LISREL 7: A guide to the program and applications (2nd ed.). Chicago, IL: SPSS Inc.

15.

Kline

R. B.

(1998). Principles and practice of structural equation modeling. New York, NY: Guilford Press.

16.

La Grow

Alpass

Stephens

Towers

. (2011). Factors affecting perceived quality of life of older persons with self-reported visual disability. Quality of Life Research, 20, 407-413.

17.

La Grow

Yeung

Towers

Alpass

Stephens

(2011). Determinants of overall quality of life among older persons who have difficulty seeing: The importance of the ability to get around. Journal of Visual Impairment & Blindness, 105, 720-730.

18.

Levasseur

Desrosiers

St-Cyr Tribble

(2008). Subjective quality-of-life predictors for older adults with physical disabilities. American Journal of Physical Medical Rehabilitation, 87, 830-841.

19.

Liu

C-J.

Latham

N. K.

(2009). Progressive resistance strength training for improving physical function in older adults. Cochrane Database of Systematic Reviews 2009, Issue 3. Art. No.: CD002759. doi:10.1002/14651858.CD002759.pub2.

20.

McCreadie

Tinker

(2005). The acceptability of assistive technology to older people. Ageing & Society, 25, 91-10. doi:10.1017/S0144686X0400248X

21.

Murray

Brett

C. E.

Starr

J. M.

Deary

I. J.

(2011). Which aspects of subjectively reported quality of life are important in predicting mortality beyond known risk factors? The Lothian Birth Cohort 1921 Study. Quality of Life Research, 20, 81-90.

22.

Parker

M. J.

Gurusamy

K. S.

Azegami

(2010). Arthroplasties (with and without bone cement) for proximal femoral fractures in adults. Cochrane Database of Systematic Reviews 2010, Issue 6. Art. No.: CD001706. doi:10.1002/14651858.CD001706.pub4.

23.

Rosenberg

Kerr

Sallis

J. F.

Patrick

Moore

D. J.

King

(2009). Feasinility and outcomes of a multilevel place-based walking intervention for seniors: A pilot study. Health & Place, 15, 173-179. doi:10.1016/j.healthplace.2008.03.010

24.

Salkeld

Cameron

I. D.

Cumming

R. G.

Easter

Seymour

Kurrle

S. E.

Quine

(2000). Quality of life related to fear of falling and hip fracture in older women: A time trade off study. British Medical Journal, 320, 241-246.

25.

Scanaill

C. N.

Carew

Barralon

Noury

Lyons

G. M.

(2006). A review of approaches to mobility telemonitoring of the elderly in their living environment. Annals of Biomedical Engineering, 34, 547-563.

26.

Skevington

Lofty

O’Connell

(2004). The World Health Organization’s WHOQOL-BREF quality of life assessment: Psychometric properties and results of the international field trial. A report from the WHOQOL Group. Quality of Life Research, 13, 299-310.

27.

Steiger

J. H.

(1990). Structural model evaluation and modification: An interval estimation approach. Multivariate Behavioral Research, 25(2), 173-180.

28.

Steinman

B. A.

Allen

S. M.

(2012). Self-reported vision impairment and its contribution to disability among older adults. Journal of Aging and Health, 24, 307-322.

29.

Tabah

Philippart

Timsit

J. F.

Willems

Fracais

Leplege

Garrouste-Orgeas

(2010). Quality of life in patients aged 80 or over after ICU discharge. Critical Care, 14, R2. Retrieved from http://ccforum.com/content/14/1/R2

30.

Thompson

(2000). Ten commandments of structural equation modeling. In Grimm

L. G.

Yarnold

P. R.

(Eds.), Reading and understanding more multivariate statistics (pp. 261-283). Washington, DC: American Psychological Association.

31.

Vogel

Brechat

P-H.

Lepretre

P-M.

Kaltenbach

Berthel

Lonsdorfer

(2009). Health benefits of physical activity in older adults: A review. International Journal of Clinical Practice, 63, 303-320.

32.

Wang

D-A.

Wang

T-L.

Lin

Y-C

Chiang

M-U.

(2009). A movable steering-wheel system for car driving at older ages. Gerotechology, 8, 104-108. doi:10.4017/gt.2009.08.02.003.00.

33.

West

S. G.

Finch

J. F.

Curran

P. J.

(1995). Structural equation models with nonnormal variables: Problems and remedies. In Hoyle

R. H.

(Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 56-75). Thousand Oaks: CA: SAGE.

34.

World Health Organisation (WHO, 2002). Active ageing—A policy framework. Geneva, Switzerland: Author.

35.

World Health Organization (WHO, 2007). Global age friendly cities: A guide. Geneva, Switzerland: Author.

36.

Yeung

La Grow

Towers

Alpass

Stephens

(2011). The centrality of O&M in rehabilitation programs designed to enhance quality of life: A structural equation modelling analysis. International Journal of Orientation & Mobility, 4(1), 10-20.