Archives of Gerontology and Geriatrics 58 (2014) 263–268
Contents lists available at ScienceDirect
Archives of Gerontology and Geriatrics journal homepage: www.elsevier.com/locate/archger
Development and preliminary validation of a new scale to assess functional ability of older population in India Aarti Nagarkar *, Swapnil Gadhave, Shruti Kulkarni Interdisciplinary School of Health Sciences, University of Pune, Ganesh Khind, Maharastra 411007, India
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 May 2013 Received in revised form 5 October 2013 Accepted 9 October 2013 Available online 23 October 2013
Identifying the decline in functional ability and preventing disability is the critical element of the quality of life of an old age. However, the lack of contextual scale to assess the decline in functional capacity is a major issue. Objective of this study is to design the functional ability assessment scale for elderly people in India and test its psychometric properties. Random sample of 659 individuals above 60 years of age from western part of India was recruited. This paper outlines the construction, reliability and validity of a newly developed 14 item scale named as Pune-Functional Ability Assessment Tool (Pune-FAAT). The factors were extracted using the principal component analysis. Two-factor-structure of scale was accepted after applying the K1 rule, scree plot and parallel analysis method. The two factor structure yielded variance of 64.4%. The psychometric properties of the scale were examined using confirmatory factor analysis. The scale has an excellent reliability (Cronbach’s a 0.928) and very good test–retest reliability (r = 0.884). Each subscale demonstrated good internal consistency (Subscale I – Cronbach’s a 0.938 and Subscale II – Cronbach’s a 0.762). Excellent convergent validity with Standford’s health assessment questionnaire (r = 0.959). Discriminant validity was very good as FAAT index showed significant difference in young adults (mean SD 1.11 0.24) and older adults (mean SD 1.69 0.70). This new measure is a potentially valuable research tool for investigating older adult’s functional ability to perform basic and complex daily activities. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Functional ability tool Validity Reliability Elderly India Pune-FAAT
1. Introduction Functional ability means ability to cope with daily-life activities, in a broader sense it indicates health status or quality of life (Wang, 2004; Bowling, 2005). Changes in functional ability are noticeable as age advances (Badiger, Kamath, & Ashalatha, 2010). Disability in old age leads to functional dependence and associated deterioration in health and quality of life (Paterson & Warburton, 2010). Morbidity has significant influence on older individual’s physical functioning and over all well-being. Impairment and disabilities are associated with increased need for social and health care services (Fried & Bush, 1988), as well as hospitalization (Courtney et al., 2011). For last two decades or more, a growing body of research has addressed various concerns namely; measurement of disability (Avlund, Kreiner, & SchultzLarsen, 1996), predictors of disability, (McCusker, Kakuma, & Abrahamowicz, 2002) consequences of disability, and potential preventive interventions (Guralnik, Fried, & Salive, 1996). However,
* Corresponding author. Tel.: +91 020 25691758; fax: +91 020 25690174. E-mail addresses: [email protected], [email protected] (A. Nagarkar), [email protected] (S. Gadhave), [email protected] (S. Kulkarni). 0167-4943/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.archger.2013.10.002
quantifying limitations in daily activities or functional ability of older adults staying in community setting happens infrequently. (Abas, Punpuing, Jirapramupitak, Tangchonlatip, & Leese, 2009; Nascimento et al., 2012; Vass, Avlund, Lauridsen, & Hendriksen, 2005). India is the second largest country in the Asian continent and faces challenge of population aging, increase in the number of people aged 60 and over. The growth of population above 60 years is expected to be to 12.4 percent by the year 2026 from 7.4 percent in year 2001. Average expectancy of life at age 60 is about 17 years (16.7 for males, 18.9 for females) and that at age 70 is around 12 years (10.9 for males and 12.4 for females) in India (CS Office GOI, 2011). Meaning thereby, a large number of elderly will live a longer life. Demand for preventive, promotive and curative care and rehabilitative services will increase phenomenally and will have a significant impact on resource allocation in public health system. Therefore research on various aspects of quality of life of elderly would be the first and most appropriate step for their welfare. One such area of importance is disability. Decline in functional capacity of fourth segment of the population will soon become a public health problem. Functional ability in daily activities can serve as a significant indicator of quality of life of old population in India. (Jotheeswaran, Joseph, & Prince, 2010)
264
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268
In this research, an attempt is made to construct and validate a tool, which measures the declining functional performance with regards to basic and instrumental activities of daily living. Investigators have initially attempted to adapt already existing tools; Katz’s index (Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963), Lawton and Brody’s scale (1969), Barthel index (Collin, Wade, Davies, & Horne, 1988). Health assessment questionnaire (HAQ 20), Fries, Spitz, Kraines, & Holman (1980) assesses a patient’s level of functional ability and includes questions of fine movements of the upper extremity, locomotors activities of the lower extremity, and activities that involve both upper and lower extremities. There are 20 questions in eight categories of functioning, which represent a comprehensive set of functional activities (Bruce & Fries, 2003). Researcher experienced limitations while using this and other instruments in India as most of the tools have been developed in the Western countries. As a result, researchers have attempted to develop a tool and validate it to get reliable estimates of the burden of declining functional performance and identify the specific areas (activity domains) where intervention is necessary. 2. Methods 2.1. Organization of the study and participants A cross-sectional study conducted for the period of two years between year 2010 and 2012 among individuals aged 60 and older. The study was carried out in a Pune district of the state of Maharashtra located in Western parts of India. Validation of a tool was carried out using a random sample of 659 individuals above 60 years of age. These individuals were selected from a list of participants in a longitudinal study of the department, entitled ‘predictors of functional ability among elderly’ which was being conducted among random sample of 1890 individuals above 60 years of age. Sample in the original study was drawn using probability proportional to size (PPS) in a multistage sampling design. The tool was administered to a subsample (659) for determining reliability and validating the structure of the scale. Sixty six participants were subjected to a retest between 6 and 8 weeks after an initial visit. Seventy individuals in the age group of 40–55 years were randomly selected and assessed using the same tool for Discriminant validity. Sixty four respondents participated in convergent validity study and filled in Health Assessment Questionnaire along with the new tool. The study was approved by the University of Pune’s Ethics Committee, India. A written informed consent was obtained from all participants prior to the data collection.
analysis (CFA). KMO test for sampling adequacy and Bartlett’s test of sphericity was conducted to check weather sample is adequate and suitable for factor analysis. After first cycle of PCA with 15 variables, the variables with communalities above 0.4 were retained for further analysis (Costello & Osborne, 2005). The number of factors to be retained was determined by applying three different criteria; Kaiser criterion (K1), scree plot and parallel analysis (PA). Factors with Eigen value greater than 1 were retained as per K1 criteria (Hayton, Allen, & Scarpello, 2004). In this study two factors fulfill the criteria of Eigen value greater than 1; first factor with eleven items and the second with three items. Cattell’s scree plot method was used for further verification of number of factors that resulted in a two factor solution. Hence, a sophisticated method of parallel analysis was conducted as per the procedure described by Hayton and colleagues (2004). Structure of the scale was finalized using CFA. The scale is henceforth referred to as Pune-FAAT. Reliability analysis was carried out on the final scale and subscales individually. The test retest reliability and convergent validity of Pune-FAAT against Health Assessment Questionnaire (HAQ) was examined using Spearman’s rank correlation coefficient. HAQ has been developed by Stanford University. HAQ has been used in several studies to predict changes in functional status and normal aging (Fries et al., 1980). Discriminant validity was determined using mean index scores and compared using independent ‘t’ test. The analysis was carried out using SPSS version 19. 3. Results 3.1. Initial design of the scale
Rigorous, standardized protocols were followed while administrating the questionnaire. The questionnaire was divided into three sections; demographic information, health and disease history and functional ability assessment tool. The filled in schedules were reviewed periodically for completeness, ambiguities, or inconsistencies in order to maintain the quality of the data. In case of an incomplete data, the participants were approached again. The Functional ability was scored according to the respondent’s current level of ability without use of any aid. The same procedure was followed for the retest interviews. All the interviews were conducted in a local language and were preceded by filling up informed consent form.
A six member team comprised of experts in the field of gerontology, social science, psychiatry, and medicine discussed daily activities, which are necessary to lead an independent life in a community. A focus group involving twelve men and women, aged sixty and above was conducted to discuss their daily life activities. The discussion yielded list of 22 commonly performed activities. During the second meeting the group revisited the list, and based on the frequency of activities 15 were finalized. These included 11 daily living and four instrumental activities. A set of eleven activities was comprised of Lifting (weight up to 1–2 kg), bending, squatting, walking (up to 1 km at a time), climbing (1–2 flights of stairs), arising (from bed or chair), handgrip, self cleaning (bathing, shaving,) toilet use, dressing, eating. The instrumental activities were household work (cooking, cleaning, etc.), outside work (shopping), use of public transport and participation in the social gatherings. Each activity was scored from 1 to 4 based on the level of difficulty in performance without the use of any physical aid. Score of 1 was used for ‘without any difficulty’, 2 for ‘with some difficulty’, 3 for ‘with most difficulty’, and 4 for ‘cannot do it’. Score was not computed when the respondents provided answers to less than one third of the items. An index score for each activity was then calculated by summing the scores for each activity and dividing it by the total number of activities answered. The present study includes the scores acquired without use of any assistive device to understand their underlying disability. Average time required for completion of the questionnaire was about 20 min while the time required for administering FAAT was 6–8 min. None of the participants reported difficulties in understanding the instructions or items included, as participants in the study were able to answer all questions.
2.3. Statistical analysis
3.2. Structure of the scale
Structure of a tool was determined by extracting factors using principal component analysis (PCA) and confirmatory factor
Principal component analysis revealed a two factor structure, with factors corresponding to (1) eleven items pertaining to the
2.2. Administration of the tool
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268 Table 1 Pune-FAAT scale variable loading on each Factor.
265
3.3. Factor label and factor explanation
No.
Items
Components 1
2
1 2 3 4 5 6 7 8 9 11 12 10 13 14
Lifting Bending Squatting Walking Climbing Household work Outside work Use of public transport Social gathering Arising from bed/chair Toilet use Self cleaning Dressing Eating food
0.716 0.692 0.719 0.869 0.825 0.815 0.851 0.863 0.841 0.508 0.608 0.168 0.254 0.085
0.298 0.182 0.147 0.142 0.12 0.258 0.173 0.129 0.214 0.441 0.226 0.805 0.78 0.816
daily functions as well as complex activities and (2) three items explaining basic functions of daily life. The item assignments and the communalities of the emerged models can be seen in Table 1. The table shows factor loading of each variable. The component matrix shown in Table 1 does not include items with communalities less than 0.4. Hand Grip function found to be with lower communality value (0.366), hence eliminated from the final analysis. PCA with varimax rotation was extracted with remaining 14 variables. The two-factor model resulted in the simplest structure and explained 64.4% of the variance. The number of factor retained was determined by applying three different criteria; K1 criterion, scree plot and Parallel analysis (PA) based on the eigenvalues of the factors. First factor had eigenvalue of 7.397 it explained 52.8 percent of variance. Second factor demonstrated eigenvalue of 1.620 and explained 11.5 percent of variance together explaining more than 64 percent of variance. The K1 criterion provided evidence for two factor model wherein all factors with eigenvalue greater than 1 should be retained. Cattell’s scree plot method, shown in Fig. 1, gave two factor solution. Parallel analysis of 50 random samples of same sample size also gave two factor solution as seen from Fig. 2. Prior to the PCA, KMO and Barlett’s test was conducted. KMO was found to be 0.936 and Barlett’s test of sphericity was significant with p-value < 0.00. Results indicated sufficiency of sample to carry out factor analysis.
Factor one was consisted of 11 items and each item described different function. Lifting involves movement of hands, bending requires movements of waist. Squatting (0.719), walking (0.869), climbing (0.825) with high communality value loaded on factor one, are important activities for independent living in community set up. Household work, work outside the household, use of public transport and participation in social gathering showed high communality value. Arising from bed or chair and toilet use are essential movements in a daily routine to maintain independence. Three items namely self care, dressing and eating food are basic activities of daily living, primarily loaded on factor two. 3.4. Confirmatory factor analysis (CFA) Correlation between first factor score with second factor score was calculated (r = 0; p-value > 0.05) prior to CFA, which indicated that the two factors are distinct and they measure different dimension of the same construct. Model fit for two factor structure was tested and compared with one factor structure solution using confirmatory factor analysis by structural equation modeling (SEM), performed using STATA version 13. Confirmatory factor analysis produced a two factor model of good fit (x2 = (76) 633.76, P < 0.001, RMSEA = 0.103, SRMR = 0.055, CFI = 0.910, TLI = 0.893). The one factor structure model was not as good fit to data as two factor model (x2 = (77) 998.28, P < .001, RMSEA = 0.135, SRMR = 0.081, CFI = 0.852, TLI = 0.825). Comparative fit indices, Akaike information criterion (AIC) or Bayesian information criterion (BIC) along with coefficient of determination (CD) were calculated for both models and compared. One factor structure model values (AIC = 19,030 and BIC = 19,232) were greater than two factor model (AIC = 17,404 and BIC = 17,597). Coefficient of determinant (CD) for two factor structure which was higher (CD 0.987) than for one factor structure (CD 0.955) indicating two factor structure was better fit model than one factor structure model. Author’s inference for better suitability of two factor model is based on the values obtained for indices like SRMR = 0.055 and CFI = 0.910 based on Hu and Bentler (1999) two index presentation strategy. Though CFI value in two factor model did not fit the cut off, it is fairly close to the cut off value as compared to the one factor structure model. Thus the two factors (i) activities of daily living and (ii) basic activities of daily living model was considered as a final model.
Table 2 Reliability analysis of Pune-FAAT. Mean (SD)
Corrected item-total correlation
Cronbach’s a if item deleted
Factor I (ADL) Cronbach’s a = 0.938 1 Lifting 2 Bending 3 Squatting 4 Walking 5 Climbing 6 Household work 7 Outside work 8 Use of public transport 9 Social gathering 10 Arising from bed/chair 11 Toilet use
1.71 (0.968) 1.86 (1.068) 2.00 (1.12) 1.9 (1.131) 2.17 (1.180) 1.73 (1.076) 1.94 (1.22) 2.04 (1.233) 1.97 (1.236) 1.33 (0.644) 1.83 (1.053)
0.720 0.664 0.683 0.836 0.776 0.812 0.824 0.824 0.825 0.558 0.592
0.934 0.936 0.935 0.928 0.931 0.930 0.929 0.929 0.929 0.940 0.938
Factor II (basic ADL) Cronbach’s a = 0.762 1 Self cleaning 2 Dressing 3 Eating
1.10 (0.401) 1.10 (0.356) 1.05 (0.302)
0.601 0.617 0.587
0.685 0.653 0.701
No.
Item
[(Fig._1)TD$IG]
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268
266
Fig. 1. Scree plot of Pune-FAAT.
3.5. Reliability and validity of Pune-FAAT Pune-FAAT was further tested for various criteria of reliability and validity. Cronbach’s a = 0.938 was obtained for first factor, while Cronbach’s a = 0.762 for second factor indicating a high degree of internal consistency, as seen in Table 2. Individual scale item statistics confirmed this finding. All Items exhibit high itemto-scale correlations as seen from Table 2. Corrected item to total correlations ranged from r = 0.55 to r = 0.83 for subscale I, and r = 0.58 to r = 0.61 for subscale II. The computed value of r = 0.884 (P < 0.01) indicated high test-retest reliability. Convergent validity: There is no benchmark scale to measure functional ability for Indian elderly. Hence the convergent validity of the scale was compared with Health Assessment Questionnaire. We used Spearman’s rank correlation to compare the corresponding values of Pune-FAAT and HAQ which (r = 0.959; P < 0.01), suggested higher correlation. Discriminant validity: Pune-FAAT was administered in young adult population who were between 40 and 55years of age. Null hypothesis was formulated for testing the scores of young and old population. Significant difference (P < 0.01) in mean scores of FAAT index between participants aged 40 to 55 (mean SD
[(Fig._2)TD$IG]
8 7
Eigen value
6 5 Actual dataset
4
Mean
3
95th percenle 2 1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Factors Fig. 2. Plot of actual versus randomly generated eigen values (mean and 95th percentile) for Pune-FAAT scale.
1.11 0.24) and above 60 years (mean SD 1.69 0.70) disproved the null hypothesis. The FAAT index was able to differentiate between young adults and older adults’ functional status. 4. Discussion Pune-FAAT is a potentially valuable tool in an aging individual to identify the risk of functional decline. It includes various culture specific activities. Squatting and bending are of importance as most of the activities are carried out at the ground level, in India. We decided to include the activity of ‘moving around using public transport’ as a variable because it triggers a series of complex tasks such as finding an appropriate bus route, or alternative public transport vehicles, paying the fare, getting in and out of the vehicle at desired destination. Thus the scale attempts to relate with the day to day tasks of the participants. This paper presented a psychometrically strong, 14 item functional ability assessment tool in the context of Indian society. To date, several ADL and IADL measures have been developed in industrialized nations and are routinely used for assessment of functional ability. Disease specific functional assessment instruments for demented patients (Fillenbaum et al., 1999), arthritis patients (Kumar, Malaviya, Pandhi, & Singh, 2002), osteoarthritis patients (Batra et al., 2012) are developed for patients in India. However, attempts to develop context specific instruments for community dwelling older individuals are infrequent. A limitation of the paper is except SRMR and CFI values other fit indices obtained by CFA are not able to fulfill the criteria for acceptance of a scale. Other comparative indices have suggested two factor structure model than one factor structure model. Higher CD value for two factor structure model and totally uncorrelated factor scores indicate that two factor structure model better explains functional ability phenomenon than one factor structure model. Higher reliability of the scale can be contributed to misfit of x2 and RMSEA values (Miles & Shevlin, 2007). Authors would like to clarify that the present study is exploratory in nature and not based on any priori or theoretical construct. CFA is most essential technique when research begins with a priory hypothesis or theoretical base (Hurley et al., 1997). Lack of theoretical construct in this research suggests that results from CFA in this study should be interpreted cautiously before rejecting the overall validity of the scale. Therefore, refusing merit of this tool on the basis of CFA results would not be justifiable.
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268
Cronbach’s alpha (0.928) of the Pune-FAAT indicates a higher value than conventionally expected value while, there are different reports about the acceptable values of alpha, ranging from 0.70 to 0.95 (Bland & Altman, 1997; DeVellis, 2003; Nunnally & Bernstein, 1994). Multidimensional test does not necessarily show lower alpha than single dimensional test (Tavakol & Dennick, 2011). The proposed scale warrants consideration from the perspective of functional performance of the healthy elderly in community. Its validity is not tested in measuring disease specific functional decline (for example as found in arthritis, gout, asthma, or stroke). Our results demonstrate that the Pune-FAAT is effective in identifying functional level or activity performance among normal subjects, and demonstrate high test–retest reliability and internal consistency and discriminant validity. Resources and time required to complete the questionnaire are much less and can be easily used in a household survey with population having lower levels of education. 5. Conclusions We have demonstrated that a new tool, Pune-FAAT is contextual and is better than existing tools in identifying decline in functional ability. Use of this tool can be recommended in the community dwelling geriatric health screening program.
267
Appendix A (Continued ) No.
Item
7.
Household work Outside work Use of public transport Social gathering Self cleaning (Bathing) Toilet use Dressing Eating food A. Sum of the total B. No. of Item answered Pune-FAAT index = (A/B)
8. 9. 10. 11. 12. 13. 14.
No difficulty performing (1)
Somewhat difficult to perform (2)
Most difficult to perform (3)
Unable to perform (4)
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.archger.2013.10.002.
Author’s contribution All authors made substantial contributions to the paper. AN who is PI of the study, had main responsibility of research design, executing and making first through final draft of paper by working on each version. SG had main responsibility for data management; data entry, statistical analysis and contributing to all versions of the manuscript. SK had main responsibility for acquisition of data, entry and preliminary analysis. SK contributed to the development of tool by way of review, initial tool development and baseline testing. All authors were involved in revising the manuscript critically and have given final approval of the version to be published. Conflict of interest The authors declare that they have no competing or financial interests. Funding source This study is supported by the departmental research funding received from the University. Appendix A Name of the respondent: Form No Please indicate your ability to perform following tasks without the use of any assistive device or physical help. No.
Item
1.
Lifting weight (up to 1–2 kg) Bending Squatting Walking (up to 1–2 km) Climbing Arising from bed/chair
2. 3. 4. 5. 6.
No difficulty performing (1)
Somewhat difficult to perform (2)
Most difficult to perform (3)
Unable to perform (4)
References Abas, M. A., Punpuing, S., Jirapramupitak, T., Tangchonlatip, K., & Leese, M. (2009). Psychological wellbeing, physical impairments and rural aging in a developing country setting. Health and Quality of Life Outcomes, 7, 66. Avlund, K., Kreiner, S., & Schultz-Larsen, K. (1996). Functional ability scales for the elderly: A validation study. European Journal of Public Health, 6, 35–42. Badiger, M. S., Kamath, R., & Ashalatha, K. V. (2010). Correlates of functional ability among the elderly. Karnataka Journal of Agricultural Sciences, 23, 480–484. Bland, J., & Altman, D. (1997). Statistics notes: Cronbach’s alpha. British Medical Journal, 314, 275. Batra, V., Sharma, V. P., Batra, M., Sharma, V., Agarwal, G. G., Singh, V. K., et al. (2012). Measurer of functional disability status in knee osteoarthritis. Internet Journal of Medical Update, 7, 47–51. Bowling, A. (2005). Measuring health. A review of quality of life measurement scales (3rd ed.). Maidenhead, England: Open University Press. Bruce, B., & Fries, J. F. (2003). The Stanford Health Assessment Questionnaire: A review of its history, issues, progress, and documentation. Journal of Rheumatology, 30, 167–178. Central Statistics Office Ministry of Statistics & Programme Implementation, (2011) Situation Analysis of the Elderly in India Government of India. Retrieved May 11, 2013, from http://mospi.nic.in/mospi_new/upload/elderly_in_india.pdf. Collin, C., Wade, D. T., Davies, S., & Horne, V. (1988). The Barthel ADL index: A reliability study. International disability studies, 10(2), 61–63. Costello, A. B., & Osborne, J. W. (2005). Best Practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical assessment. Research & Evaluation, 10(7), 1–9. Courtney, M. D., Edwards, H. E., Anne, M. C., Parker, A. W., Kathleen, F., & Hamilton, K. (2011). A randomised controlled trial to prevent hospital readmissions and loss of functional ability in high risk older adults: A study protocol. BMC Health Services Research, 11, 202. DeVellis, R. (2003). Scale development: Theory and applications: Theory and application. Thousand Oaks, CA: Sage. Fillenbaum, G. G., Chandra, V., Ganguli, M., Pandav, R., Gilby, J. E., Seaberg, E. C., et al. (1999). Development of activities of daily living scale to screen for dementia in an illiterate rural older population in India. Age and Ageing, 28, 161–168. Fried, L. P., & Bush, T. L. (1988). Morbidity as a focus of preventive healthcare in the elderly. Epidemiologic Reviews, 10(1), 48–64. Fries, J. F., Spitz, P., Kraines, R. G., & Holman, H. R. (1980). Measurement of patient outcomes in arthritis. Arthritis and Rheumatism, 23, 137–145. Guralnik, J. M., Fried, L. P., & Salive, M. E. (1996). Disability as a public health outcome in the aging. Population Annual Review of Public Health, 17, 25–46. Hayton, J. C., Allen, D. G., & Scarpello, V. (2004). Factor retention decisions in exploratory factor analysis: A tutorial on parallel analysis. Organizational Research Methods, 7(2), 191–205. Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. Hurley, A. E., Scandura, T. A., Schriesheim, C. A., Brannick, M. T., Seers, A., Vandenberg, R. J., et al. (1997). Exploratory and confirmatory factor analysis: Guidelines, issues, and alternatives. Journal of Organizational Behavior, 18, 667–683.
268
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268
Jotheeswaran, A. T., Joseph, D. W., & Prince, M. J. (2010). Predictors of mortality among elderly people living in a south Indian urban community; a 10/66 Dementia research group prospective population-based Cohort study. BMC Public Health, 10, 366. Katz, S., Ford, A. B., Moskowitz, R. W., Jackson, B. A., & Jaffe, M. W. (1963). Studies of illness in the aged. The index of ADL: A standardized measure of biological and psychosocial function. Journal of American Medical Association, 185, 914–919. Kumar, A., Malaviya, A. N., Pandhi, A., & Singh, R. (2002). Validation of an Indian version of the health assessment questionnaire in patients with rheumatoid arthritis. Rheumatology, 41, 1457–1459. Lawton, M. P., & Brody, E. M. (1969). Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist, 9, 179–186. McCusker, J., Kakuma, R., & Abrahamowicz, M. (2002). Predictors of functional decline in hospitalized elderly patients: A systematic review. Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 57, 569–577. Miles, J., & Shevlin, M. (2007). A time and a place for incremental fit indices. Personality and Individual Differences, 42(5), 869–874.
Nascimento, C. M., Ribeiro, A. Q., Cotta, R. M., Acurcio, F. A., Peixoto, S. V., Priore, S. E., et al. (2012). Factors associated with functional ability in Brazilian elderly. Archives of Gerontology and Geriatrics, 54, 89–94. Nunnally, J., & Bernstein, L. (1994). Psychometric theory. New York: McGraw-Hill Higher. Paterson, D. H., & Warburton, E. R. (2010). Physical activity and functional limitations in older adults: A systematic review related to Canada’s physical activity guidelines. International Journal of Behavioral Nutrition and Physical Activity, 7, 38. Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach’s alpha. International Journal of Medical Education, 2, 53–55. Vass, M., Avlund, K., Lauridsen, J., & Hendriksen, C. (2005). Feasible model for prevention of functional decline in older people: Municipality-randomized, controlled trial. Journal of American Geriatric Society, 53, 563–568. Wang, T. J. (2004). Concept analysis of functional status. International Journal of Nursing Studies, 41, 457–462.
Contents lists available at ScienceDirect
Archives of Gerontology and Geriatrics journal homepage: www.elsevier.com/locate/archger
Development and preliminary validation of a new scale to assess functional ability of older population in India Aarti Nagarkar *, Swapnil Gadhave, Shruti Kulkarni Interdisciplinary School of Health Sciences, University of Pune, Ganesh Khind, Maharastra 411007, India
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 May 2013 Received in revised form 5 October 2013 Accepted 9 October 2013 Available online 23 October 2013
Identifying the decline in functional ability and preventing disability is the critical element of the quality of life of an old age. However, the lack of contextual scale to assess the decline in functional capacity is a major issue. Objective of this study is to design the functional ability assessment scale for elderly people in India and test its psychometric properties. Random sample of 659 individuals above 60 years of age from western part of India was recruited. This paper outlines the construction, reliability and validity of a newly developed 14 item scale named as Pune-Functional Ability Assessment Tool (Pune-FAAT). The factors were extracted using the principal component analysis. Two-factor-structure of scale was accepted after applying the K1 rule, scree plot and parallel analysis method. The two factor structure yielded variance of 64.4%. The psychometric properties of the scale were examined using confirmatory factor analysis. The scale has an excellent reliability (Cronbach’s a 0.928) and very good test–retest reliability (r = 0.884). Each subscale demonstrated good internal consistency (Subscale I – Cronbach’s a 0.938 and Subscale II – Cronbach’s a 0.762). Excellent convergent validity with Standford’s health assessment questionnaire (r = 0.959). Discriminant validity was very good as FAAT index showed significant difference in young adults (mean SD 1.11 0.24) and older adults (mean SD 1.69 0.70). This new measure is a potentially valuable research tool for investigating older adult’s functional ability to perform basic and complex daily activities. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Functional ability tool Validity Reliability Elderly India Pune-FAAT
1. Introduction Functional ability means ability to cope with daily-life activities, in a broader sense it indicates health status or quality of life (Wang, 2004; Bowling, 2005). Changes in functional ability are noticeable as age advances (Badiger, Kamath, & Ashalatha, 2010). Disability in old age leads to functional dependence and associated deterioration in health and quality of life (Paterson & Warburton, 2010). Morbidity has significant influence on older individual’s physical functioning and over all well-being. Impairment and disabilities are associated with increased need for social and health care services (Fried & Bush, 1988), as well as hospitalization (Courtney et al., 2011). For last two decades or more, a growing body of research has addressed various concerns namely; measurement of disability (Avlund, Kreiner, & SchultzLarsen, 1996), predictors of disability, (McCusker, Kakuma, & Abrahamowicz, 2002) consequences of disability, and potential preventive interventions (Guralnik, Fried, & Salive, 1996). However,
* Corresponding author. Tel.: +91 020 25691758; fax: +91 020 25690174. E-mail addresses: [email protected], [email protected] (A. Nagarkar), [email protected] (S. Gadhave), [email protected] (S. Kulkarni). 0167-4943/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.archger.2013.10.002
quantifying limitations in daily activities or functional ability of older adults staying in community setting happens infrequently. (Abas, Punpuing, Jirapramupitak, Tangchonlatip, & Leese, 2009; Nascimento et al., 2012; Vass, Avlund, Lauridsen, & Hendriksen, 2005). India is the second largest country in the Asian continent and faces challenge of population aging, increase in the number of people aged 60 and over. The growth of population above 60 years is expected to be to 12.4 percent by the year 2026 from 7.4 percent in year 2001. Average expectancy of life at age 60 is about 17 years (16.7 for males, 18.9 for females) and that at age 70 is around 12 years (10.9 for males and 12.4 for females) in India (CS Office GOI, 2011). Meaning thereby, a large number of elderly will live a longer life. Demand for preventive, promotive and curative care and rehabilitative services will increase phenomenally and will have a significant impact on resource allocation in public health system. Therefore research on various aspects of quality of life of elderly would be the first and most appropriate step for their welfare. One such area of importance is disability. Decline in functional capacity of fourth segment of the population will soon become a public health problem. Functional ability in daily activities can serve as a significant indicator of quality of life of old population in India. (Jotheeswaran, Joseph, & Prince, 2010)
264
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268
In this research, an attempt is made to construct and validate a tool, which measures the declining functional performance with regards to basic and instrumental activities of daily living. Investigators have initially attempted to adapt already existing tools; Katz’s index (Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963), Lawton and Brody’s scale (1969), Barthel index (Collin, Wade, Davies, & Horne, 1988). Health assessment questionnaire (HAQ 20), Fries, Spitz, Kraines, & Holman (1980) assesses a patient’s level of functional ability and includes questions of fine movements of the upper extremity, locomotors activities of the lower extremity, and activities that involve both upper and lower extremities. There are 20 questions in eight categories of functioning, which represent a comprehensive set of functional activities (Bruce & Fries, 2003). Researcher experienced limitations while using this and other instruments in India as most of the tools have been developed in the Western countries. As a result, researchers have attempted to develop a tool and validate it to get reliable estimates of the burden of declining functional performance and identify the specific areas (activity domains) where intervention is necessary. 2. Methods 2.1. Organization of the study and participants A cross-sectional study conducted for the period of two years between year 2010 and 2012 among individuals aged 60 and older. The study was carried out in a Pune district of the state of Maharashtra located in Western parts of India. Validation of a tool was carried out using a random sample of 659 individuals above 60 years of age. These individuals were selected from a list of participants in a longitudinal study of the department, entitled ‘predictors of functional ability among elderly’ which was being conducted among random sample of 1890 individuals above 60 years of age. Sample in the original study was drawn using probability proportional to size (PPS) in a multistage sampling design. The tool was administered to a subsample (659) for determining reliability and validating the structure of the scale. Sixty six participants were subjected to a retest between 6 and 8 weeks after an initial visit. Seventy individuals in the age group of 40–55 years were randomly selected and assessed using the same tool for Discriminant validity. Sixty four respondents participated in convergent validity study and filled in Health Assessment Questionnaire along with the new tool. The study was approved by the University of Pune’s Ethics Committee, India. A written informed consent was obtained from all participants prior to the data collection.
analysis (CFA). KMO test for sampling adequacy and Bartlett’s test of sphericity was conducted to check weather sample is adequate and suitable for factor analysis. After first cycle of PCA with 15 variables, the variables with communalities above 0.4 were retained for further analysis (Costello & Osborne, 2005). The number of factors to be retained was determined by applying three different criteria; Kaiser criterion (K1), scree plot and parallel analysis (PA). Factors with Eigen value greater than 1 were retained as per K1 criteria (Hayton, Allen, & Scarpello, 2004). In this study two factors fulfill the criteria of Eigen value greater than 1; first factor with eleven items and the second with three items. Cattell’s scree plot method was used for further verification of number of factors that resulted in a two factor solution. Hence, a sophisticated method of parallel analysis was conducted as per the procedure described by Hayton and colleagues (2004). Structure of the scale was finalized using CFA. The scale is henceforth referred to as Pune-FAAT. Reliability analysis was carried out on the final scale and subscales individually. The test retest reliability and convergent validity of Pune-FAAT against Health Assessment Questionnaire (HAQ) was examined using Spearman’s rank correlation coefficient. HAQ has been developed by Stanford University. HAQ has been used in several studies to predict changes in functional status and normal aging (Fries et al., 1980). Discriminant validity was determined using mean index scores and compared using independent ‘t’ test. The analysis was carried out using SPSS version 19. 3. Results 3.1. Initial design of the scale
Rigorous, standardized protocols were followed while administrating the questionnaire. The questionnaire was divided into three sections; demographic information, health and disease history and functional ability assessment tool. The filled in schedules were reviewed periodically for completeness, ambiguities, or inconsistencies in order to maintain the quality of the data. In case of an incomplete data, the participants were approached again. The Functional ability was scored according to the respondent’s current level of ability without use of any aid. The same procedure was followed for the retest interviews. All the interviews were conducted in a local language and were preceded by filling up informed consent form.
A six member team comprised of experts in the field of gerontology, social science, psychiatry, and medicine discussed daily activities, which are necessary to lead an independent life in a community. A focus group involving twelve men and women, aged sixty and above was conducted to discuss their daily life activities. The discussion yielded list of 22 commonly performed activities. During the second meeting the group revisited the list, and based on the frequency of activities 15 were finalized. These included 11 daily living and four instrumental activities. A set of eleven activities was comprised of Lifting (weight up to 1–2 kg), bending, squatting, walking (up to 1 km at a time), climbing (1–2 flights of stairs), arising (from bed or chair), handgrip, self cleaning (bathing, shaving,) toilet use, dressing, eating. The instrumental activities were household work (cooking, cleaning, etc.), outside work (shopping), use of public transport and participation in the social gatherings. Each activity was scored from 1 to 4 based on the level of difficulty in performance without the use of any physical aid. Score of 1 was used for ‘without any difficulty’, 2 for ‘with some difficulty’, 3 for ‘with most difficulty’, and 4 for ‘cannot do it’. Score was not computed when the respondents provided answers to less than one third of the items. An index score for each activity was then calculated by summing the scores for each activity and dividing it by the total number of activities answered. The present study includes the scores acquired without use of any assistive device to understand their underlying disability. Average time required for completion of the questionnaire was about 20 min while the time required for administering FAAT was 6–8 min. None of the participants reported difficulties in understanding the instructions or items included, as participants in the study were able to answer all questions.
2.3. Statistical analysis
3.2. Structure of the scale
Structure of a tool was determined by extracting factors using principal component analysis (PCA) and confirmatory factor
Principal component analysis revealed a two factor structure, with factors corresponding to (1) eleven items pertaining to the
2.2. Administration of the tool
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268 Table 1 Pune-FAAT scale variable loading on each Factor.
265
3.3. Factor label and factor explanation
No.
Items
Components 1
2
1 2 3 4 5 6 7 8 9 11 12 10 13 14
Lifting Bending Squatting Walking Climbing Household work Outside work Use of public transport Social gathering Arising from bed/chair Toilet use Self cleaning Dressing Eating food
0.716 0.692 0.719 0.869 0.825 0.815 0.851 0.863 0.841 0.508 0.608 0.168 0.254 0.085
0.298 0.182 0.147 0.142 0.12 0.258 0.173 0.129 0.214 0.441 0.226 0.805 0.78 0.816
daily functions as well as complex activities and (2) three items explaining basic functions of daily life. The item assignments and the communalities of the emerged models can be seen in Table 1. The table shows factor loading of each variable. The component matrix shown in Table 1 does not include items with communalities less than 0.4. Hand Grip function found to be with lower communality value (0.366), hence eliminated from the final analysis. PCA with varimax rotation was extracted with remaining 14 variables. The two-factor model resulted in the simplest structure and explained 64.4% of the variance. The number of factor retained was determined by applying three different criteria; K1 criterion, scree plot and Parallel analysis (PA) based on the eigenvalues of the factors. First factor had eigenvalue of 7.397 it explained 52.8 percent of variance. Second factor demonstrated eigenvalue of 1.620 and explained 11.5 percent of variance together explaining more than 64 percent of variance. The K1 criterion provided evidence for two factor model wherein all factors with eigenvalue greater than 1 should be retained. Cattell’s scree plot method, shown in Fig. 1, gave two factor solution. Parallel analysis of 50 random samples of same sample size also gave two factor solution as seen from Fig. 2. Prior to the PCA, KMO and Barlett’s test was conducted. KMO was found to be 0.936 and Barlett’s test of sphericity was significant with p-value < 0.00. Results indicated sufficiency of sample to carry out factor analysis.
Factor one was consisted of 11 items and each item described different function. Lifting involves movement of hands, bending requires movements of waist. Squatting (0.719), walking (0.869), climbing (0.825) with high communality value loaded on factor one, are important activities for independent living in community set up. Household work, work outside the household, use of public transport and participation in social gathering showed high communality value. Arising from bed or chair and toilet use are essential movements in a daily routine to maintain independence. Three items namely self care, dressing and eating food are basic activities of daily living, primarily loaded on factor two. 3.4. Confirmatory factor analysis (CFA) Correlation between first factor score with second factor score was calculated (r = 0; p-value > 0.05) prior to CFA, which indicated that the two factors are distinct and they measure different dimension of the same construct. Model fit for two factor structure was tested and compared with one factor structure solution using confirmatory factor analysis by structural equation modeling (SEM), performed using STATA version 13. Confirmatory factor analysis produced a two factor model of good fit (x2 = (76) 633.76, P < 0.001, RMSEA = 0.103, SRMR = 0.055, CFI = 0.910, TLI = 0.893). The one factor structure model was not as good fit to data as two factor model (x2 = (77) 998.28, P < .001, RMSEA = 0.135, SRMR = 0.081, CFI = 0.852, TLI = 0.825). Comparative fit indices, Akaike information criterion (AIC) or Bayesian information criterion (BIC) along with coefficient of determination (CD) were calculated for both models and compared. One factor structure model values (AIC = 19,030 and BIC = 19,232) were greater than two factor model (AIC = 17,404 and BIC = 17,597). Coefficient of determinant (CD) for two factor structure which was higher (CD 0.987) than for one factor structure (CD 0.955) indicating two factor structure was better fit model than one factor structure model. Author’s inference for better suitability of two factor model is based on the values obtained for indices like SRMR = 0.055 and CFI = 0.910 based on Hu and Bentler (1999) two index presentation strategy. Though CFI value in two factor model did not fit the cut off, it is fairly close to the cut off value as compared to the one factor structure model. Thus the two factors (i) activities of daily living and (ii) basic activities of daily living model was considered as a final model.
Table 2 Reliability analysis of Pune-FAAT. Mean (SD)
Corrected item-total correlation
Cronbach’s a if item deleted
Factor I (ADL) Cronbach’s a = 0.938 1 Lifting 2 Bending 3 Squatting 4 Walking 5 Climbing 6 Household work 7 Outside work 8 Use of public transport 9 Social gathering 10 Arising from bed/chair 11 Toilet use
1.71 (0.968) 1.86 (1.068) 2.00 (1.12) 1.9 (1.131) 2.17 (1.180) 1.73 (1.076) 1.94 (1.22) 2.04 (1.233) 1.97 (1.236) 1.33 (0.644) 1.83 (1.053)
0.720 0.664 0.683 0.836 0.776 0.812 0.824 0.824 0.825 0.558 0.592
0.934 0.936 0.935 0.928 0.931 0.930 0.929 0.929 0.929 0.940 0.938
Factor II (basic ADL) Cronbach’s a = 0.762 1 Self cleaning 2 Dressing 3 Eating
1.10 (0.401) 1.10 (0.356) 1.05 (0.302)
0.601 0.617 0.587
0.685 0.653 0.701
No.
Item
[(Fig._1)TD$IG]
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268
266
Fig. 1. Scree plot of Pune-FAAT.
3.5. Reliability and validity of Pune-FAAT Pune-FAAT was further tested for various criteria of reliability and validity. Cronbach’s a = 0.938 was obtained for first factor, while Cronbach’s a = 0.762 for second factor indicating a high degree of internal consistency, as seen in Table 2. Individual scale item statistics confirmed this finding. All Items exhibit high itemto-scale correlations as seen from Table 2. Corrected item to total correlations ranged from r = 0.55 to r = 0.83 for subscale I, and r = 0.58 to r = 0.61 for subscale II. The computed value of r = 0.884 (P < 0.01) indicated high test-retest reliability. Convergent validity: There is no benchmark scale to measure functional ability for Indian elderly. Hence the convergent validity of the scale was compared with Health Assessment Questionnaire. We used Spearman’s rank correlation to compare the corresponding values of Pune-FAAT and HAQ which (r = 0.959; P < 0.01), suggested higher correlation. Discriminant validity: Pune-FAAT was administered in young adult population who were between 40 and 55years of age. Null hypothesis was formulated for testing the scores of young and old population. Significant difference (P < 0.01) in mean scores of FAAT index between participants aged 40 to 55 (mean SD
[(Fig._2)TD$IG]
8 7
Eigen value
6 5 Actual dataset
4
Mean
3
95th percenle 2 1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Factors Fig. 2. Plot of actual versus randomly generated eigen values (mean and 95th percentile) for Pune-FAAT scale.
1.11 0.24) and above 60 years (mean SD 1.69 0.70) disproved the null hypothesis. The FAAT index was able to differentiate between young adults and older adults’ functional status. 4. Discussion Pune-FAAT is a potentially valuable tool in an aging individual to identify the risk of functional decline. It includes various culture specific activities. Squatting and bending are of importance as most of the activities are carried out at the ground level, in India. We decided to include the activity of ‘moving around using public transport’ as a variable because it triggers a series of complex tasks such as finding an appropriate bus route, or alternative public transport vehicles, paying the fare, getting in and out of the vehicle at desired destination. Thus the scale attempts to relate with the day to day tasks of the participants. This paper presented a psychometrically strong, 14 item functional ability assessment tool in the context of Indian society. To date, several ADL and IADL measures have been developed in industrialized nations and are routinely used for assessment of functional ability. Disease specific functional assessment instruments for demented patients (Fillenbaum et al., 1999), arthritis patients (Kumar, Malaviya, Pandhi, & Singh, 2002), osteoarthritis patients (Batra et al., 2012) are developed for patients in India. However, attempts to develop context specific instruments for community dwelling older individuals are infrequent. A limitation of the paper is except SRMR and CFI values other fit indices obtained by CFA are not able to fulfill the criteria for acceptance of a scale. Other comparative indices have suggested two factor structure model than one factor structure model. Higher CD value for two factor structure model and totally uncorrelated factor scores indicate that two factor structure model better explains functional ability phenomenon than one factor structure model. Higher reliability of the scale can be contributed to misfit of x2 and RMSEA values (Miles & Shevlin, 2007). Authors would like to clarify that the present study is exploratory in nature and not based on any priori or theoretical construct. CFA is most essential technique when research begins with a priory hypothesis or theoretical base (Hurley et al., 1997). Lack of theoretical construct in this research suggests that results from CFA in this study should be interpreted cautiously before rejecting the overall validity of the scale. Therefore, refusing merit of this tool on the basis of CFA results would not be justifiable.
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268
Cronbach’s alpha (0.928) of the Pune-FAAT indicates a higher value than conventionally expected value while, there are different reports about the acceptable values of alpha, ranging from 0.70 to 0.95 (Bland & Altman, 1997; DeVellis, 2003; Nunnally & Bernstein, 1994). Multidimensional test does not necessarily show lower alpha than single dimensional test (Tavakol & Dennick, 2011). The proposed scale warrants consideration from the perspective of functional performance of the healthy elderly in community. Its validity is not tested in measuring disease specific functional decline (for example as found in arthritis, gout, asthma, or stroke). Our results demonstrate that the Pune-FAAT is effective in identifying functional level or activity performance among normal subjects, and demonstrate high test–retest reliability and internal consistency and discriminant validity. Resources and time required to complete the questionnaire are much less and can be easily used in a household survey with population having lower levels of education. 5. Conclusions We have demonstrated that a new tool, Pune-FAAT is contextual and is better than existing tools in identifying decline in functional ability. Use of this tool can be recommended in the community dwelling geriatric health screening program.
267
Appendix A (Continued ) No.
Item
7.
Household work Outside work Use of public transport Social gathering Self cleaning (Bathing) Toilet use Dressing Eating food A. Sum of the total B. No. of Item answered Pune-FAAT index = (A/B)
8. 9. 10. 11. 12. 13. 14.
No difficulty performing (1)
Somewhat difficult to perform (2)
Most difficult to perform (3)
Unable to perform (4)
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.archger.2013.10.002.
Author’s contribution All authors made substantial contributions to the paper. AN who is PI of the study, had main responsibility of research design, executing and making first through final draft of paper by working on each version. SG had main responsibility for data management; data entry, statistical analysis and contributing to all versions of the manuscript. SK had main responsibility for acquisition of data, entry and preliminary analysis. SK contributed to the development of tool by way of review, initial tool development and baseline testing. All authors were involved in revising the manuscript critically and have given final approval of the version to be published. Conflict of interest The authors declare that they have no competing or financial interests. Funding source This study is supported by the departmental research funding received from the University. Appendix A Name of the respondent: Form No Please indicate your ability to perform following tasks without the use of any assistive device or physical help. No.
Item
1.
Lifting weight (up to 1–2 kg) Bending Squatting Walking (up to 1–2 km) Climbing Arising from bed/chair
2. 3. 4. 5. 6.
No difficulty performing (1)
Somewhat difficult to perform (2)
Most difficult to perform (3)
Unable to perform (4)
References Abas, M. A., Punpuing, S., Jirapramupitak, T., Tangchonlatip, K., & Leese, M. (2009). Psychological wellbeing, physical impairments and rural aging in a developing country setting. Health and Quality of Life Outcomes, 7, 66. Avlund, K., Kreiner, S., & Schultz-Larsen, K. (1996). Functional ability scales for the elderly: A validation study. European Journal of Public Health, 6, 35–42. Badiger, M. S., Kamath, R., & Ashalatha, K. V. (2010). Correlates of functional ability among the elderly. Karnataka Journal of Agricultural Sciences, 23, 480–484. Bland, J., & Altman, D. (1997). Statistics notes: Cronbach’s alpha. British Medical Journal, 314, 275. Batra, V., Sharma, V. P., Batra, M., Sharma, V., Agarwal, G. G., Singh, V. K., et al. (2012). Measurer of functional disability status in knee osteoarthritis. Internet Journal of Medical Update, 7, 47–51. Bowling, A. (2005). Measuring health. A review of quality of life measurement scales (3rd ed.). Maidenhead, England: Open University Press. Bruce, B., & Fries, J. F. (2003). The Stanford Health Assessment Questionnaire: A review of its history, issues, progress, and documentation. Journal of Rheumatology, 30, 167–178. Central Statistics Office Ministry of Statistics & Programme Implementation, (2011) Situation Analysis of the Elderly in India Government of India. Retrieved May 11, 2013, from http://mospi.nic.in/mospi_new/upload/elderly_in_india.pdf. Collin, C., Wade, D. T., Davies, S., & Horne, V. (1988). The Barthel ADL index: A reliability study. International disability studies, 10(2), 61–63. Costello, A. B., & Osborne, J. W. (2005). Best Practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical assessment. Research & Evaluation, 10(7), 1–9. Courtney, M. D., Edwards, H. E., Anne, M. C., Parker, A. W., Kathleen, F., & Hamilton, K. (2011). A randomised controlled trial to prevent hospital readmissions and loss of functional ability in high risk older adults: A study protocol. BMC Health Services Research, 11, 202. DeVellis, R. (2003). Scale development: Theory and applications: Theory and application. Thousand Oaks, CA: Sage. Fillenbaum, G. G., Chandra, V., Ganguli, M., Pandav, R., Gilby, J. E., Seaberg, E. C., et al. (1999). Development of activities of daily living scale to screen for dementia in an illiterate rural older population in India. Age and Ageing, 28, 161–168. Fried, L. P., & Bush, T. L. (1988). Morbidity as a focus of preventive healthcare in the elderly. Epidemiologic Reviews, 10(1), 48–64. Fries, J. F., Spitz, P., Kraines, R. G., & Holman, H. R. (1980). Measurement of patient outcomes in arthritis. Arthritis and Rheumatism, 23, 137–145. Guralnik, J. M., Fried, L. P., & Salive, M. E. (1996). Disability as a public health outcome in the aging. Population Annual Review of Public Health, 17, 25–46. Hayton, J. C., Allen, D. G., & Scarpello, V. (2004). Factor retention decisions in exploratory factor analysis: A tutorial on parallel analysis. Organizational Research Methods, 7(2), 191–205. Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. Hurley, A. E., Scandura, T. A., Schriesheim, C. A., Brannick, M. T., Seers, A., Vandenberg, R. J., et al. (1997). Exploratory and confirmatory factor analysis: Guidelines, issues, and alternatives. Journal of Organizational Behavior, 18, 667–683.
268
A. Nagarkar et al. / Archives of Gerontology and Geriatrics 58 (2014) 263–268
Jotheeswaran, A. T., Joseph, D. W., & Prince, M. J. (2010). Predictors of mortality among elderly people living in a south Indian urban community; a 10/66 Dementia research group prospective population-based Cohort study. BMC Public Health, 10, 366. Katz, S., Ford, A. B., Moskowitz, R. W., Jackson, B. A., & Jaffe, M. W. (1963). Studies of illness in the aged. The index of ADL: A standardized measure of biological and psychosocial function. Journal of American Medical Association, 185, 914–919. Kumar, A., Malaviya, A. N., Pandhi, A., & Singh, R. (2002). Validation of an Indian version of the health assessment questionnaire in patients with rheumatoid arthritis. Rheumatology, 41, 1457–1459. Lawton, M. P., & Brody, E. M. (1969). Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist, 9, 179–186. McCusker, J., Kakuma, R., & Abrahamowicz, M. (2002). Predictors of functional decline in hospitalized elderly patients: A systematic review. Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 57, 569–577. Miles, J., & Shevlin, M. (2007). A time and a place for incremental fit indices. Personality and Individual Differences, 42(5), 869–874.
Nascimento, C. M., Ribeiro, A. Q., Cotta, R. M., Acurcio, F. A., Peixoto, S. V., Priore, S. E., et al. (2012). Factors associated with functional ability in Brazilian elderly. Archives of Gerontology and Geriatrics, 54, 89–94. Nunnally, J., & Bernstein, L. (1994). Psychometric theory. New York: McGraw-Hill Higher. Paterson, D. H., & Warburton, E. R. (2010). Physical activity and functional limitations in older adults: A systematic review related to Canada’s physical activity guidelines. International Journal of Behavioral Nutrition and Physical Activity, 7, 38. Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach’s alpha. International Journal of Medical Education, 2, 53–55. Vass, M., Avlund, K., Lauridsen, J., & Hendriksen, C. (2005). Feasible model for prevention of functional decline in older people: Municipality-randomized, controlled trial. Journal of American Geriatric Society, 53, 563–568. Wang, T. J. (2004). Concept analysis of functional status. International Journal of Nursing Studies, 41, 457–462.
