Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Developing an Assessment Method of Active Aging: University of Jyvaskyla Active Aging Scale

Developing an Assessment Method of Active Aging: University of Jyvaskyla Active Aging Scale Objective: To develop an assessment method of active aging for research on older people. Method: A multiphase process that included drafting by an expert panel, a pilot study for item analysis and scale validity, a feedback study with focus groups and questionnaire respondents, and a test–retest study. Altogether 235 people aged 60 to 94 years provided responses and/ or feedback. Results: We developed a 17-item University of Jyvaskyla Active Aging Scale with four aspects in each item (goals, ability, opportunity, and activity; range 0-272). The psychometric and item properties are good and the scale assesses a unidimensional latent construct of active aging. Discussion: Our scale assesses older people’s striving for well-being through activities pertaining to their goals, abilities, and opportunities. The University of Jyvaskyla Active Aging Scale provides a quantifiable measure of active aging that may be used in postal questionnaires or interviews in research and practice. Keywords aging, well-being, activity, participation, psychometric properties University of Jyvaskyla, Finland Corresponding Author: Taina Rantanen, Gerontology Research Center, Faculty of Sport and Health Sciences, University of Jyvaskyla, P.O. Box 35, 40014 Jyvaskyla, Finland. Email: taina.rantanen@jyu.fi Rantanen et al. 1003 Introduction To ensure that the increasingly longer lives of older people are worth living, attention needs to be paid to maintaining the positive sides of life during aging. The contributions of older persons to their own well-being and the well-being of their families and other people are of central interest in societ- ies with a growing proportion of older people. World Health Organization (WHO) defined the policy goal active aging in 2002 as follows: “Active aging is the process of optimizing opportunities for health, participation and security in order to enhance quality of life as people age” (WHO, 2002). The same document also states, “these policies and programmes should be based on the rights, needs, preferences and capacities of older people.” During the recent years, diverse use has been made of the term active aging. Some researchers have sought to categorize individuals and define an active aging phenotype based on the WHO definition. For example, the main- tenance of positive subjective well-being; good physical, social, and mental health; good functional ability and fitness; and continued involvement in one’s family, peer group, and community throughout the aging process have been proposed as aspects that describe an active aging phenotype (Fernández- Ballesteros, Robine, Walker, & Kalache, 2013). Some attempts have been made to validate an empirical model of an active aging phenotype based on the WHO definition by using, for example, latent factor analysis in survey data. However, the authors concluded that it was not possible to create a sta- tistical model that would fit their data (e.g., Bélanger, Ahmed, Filiatrault, Yu, & Zunzunegui, 2017; Paúl, Ribeiro, & Teixeira, 2012). The WHO definition of active aging is a goal for societies aimed at spur- ring them to make provision for the participation of older individuals in val- ued activities through such means as social security schemes, environmental and urban planning, health services, civil society, and legislation. An example of how active aging is assessed at the societal level is the Active Aging Index which ranks countries according to different societal indicators such as par- ticipation of older people in the workforce or life expectancy (“Active Ageing Index Home,” 2017). However, the WHO definition does not refer to active aging as a state of well-being or as good quality of life of individuals. It is not a phenotypic definition that can be applied to individuals or groups of indi- viduals. To assess the active aging of individuals, we first need a definition that centers on the individual. To this end, we suggest the following definition: the striving for elements of well-being through activities relating to a person’s goals, functional capac- ities and opportunities. This definition encompasses four central sides in the active aging of individuals: their goals (what they want to do), their functional 1004 Journal of Aging and Health 31(6) capacity (what they are able to do), their autonomy (perceived opportunities to do the valued activities), and their activities (what they actually do). We hypothesize that active aging maintains or promotes well-being. In reviewing earlier suggestions for active aging assessment scales, we found a scale that had been developed through a multiphase process, and which assesses active aging as a way to promote well-being. In Thailand, researchers developed a seven-factor 36-item scale containing the following factors: being self-reliant, being actively engaged with society, developing spiritual wisdom, building up financial security, maintaining a healthy life- style, engaging in active learning, and strengthening family ties to ensure care in later life (Thanakwang, Isaramalai, & Hatthakit, 2014). The authors conclude that although the content of some factors may be culture-specific, the factors included in this scale describe the actions that, within their culture, older people themselves initiate to promote their own well-being. In this lat- ter respect, the idea underlying our scale resembles that of the Thai-scale. The goal of this study was to develop a quantitative scale for assessing active aging among older people that can be used in research and practice. This article reports the process of developing the University of Jyvaskyla Active Aging Scale (UJACAS) and describes the final scale. Method The Process of Developing the UJACAS The UJACAS scale development process is described in Table 1. First, an expert panel consisting of six members was convened that included represen- tatives of gerontology, psychology, education, physiotherapy, sport sciences, statistics, and public health. The panel’s goal was to create a draft version of the assessment scale and throughout the scale development process decide on the phrasing of the items and which items to include in the final scale. The starting point for developing the scale items were the Activities and Participation categories of the International Classification of Functioning, Disability and Health (“Towards a Common Language for Functioning, Disability and Health, ICF,” 2002) which provide a valid, well-thought-out classification of essential life areas. The expert panel studied scales devel- oped for similar or closely related topics, created potential scale items, and produced a 24-item draft questionnaire that was then tested in a pilot study. The pilot study goals were to assess the item properties, to condense the scale, and to assess its validity and reliability. The resulting condensed pilot scale with 15 items was exposed to two focus groups of older people consist- ing of four and five people to gain their input and views in the feedback study 1005 Table 1. Subsequent Study Phases and Procedures in Each Phase of Scale Development. Drafting phase Pilot study Feedback study Test–retest study Expert panel convened Convenience sample recruited Two focus groups recruited Convenience sample Literature search and review from different sources Focus groups gave feedback on recruited from different Brain-storming to find Occupational therapist evaluated the pilot scale contexts relevant items and creation participants Focus group suggested items Participants responded to of the initial scale Participants responded to they perceived as missing the final 17-item scale Condensing of the initial draft interview or postal questionnaire from the pilot scale by postal questionnaire scale Participant and interviewer Expert panel analyzed twice 2 weeks apart Test interviews among feedback collected interviewer and participant Test–retest reliability volunteers Validation against external feedback from focus groups assessed Expert panel created a 24- evaluation and self-reports and pilot study The 17-item final scale item draft scale Item response analysis Expert panel agreed on the remained unchanged The draft scale condensed to 15 final 17-item scale items (pilot scale) 1006 Journal of Aging and Health 31(6) phase. The participant feedback on the scale led to the addition of two further items. The resulting final active aging scale comprised 17 items. The final phase of the scale development was test–retest analysis to determine the sta- bility and reliability over time of the final UJACAS. Ethics This project was approved by the Ethical Committee of the University of Jyvaskyla, Finland. The study was conducted according to the guidelines for good scientific and clinical practice laid down by the Declaration of Helsinki. Participants were informed about the research verbally and in writing. In the pilot study and test–retest study, the participants consented by returning the questionnaire. Focus group members and interview participants signed an informed consent form prior to the group meeting or interview. Pilot Study One hundred fifty-nine participants were recruited from different sources (a health center flue vaccination campaign, a rehabilitation trial, assisted living facilities, University of the Third Age, and through word of mouth) to guar- antee a sufficiently heterogeneous convenience sample. Interview with occupational therapist. In the absence of a true golden standard for rating of active aging, we considered that a trained occupational therapist could serve as an external evaluator for active aging. The occupational thera- pist evaluated the 45 participants’ activity in an interview based on the “Cana- dian Occupational Performance Measure” (COPM; 2017). Each interview lasted for 30 min and was started by asking the respondent to describe their daily routines in a typical week. The occupational therapist aimed at finding out about the participant’s meaningful activities and whether there were activ- ities that he or she wanted to do but could not do or for some other reason did not do. Participants were also asked to describe their functional ability in their own words. Based on the interview, the occupational therapist gave the par- ticipant a score ranging from 0 (not active at all) to 10 (the highest level of activity and participation in meaningful activities). The score was not revealed to the participants, who subsequently took part in a separate interview focused on their own assessments of their own active aging and answered the pilot version of the active aging scale and the validation questions. Active aging pilot scale and validation questions. The pilot scale assessing active aging comprised 24 items in four sides, each forming a subscale. Questions Rantanen et al. 1007 were phrased as follows and were answered on a five-point Likert-type scale. The four sides with the framing questions were as follows: Goals: “During the previous four weeks, I have wanted to . . .” Ability: “During the previous four weeks, I have or would have been able to . . .” Autonomy: “During the previous four weeks, my opportunity to . . .” Activity: “During the previous four weeks, I did . . .” Response options were worded to suit the item and scored from 0 (lowest, for example, least active) to 4 (highest, for example, most active). Scores were computed by summing the scores of the individual items of each side to form a subscale and of the total questionnaire to form the total score. A maximum of two missing items was allowed for each subscale and eight for the total sum score. To correct for missing items, we used the following formulas: (sum score / items responded to) × items offered. The pilot questionnaire included validation questions on socio-demo- graphic background, health, quality of life, life-space mobility, physical activity, participation and autonomy, and opportunities to provide feedback about the study. To gain an idea of their goals, people were asked how strongly they agreed with the claim “I have special interests in my life” with five-point Likert-type scale response options ranging from strongly agree to strongly disagree. Poor health as a barrier to active aging was elicited with a question “To what extent has your health or functional ability prevented you from doing activities that you wanted to do during the four previous weeks?” The opportunity for active aging was elicited with a question “To what extent have other matters related to your life or environment prevented you from the activities that you wanted to do during the past four weeks?” For both questions, responses were given on a five-point Likert-type scale ranging from very much to not at all. Autonomy was assessed with a question “Do you feel that you have sufficient control over your own life? Do you think that your opportunities to live the way you wish are very good, good, fair, bad, or very bad?” (Cardol, de Haan, de Jong, van den Bos, & de Groot, 2001). Self-rated health was elicited with a question “How do you perceive your health at the moment?” on a five-point response scale ranging from very good to very poor (Subramanian, Huijts, & Avendano, 2010). Quality of life was elicited with a question “How do you rate your quality of life?” using a similar response scale (The World Health Organization Quality of Life [WHOQOL] Group, 1998). To evaluate the community mobility of the 1008 Journal of Aging and Health 31(6) participants, the questionnaire included two questions drawn from the University of Alabama at Birmingham Life-Space Assessment (Baker, Bodner, & Allman, 2003). Question LS5 was “Have you been outside your own municipality during the last for weeks?” Question LS4 was as follows: “Have you been outside your own neighborhood during the last for weeks?” The response options for both were as follows: No (scored 0); Yes, less than once a week (scored 1); Yes, one to three times a week (scored 2); Yes, four to six times a week (scored 3); Yes, daily (scored 4). Based on these responses, a life-space mobility score ranging from 0 to 36 was calculated as follows: sum score = (4 × LS4 score) + (5 × LS5 score). Feedback Study Two focus groups of older people (five and four members, respectively) gave their perceptions of the relevance of the items comprising the 15-item pilot scale. The focus groups were recruited from the University of the Third Age and through word of mouth. Each focus group discussion lasted 2 hr and was led by an expert panel member while other researchers made notes of the comments made. In the discussion, each item on the scale was discussed for its relevance, clarity, and possible interpretation from a respondent’s perspec- tive. An important discussion topic concerned potential aspects of active aging that were missing from the scale. The expert panel considered the participants’ feedback from the focus groups, the pilot study interviews, and questionnaires along with the feed- back received from the interviewers. The expert panel continued their discus- sions until a consensus was reached on the number of items and their phrasing. Test–Retest Study The final UJACAS that was examined for test–retest reliability comprised 17 items. The scores were calculated as in the pilot study. A maximum of two missing items from each subscale were allowed. Statistical Methods Single-item properties were described by using the logit measures from one- parameter logistic (Rasch) model. Rasch modeling was conducted for sub- scores and the total score. Rasch model assumes that items capture different levels of an essentially unidimensional underlying trait. In the model, each item receives fit index estimates indicating how plausible the model is with Rantanen et al. 1009 any given item. We report infit mean-square values, which indicate how pre- dictable the item is. Values in the range of 0.6 to 1.4 indicate optimal predict- ability, whereas higher values suggest random noise and lower values too high a level of predictability. Finally, we calculated point-biserial correla- tions to estimate item-total correlations. Low correlations (less than 0.20) may indicate poor item discrimination. We report Cronbach reliability coef- ficient alphas as estimates of the lower bound of reliability for subscales scores and total score. We calculated person and item separation reliabilities as indices of the ability to separate the active aging levels of persons and items. We assessed scale dimensionality by plotting eigenvalues of the cor- relation matrix into Cattell’s scree plot. Parallel analysis was conducted to estimate eigenvalues resulting from the modeling of random noise data. Observed data eigenvalues exceeding eigenvalues from the parallel analysis were considered material. We assessed the validity of the 15-item pilot UJACAS scale against the occupational therapist’s assessment of active aging, against the participants’ own assessments of their active aging level and against separate self-reported questions by calculating correlation coefficients. We assessed the test–retest reliability of the subscales and total scale with intra-class correlation coeffi- cients and t-tests for repeated measures and for individual items with kappa coefficients. Results The pilot study comprised a convenience sample of 159 participants. Of these, 45 attended the occupational therapist’s evaluation that formed an external criterion for active aging. Altogether 114 people responded indepen- dently to the pilot scale in the postal questionnaire and returned it to us by mail. The participants were 65 to 94 years old (mean age 73.4 years, SD 6.8), 68% were women, 62% rated their health as good or very good, 32% as aver- age, and 6% as poor or very poor. In all, 75% reported no difficulty walking a distance of 2 km, whereas the rest reported minor or major difficulties or inability to walk such a distance. Tertiary education was reported by 64%. Based on the Rasch model and feedback from the participants, we identi- fied nine items in the 24-item pilot scale that had low variability, low infit or low person or item separation, or showed random effects in the responses. These items were omitted (shown in Table 2). The resulting pilot scale con- tained 15 items. Table 3 shows the item properties of these items in the pilot data. The reliability indices for all the subscales and the total scale were ≥0.68, and thus good. The Rasch modeling of pilot scale showed that the total score consisting of the four sub-scores captures a unidimensional latent con- struct of active aging. 1010 Table 2. The Items of the UJACAS at the Different Study Phases. Item abbreviation Draft and pilot study items Comments Final scale items Events Taking part in different Unsure what “events” Taking part in various events or activities to do events refer to; inconsistent with studying or with clubs or associations interpretation by participants Nature Observing nature’s Low variability; Randomness Going outside and enjoying nature phenomena in activity dimension Exercise Physical activity Rephrased to reduce Practicing keeping fit variability in interpretation Cognitive training Training my intellect or Perceived as unclear; Consciously exercising my mind or memory memory Randomness in activity dimension Help others Helping others Perceived as referring to Helping or supporting other people or people physical help close to me Maintain Acting to maintain OK Doing things to maintain my relationships with relationships relationships other people Meet new people Acting to meet new people OK Taking actions to make new acquaintances Promote own Taking responsibility to Perceived as vague Taking responsibility to furthering matters matters promote matters relating to my own life Societal activity Promoting societal or Political was considered Taking responsibility for promoting societal or political affairs peripheral public matters of importance to me Technology Following technological Perceived as unclear Using a computer or pad development (continued) 1011 Table 2. (continued) Item abbreviation Draft and pilot study items Comments Final scale items Make days Doing things that would OK Doing things that would make my days more interesting make my days more interesting or delightful interesting or delightful Make home cozy Making home cozy and OK Improving or maintaining the coziness of my comfortable home Appearance Taking care of my external OK Taking care of my external appearance appearance Economic balance Seeing to personal OK Ensuring that my financial affairs are in order economic balance Spirituality Acting according to Religiosity too prominent Taking actions to promote matters in spiritual or religious accordance with my faith or worldview conviction Items added based on participant feedback Crafting Added based on participant Crafting, DIY or other pastimes requiring feedback manual skills Artistic pursuit Added based on participant Drawing, singing, playing a musical instrument, feedback writing or practicing some other artistic pursuit (continued) 1012 Table 2. (continued) Rejected items Errands Running errands outside Overlap with taking responsibility to promote matters in own life home News Following news and Low variability; No active agency current issues Books Reading fiction and non- Differentiation between reading books or journals questioned; No active agency fiction books New things Doing new things Perceived as vague and irrelevant by participants Discussions Meaningful discussions Randomness in the activity and ability dimensions; little unique contribution to scale Problems Solving problems Unclear what problems refer to; Perceived irrelevant Eating Promoting pleasant Little unique contribution to scale; Overlap especially with making days more mealtimes interesting Advising Teaching and advising Considered socially problematic and vague; Redundant after rephrasing item on others helping and supporting others Promote health Promoting own health Little unique contribution to scale; Overlap with other variables; Can also be a predictor of active aging Note. Comments on items in the draft and pilot scales are shown. The final items of UJACAS are bolded. The same item abbreviations are used in Table 3. UJACAS = University of Jyvaskyla Active Aging Scale. 1013 Table 3. Measurement Properties of the University of Jyvaskyla Active Aging Pilot Scale, Subscales Including Measure Est, Its SE, Infit, and Pbc. Complete scale statistics Subscale statistics Item Estimate SE Infit Pbc Estimate SE Infit Pbc Goals Events 0.57 .10 0.84 .40 0.03 .11 0.79 .49 Nature −0.32 .12 0.79 .51 −1.06 .13 0.81 .55 Exercise −0.02 .11 0.88 .52 −0.69 0.12 1.06 .46 Cognitive training 0.41 .10 1.14 .48 −0.17 0.11 1.17 .54 Help others 0.50 .10 0.84 .47 −0.05 .11 0.94 .46 Maintain relationships 0.39 .10 0.85 .40 −0.19 .11 0.77 .53 Meet new people 1.19 .09 0.92 .44 0.82 .10 0.91 .51 Promote own matters 1.01 .09 0.66 .59 0.59 .10 0.75 .59 Societal activity 2.16 .08 1.38 .35 2.04 .09 1.50 .37 Follow technology 0.94 .09 0.85 .61 0.50 .10 1.06 .55 Make days interesting 0.19 .10 0.74 .62 −0.44 .11 0.73 .68 Make home cozy −0.21 .11 0.96 .48 −0.94 .12 0.99 .52 Smart appearance 0.29 .10 0.81 .43 −0.34 .12 0.88 .45 Economic balance −0.10 .11 0.90 .57 −0.81 .12 0.97 .59 Spirituality 1.11 .09 1.38 .33 0.70 .10 1.47 .38 Subscale alpha reliability 0.86 Person separation reliability 0.84 Item separation reliability 0.98 (continued) 1014 Table 3. (continued) Complete scale statistics Subscale statistics Item Estimate SE Infit Pbc Estimate SE Infit Pbc Ability Events −1.03 .14 1.52 .62 0.12 .15 0.87 .75 Observe nature −3.11 .32 1.13 .26 −2.17 .34 1.25 .20 Exercise −0.68 .13 1.49 .65 0.51 .13 0.77 .77 Cognitive training −1.50 .17 1.06 .33 −0.38 .18 1.26 .24 Help others −0.86 .14 1.50 .69 0.29 .14 0.90 .74 Maintain relationships −1.63 .18 1.10 .51 −0.53 .18 1.06 .51 Meet new people −1.26 .16 1.53 .44 −0.14 .16 1.42 .46 Promote own matters −0.89 .14 1.09 .55 0.25 .14 0.89 .58 Societal activity 0.07 .11 2.35 .57 1.23 .12 1.62 .60 Follow technology 0.38 .10 1.21 .61 1.63 .11 1.06 .62 Make days interesting −1.07 .15 0.92 .54 0.07 .15 0.73 .60 Make home cozy −0.98 .14 1.13 .59 0.17 .15 0.82 .65 Smart appearance −1.70 .19 1.29 .41 −0.59 .19 1.11 .51 Economic balance −1.27 .16 1.58 .56 −0.11 .16 1.12 .66 Spirituality −1.48 .17 1.53 .56 −0.34 .18 1.32 .56 Subscale alpha reliability .91 Person separation reliability .68 Item separation reliability .95 Opportunity Events −0.20 .11 0.98 .60 −0.24 .14 1.16 .63 Observe nature −0.93 .14 1.10 .41 −1.25 .16 1.32 .43 (continued) 1015 Table 3. (continued) Complete scale statistics Subscale statistics Item Estimate SE Infit Pbc Estimate SE Infit Pbc Exercise −0.23 .11 1.17 .63 −0.27 .14 1.52 .62 Cognitive training −0.58 .13 1.05 .54 −0.78 .15 1.49 .46 Help others 0.27 .10 0.73 .64 0.45 .12 0.96 .61 Maintain relationships −0.11 .11 0.80 .53 −0.11 .13 0.95 .55 Meet new people 0.42 .10 0.73 .48 0.67 .12 0.92 .49 Promote own matters 0.41 .10 0.74 .50 0.65 .12 0.85 .55 Societal activity 1.03 .09 0.98 .39 1.55 .11 1.25 .40 Follow technology 0.56 .10 0.83 .53 0.87 .12 0.87 .61 Make days interesting −0.16 .11 0.64 .64 −0.18 .13 0.64 .73 Make home cozy −0.28 .12 0.77 .55 −0.35 .14 0.89 .60 Smart appearance −0.55 .13 0.78 .43 −0.73 .15 0.89 .49 Economic balance −0.31 .12 0.88 .52 −0.39 .14 1.05 .55 Spirituality 0.04 .11 0.90 .49 0.11 .13 1.02 .55 Subscale alpha reliability: .89 Person separation reliability: .85 Item separation reliability: .96 Activity Events 1.15 .09 1.06 .32 0.49 .09 0.93 .27 Observe nature −1.95 .20 2.21 .04 −2.49 .20 2.07 .05 Exercise −0.27 .12 1.60 .44 −0.87 .11 1.51 .36 (continued) 1016 Table 3. (continued) Complete scale statistics Subscale statistics Item Estimate SE Infit Pbc Estimate SE Infit Pbc Cognitive training −0.03 .11 2.06 .43 −0.64 .11 1.84 .41 Help others 1.06 .09 1.35 .45 0.41 .09 1.24 .38 Maintain relationships 0.49 .10 0.69 .51 −0.15 .10 0.62 .50 Meet new people 1.78 .08 0.91 .37 1.10 .08 0.72 .41 Promote own matters 1.10 .09 0.99 .49 0.45 .09 0.85 .49 Societal activity 2.87 .09 1.35 .32 2.14 .09 1.09 .34 Follow technology 1.34 .09 0.99 .60 0.68 .08 0.98 .51 Make days interesting 0.37 .10 0.70 .62 −0.26 .10 0.67 .57 Make home cozy 0.08 .11 0.93 .58 −0.54 .10 0.93 .49 Smart appearance 0.22 .10 0.69 .52 −0.41 .10 0.69 .43 Economic balance 0.07 .11 1.26 .45 −0.55 .11 1.12 .44 Spirituality 1.31 .09 1.12 .42 0.65 .09 0.93 .42 Subscale alpha reliability .80 Person separation reliability .76 Item separation reliability .99 Complete scale alpha reliability .95 Person separation reliability .94 Item separation reliability .98 Note. Reliability indices are shown for subscale and total scale scores. See Table 2 for explanations of item abbreviations. Est = estimate; Infit = mean- square fit estimates; Pbc = point-biserial correlation. Rantanen et al. 1017 Table 4. Pearson Correlation Coefficients Between the University of Jyvaskyla Active Aging Scale Scores and the Validation Variables (N = 155, Except for the Occupation Therapist’s Assessment N = 44-45). Goals Activity Ability Opportunity Sum score r r r r r AA by OT (N = 44-45) .476 .549 .681 .553 .658 AA self-assessed .365 .500 .537 .463 .549 Interests .421 .508 .480 .366 .522 Health as a barrier for AA −.325 −.479 −.705 −.420 −.561 Poor opportunity for AA −.162* −.240** −.405 −.500 −.371 Autonomy .314 .475 .522 .612 .557 Self-rated health .285 .353 .516 .449 .470 Quality of life .268 .368 .422 .503 .455 Life-space mobility .308 .371 .415 .296 .414 Note. AA by OT = Active aging assessed by occupational therapist; AA self-assessed = self- rated active aging; Interests = I have special interests in my life; Health as a barrier for AA = Poor health prevents me from activities I would like to do; Poor opportunity for AA = Other conditions in my life or environment prevent me from activities I would like to do. p < .001 for all except *p < .05 and **p < .01. Scale Validity The mean value for participants’ own assessment of active aging was 7.77 (SD 1.70), and for occupational therapist’s assessment, 8.71 (SD 1.74). The correla- tion between these two assessments was r = .508 (p < .001, n = 44). The higher were the occupational therapist’s assessment and the participants’ own assess- ment of their active aging, the higher were the scores in the UJACAS scale (Table 4). The occupational therapist’s evaluation correlated most closely correlated with the ability dimension of UJACAS while the self-assessments captured the four dimensions more uniformly. If people reported barriers to active aging, the UJACAS scores were lower, whereas good self-rated health, quality of life, autonomy, and high life-space mobility correlated with higher UJACAS scores, indicating a higher level of active aging. Parallel correlations were observed for interviewees and postal questionnaire respondents. Similarly, no material gender differences were observed in the correlations; however, the correlations of UJACAS scores and self-rated health were somewhat weaker for men. Focus Group Interviews The focus group members were aged 60 to 84 years and included both men and women. The meetings were held at the Gerontology Research Center 1018 Journal of Aging and Health 31(6) where the participants arrived independently. The focus groups considered the 15 items in the pilot scale to be relevant for their life. The most impor- tant concern raised by focus group members was social desirability in responses. In the pilot scale, the frequency of performing the activity was the first question asked. The participants felt that they were expected to engage in all the activities to a high degree. The respondents emphasized that the relevance of one’s own goals should be the starting point, and that engaging in different activities as much as they found personally satisfying should be sufficient. Based on this feedback, we revised the order of the dimensions so that the first question concerned the goal dimension. Finally, based on the focus group comments on what was felt to be missing from the scale, we added two new items related to the hobbies in the scale, as these were considered essential for active aging (see Table 2). The focus groups also suggested including an item about nutrition, but we decided against this, as the pilot study had shown low variability in the item and led to its rejection. Test–Retest Repeatability The final version of the UJACAS comprised 17 items each with four dimen- sions. The total score ranges from 0 to 272, with higher scores indicating more active aging. For the test–retest study, a convenience sample of 67 respondents answered the questions twice approximately 2 weeks apart. The participants were 65 to 86 years old (mean age 72.9 years, SD, 5.6); 82% were women and 83% reported no difficulty walking a distance of 2 km. The means and intra-class correlations for the test–retest analyses are shown in Table 5. The smallest potential value for all subscales is 0 and the highest 68. For the total score, the corresponding limits are 0 and 272, respectively. The responses showed no apparent clustering toward mini- mum or maximum values, except for the ability dimension where 12 people (17%) received the maximum score. However, floor and ceiling effects were not observed for the total score. The intra-class correlation coeffi- cients between the two surveys carried out approximately 2 weeks apart were high for all sub-scores and the total score (for all ICC > .886, p < .001), and the results for men and women were practically parallel. The kappa coefficient was computed for each individual item and indicated that the agreement found over time was significantly greater than by chance; kappa coefficients were >.21 for each item in the subscales goals (.21-.62), ability (.24-.80), and activity (.24-.61). For the opportunity subscale, the kappa coefficients ranged from .17 and .60, but the item on societal matters had agreement equal to chance (kappa = .057). Rantanen et al. 1019 Table 5. Means and Standard Deviations of University of Jyvaskyla Active Aging Subscale Cores and Total Score at First (T1) and Second (T2) Assessments. ICC Coefficients Between the Assessments (N = 68). Scale Mean SD Mean SD ICC T1 T1 T2 T2 Goals 48.3 6.5 47.3 7.6 .894 Activity 45.9 7.2 45.6 8.2 .885 Ability 61.5 6.3 61.1 6.6 .928 Opportunity 55.5 7.7 54.6 8.0 .886 Total 211.2 22.6 208.7 23.7 .915 Note. ICC = intra-class correlation. p < .001 for all. Discussion We have developed a novel scale to assess active aging from the point of view of older individuals that can be used in research or practice. The crucial difference between this and previous studies arises from the realization that before we can assess the active aging of individuals, we need to define what active aging entails. We defined active aging at the level of the individual as the striving for elements of well-being through activities relating to a per- son’s goals, functional capacities, and opportunities. The UJACAS question- naire comprises 17 items each of which is assessed for four sides yielding four sub-scores. The minimum total score is 0 and the maximum 272. UJACAS can be administered as a questionnaire or as an interview, and both the sub-scores and total score are valid and reliable. The item analyses showed that the UJACAS sub-scores capture an underlying single construct that reflects active aging and that the total score describes active aging in a condensed form and reflects a unilateral construct. We are not aware of any earlier scales that have been developed by applying an approach correspond- ing to that of the UJACAS. We consider that what the older people truly want to do is essential for active aging. Most of the earlier scales related to aspects of active aging have simply addressed the ability of people to do different activities (e.g., Kempen, Miedema, Ormel, & Molenaar, 1996; Saito, Izawa, Omori, & Watanabe, 2016) while some have also addressed the actual activities done (e.g., Jette, Haley, Ni, Olarsch, & Moed, 2008) or autonomy in life situa- tions (Wilkie, Peat, Thomas, Hooper, & Croft, 2005). However, knowledge of what older people’s goals are remains unclear in these cases. Goals inspire people to act according to those goals (Ryan & Deci, 2000) and may help older people to maintain the activities that they value most (Baltes, 1020 Journal of Aging and Health 31(6) 1997), even in the face of functional decline. Research has indicated that goal pursuit relates to higher activity participation in terms of engagement in physical activity and higher life-space mobility (Saajanaho, Rantakokko, et al., 2015; Saajanaho, Viljanen, et al., 2014) Therefore, it is presumable that goal setting is a primary element also in active aging. Knowledge about people’s goals allows for analysis of potential congruence or discrepancy between goals and activities performed or between goals and the ability or opportunity to realize them. In the future, UJACAS will enable analysis of what people actually wish to do and what they actually do, and what they are able or have the have opportunity to do, thus providing a more compre- hensive picture of active aging. Developing the scale was a multifaceted process in which we placed high importance on participant feedback and on quantitatively analyzing item properties. We started with a draft scale produced by an expert panel, col- lected a pilot dataset for item response analyses and for testing validity, con- ducted two focus group discussions, and tested the scale’s reliability in a test–retest study. Consequently, the number of items in the pilot study version of the UJACAS and the test–retest study of UJACAS differ, and for some items, wording differed across the different study phases. In sum, we believe that the validity analysis conducted in the pilot sample is reliable. It is pos- sible that the score distributions in the pilot sample were slightly truncated as the pilot sample included rather high functioning and highly educated people. Consequently, the correlations may be underestimates rather than overesti- mates of the associations of the UJACAS scores with the indicators of activ- ity and well-being, suggesting that had there been more variability, the correlations might have been higher. Nevertheless, all the indicators of valid- ity and reliability are good. User involvement in aging research has been emphasized for more than two decades and has gradually become increasingly prevalent (Walker, 2007). Its benefits of it are, first of all, that older people, as is their right, are able to influence the research that is being conducted on them. Second, the chance of producing findings of relevance for the well-being of older people is likely to be higher if they have been invited to contribute their own under- standings of the issue. However, in the final scale, we did not include an item they frequently brought up, that is, eating and nutrition. The item was included in the pilot study, where the item analysis showed that practically all the respondents selected the highest response option, and thus the variable did not contribute to scale variability. In this case, participants’ views and the item analysis pointed in opposite directions regarding the inclusion of the item in the UJACAS scale. Overall, the content of the scale was perceived essential for meeting its objectives and the questionnaire was considered Rantanen et al. 1021 acceptable. Participant burden was perceived reasonable with interviews tak- ing from 10 to 25 min in most cases. The scale is intended to be used among older people able to respond inde- pendently and is not limited to a specific segment of the older population. We have successfully interviewed people ranging from residents of an assisted living facility to very vigorous community-living older people. We set the recall period for 4 weeks so that it would be long enough to smooth out pos- sible exceptional situations during the previous week, such as unusual weather conditions, sickness, travel, or a visit. The scale discriminates between more and less active respondents. The validation analyses suggest that higher scores are more favorable than lower scores, but whether the scores have a linear or, for example, a curved association with the indicators of well-being remains to be investigated. Study Limitations As part of the multifaceted study process, two different convenience samples of older people were recruited. Convenience samples are nonprobability samples that may not fully represent the characteristics of their age peers in the popula- tion. We recruited the participants from different sources to guarantee variabil- ity in terms of active aging. When studying the associations between variables, it is more important to have a broad range of responses than sample represen- tativeness. We report mean values only for the final 17-item UJACAS, which was used in the test–retest study. The test–retest study participants were rela- tively young (average age 73 years) and high functioning with 83% having no difficulty even in an advanced mobility task. We believe that the reported means are therefore rather high and need to be confirmed by a future study among a representative sample of older people. However, we do not believe that this essentially influenced the reliability as evaluated by test–retest intra- class correlations. Consequently, reference values for the UJACAS scores are not yet available and future studies will be needed to produce them. The indi- vidual items of the UJACAS may not be fully reliable when used separately; however, absolute agreement is typically not expected for individual items of a scale. The lower end of kappa coefficients observed for individual items except one was .21, which indicates fair agreement, whereas the higher end observed (.80) suggests a strong agreement (Landis & Koch, 1977). It is also worth not- ing that the more categories the items have, the lower the kappa values tend to be. The sub-scores and total score showed high test–retest agreement and indi- cate that overall active aging can be reliably assessed. Another potential limita- tion of the UJACAS is its unknown cultural generalizability. We consider the scale to be generalizable to the Northern European and other Western countries. 1022 Journal of Aging and Health 31(6) Although the items are rather universal and even have some similarities with the previous scale developed in Thailand (Thanakwang et al., 2014), the scale requires validation when used in a different cultural or societal context. The strengths of the study are the definition of active aging at the level of the individual that was used as the foundation for its development, the nov- elty of developing a scale for assessing active aging as a quantifiable con- struct, the item response analyses, and the participant involvement. Participant burden in responding to the UJACAS was rather low and the focus groups suggested that the clarity of the items was good or acceptable. Finally, an important strength of this study is its potential for regenerating the study of active aging, which has thus far been hindered by the lack of a valid method of assessment. The UJACAS may expand the epidemiology of aging by introducing active aging as a quantifiable entity that may be either low or high, and thus, it enables the analysis of, for example, individual physical and psychological characteristics as well as environmental and social factors as determinants or modifiers of active aging. In our upcoming cohort study, we plan to assess, for example, whether the four sides of active aging have different predictors, whether active aging changes with age, and whether active aging helps alle- viate decline in well-being when people face functional decline and other losses. The UJACAS may be used to monitor changes and to evaluate the effectiveness of different interventions and technological solutions in pro- moting active aging. Eventually, the UJACAS may be used for the evaluation of implementation research and policy formulation. The UJACAS is avail- able at www.gerec.fi/en/agnes/ujacas. Acknowledgments We wish to thank Dr. Timo Hinrichs for his contribution to the drafting phase of the scale. 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: The study is funded by the European Research Council (ERC Advanced Grant 693045 to TR). ORCID iD Taina Rantanen http://orcid.org/0000-0002-1604-1945 Rantanen et al. 1023 References Active Ageing Index Home. (2017, August 22). Retrieved from https://statswiki. unece.org/display/AAI Baker, P. S., Bodner, E. V., & Allman, R. M. (2003). Measuring life-space mobility in community-dwelling older adults. Journal of the American Geriatric Society, 51, 1610-1614. doi:10.1046/j.1532-5415.2003.51512.x Baltes, P. B. (1997). On the incomplete architecture of human ontogeny: Selection, optimization, and compensation as foundation of developmental theory. American Psychologist, 52, 366-380. doi:10.1037/0003-066X.52.4.366 Bélanger, E., Ahmed, T., Filiatrault, J., Yu, H. T., & Zunzunegui, M. V. (2017). An empirical comparison of different models of active aging in Canada: The inter- national mobility in aging study. The Gerontologist, 57, 197-205. doi:10.1093/ geront/gnv126 Canadian Occupational Performance Measure. (2017, January 1). Retrieved from http://neuroreha.cz/sites/default/files/materialy/COPM.pdf Cardol, M., de Haan, R. J., de Jong, B. A., van den Bos, G. A., & de Groot, I. J. (2001). Psychometric properties of the impact on participation and autonomy questionnaire. Archives of Physical Medicine and Rehabilitation, 82, 210-216. doi:10.1053/apmr.2001.18218 Fernández-Ballesteros, R., Robine, J. M., Walker, A., & Kalache, A. (2013). Active aging: A global goal. Current Gerontology & Geriatrics Research, 2013, Article ID 298012. doi:10.1155/2013/298012 Jette, A. M., Haley, S. M., Ni, P., Olarsch, S., & Moed, R. (2008). Creating a com- puter adaptive test version of the late-life function & disability instrument. The Journals of Gerontology, Series A: Biological Sciences & Medical Sciences, 63, 1246-1256. Kempen, G. I., Miedema, I., Ormel, J., & Molenaar, W. (1996). The assessment of disability with the Groningen Activity Restriction Scale. Conceptual framework and psychometric properties. Social Science in Medicine, 43, 1601-1610. Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159-174. doi:10.2307/2529310 Paúl, C., Ribeiro, O., & Teixeira, L. (2012). Active ageing: An empirical approach to the WHO model. Current Gerontology & Geriatric Research, 2012, Article ID 382972. doi:10.1155/2012/382972 Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68-78. doi:10.1037/0003-066X.55.1.68 Saajanaho, M., Rantakokko, M., Portegijs, E., Törmäkangas, T., Eronen, J., Tsai, L.-T., . . . Rantanen, T. (2015). Personal goals and changes in life-space mobil- ity among older people. Preventive Medicine, 81, 163-167. doi:10.1016/j. ypmed.2015.08.015 Saajanaho, M., Viljanen, A., Read, S., Rantakokko, M., Tsai, L.-T., Kaprio, J., . . . Rantanen, T. (2014). Older women’s personal goals and exercise activity: An eight-year follow-up. Journal of Aging and Physical Activity, 22, 386-392. doi:10.1123/JAPA.2012-0339 1024 Journal of Aging and Health 31(6) Saito, T., Izawa, K. P., Omori, Y., & Watanabe, S. (2016). Functional Independence and Difficulty Scale: Instrument development and validity evaluation. Geriatric & Gerontology International, 16, 1127-1137. doi:10.1111/ggi.12605 Subramanian, S. V., Huijts, T., & Avendano, M. (2010). Self-reported health assess- ments in the 2002 world health survey: How do they correlate with educa- tion? Bulletin of the World Health Organization, 88, 131-138. doi:10.2471/ BLT.09.067058 Thanakwang, K., Isaramalai, S. A., & Hatthakit, U. (2014). Development and psycho- metric testing of the active aging scale for Thai adults. Clinical Interventions in Aging, 9, 1211-1221. doi:10.2147/CIA.S66069 Towards a common language for functioning, disability and health, ICF. (2002). Geneva, Switzerland: World Health Organization. Walker, A. (2007). Why involve older people in research? Age and Ageing, 36, 481-483. doi:10.1093/ageing/afm100 The WHOQOL Group. (1998). Development of the world health organization WHOQOL-BREF quality of life assessment. Psychological Medicine, 28, 551-558. Wilkie, R., Peat, G., Thomas, E., Hooper, H., & Croft, P. R. (2005). The Keele Assessment of Participation: A new instrument to measure participation restric- tion in population studies. Combined qualitative and quantitative examina- tion of its psychometric properties. Quality of Life Research, 14, 1889-1899. doi:10.1007/s11136-005-4325-2 World Health Organization. (2002). Active ageing: A policy framework. Geneva, Switzerland. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Aging and Health Pubmed Central

Developing an Assessment Method of Active Aging: University of Jyvaskyla Active Aging Scale

Journal of Aging and Health , Volume 31 (6) – Jan 1, 2018

Loading next page...
 
/lp/pubmed-central/developing-an-assessment-method-of-active-aging-university-of-iqQbMjO5Jh

References (23)

Publisher
Pubmed Central
Copyright
© The Author(s) 2018
ISSN
0898-2643
eISSN
1552-6887
DOI
10.1177/0898264317750449
Publisher site
See Article on Publisher Site

Abstract

Objective: To develop an assessment method of active aging for research on older people. Method: A multiphase process that included drafting by an expert panel, a pilot study for item analysis and scale validity, a feedback study with focus groups and questionnaire respondents, and a test–retest study. Altogether 235 people aged 60 to 94 years provided responses and/ or feedback. Results: We developed a 17-item University of Jyvaskyla Active Aging Scale with four aspects in each item (goals, ability, opportunity, and activity; range 0-272). The psychometric and item properties are good and the scale assesses a unidimensional latent construct of active aging. Discussion: Our scale assesses older people’s striving for well-being through activities pertaining to their goals, abilities, and opportunities. The University of Jyvaskyla Active Aging Scale provides a quantifiable measure of active aging that may be used in postal questionnaires or interviews in research and practice. Keywords aging, well-being, activity, participation, psychometric properties University of Jyvaskyla, Finland Corresponding Author: Taina Rantanen, Gerontology Research Center, Faculty of Sport and Health Sciences, University of Jyvaskyla, P.O. Box 35, 40014 Jyvaskyla, Finland. Email: taina.rantanen@jyu.fi Rantanen et al. 1003 Introduction To ensure that the increasingly longer lives of older people are worth living, attention needs to be paid to maintaining the positive sides of life during aging. The contributions of older persons to their own well-being and the well-being of their families and other people are of central interest in societ- ies with a growing proportion of older people. World Health Organization (WHO) defined the policy goal active aging in 2002 as follows: “Active aging is the process of optimizing opportunities for health, participation and security in order to enhance quality of life as people age” (WHO, 2002). The same document also states, “these policies and programmes should be based on the rights, needs, preferences and capacities of older people.” During the recent years, diverse use has been made of the term active aging. Some researchers have sought to categorize individuals and define an active aging phenotype based on the WHO definition. For example, the main- tenance of positive subjective well-being; good physical, social, and mental health; good functional ability and fitness; and continued involvement in one’s family, peer group, and community throughout the aging process have been proposed as aspects that describe an active aging phenotype (Fernández- Ballesteros, Robine, Walker, & Kalache, 2013). Some attempts have been made to validate an empirical model of an active aging phenotype based on the WHO definition by using, for example, latent factor analysis in survey data. However, the authors concluded that it was not possible to create a sta- tistical model that would fit their data (e.g., Bélanger, Ahmed, Filiatrault, Yu, & Zunzunegui, 2017; Paúl, Ribeiro, & Teixeira, 2012). The WHO definition of active aging is a goal for societies aimed at spur- ring them to make provision for the participation of older individuals in val- ued activities through such means as social security schemes, environmental and urban planning, health services, civil society, and legislation. An example of how active aging is assessed at the societal level is the Active Aging Index which ranks countries according to different societal indicators such as par- ticipation of older people in the workforce or life expectancy (“Active Ageing Index Home,” 2017). However, the WHO definition does not refer to active aging as a state of well-being or as good quality of life of individuals. It is not a phenotypic definition that can be applied to individuals or groups of indi- viduals. To assess the active aging of individuals, we first need a definition that centers on the individual. To this end, we suggest the following definition: the striving for elements of well-being through activities relating to a person’s goals, functional capac- ities and opportunities. This definition encompasses four central sides in the active aging of individuals: their goals (what they want to do), their functional 1004 Journal of Aging and Health 31(6) capacity (what they are able to do), their autonomy (perceived opportunities to do the valued activities), and their activities (what they actually do). We hypothesize that active aging maintains or promotes well-being. In reviewing earlier suggestions for active aging assessment scales, we found a scale that had been developed through a multiphase process, and which assesses active aging as a way to promote well-being. In Thailand, researchers developed a seven-factor 36-item scale containing the following factors: being self-reliant, being actively engaged with society, developing spiritual wisdom, building up financial security, maintaining a healthy life- style, engaging in active learning, and strengthening family ties to ensure care in later life (Thanakwang, Isaramalai, & Hatthakit, 2014). The authors conclude that although the content of some factors may be culture-specific, the factors included in this scale describe the actions that, within their culture, older people themselves initiate to promote their own well-being. In this lat- ter respect, the idea underlying our scale resembles that of the Thai-scale. The goal of this study was to develop a quantitative scale for assessing active aging among older people that can be used in research and practice. This article reports the process of developing the University of Jyvaskyla Active Aging Scale (UJACAS) and describes the final scale. Method The Process of Developing the UJACAS The UJACAS scale development process is described in Table 1. First, an expert panel consisting of six members was convened that included represen- tatives of gerontology, psychology, education, physiotherapy, sport sciences, statistics, and public health. The panel’s goal was to create a draft version of the assessment scale and throughout the scale development process decide on the phrasing of the items and which items to include in the final scale. The starting point for developing the scale items were the Activities and Participation categories of the International Classification of Functioning, Disability and Health (“Towards a Common Language for Functioning, Disability and Health, ICF,” 2002) which provide a valid, well-thought-out classification of essential life areas. The expert panel studied scales devel- oped for similar or closely related topics, created potential scale items, and produced a 24-item draft questionnaire that was then tested in a pilot study. The pilot study goals were to assess the item properties, to condense the scale, and to assess its validity and reliability. The resulting condensed pilot scale with 15 items was exposed to two focus groups of older people consist- ing of four and five people to gain their input and views in the feedback study 1005 Table 1. Subsequent Study Phases and Procedures in Each Phase of Scale Development. Drafting phase Pilot study Feedback study Test–retest study Expert panel convened Convenience sample recruited Two focus groups recruited Convenience sample Literature search and review from different sources Focus groups gave feedback on recruited from different Brain-storming to find Occupational therapist evaluated the pilot scale contexts relevant items and creation participants Focus group suggested items Participants responded to of the initial scale Participants responded to they perceived as missing the final 17-item scale Condensing of the initial draft interview or postal questionnaire from the pilot scale by postal questionnaire scale Participant and interviewer Expert panel analyzed twice 2 weeks apart Test interviews among feedback collected interviewer and participant Test–retest reliability volunteers Validation against external feedback from focus groups assessed Expert panel created a 24- evaluation and self-reports and pilot study The 17-item final scale item draft scale Item response analysis Expert panel agreed on the remained unchanged The draft scale condensed to 15 final 17-item scale items (pilot scale) 1006 Journal of Aging and Health 31(6) phase. The participant feedback on the scale led to the addition of two further items. The resulting final active aging scale comprised 17 items. The final phase of the scale development was test–retest analysis to determine the sta- bility and reliability over time of the final UJACAS. Ethics This project was approved by the Ethical Committee of the University of Jyvaskyla, Finland. The study was conducted according to the guidelines for good scientific and clinical practice laid down by the Declaration of Helsinki. Participants were informed about the research verbally and in writing. In the pilot study and test–retest study, the participants consented by returning the questionnaire. Focus group members and interview participants signed an informed consent form prior to the group meeting or interview. Pilot Study One hundred fifty-nine participants were recruited from different sources (a health center flue vaccination campaign, a rehabilitation trial, assisted living facilities, University of the Third Age, and through word of mouth) to guar- antee a sufficiently heterogeneous convenience sample. Interview with occupational therapist. In the absence of a true golden standard for rating of active aging, we considered that a trained occupational therapist could serve as an external evaluator for active aging. The occupational thera- pist evaluated the 45 participants’ activity in an interview based on the “Cana- dian Occupational Performance Measure” (COPM; 2017). Each interview lasted for 30 min and was started by asking the respondent to describe their daily routines in a typical week. The occupational therapist aimed at finding out about the participant’s meaningful activities and whether there were activ- ities that he or she wanted to do but could not do or for some other reason did not do. Participants were also asked to describe their functional ability in their own words. Based on the interview, the occupational therapist gave the par- ticipant a score ranging from 0 (not active at all) to 10 (the highest level of activity and participation in meaningful activities). The score was not revealed to the participants, who subsequently took part in a separate interview focused on their own assessments of their own active aging and answered the pilot version of the active aging scale and the validation questions. Active aging pilot scale and validation questions. The pilot scale assessing active aging comprised 24 items in four sides, each forming a subscale. Questions Rantanen et al. 1007 were phrased as follows and were answered on a five-point Likert-type scale. The four sides with the framing questions were as follows: Goals: “During the previous four weeks, I have wanted to . . .” Ability: “During the previous four weeks, I have or would have been able to . . .” Autonomy: “During the previous four weeks, my opportunity to . . .” Activity: “During the previous four weeks, I did . . .” Response options were worded to suit the item and scored from 0 (lowest, for example, least active) to 4 (highest, for example, most active). Scores were computed by summing the scores of the individual items of each side to form a subscale and of the total questionnaire to form the total score. A maximum of two missing items was allowed for each subscale and eight for the total sum score. To correct for missing items, we used the following formulas: (sum score / items responded to) × items offered. The pilot questionnaire included validation questions on socio-demo- graphic background, health, quality of life, life-space mobility, physical activity, participation and autonomy, and opportunities to provide feedback about the study. To gain an idea of their goals, people were asked how strongly they agreed with the claim “I have special interests in my life” with five-point Likert-type scale response options ranging from strongly agree to strongly disagree. Poor health as a barrier to active aging was elicited with a question “To what extent has your health or functional ability prevented you from doing activities that you wanted to do during the four previous weeks?” The opportunity for active aging was elicited with a question “To what extent have other matters related to your life or environment prevented you from the activities that you wanted to do during the past four weeks?” For both questions, responses were given on a five-point Likert-type scale ranging from very much to not at all. Autonomy was assessed with a question “Do you feel that you have sufficient control over your own life? Do you think that your opportunities to live the way you wish are very good, good, fair, bad, or very bad?” (Cardol, de Haan, de Jong, van den Bos, & de Groot, 2001). Self-rated health was elicited with a question “How do you perceive your health at the moment?” on a five-point response scale ranging from very good to very poor (Subramanian, Huijts, & Avendano, 2010). Quality of life was elicited with a question “How do you rate your quality of life?” using a similar response scale (The World Health Organization Quality of Life [WHOQOL] Group, 1998). To evaluate the community mobility of the 1008 Journal of Aging and Health 31(6) participants, the questionnaire included two questions drawn from the University of Alabama at Birmingham Life-Space Assessment (Baker, Bodner, & Allman, 2003). Question LS5 was “Have you been outside your own municipality during the last for weeks?” Question LS4 was as follows: “Have you been outside your own neighborhood during the last for weeks?” The response options for both were as follows: No (scored 0); Yes, less than once a week (scored 1); Yes, one to three times a week (scored 2); Yes, four to six times a week (scored 3); Yes, daily (scored 4). Based on these responses, a life-space mobility score ranging from 0 to 36 was calculated as follows: sum score = (4 × LS4 score) + (5 × LS5 score). Feedback Study Two focus groups of older people (five and four members, respectively) gave their perceptions of the relevance of the items comprising the 15-item pilot scale. The focus groups were recruited from the University of the Third Age and through word of mouth. Each focus group discussion lasted 2 hr and was led by an expert panel member while other researchers made notes of the comments made. In the discussion, each item on the scale was discussed for its relevance, clarity, and possible interpretation from a respondent’s perspec- tive. An important discussion topic concerned potential aspects of active aging that were missing from the scale. The expert panel considered the participants’ feedback from the focus groups, the pilot study interviews, and questionnaires along with the feed- back received from the interviewers. The expert panel continued their discus- sions until a consensus was reached on the number of items and their phrasing. Test–Retest Study The final UJACAS that was examined for test–retest reliability comprised 17 items. The scores were calculated as in the pilot study. A maximum of two missing items from each subscale were allowed. Statistical Methods Single-item properties were described by using the logit measures from one- parameter logistic (Rasch) model. Rasch modeling was conducted for sub- scores and the total score. Rasch model assumes that items capture different levels of an essentially unidimensional underlying trait. In the model, each item receives fit index estimates indicating how plausible the model is with Rantanen et al. 1009 any given item. We report infit mean-square values, which indicate how pre- dictable the item is. Values in the range of 0.6 to 1.4 indicate optimal predict- ability, whereas higher values suggest random noise and lower values too high a level of predictability. Finally, we calculated point-biserial correla- tions to estimate item-total correlations. Low correlations (less than 0.20) may indicate poor item discrimination. We report Cronbach reliability coef- ficient alphas as estimates of the lower bound of reliability for subscales scores and total score. We calculated person and item separation reliabilities as indices of the ability to separate the active aging levels of persons and items. We assessed scale dimensionality by plotting eigenvalues of the cor- relation matrix into Cattell’s scree plot. Parallel analysis was conducted to estimate eigenvalues resulting from the modeling of random noise data. Observed data eigenvalues exceeding eigenvalues from the parallel analysis were considered material. We assessed the validity of the 15-item pilot UJACAS scale against the occupational therapist’s assessment of active aging, against the participants’ own assessments of their active aging level and against separate self-reported questions by calculating correlation coefficients. We assessed the test–retest reliability of the subscales and total scale with intra-class correlation coeffi- cients and t-tests for repeated measures and for individual items with kappa coefficients. Results The pilot study comprised a convenience sample of 159 participants. Of these, 45 attended the occupational therapist’s evaluation that formed an external criterion for active aging. Altogether 114 people responded indepen- dently to the pilot scale in the postal questionnaire and returned it to us by mail. The participants were 65 to 94 years old (mean age 73.4 years, SD 6.8), 68% were women, 62% rated their health as good or very good, 32% as aver- age, and 6% as poor or very poor. In all, 75% reported no difficulty walking a distance of 2 km, whereas the rest reported minor or major difficulties or inability to walk such a distance. Tertiary education was reported by 64%. Based on the Rasch model and feedback from the participants, we identi- fied nine items in the 24-item pilot scale that had low variability, low infit or low person or item separation, or showed random effects in the responses. These items were omitted (shown in Table 2). The resulting pilot scale con- tained 15 items. Table 3 shows the item properties of these items in the pilot data. The reliability indices for all the subscales and the total scale were ≥0.68, and thus good. The Rasch modeling of pilot scale showed that the total score consisting of the four sub-scores captures a unidimensional latent con- struct of active aging. 1010 Table 2. The Items of the UJACAS at the Different Study Phases. Item abbreviation Draft and pilot study items Comments Final scale items Events Taking part in different Unsure what “events” Taking part in various events or activities to do events refer to; inconsistent with studying or with clubs or associations interpretation by participants Nature Observing nature’s Low variability; Randomness Going outside and enjoying nature phenomena in activity dimension Exercise Physical activity Rephrased to reduce Practicing keeping fit variability in interpretation Cognitive training Training my intellect or Perceived as unclear; Consciously exercising my mind or memory memory Randomness in activity dimension Help others Helping others Perceived as referring to Helping or supporting other people or people physical help close to me Maintain Acting to maintain OK Doing things to maintain my relationships with relationships relationships other people Meet new people Acting to meet new people OK Taking actions to make new acquaintances Promote own Taking responsibility to Perceived as vague Taking responsibility to furthering matters matters promote matters relating to my own life Societal activity Promoting societal or Political was considered Taking responsibility for promoting societal or political affairs peripheral public matters of importance to me Technology Following technological Perceived as unclear Using a computer or pad development (continued) 1011 Table 2. (continued) Item abbreviation Draft and pilot study items Comments Final scale items Make days Doing things that would OK Doing things that would make my days more interesting make my days more interesting or delightful interesting or delightful Make home cozy Making home cozy and OK Improving or maintaining the coziness of my comfortable home Appearance Taking care of my external OK Taking care of my external appearance appearance Economic balance Seeing to personal OK Ensuring that my financial affairs are in order economic balance Spirituality Acting according to Religiosity too prominent Taking actions to promote matters in spiritual or religious accordance with my faith or worldview conviction Items added based on participant feedback Crafting Added based on participant Crafting, DIY or other pastimes requiring feedback manual skills Artistic pursuit Added based on participant Drawing, singing, playing a musical instrument, feedback writing or practicing some other artistic pursuit (continued) 1012 Table 2. (continued) Rejected items Errands Running errands outside Overlap with taking responsibility to promote matters in own life home News Following news and Low variability; No active agency current issues Books Reading fiction and non- Differentiation between reading books or journals questioned; No active agency fiction books New things Doing new things Perceived as vague and irrelevant by participants Discussions Meaningful discussions Randomness in the activity and ability dimensions; little unique contribution to scale Problems Solving problems Unclear what problems refer to; Perceived irrelevant Eating Promoting pleasant Little unique contribution to scale; Overlap especially with making days more mealtimes interesting Advising Teaching and advising Considered socially problematic and vague; Redundant after rephrasing item on others helping and supporting others Promote health Promoting own health Little unique contribution to scale; Overlap with other variables; Can also be a predictor of active aging Note. Comments on items in the draft and pilot scales are shown. The final items of UJACAS are bolded. The same item abbreviations are used in Table 3. UJACAS = University of Jyvaskyla Active Aging Scale. 1013 Table 3. Measurement Properties of the University of Jyvaskyla Active Aging Pilot Scale, Subscales Including Measure Est, Its SE, Infit, and Pbc. Complete scale statistics Subscale statistics Item Estimate SE Infit Pbc Estimate SE Infit Pbc Goals Events 0.57 .10 0.84 .40 0.03 .11 0.79 .49 Nature −0.32 .12 0.79 .51 −1.06 .13 0.81 .55 Exercise −0.02 .11 0.88 .52 −0.69 0.12 1.06 .46 Cognitive training 0.41 .10 1.14 .48 −0.17 0.11 1.17 .54 Help others 0.50 .10 0.84 .47 −0.05 .11 0.94 .46 Maintain relationships 0.39 .10 0.85 .40 −0.19 .11 0.77 .53 Meet new people 1.19 .09 0.92 .44 0.82 .10 0.91 .51 Promote own matters 1.01 .09 0.66 .59 0.59 .10 0.75 .59 Societal activity 2.16 .08 1.38 .35 2.04 .09 1.50 .37 Follow technology 0.94 .09 0.85 .61 0.50 .10 1.06 .55 Make days interesting 0.19 .10 0.74 .62 −0.44 .11 0.73 .68 Make home cozy −0.21 .11 0.96 .48 −0.94 .12 0.99 .52 Smart appearance 0.29 .10 0.81 .43 −0.34 .12 0.88 .45 Economic balance −0.10 .11 0.90 .57 −0.81 .12 0.97 .59 Spirituality 1.11 .09 1.38 .33 0.70 .10 1.47 .38 Subscale alpha reliability 0.86 Person separation reliability 0.84 Item separation reliability 0.98 (continued) 1014 Table 3. (continued) Complete scale statistics Subscale statistics Item Estimate SE Infit Pbc Estimate SE Infit Pbc Ability Events −1.03 .14 1.52 .62 0.12 .15 0.87 .75 Observe nature −3.11 .32 1.13 .26 −2.17 .34 1.25 .20 Exercise −0.68 .13 1.49 .65 0.51 .13 0.77 .77 Cognitive training −1.50 .17 1.06 .33 −0.38 .18 1.26 .24 Help others −0.86 .14 1.50 .69 0.29 .14 0.90 .74 Maintain relationships −1.63 .18 1.10 .51 −0.53 .18 1.06 .51 Meet new people −1.26 .16 1.53 .44 −0.14 .16 1.42 .46 Promote own matters −0.89 .14 1.09 .55 0.25 .14 0.89 .58 Societal activity 0.07 .11 2.35 .57 1.23 .12 1.62 .60 Follow technology 0.38 .10 1.21 .61 1.63 .11 1.06 .62 Make days interesting −1.07 .15 0.92 .54 0.07 .15 0.73 .60 Make home cozy −0.98 .14 1.13 .59 0.17 .15 0.82 .65 Smart appearance −1.70 .19 1.29 .41 −0.59 .19 1.11 .51 Economic balance −1.27 .16 1.58 .56 −0.11 .16 1.12 .66 Spirituality −1.48 .17 1.53 .56 −0.34 .18 1.32 .56 Subscale alpha reliability .91 Person separation reliability .68 Item separation reliability .95 Opportunity Events −0.20 .11 0.98 .60 −0.24 .14 1.16 .63 Observe nature −0.93 .14 1.10 .41 −1.25 .16 1.32 .43 (continued) 1015 Table 3. (continued) Complete scale statistics Subscale statistics Item Estimate SE Infit Pbc Estimate SE Infit Pbc Exercise −0.23 .11 1.17 .63 −0.27 .14 1.52 .62 Cognitive training −0.58 .13 1.05 .54 −0.78 .15 1.49 .46 Help others 0.27 .10 0.73 .64 0.45 .12 0.96 .61 Maintain relationships −0.11 .11 0.80 .53 −0.11 .13 0.95 .55 Meet new people 0.42 .10 0.73 .48 0.67 .12 0.92 .49 Promote own matters 0.41 .10 0.74 .50 0.65 .12 0.85 .55 Societal activity 1.03 .09 0.98 .39 1.55 .11 1.25 .40 Follow technology 0.56 .10 0.83 .53 0.87 .12 0.87 .61 Make days interesting −0.16 .11 0.64 .64 −0.18 .13 0.64 .73 Make home cozy −0.28 .12 0.77 .55 −0.35 .14 0.89 .60 Smart appearance −0.55 .13 0.78 .43 −0.73 .15 0.89 .49 Economic balance −0.31 .12 0.88 .52 −0.39 .14 1.05 .55 Spirituality 0.04 .11 0.90 .49 0.11 .13 1.02 .55 Subscale alpha reliability: .89 Person separation reliability: .85 Item separation reliability: .96 Activity Events 1.15 .09 1.06 .32 0.49 .09 0.93 .27 Observe nature −1.95 .20 2.21 .04 −2.49 .20 2.07 .05 Exercise −0.27 .12 1.60 .44 −0.87 .11 1.51 .36 (continued) 1016 Table 3. (continued) Complete scale statistics Subscale statistics Item Estimate SE Infit Pbc Estimate SE Infit Pbc Cognitive training −0.03 .11 2.06 .43 −0.64 .11 1.84 .41 Help others 1.06 .09 1.35 .45 0.41 .09 1.24 .38 Maintain relationships 0.49 .10 0.69 .51 −0.15 .10 0.62 .50 Meet new people 1.78 .08 0.91 .37 1.10 .08 0.72 .41 Promote own matters 1.10 .09 0.99 .49 0.45 .09 0.85 .49 Societal activity 2.87 .09 1.35 .32 2.14 .09 1.09 .34 Follow technology 1.34 .09 0.99 .60 0.68 .08 0.98 .51 Make days interesting 0.37 .10 0.70 .62 −0.26 .10 0.67 .57 Make home cozy 0.08 .11 0.93 .58 −0.54 .10 0.93 .49 Smart appearance 0.22 .10 0.69 .52 −0.41 .10 0.69 .43 Economic balance 0.07 .11 1.26 .45 −0.55 .11 1.12 .44 Spirituality 1.31 .09 1.12 .42 0.65 .09 0.93 .42 Subscale alpha reliability .80 Person separation reliability .76 Item separation reliability .99 Complete scale alpha reliability .95 Person separation reliability .94 Item separation reliability .98 Note. Reliability indices are shown for subscale and total scale scores. See Table 2 for explanations of item abbreviations. Est = estimate; Infit = mean- square fit estimates; Pbc = point-biserial correlation. Rantanen et al. 1017 Table 4. Pearson Correlation Coefficients Between the University of Jyvaskyla Active Aging Scale Scores and the Validation Variables (N = 155, Except for the Occupation Therapist’s Assessment N = 44-45). Goals Activity Ability Opportunity Sum score r r r r r AA by OT (N = 44-45) .476 .549 .681 .553 .658 AA self-assessed .365 .500 .537 .463 .549 Interests .421 .508 .480 .366 .522 Health as a barrier for AA −.325 −.479 −.705 −.420 −.561 Poor opportunity for AA −.162* −.240** −.405 −.500 −.371 Autonomy .314 .475 .522 .612 .557 Self-rated health .285 .353 .516 .449 .470 Quality of life .268 .368 .422 .503 .455 Life-space mobility .308 .371 .415 .296 .414 Note. AA by OT = Active aging assessed by occupational therapist; AA self-assessed = self- rated active aging; Interests = I have special interests in my life; Health as a barrier for AA = Poor health prevents me from activities I would like to do; Poor opportunity for AA = Other conditions in my life or environment prevent me from activities I would like to do. p < .001 for all except *p < .05 and **p < .01. Scale Validity The mean value for participants’ own assessment of active aging was 7.77 (SD 1.70), and for occupational therapist’s assessment, 8.71 (SD 1.74). The correla- tion between these two assessments was r = .508 (p < .001, n = 44). The higher were the occupational therapist’s assessment and the participants’ own assess- ment of their active aging, the higher were the scores in the UJACAS scale (Table 4). The occupational therapist’s evaluation correlated most closely correlated with the ability dimension of UJACAS while the self-assessments captured the four dimensions more uniformly. If people reported barriers to active aging, the UJACAS scores were lower, whereas good self-rated health, quality of life, autonomy, and high life-space mobility correlated with higher UJACAS scores, indicating a higher level of active aging. Parallel correlations were observed for interviewees and postal questionnaire respondents. Similarly, no material gender differences were observed in the correlations; however, the correlations of UJACAS scores and self-rated health were somewhat weaker for men. Focus Group Interviews The focus group members were aged 60 to 84 years and included both men and women. The meetings were held at the Gerontology Research Center 1018 Journal of Aging and Health 31(6) where the participants arrived independently. The focus groups considered the 15 items in the pilot scale to be relevant for their life. The most impor- tant concern raised by focus group members was social desirability in responses. In the pilot scale, the frequency of performing the activity was the first question asked. The participants felt that they were expected to engage in all the activities to a high degree. The respondents emphasized that the relevance of one’s own goals should be the starting point, and that engaging in different activities as much as they found personally satisfying should be sufficient. Based on this feedback, we revised the order of the dimensions so that the first question concerned the goal dimension. Finally, based on the focus group comments on what was felt to be missing from the scale, we added two new items related to the hobbies in the scale, as these were considered essential for active aging (see Table 2). The focus groups also suggested including an item about nutrition, but we decided against this, as the pilot study had shown low variability in the item and led to its rejection. Test–Retest Repeatability The final version of the UJACAS comprised 17 items each with four dimen- sions. The total score ranges from 0 to 272, with higher scores indicating more active aging. For the test–retest study, a convenience sample of 67 respondents answered the questions twice approximately 2 weeks apart. The participants were 65 to 86 years old (mean age 72.9 years, SD, 5.6); 82% were women and 83% reported no difficulty walking a distance of 2 km. The means and intra-class correlations for the test–retest analyses are shown in Table 5. The smallest potential value for all subscales is 0 and the highest 68. For the total score, the corresponding limits are 0 and 272, respectively. The responses showed no apparent clustering toward mini- mum or maximum values, except for the ability dimension where 12 people (17%) received the maximum score. However, floor and ceiling effects were not observed for the total score. The intra-class correlation coeffi- cients between the two surveys carried out approximately 2 weeks apart were high for all sub-scores and the total score (for all ICC > .886, p < .001), and the results for men and women were practically parallel. The kappa coefficient was computed for each individual item and indicated that the agreement found over time was significantly greater than by chance; kappa coefficients were >.21 for each item in the subscales goals (.21-.62), ability (.24-.80), and activity (.24-.61). For the opportunity subscale, the kappa coefficients ranged from .17 and .60, but the item on societal matters had agreement equal to chance (kappa = .057). Rantanen et al. 1019 Table 5. Means and Standard Deviations of University of Jyvaskyla Active Aging Subscale Cores and Total Score at First (T1) and Second (T2) Assessments. ICC Coefficients Between the Assessments (N = 68). Scale Mean SD Mean SD ICC T1 T1 T2 T2 Goals 48.3 6.5 47.3 7.6 .894 Activity 45.9 7.2 45.6 8.2 .885 Ability 61.5 6.3 61.1 6.6 .928 Opportunity 55.5 7.7 54.6 8.0 .886 Total 211.2 22.6 208.7 23.7 .915 Note. ICC = intra-class correlation. p < .001 for all. Discussion We have developed a novel scale to assess active aging from the point of view of older individuals that can be used in research or practice. The crucial difference between this and previous studies arises from the realization that before we can assess the active aging of individuals, we need to define what active aging entails. We defined active aging at the level of the individual as the striving for elements of well-being through activities relating to a per- son’s goals, functional capacities, and opportunities. The UJACAS question- naire comprises 17 items each of which is assessed for four sides yielding four sub-scores. The minimum total score is 0 and the maximum 272. UJACAS can be administered as a questionnaire or as an interview, and both the sub-scores and total score are valid and reliable. The item analyses showed that the UJACAS sub-scores capture an underlying single construct that reflects active aging and that the total score describes active aging in a condensed form and reflects a unilateral construct. We are not aware of any earlier scales that have been developed by applying an approach correspond- ing to that of the UJACAS. We consider that what the older people truly want to do is essential for active aging. Most of the earlier scales related to aspects of active aging have simply addressed the ability of people to do different activities (e.g., Kempen, Miedema, Ormel, & Molenaar, 1996; Saito, Izawa, Omori, & Watanabe, 2016) while some have also addressed the actual activities done (e.g., Jette, Haley, Ni, Olarsch, & Moed, 2008) or autonomy in life situa- tions (Wilkie, Peat, Thomas, Hooper, & Croft, 2005). However, knowledge of what older people’s goals are remains unclear in these cases. Goals inspire people to act according to those goals (Ryan & Deci, 2000) and may help older people to maintain the activities that they value most (Baltes, 1020 Journal of Aging and Health 31(6) 1997), even in the face of functional decline. Research has indicated that goal pursuit relates to higher activity participation in terms of engagement in physical activity and higher life-space mobility (Saajanaho, Rantakokko, et al., 2015; Saajanaho, Viljanen, et al., 2014) Therefore, it is presumable that goal setting is a primary element also in active aging. Knowledge about people’s goals allows for analysis of potential congruence or discrepancy between goals and activities performed or between goals and the ability or opportunity to realize them. In the future, UJACAS will enable analysis of what people actually wish to do and what they actually do, and what they are able or have the have opportunity to do, thus providing a more compre- hensive picture of active aging. Developing the scale was a multifaceted process in which we placed high importance on participant feedback and on quantitatively analyzing item properties. We started with a draft scale produced by an expert panel, col- lected a pilot dataset for item response analyses and for testing validity, con- ducted two focus group discussions, and tested the scale’s reliability in a test–retest study. Consequently, the number of items in the pilot study version of the UJACAS and the test–retest study of UJACAS differ, and for some items, wording differed across the different study phases. In sum, we believe that the validity analysis conducted in the pilot sample is reliable. It is pos- sible that the score distributions in the pilot sample were slightly truncated as the pilot sample included rather high functioning and highly educated people. Consequently, the correlations may be underestimates rather than overesti- mates of the associations of the UJACAS scores with the indicators of activ- ity and well-being, suggesting that had there been more variability, the correlations might have been higher. Nevertheless, all the indicators of valid- ity and reliability are good. User involvement in aging research has been emphasized for more than two decades and has gradually become increasingly prevalent (Walker, 2007). Its benefits of it are, first of all, that older people, as is their right, are able to influence the research that is being conducted on them. Second, the chance of producing findings of relevance for the well-being of older people is likely to be higher if they have been invited to contribute their own under- standings of the issue. However, in the final scale, we did not include an item they frequently brought up, that is, eating and nutrition. The item was included in the pilot study, where the item analysis showed that practically all the respondents selected the highest response option, and thus the variable did not contribute to scale variability. In this case, participants’ views and the item analysis pointed in opposite directions regarding the inclusion of the item in the UJACAS scale. Overall, the content of the scale was perceived essential for meeting its objectives and the questionnaire was considered Rantanen et al. 1021 acceptable. Participant burden was perceived reasonable with interviews tak- ing from 10 to 25 min in most cases. The scale is intended to be used among older people able to respond inde- pendently and is not limited to a specific segment of the older population. We have successfully interviewed people ranging from residents of an assisted living facility to very vigorous community-living older people. We set the recall period for 4 weeks so that it would be long enough to smooth out pos- sible exceptional situations during the previous week, such as unusual weather conditions, sickness, travel, or a visit. The scale discriminates between more and less active respondents. The validation analyses suggest that higher scores are more favorable than lower scores, but whether the scores have a linear or, for example, a curved association with the indicators of well-being remains to be investigated. Study Limitations As part of the multifaceted study process, two different convenience samples of older people were recruited. Convenience samples are nonprobability samples that may not fully represent the characteristics of their age peers in the popula- tion. We recruited the participants from different sources to guarantee variabil- ity in terms of active aging. When studying the associations between variables, it is more important to have a broad range of responses than sample represen- tativeness. We report mean values only for the final 17-item UJACAS, which was used in the test–retest study. The test–retest study participants were rela- tively young (average age 73 years) and high functioning with 83% having no difficulty even in an advanced mobility task. We believe that the reported means are therefore rather high and need to be confirmed by a future study among a representative sample of older people. However, we do not believe that this essentially influenced the reliability as evaluated by test–retest intra- class correlations. Consequently, reference values for the UJACAS scores are not yet available and future studies will be needed to produce them. The indi- vidual items of the UJACAS may not be fully reliable when used separately; however, absolute agreement is typically not expected for individual items of a scale. The lower end of kappa coefficients observed for individual items except one was .21, which indicates fair agreement, whereas the higher end observed (.80) suggests a strong agreement (Landis & Koch, 1977). It is also worth not- ing that the more categories the items have, the lower the kappa values tend to be. The sub-scores and total score showed high test–retest agreement and indi- cate that overall active aging can be reliably assessed. Another potential limita- tion of the UJACAS is its unknown cultural generalizability. We consider the scale to be generalizable to the Northern European and other Western countries. 1022 Journal of Aging and Health 31(6) Although the items are rather universal and even have some similarities with the previous scale developed in Thailand (Thanakwang et al., 2014), the scale requires validation when used in a different cultural or societal context. The strengths of the study are the definition of active aging at the level of the individual that was used as the foundation for its development, the nov- elty of developing a scale for assessing active aging as a quantifiable con- struct, the item response analyses, and the participant involvement. Participant burden in responding to the UJACAS was rather low and the focus groups suggested that the clarity of the items was good or acceptable. Finally, an important strength of this study is its potential for regenerating the study of active aging, which has thus far been hindered by the lack of a valid method of assessment. The UJACAS may expand the epidemiology of aging by introducing active aging as a quantifiable entity that may be either low or high, and thus, it enables the analysis of, for example, individual physical and psychological characteristics as well as environmental and social factors as determinants or modifiers of active aging. In our upcoming cohort study, we plan to assess, for example, whether the four sides of active aging have different predictors, whether active aging changes with age, and whether active aging helps alle- viate decline in well-being when people face functional decline and other losses. The UJACAS may be used to monitor changes and to evaluate the effectiveness of different interventions and technological solutions in pro- moting active aging. Eventually, the UJACAS may be used for the evaluation of implementation research and policy formulation. The UJACAS is avail- able at www.gerec.fi/en/agnes/ujacas. Acknowledgments We wish to thank Dr. Timo Hinrichs for his contribution to the drafting phase of the scale. 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: The study is funded by the European Research Council (ERC Advanced Grant 693045 to TR). ORCID iD Taina Rantanen http://orcid.org/0000-0002-1604-1945 Rantanen et al. 1023 References Active Ageing Index Home. (2017, August 22). Retrieved from https://statswiki. unece.org/display/AAI Baker, P. S., Bodner, E. V., & Allman, R. M. (2003). Measuring life-space mobility in community-dwelling older adults. Journal of the American Geriatric Society, 51, 1610-1614. doi:10.1046/j.1532-5415.2003.51512.x Baltes, P. B. (1997). On the incomplete architecture of human ontogeny: Selection, optimization, and compensation as foundation of developmental theory. American Psychologist, 52, 366-380. doi:10.1037/0003-066X.52.4.366 Bélanger, E., Ahmed, T., Filiatrault, J., Yu, H. T., & Zunzunegui, M. V. (2017). An empirical comparison of different models of active aging in Canada: The inter- national mobility in aging study. The Gerontologist, 57, 197-205. doi:10.1093/ geront/gnv126 Canadian Occupational Performance Measure. (2017, January 1). Retrieved from http://neuroreha.cz/sites/default/files/materialy/COPM.pdf Cardol, M., de Haan, R. J., de Jong, B. A., van den Bos, G. A., & de Groot, I. J. (2001). Psychometric properties of the impact on participation and autonomy questionnaire. Archives of Physical Medicine and Rehabilitation, 82, 210-216. doi:10.1053/apmr.2001.18218 Fernández-Ballesteros, R., Robine, J. M., Walker, A., & Kalache, A. (2013). Active aging: A global goal. Current Gerontology & Geriatrics Research, 2013, Article ID 298012. doi:10.1155/2013/298012 Jette, A. M., Haley, S. M., Ni, P., Olarsch, S., & Moed, R. (2008). Creating a com- puter adaptive test version of the late-life function & disability instrument. The Journals of Gerontology, Series A: Biological Sciences & Medical Sciences, 63, 1246-1256. Kempen, G. I., Miedema, I., Ormel, J., & Molenaar, W. (1996). The assessment of disability with the Groningen Activity Restriction Scale. Conceptual framework and psychometric properties. Social Science in Medicine, 43, 1601-1610. Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159-174. doi:10.2307/2529310 Paúl, C., Ribeiro, O., & Teixeira, L. (2012). Active ageing: An empirical approach to the WHO model. Current Gerontology & Geriatric Research, 2012, Article ID 382972. doi:10.1155/2012/382972 Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68-78. doi:10.1037/0003-066X.55.1.68 Saajanaho, M., Rantakokko, M., Portegijs, E., Törmäkangas, T., Eronen, J., Tsai, L.-T., . . . Rantanen, T. (2015). Personal goals and changes in life-space mobil- ity among older people. Preventive Medicine, 81, 163-167. doi:10.1016/j. ypmed.2015.08.015 Saajanaho, M., Viljanen, A., Read, S., Rantakokko, M., Tsai, L.-T., Kaprio, J., . . . Rantanen, T. (2014). Older women’s personal goals and exercise activity: An eight-year follow-up. Journal of Aging and Physical Activity, 22, 386-392. doi:10.1123/JAPA.2012-0339 1024 Journal of Aging and Health 31(6) Saito, T., Izawa, K. P., Omori, Y., & Watanabe, S. (2016). Functional Independence and Difficulty Scale: Instrument development and validity evaluation. Geriatric & Gerontology International, 16, 1127-1137. doi:10.1111/ggi.12605 Subramanian, S. V., Huijts, T., & Avendano, M. (2010). Self-reported health assess- ments in the 2002 world health survey: How do they correlate with educa- tion? Bulletin of the World Health Organization, 88, 131-138. doi:10.2471/ BLT.09.067058 Thanakwang, K., Isaramalai, S. A., & Hatthakit, U. (2014). Development and psycho- metric testing of the active aging scale for Thai adults. Clinical Interventions in Aging, 9, 1211-1221. doi:10.2147/CIA.S66069 Towards a common language for functioning, disability and health, ICF. (2002). Geneva, Switzerland: World Health Organization. Walker, A. (2007). Why involve older people in research? Age and Ageing, 36, 481-483. doi:10.1093/ageing/afm100 The WHOQOL Group. (1998). Development of the world health organization WHOQOL-BREF quality of life assessment. Psychological Medicine, 28, 551-558. Wilkie, R., Peat, G., Thomas, E., Hooper, H., & Croft, P. R. (2005). The Keele Assessment of Participation: A new instrument to measure participation restric- tion in population studies. Combined qualitative and quantitative examina- tion of its psychometric properties. Quality of Life Research, 14, 1889-1899. doi:10.1007/s11136-005-4325-2 World Health Organization. (2002). Active ageing: A policy framework. Geneva, Switzerland.

Journal

Journal of Aging and HealthPubmed Central

Published: Jan 1, 2018

There are no references for this article.