Assessment of Readiness for Mobility Transition (ARMT)

Abstract

Individualized assessment is important when counseling older adults concerning the transition from driving to nondriving mobility. This study validated a measure of emotional and attitudinal readiness in support of mobility transition counseling (MTC). Items derived from a mixed-methods approach were administered by mailed questionnaire to community-dwelling adults (n = 297; ages 57-95). Factor analysis was employed to form the 24-item Assessment of Readiness for Mobility Transition (ARMT). The ARMT–Total Score (ARMT-TS) demonstrated sound internal consistency and split-half reliability (.88 each). The ARMT-TS correlated as hypothesized with validity measures, including self-reported physical functioning, mental health, and openness to experience. High scorers, who evidenced strong self-reliance and an unwillingness to be a burden on others, are considered to be at risk when faced with a significant mobility transition. An appreciation for such differences can allow for personalized, tailored discussion and planning for when it is time to “hang up the keys.”

Keywords

assessment mobility counseling planning individual differences

Introduction

Personal mobility involves moving at will from one place to another. One’s chosen mode of transportation (e.g., walking unaided, walking with an assistive device, cycling, riding in a bus or car, driving a car) will depend on many factors, including the purpose of the trip, distance to be traveled, level of functional independence, available resources, and personal attitudes and preferences. Losses in vision, balance, muscle strength, stamina, reaction time, and other functional abilities are common in advancing age and may affect on individual mobility options. Of particular concern to community-dwelling older adults is the loss of personal independence associated with the transition from driving to nondriving mobility (Adler & Rottunda, 2006). Knowing how older adults in the community understand and approach such transitions is relevant for successful mobility counseling and planning, and such knowledge can both facilitate and promote successful mobility transitions later on (Kaiser, 2009; Suen & Sen, 2004).

A mobility transition occurs when one or more mobility options, such as driving a personal vehicle, become unsafe, impractical or impossible due to changing health, function, and/or other circumstances of life. A host of personal and situational factors can influence how a given individual may cope and adapt. Increasingly, health and social service professionals are asked to get involved in such transitions by assisting older adults and their family members to (a) recognize the realities of age-related mobility loss (present and/or anticipated); (b) address feelings of worry, threat, grief, uncertainty, and disruption that naturally arise; and (c) formulate viable plans that balance available resources with personal preferences and abilities. This process is known as mobility transition counseling (MTC; MacLean, Berg-Weger, Meuser, & Carr, 2007).

Any professional with training and/or experience concerning the aging process and mobility planning may conduct mobility transition counseling. An older adult’s readiness to engage in MTC and planning will depend, in part, on his or her level of awareness and felt concern about mobility change. Coping theorists suggest that threat appraisal—how a person perceives a new challenge—is a critical early motivator in the adaptive process (Lazarus & Folkman, 1984; Witte & Allen, 2000). A primary appraisal of risk to self (or others) that involves tangible worry, fear, or threat serves to activate secondary appraisals aimed at marshalling resources and taking action to deal with the situation at hand (see Figure 1).

Figure 1.

Appraisal, coping, and adaption process in response to a life stressor.

Although fear can be a helpful motivator, too much of it can interfere with this process and act as a barrier to productive action (Ellsworth & Scherer, 2003; Lewis, Watson, White, & Tay, 2007). On an individual level, threats to driving mobility may be viewed as something to be planned for now (or soon), as distant concerns to be considered months or years in the future, or as too worrisome to contemplate even in the face of high present risk (King et al., 2011). Readiness to manage a mobility transition starts with the individual, notably with the personal attitudes and emotions that may influence the appraisal process and subsequent adaptation. Few studies, to date, have examined how individual differences in these antecedent characteristics impact on mobility planning. This is an important area for research.

Individual Differences and Mobility Counseling

Although growing as a domain for social service practice, MTC, as practiced today, is still largely a function of individual professional initiative and formulation. Tools and guides exist to address specific needs and tasks in mobility planning (see http://seniortransportation.easterseals.com; http://www.beverlyfoundation.org). Many of these resources focus on instrumental decisions and transportation options in the transition to nondriving mobility and less so on underlying attitudes, meanings, and emotions of the individual client. Personal mobility is often measured in terms of self-reported driving behaviors (Classen et al., 2010) and miles traveled on a daily basis and mode of transportation (Marottoli, Ostfield, Merrill, Perlman, & Cooney, 1993). One of the few related individual difference measures in the literature, the Life Space Questionnaire (LSQ), quantifies personal mobility by combining definitions of home space and extent of regular travel (Stalvey, Owsley, Sloane, & Ball, 1999). The LSQ is a measure of current or preferred mobility but does not quantify emotional or attitudinal antecedents of such preferences.

Similarly, several studies have investigated the educational needs and decisional aspects of behavior change with respect to mobility transitions (Bamberg, 2007; Stalvey & Owsley, 2003; Tuokko, Gabriel, & Rhodes, 2007), but none have explored underlying attitudes and emotions (i.e., “readiness factors”) that may influence how individuals understand a mobility transition and choose to cope with it. One often overlooked readiness factor concerns as dispositional responses to life challenges. Coping theorist Richard Lazarus explained it this way:

An approach that doesn’t supplement contextual measurement of coping with an attempt at synthesis into a whole person is bound to be too limited . . . The aspect of personality that is most apt to be missed in such an approach is motivational, that is, it consists of general goals and situational intentions that mobilize and direct the choice of the coping strategies employed. (Lazarus, 1993, p. 243)

As a mobility counselor, it is not enough to merely offer a list of transportation alternatives and encourage the older client to adopt them, but rather professionals must also take into account individual readiness and motivation if interventions for mobility transitions are to achieve maximal benefit (Dickerson et al., 2007). Insights concerning the “whole person” and readiness are vital for the translation of mobility resources to the level of the individual in need (Leary, 2008).

A Clinical Scale to Assess Readiness for Mobility Transition

This article details findings from the second of a three-phase, qualitative-quantitative study to (a) understand the ways in which community-dwelling adults, age 55 and older, view mobility loss and transition in the context of advancing age and (b) develop a clinical tool to measure emotional and attitudinal readiness in support of MTC.

Focus groups in Phase 1 revealed diverse attitudes concerning mobility losses and transitions in aging (King et al., 2011). Participants ranged in age from 57 to 92 and self-identified along a continuum of disability and mobility function, ranging from fully functional (i.e., distant travel from home, full ambulation, safe driving) to moderately disabled (close-to-home travel only, difficulty with ambulation, partial to full compromise in driving ability). Themes and statements derived from these discussions formed the item pool for scale development and initial validation in Phase 2, the focus of his article. The new scale—the Assessment of Readiness for Mobility Transition (ARMT)—is now the subject of further validation study in Phase 3.

Focus group participants defined optimal mobility in the context of driving a motor vehicle and consequently viewed nondriving choices as suboptimal and a cause for personal concern (i.e., “how will I get around if I am not able to drive?”). Regardless of self-rated disability status, if a participant viewed him- or herself as a safe driver, this was enough to consider one’s mobility status as still within the optimal range. This finding is similar to feedback from another focus group study on mobility involving community-dwelling elders from upstate New York (Glasgow & Blakely, 2000). The thought of transitioning to a nondriving status was foreign to such participants and elicited various “not me” responses. Although the participant might have reported trouble in walking, still being able to drive successfully was the key factor in their self-definitions. In contrast, those reporting significant challenges to health or function (including their driving ability) struggled to adapt and remain mobile and so evidenced a greater awareness, if not acceptance, of nondriving options. Circumstances of life forced them to view driving as one choice along a continuum of mobility options. Lived experiences of disability and personal limitation can also influence readiness to cope with a mobility transition; some individuals need clear evidence before primary appraisals will favor an active, engaged coping response.

For those providing MTC services, it would be advantageous to know about the older adult’s emotional and attitudinal readiness before active planning begins. A paper-and-pencil measure of such readiness would allow the counselor to understand the elder on a personal level and tailor interventions for maximal potential benefit. A readiness measure could also serve as an educational tool to raise awareness and foster discussion on a group level (e.g., within the context of a safe driving course, other community education program, or a support group).

Design and Method

Item Development

Following a grounded qualitative method employed by Meuser and Marwit (2001; Marwit & Meuser 2002), focus group transcripts from Phase I were distributed to a review panel for detailed examination, thematic analysis, and later, scale item generation. Six investigators (authors and other professional volunteers) and four trained graduate student assistants served as reviewers. In addition to examining overall themes, reviewers were instructed to look for indicators of emotional and attitudinal readiness for mobility change. A full-day qualitative analysis retreat and consensus conference was convened, during which primary and secondary themes were identified (see description in King et al., 2011). One hundred fifteen readiness items (statements) were drafted for response on a 5-point Likert-type scale (5 = strongly agree; 1 = strongly disagree). Items with double or overlapping meanings were edited, combined, or dropped. A set of 87 items was determined by subsequent consensus.

As a final step, all participants from the original focus groups were invited back for a special debriefing session; 15 of 30 attended. Main findings were presented, and the 87 readiness items were reviewed and participants were invited to comment. Based on this feedback, ~10% of items were revised to enhance readability, comprehension, and/or clarity of meaning. This final step is rare in focus group research, yet served a constructive purpose here to ensure the veracity and face validity of items for the target population (see samples in Table 1).

Table 1.

Sample Attitudinal and Emotional Readiness Items.

Item number (from Phase 2 Questionnaire)	Item for response on 5-point Likert-type scale (5 = strongly agree)
21	I wish others would stop talking to me about my mobility.^a
22	My life can have meaning even if I rarely leave my home.
23	Thinking a lot about mobility loss can be like a self-fulfilling prophecy; by worrying, you risk making things worse.
24	Change comes easy for me, so I don’t worry too much about my future mobility.
25	I never thought about mobility change until filling out this questionnaire.
26	I would like to speak with a mobility specialist about a personal plan.
27	Asking for a ride creates an inconvenience for others.^a
28	It’s a waste of time to plan for the limitations of getting older, since you never know what is going to happen.
29	Other people simply don’t understand what it’s like to have limited mobility.
30	Mobility loss can be sudden or progressive, but it is always devastating.^a
31	I can avoid problems with mobility by taking care of myself.
32	I can continue contributing to my community even if my mobility is limited.
33	I would like to talk with someone about a mobility plan but I am afraid to bring it up.

Items included in the ARMT.

Procedure and Instrumentation

This study was approved by the Institutional Review Board at the University of Missouri–St. Louis. Recruitment for Phase 2 was accomplished through word-of-mouth and distribution of a flyer to social service professionals, independent living and retirement communities, and public bulletin boards. Once written consent was obtained, volunteers, age 55 and older, were contacted by telephone by a trained graduate student assistant and administered a brief prequestionnaire focused on demographics, current mobility status (e.g., driving vs. nondriving, satisfaction with places visited), and self-rated disability and mental health. Administration of the prequestionnaire provided an opportunity to monitor sample diversity, orient each volunteer to the topic, and motivate continued participation.

A 9-page questionnaire was distributed by mail to all volunteers. Of the 360 questionnaires mailed, 320 were returned resulting in a response rate of 89%. Twenty-three questionnaires were excluded from analysis due to excessive missing items. The mailed questionnaire included the same mobility status items as collected by telephone, the 87 readiness items, and a number of standardized validity measures:

SF-12v2 Health Questionnaire: Administered by telephone, the SF-12v2 is a 12-item scale developed for epidemiological research to measure self-rated physical functioning (disability) and mental health (Ware, Kosinski, Turner-Bowker, & Gandek, 2002; Ware & Sherbourne, 1992). The SF-12v2 is scored by computer; higher scores indicate better functioning in each domain. Positive associations between each score and readiness were hypothesized. In other words, respondents who viewed themselves as more physically capable and mentally healthy would evidence higher overall readiness to cope with a mobility transition.

Life Space Questionnaire (LSQ): Administered by telephone, the LSQ is a 9-item scale measuring places visited recently (within 14 days for this study) starting in one’s bedroom and advancing outward in concentric circles up to interstate and national travel (Stalvey et al., 1999). Higher scores indicate broader desire and preference for travel (2 = travel to one’s porch, deck or garage; 4 = travel in immediate neighborhood; 6 = travel over town border; 8 = travel outside of one’s state). A positive association between LSQ score and readiness was hypothesized.

Geriatric Depression Scale–Short Form (GDS-SF): Administered in writing, the GDS-SF is a 15-item yes–no screening tool designed to measure symptoms of clinical depression as manifested by older adults (scores in the 0-5 range indicate normal mood; 6-15 suggest possible depression warranting further evaluation; Sheikh & Yesavage, 1986). A negative association between GDS-SF score and readiness was hypothesized.

Well-Being Scale, Basic Needs subscale (WBS): Administered in writing, the WBS is a 22-item strengths-based measure of needs (e.g., feeling loved, having meaning in your life, adequate financial resources) as met in the respondent’s current life (Berg-Weger, Rubio, & Tebb, 2000; Rubio, Berg-Weger, & Tebb, 1999; Tebb, 1995). Possible scores range from 22 to 110. Higher scores indicate greater felt well-being across a range of resource and interpersonal indicators. A positive association between WBS score and readiness was hypothesized.

Openness to Experience Scale (OES): Administered in writing, the OES is a 20-item scale measuring trait-based cognitive flexibility and a willingness to accept uncertainty in daily-life situations (Goldberg et al., 2006). The OES is part of the International Personality Item Pool (see http://ipip.ori.org/ipip/) based on the Openness Scale of NEO Personality Inventory (Costa & McCrae, 1992). Possible scores range from 20 to 100. Higher scores indicate greater openness and flexibility. A positive association between OES score and readiness was hypothesized.

Satisfaction With Places Visited Regularly (SPVR): Administered by telephone and in writing, the SPVR is a single-item question (“How satisfied are you concerning the places you visit on a regular basis?”) answered on a 5-point Likert-type scale (5 = very satisfied). The SVPR was administered twice, a few weeks apart, and scored as a sum (range 2-10) with higher scores indicating greater expressed satisfaction. A negative association between SPVR score and readiness was hypothesized. In other words, those satisfied in the present may have less felt motivation and readiness to embrace change in the future.

Results

Participant Sample

Participants (n = 297) were 78% female with a mean age of 71 years (range 55-95) and mean education of 15 years (range 1-27 years). Most self-identified as White (77%) or African American (19%). One third (37%) reported themselves to be presently married or partnered, 27% widowed, 15% single, and 21% divorced. Respondents who chose “single” may have been married in the past; this was not clarified sufficiently at the time of the telephone interview. The majority (66%) reported living in single family homes. Most (76%) reported driving a car within the 2 weeks prior to completing the written questionnaire.

Summary data for the validity scales are shown in Table 2. A range of responses was obtained for each scale, with three exceptions: the LSQ score was negatively skewed (i.e., the majority of respondents traveled quite far from home in the past 14 days); the SPVR score was similarly skewed (i.e., the majority of respondents were quite satisfied with the places they visited regularly); and the GDS scores were negatively skewed (i.e., most scored below the cutoff for depression). These findings suggest that the majority of this sample was active, content, and satisfied travelers.

Table 2.

Descriptive Statistics for Validity Measures.

Validity measures	M	SD	Minimum	Maximum	α	Hypothesized correlations to ARMT-TS^a
SF-12v2 physical function	43.7	11.4	9.4	63.8	NA	−
SF-12v2 mental health	52.8	9.9	18.8	71.8	NA	−
Life space	6.77	1.2	2	9	.63	−
Geriatric depression	1.9	2.6	0	11	.83	+
Well-being	79	14.2	23	105	.92	−
Openness to experience	74.6	11.7	36	99	.85	−
Satisfaction with places visited	8.5	1.7	2	10	.68	+

Higher emotional and attitudinal readiness is represented by a lower ARMT–Total Score (ARMT-TS). For example, those scoring high on depression were hypothesized to demonstrate less overall readiness, hence the positive correlation shown in the table above.

Factor Analysis of Readiness Items

Step 1

The 87 readiness items were examined for extreme skewness (i.e., significant nonnormality of the response distribution) which can attenuate correlations and so reduce reliability. A skewness metric was calculated (absolute value/standard error), and items showing scores ≥ 4 were dropped. The mean skewness metric of the 42 dropped items was 7.3 (SD = 2.6); the mean item score ranged from 1.4 to 4.4 (M = 3.3, SD = 0.94; see examples in Table 2). In contrast, the 45 items retained for further analysis had a mean skewness metric of 2.1 (SD = 1.2); the mean item score ranged from 2.3 to 3.5 (M = 3.0, SD = 0.37).

Step 2—Missing values replacement

The 45 retained items had a mean percentage of missing values of 3.2%. Two thirds (62%) of these items had less than 3% missing values. Missing values were replaced via mean item substitution.

Step 3—Factor analysis

The 45 retained items were analyzed using principal components (PC) factor analysis (varimax rotation, rotation of factors with eigenvalue > 1, interpretation of loadings ≥ 0.32). The first PC analysis yielded 13 factors that explained 61% of the variance, including 5 relatively large factors explaining 35% of the variance. Only one to two items loaded on factors 6 to 13. Nevertheless, this analysis yielded an acceptable Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy of 0.83. Bartlett’s test of sphericity was also significant, χ²(990) = 3,990, p < .001, rejecting the null hypothesis that the variables in the data set were uncorrelated.

Items were then eliminated from the model based on several criteria: (a) item did not load ≥0.32 on any factor; (b) item had relatively low loadings (< 0.40) across ≥3 factors; (c) item loaded singly on a factor, beginning with the lowest loading items on the smallest factors. Two items were dropped following these rules until no more items qualified. With each PC reanalysis, the KMO statistic either remained the same or increased and Bartlett’s test was always significant (p < .001).

Final 24-Item ARMT Solution

The 24 readiness items comprising the ARMT instrument yielded five factors that explained 54% of the variance (KMO = 0.88; Bartlett’s χ²(276) = 2,103, p < .001). Factors 4 and 5 included just three items in each, and these were found to be internally unreliable (α = .54 and .44, respectively). These six items appeared to have similar clinical meaning, however, and were thus combined to form a single fourth factor as described below.

Table 3 depicts the content labels and item loadings for the four factors. Factor 1 (anticipatory anxiety) represents anxiety and worry about a loss of independence, with an emphasis on personal implications and emotional responses. Factor 2 (perceived burden) represents felt concern associated with becoming overly dependent and a burden on others. Factor 3 (avoidance) represents a general resistance to address the topic of mobility loss (i.e., “not me” attitude). Factor 4 (adverse situation) represents a general perception of mobility loss as harmful to individual well-being and quality of life.

Table 3.

Loadings in Final Rotated Factor Matrix and ARMT Scale Content.

ARMT item (Number in administration order)	Anticipatory anxiety	Perceived burden	Avoidance	Adverse situation
	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5
I feel depressed at the thought of being limited in my mobility (10)	0.72
My future independence hinges on my ability to get myself around (18)	0.68
Loss of mobility is very isolating and depressing (15)	0.67
When I see older people with significant limitations in mobility, I fear that I will end up like that too (12)	0.65
I feel angry when I think about losing my mobility (22)	0.63
I shudder to think of a time when I am less mobile than I am now (16)	0.62
It really frustrates me when I have difficulty getting around (21)	0.60
A big loss of mobility would really hurt my self-esteem (14)	0.58
Asking others for help with mobility means that I am losing my independence (2)	0.53
It is not easy for me to ask for help with transportation when I need it (24)		0.79
I am a burden if I ask others for help with transportation (3)		0.76
Asking for a ride creates an inconvenience for others (6)		0.73
I do not like to ask others for a ride (9)		0.70
I feel self-conscious when my mobility needs become a concern for others (23)		0.46
I have not thought much about my future mobility before today (19)			0.76
There is no way to plan for loss of mobility in aging (13)			0.64
I wish others would stop talking to me about my mobility (5)			0.59
I avoid thinking about losing my mobility (4)			0.58
Other people simply don’t understand what it’s like to have limited mobility (7)				0.69
It is devastating for older people to have someone take away their car keys (8)				0.57
Mobility loss can be sudden or progressive, but it is always devastating (1)				0.50
I’ve seen others become frail and immobile in older age, and I am determined to avoid this fate at whatever cost (20)					0.72
Moving to a retirement community is too restrictive for my desired mobility (11)					0.58
I refuse to accept that I might lose my mobility in the future (17)					0.36

Table 4 shows descriptive statistics for the ARMT–Total Score (ARMT-TS) and four factor (subscale) scores. As the number of items differs between subscales, mean scoring was used. The directionality of ARMT scores merits clarification here. Although responses to individual items were made via a 5-point Likert-type scale (5 = strongly agree), the concept of readiness is interpreted in the other direction. High total and subscale scores indicate less readiness (i.e., excessive worry and resistance) in context of a mobility transition, whereas low scores indicate greater readiness (i.e., more awareness, comfort and adaptability) for such a transition.

Table 4.

Descriptive Statistics for ARMT Total and Factor (Subscale) Scores.

	M	SD	α	Guttman split-half	Number of items
ARMT–Total score^a	2.93	.64	.88	.88	24
F1—Anticipatory anxiety	2.93	.88	.87	.88	9
F2—Perceived burden	2.76	.91	.79	.72	5
F3—Avoidance	2.77	.89	.62	.58	4
F4—Adverse situation	3.12	.72	.63	.55	6

Mean scoring is used.The range is 1 to 5 points in each case.

Scale Reliability

Internal consistency (Cronbach’s alpha) and Guttman split-half reliability measures were sound for the total score and first two factors (anticipatory anxiety, perceived burden) but marginal for the remainder (see Table 4). More research is needed to confirm the reliability, validity, and utility of the final two factors.

Data from a recent student research project (volunteer sample; n = 69; age range = 61-93; 76% female; 93% White) conducted by Stowe (2011) yielded essentially identical internal consistency and split-half values for the ARMT-TS at first administration (0.88 and 0.84, respectively). Values for the four-factors scores were also much the same as those in the validation sample. Stowe (2011) presented vignettes about mobility change to motivate discussion and self-reflection in half of his sample; the remaining served as controls. He administered the ARMT, again, to all participants after a 2-month interval. His control sample (n = 36) provided an opportunity to demonstrate test-retest reliability. The ARMT-TS showed high test-retest reliability with a correlation coefficient of 0.84 (p < .01) suggesting that the construct of readiness is relatively stable in persons over time. Individual factor scores were less stable, yielding coefficients of 0.54 (p < .01) for Factor 1 (Anticipatory Anxiety) and Factor 2 (Perceived Burden), 0.36 (p < .03) for Factor 3 (Avoidance), and 0.27 (ns) for Factor 4 (Adverse Situation).

Taken together, findings from the primary validation sample and this additional student-developed sample suggest that the overall construct of readiness, as represented in the ARMT-TS, can be measured reliably and is also relatively stable over time.

Associations of Scale Scores

Correlations among the ARMT scores are shown in Table 5. As is common in individual differences measures, the ARMT-TS and factor scores were significantly intercorrelated. The factor scores all relate to a single readiness construct. Correlations for Factor 2 (Perceived Burden) are moderate and suggest sufficient uniqueness to support separate, clinical interpretation. Factor 3 (Avoidance) was least correlated among the group suggesting that it captures a more distinct component of readiness. Factor 1 (Anticipatory Anxiety) was highly correlated with the total score and so probably not distinct enough for individual interpretation.

Table 5.

Intercorrelations Among Demographic, ARMT, and Validity Scales.

	ARMT– Total Score	F1— Anticipatory anxiety	F2— Perceived burden	F3— Avoidance	F4— Adverse situation
ARMT–Total score	1	.898**	.710**	.520**	.732**
F1— Anticipatory anxiety	.898**	1	.559**	.276**	.534**
F2—Perceived burden	.710**	.559**	1	.163**	.296**
F3—Avoidance	.520**	.276**	.163**	1	.357**
F4—Adverse situations	.732**	.534**	.296**	.357**	1
Age	.037	−.009	.052	.085	.008
Years of education	−.122*	−.042	−.025	−.305**	−.093
Number of nondriving modes	−.103	−.103	−.004	−.101	−.090
Satisfaction with places visited	−.190**	−.246**	−.244**	.126*	−.056
Life Space Questionnaire	−.038	.020	−.046	−.090	−.055
SF-12 physical health composite	−.155**	−.120*	−.216**	.067	−.174**
SF-12 mental health composite	−.175**	−.284**	−.139*	.070	.000
Geriatric Depression Scale–Short Form (15)	.368**	.417**	.331**	−.010	.188**
Well-Being Scale	−.356**	−.366**	−.364**	−.096	−.133*
Openness to experience	−.245**	−.170**	−.128*	−.390**	−.103

p < .05. **p < .01.

Content Validity

As noted above, face and content validity were assured through the use of a mixed-methods approach involving substantial input from a diverse sample of community-dwelling older adults. Qualitative analysis in Phase 1 involved the application of a grounded approach (see King et al., 2011). ARMT items were derived from focus group interviews and expert consensus, thus ensuring content grounded in the views of the target population for this tool.

Construct Validity

Correlations of ARMT scores with validity measures and demographic characteristics are shown in Table 5. Most hypothesized associations between overall readiness as represented by the ARMT-TS and validity measures were confirmed (see Tables 3 and 5) with two exceptions: (a) a modest positive association was found between satisfaction with current places visited and readiness (i.e., higher satisfaction with current travel was associated with higher readiness scores) and (b) the extent of current life space was uncorrelated with readiness. In other words, no relationship was found between life space (i.e., typical radius of travel from home) and readiness as measured by both total and subscale scores.

Age was uncorrelated with ARMT scores, an important attribute for a scale intended for use across a wide-age spectrum. A different pattern emerged for years of education. Although uncorrelated with Factor 1 (Anticipatory Anxiety) and Factor 2 (Perceived Burden), education was negatively correlated with Factor 3 (Avoidance). In other words, persons with high education were somewhat less likely to endorse an avoidance response.

Mean comparisons (t test) were made between the ARMT-TS and four subscale scores and other relevant variables in the data set: gender, race (White vs. Other); marital status (single/divorced/widowed vs. married/partnered); living situation (single family home vs. apartment/other), recent driving status (driving vs. nondriving). No differences were found at a conservative p < .01 level.

Finally, a subsample of high scorers on the ARMT-TS (1+ standard deviation above the mean, n = 57) and low scorers (1– standard deviation below the mean, n = 42) was generated. As noted above, high scorers are considered less ready to manage a mobility transition. High and low scorers differed significantly on a number of validity measures (p < .01): high scorers evidenced greater self-reported depression and less well-being, openness, mental health, general happiness, and satisfaction with places visited. Item responses for the 63 readiness items not included in the ARMT were compared between high and low scorers. Significant differences (p < .01) were revealed by t tests on 45 items as shown in Table 6. High scorers were more likely to agree with items emphasizing individual responsibility, self-reliance, and a pessimistic view of future mobility. High scorers were more likely to disagree with items emphasizing reliance on others, a flexible view of mobility and personal worth, and confidence in personal coping resources.

Table 6.

Attitudinal Differences Between High and Low Scorers on Non-ARMT Readiness Items.

Low scorers more likely to disagree	Low scorers more likely to agree
I feel embarrassed when others see me struggling to get around.	My family and friends know when to step in and help me with transportation.
I slow down and this impacts my mobility significantly.	It is easy for me to accept a ride when one is offered.
I worry that I will not have the resources to arrange transportation when I need it.	There are viable ways for me to get around and remain mobile other than driving or riding in a car.
It is a shameful thing for older people to have to rely on others to help them get around.	I can downsize to a smaller space and still be happy.
I worry what others will think of me when I’m not able to get around as much as I do now.	Change comes easy for me, so I don’t worry too much about my future mobility.
I feel isolated due to my lack of mobility.	I can continue contributing to my community even if my mobility is limited.
I do not like the prospect of having to deal with mobility loss.	If I need to call a cab to get somewhere important, I will do it.
A positive attitude is not enough to cope successfully with changes in mobility.	If a friend or family member wants to help me get somewhere, I’m very open to their assistance.
Thinking a lot about mobility loss can be like a self fulfilling prophecy; by worrying, you risk making things worse.	I would be willing to use alternative transportation (e.g., taxi, shuttle) if or when driving myself is not an option.
I never thought about mobility change until filling out this questionnaire.	I can deal with changes in my mobility as I get older.
It’s a waste of time to plan for the limitations of getting older, since you never know what is going to happen.	My ability to get around doesn’t determine my quality of life.
I would like to talk with someone about a mobility plan but	I would be happy to accept rides from my family and friends.
I am afraid to bring it up.	I am confident I can respond to any mobility changes in my future.
I will look “really old” if I get onto a senior service bus.	I don’t focus on my age as much as what I can still do and enjoy.
I worry a lot about mobility loss and its impact on my life.	I am at peace about my health and the mobility that I still enjoy each day.
It is hurtful to me when others assume and act as though I need help with my mobility.	I can get where I need to go, even if I don’t drive.
I can be satisfied with my mobility as long as it includes driving a car.	I choose not to worry over changes in mobility.
If I admit to being limited in mobility, then I am giving up.	I know that I can cope with loss of mobility because I have dealt with major changes in the past.
Others have raised the issue of mobility, but I prefer not talking about it.
I am reluctant to use public transportation, whether available in my area or not.
My mobility is solely my responsibility.
If those close to me knew that I was having trouble with mobility, they would take over my life.
Older people are pushed to the side because they can’t keep up.
I am destined to end up like most older people–limited and frail.
It is other people’s fault that I cannot get around as much as I want to.
I worry that I won’t be able to afford the transportation I need in the future.
I feel worthless when I have difficulty getting around.
It is rare that people ever ask if I need help with transportation.

Note: Readiness items that were excluded due to skewness or did not load in the factor analysis.

Discussion

This study establishes the relevance and importance of measuring emotional and attitudinal antecedents (i.e., readiness factors) to adaptive coping in older adults faced with a significant mobility transition, such as retirement from driving. The result is a new clinical practice tool, the ARMT, for use by social service, health, and transportation professionals when assessing older clients and intervening to promote individualized planning. The ARMT is based on a multiphase, qualitative-quantitative research effort to identify and measure key individual differences that define the construct of readiness. The findings define the qualities of individuals along a continuum from lower readiness (higher ARMT scores) to higher readiness (lower ARMT scores).

The ARMT–Total Score (ARMT-TS) and scores for Factors 1 (Anxiety) and 2 (Perceived Burden) were found to have sound reliability in a large volunteer sample of community-dwelling older adults (n = 297). Face validity was enhanced early in the scale-development process through an item-by-item review by many of the original focus-group participants. Items were reworded based on this feedback to maximize understanding and acceptance by the target population. Content and construct validity were then confirmed through correlations with various established measures and through examination of response patterns by those scoring one standard deviation above and below the sample mean.

As hypothesized, a high ARMT-TS, indicating lower readiness and greater potential risk in a mobility transition, was associated with lower self-reported physical function, mental health, emotional and instrumental well-being, and a less open (i.e., more rigid, less adaptable) personality style. Similarly, high scorers endorsed more depressive symptoms and lower satisfaction with places visited. Low motivation is a common symptom in depressed individuals, making the former relationship logical. The latter association makes sense in the context of readiness, such that those satisfied in the present have less felt motivation to fear change in the future.

In addition to meaningful validity correlations, an equally important set of relationships speaks to the potential generalizability of the ARMT as an individual differences measure. ARMT scores were found to be unrelated to age, gender, race (white vs. non-White), marital status, current driving status, and willingness to attend a mobility-related education event. The lack of association to core demographic factors suggests that the ARMT may be applied similarly across older adult populations from the young-old to the old-old. Although the sample in this study was reasonably diverse, additional research is needed to confirm a broad lack of racial bias. Initial data are promising but not definitive on this point.

The lack of association with driving status and educational participation merits further discussion. Although loss of the driving privilege is a major concern for many older adults, it is notable that ARMT scores were found to be unrelated to driving status. Items were worded to focus on broader issues of personal mobility and not just driving. The ARMT was designed to quantify emotional and attitudinal antecedents that may apply to and influence any mobility transition. This lack of relationship to driving status supports the ARMT’s broad applicability. The lack of association with a willingness to attend an education program on mobility is similarly supportive. The decision to attend an educational program (or not) is an instrumental coping behavior—something that comes after an initial appraisal of worry or threat. Any respondent may choose, in time, to attend a program, but ARMT scores do not suggest a course of action. Instead, ARMT scores show the older client’s state of mind on the issue. Understanding the client’s emotional and attitudinal readiness is a reasonable first step in any individualized MTC intervention.

Limitations

This study, and the ARMT measure itself, have a number of strengths and limitations. Strengths include a large sample of community-dwelling adults, a broad distribution of scores on most variables (see Table 2), and confirmation of most a priori hypothesized relationships. ARMT items were grounded in the expressed views of older adults from the target population, and the final structure of the scale was confirmed through a rigorous and unbiased factor analytic process.

The construct of emotional and attitudinal readiness to manage a significant mobility transition is new to the literature with this publication. As noted earlier, much of the work in this field have focused on strategies to identify at-risk older drivers (e.g., those with vision loss, dementia, and other serious conditions), assess their fitness to drive, develop new transportation options, and encourage seniors to adopt alternatives that may work for them (Dickerson et al., 2007). The innovation in the present work is the addition of coping theory and a new construct to guide assessment and intervention on an individual, person-centered level. Although the construct of readiness appears valid from this initial work (see characteristics of high vs. low scorers in Table 6), further research is needed. The predictive value of measuring this readiness construct is still to be determined, and other investigators can assist by incorporating the ARMT into ongoing or new studies on elder mobility.

Although the sample size was adequately large, it was still a volunteer sample and so nonrandom. Substantially more women participated than men, 63% were unmarried (including widowed, divorced, and single), and most reported themselves to be reasonably happy with their current mobility status. The high percentage of unmarried participants can be attributed, in part, to the female dominance and high average age of the sample. According to the America’s Families and Living Arrangements Survey (U.S. Census, 2010), 40% of American women aged 65+ self-identify as widowed and 11% as divorced. The high proportion of divorced participants defies ready explanation. Due to a data-collection error, also, the status of those who self-identified as single cannot be determined. What is relevant for this study is the fact that older, single women often have fewer supports and choices with respect to personal mobility and transportation in comparison with their married counterparts (Bryanton, Weeks, & Lees, 2010). Mobility counselors must take this reality into account but not necessarily with respect to the readiness construct, as ARMT scores were not found to differ based on marital status.

It is notable that the distributions of other key variables, including self-rated physical functioning, mental health, well-being, and openness to experience, were broad, indicating that many types of community-dwelling adults were represented. As the ARMT is used in future research with different populations, it will be interesting to see if psychometric characteristics in this first study hold up.

From a practice perspective, a potential limitation is the strong wording of some items. Unlike survey instruments designed to measure general views in a population, the ARMT is a clinical scale intended to separate individuals into risk categories and target them for possible intervention. Scale items are quotations and paraphrases from real people in most instances, and so there is an “edge” to some of them. Administration of the ARMT may be limited in some settings to a face-to-face dialogue so that this provocative wording may be processed along the way. Early users of the scale have reported, in fact, that the items work well for fostering discussion and elaboration. The ideal administration of the ARMT may be in the context of a counseling session, in which the item-related discussion may be just as useful for person-centered planning as specific final scores.

Another limitation may lie in the desired qualifications of the professional user. A detailed understanding of the aging process and good counseling skills are essential. A working knowledge of driving and nondriving mobility options is also important. The ARMT is intended for use by social service professionals (e.g., social workers, gerontologists, counselors, psychologists, care managers), health professionals (e.g., occupational therapists, nurses), and transportation professionals (e.g., mobility planners).

Future Applications

Pending additional validation studies, it is recommended that most users of the ARMT calculate and interpret just the total score (ARMT-TS) which represents the full readiness construct. A cut score of one standard deviation above the validation sample mean is suggested (>3.57). The factor scores show different components of the readiness construct and so may be useful for professionals skilled in counseling and planning interventions. The first factor score, Anticipatory Anxiety, is highly correlated with the total score and not particularly useful as a distinct measure. The other factors, however, capture themes voiced frequently in the original focus groups (see King et al., 2011). Concerns about burdening others and fear associated with postdriving quality of life were prominent and are likely shared by many community-dwelling older adults.

As shown in Table 6, high scorers are likely to express various emotions and attitudes that may hinder effective planning. As currently conceptualized, the ARMT provides a means to understand the personal attitudes and concerns that older adults may bring to the planning encounter. High and low scorers should be treated differently. An immediate discussion on use of the local bus would not be appropriate with a high scorer, for example. That individual will need to voice his or her concerns and gradually come to view mobility needs in a more balanced way. Fear of becoming a burden on others is a major concern for some individuals; they will need time and attention to work through such fears and impulses to avoid. In contrast, low scorers may already understand the personal implications of mobility change and come to the encounter ready to form a new life plan. Knowing where on the readiness continuum the older client falls is helpful information for any mobility intervention.

Although the ARMT may have its full utility as an assessment and interviewing tool in one-on-one interventions, the measure can also support larger education and awareness-raising objectives. The ARMT can be administered, for example, in a small group or classroom setting, not so much for the final scores but as a way to engage participants in thinking about issues that, for some, will be new and unexplored.

More research is needed to confirm the predictive validity and grassroots clinical utility of the ARMT in mobility transition counseling and planning when applied in a community-based setting. This tool is a first of its kind in the literature and suggests a new approach to mobility-related planning; namely, that professionals must first assess individual feelings and attitudes and then tailor interventions to meet the expectations and capabilities of the client. Prospective users may obtain an administration copy of the ARMT by email request (meusert@umsl.edu) or online (www.umsl.edu/~mtci or http://seniortransportation.easterseals.com/site/PageServer?pagename=NCST2_about_projects_ARMT).

Footnotes

Acknowledgements

The authors wish to thank our committed volunteer participants and the many organizations that helped in the recruitment process, including the Alzheimer’s Association − St. Louis Chapter, Shepherd Center, St. Louis Naturally Occurring Retirement Community, and Washington University Knight Alzheimer’s Disease Research Center. We also wish to acknowledge other key supporters of this project: Patricia Niewoehner, OTR/L, CDRS, St. Louis VA Medical Center; Richard Yakimo, PhD, College of Nursing, University of Missouri − St. Louis; M. Denise King, PhD, School of Social Work, University of Missouri − St. Louis; and graduate students: Elizabeth Yates and Bryant Price (University of Missouri − St. Louis), and Lauren Choate (Saint Louis University).

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a grant from the National Center on Senior Transportation (NCST), Washington, DC (Meuser, T.M., Principal Investigator). The NCST is administered by Easter Seals, Inc., in partnership with the National Association of Area Agencies on Aging (N4A).

Author Biographies

Thomas M. Meuser, PhD, is a clinical psychologist and director of the Gerontology Graduate Program at the University of Missouri – St. Louis. His research interests include mobility in aging, medical fitness to drive, grief and loss issues, and gerontological education.

Marla Berg-Weger, PhD, LCSW, is a professor and director of Field Education at the Saint Louis University School of Social Work, St. Louis, Missouri. Her research interests focus on older adult mobility and transportation issues. Along with Dr. Meuser, she has also developed educational curricula for health, law, and social service professionals to integrate older adult mobility issues into their professional work.

John T. Chibnall, PhD, is professor of Neurology & Psychiatry in the School of Medicine at Saint Louis University. His research focuses on psychosocial aspects of chronic pain, including in older adults.

Annie C. Harmon, MSG, is a Gerontologist and doctoral student in health education and health behavior at Ann Arbor School of Public Health, University of Michigan. She is interested in how older adults cope with changes in mobility with aging, particularly the decision to retire from driving and the process that follows.

James D. Stowe, MS, is the coordinator of trauma injury prevention and outreach education at the Frank L. Mitchell Jr. Trauma Center, University of Missouri Hospital. He is also a doctoral candidate in human development and family studies at the University of Missouri. His interests include mobility transitions in older age and sustained safe mobility throughout the life course.

References

Adler

Rottunda

(2006). Older adults’ perspectives on driving cessation. Journal of Aging Studies, 20(3), 227-235.

Bamberg

(2007). Is a stage model a useful approach to explain care drivers’ willingness to use public transportation? Journal of Applied Social Psychology, 37, 1757-1783.

Berg-Weger

Rubio

D. M.

Tebb

(2000) The Caregiver Well-Being Scale–Revisited. Health and Social Work, 25, 255-263.

Bryanton

Weeks

L. E.

Lees

J. M.

(2010). Supporting older women in the transition to driving cessation. Activities, Adaptation & Aging, 34(3), 181-195.

Classen

Winter

S. M.

Velozo

C. A.

Bédard

Lanford

D. N.

Brumback

Lutz

B. J.

(2010). Item development and validity testing for a self-and proxy report: The Safe Driving Behavior Measure. American Journal of Occupational Therapy, 64, 296-305.

Costa

P. T

Jr. McCrae

R. R.

(1985). The NEO Personality Inventory. Odessa, FL: Psychological Assessment Resources.

Dickerson

A. E.

Molnar

L. J.

Eby

D. W.

Adler

Bédard

Berg-Weger

. . . Trujillo

(2007). Transportation and aging: A research agenda for advancing safe mobility. The Gerontologist, 47, 578-590.

Ellsworth

P. C.

Scherer

K. R.

(2003). Appraisal processes in emotion. In Davidson

R. J.

Goldsmith

Scherer

K. R.

(Eds.), Handbook of affective sciences (pp. 572-595). New York, NY: Oxford University Press.

Glasgow

Blakely

R. M.

(2000). Older nonmetropolitan residents’ evaluations of their transportation arrangements. Journal of Applied Gerontology, 19(1), 95-116.

10.

Goldberg

L. R.

Johnson

J. A.

Eber

H. W.

Hogan

Ashton

M. C.

Cloninger

C. R.

Gough

H. C.

(2006). The international personality item pool and the future of public-domain personality measures. Journal of Research in Personality, 40, 84-96.

11.

Kaiser

H. J.

(2009). Mobility in old age: Beyond the transportation perspective. Journal of Applied Gerontology, 28, 411-418.

12.

King

M. D.

Meuser

T. M.

Berg-Weger

Chibnall

J. C.

Harmon

Yakimo

(2011). Decoding the Miss Daisy syndrome: An examination of subjective responses to mobility change. Journal of Gerontological Social Work, 54(1), 29-52.

13.

Lazarus

R. S.

(1993). Coping theory and research: Past, present, and future. Psychosomatic Medicine, 55, 234-247.

14.

Lazarus

R. S.

Folkman

(1984). Stress, appraisal, & coping. New York, NY: Springer.

15.

Leary

(2008). Policy intersections or policy chasm—State elder mobility policy, practice and long-term care reform (Doctoral dissertation). George Mason University, Fairfax, VA.

16.

Lewis

I. M.

Watson

White

K. M.

Tay

(2007). Promoting public health messages: Should we move beyond fear-evoking appeals in road safety? Qualitative Health Research, 17, 61-74.

17.

MacLean

Berg-Weger

Meuser

T. M.

Carr

D. B.

(2007). Driving retirement: Help with counseling older adults. Family Practice Recertification, 29(6), 1-6.

18.

Marottoli

R. A.

Ostfield

A. M.

Merrill

S. S.

Perlman

G. D.

Cooney

L. M.

(1993). Driving cessation and changes in mileage driven among elderly individuals. Journal of Gerontology: Social Sciences, 48, 255-260.

19.

Marwit

S. J.

Meuser

T. M.

(2002). Development and Initial Validation of an Inventory to Assess Grief in Caregivers of Persons with Alzheimer’s Disease. The Gerontologist, 42(6), 751-765.

20.

Meuser

T. M.

Marwit

S. J.

(2001). A comprehensive stage-sensitive model of grief in dementia caregiving. The Gerontologist, 41, 658-700.

21.

Rubio

D. M.

Berg-Weger

Tebb

(1999). Assessing the validity and reliability of well-being and stress in family caregivers. Social Work Research, 23(1), 54-64.

22.

Sheikh

J. I.

Yesavage

J. A.

(1986). Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. In Clinical gerontology: A guide to assessment and intervention (pp. 165-173). New York, NY: Haworth.

23.

Stalvey

B. T.

Owsley

(2003). The development and efficacy of a theory-based educational curriculum to promote self-regulation among high-risk older drivers. Health Promotion Practice, 4(2), 109-119.

24.

Stalvey

B. T.

Owsley

Sloane

M. E.

Ball

(1999). The Life Space Questionnaire: A measure of the extent of mobility of older adults. Journal of Applied Gerontology, 18, 460-478.

25.

Stowe

J. D.

(2011). Heightening older adults’ awareness of mobility preparedness using multiple segment factorial vignettes. Unpublished manuscript, University of Missouri–Columbia.

26.

Suen

S. L.

Sen

(2004). Mobility options for seniors. In Transportation in an aging society: A decade of experience, Conference Proceedings 27 (pp. 97-113). Washington, DC: Transportation Research Board.

27.

Tebb

S. S.

(1995). An aid to empowerment: A caregiver well-being scale. Health and Social Work, 20(2), 87-92.

28.

Tuokko

H. A.

McGee

Gabriel

Rhodes

R. E.

(2007). Perception, attitudes and beliefs, and openness to change: Implications for older driver education. Accident Analysis and Prevention, 39, 812-817.

29.

U.S. Census (2010). American Fact Finder (http://factfinder2.census.gov/main.html). Accessed August 1, 2011.

30.

Ware

J. E.

Sherbourne

C. D.

(1992). The MOS 36-item short-form health survey (SF-36): Conceptual framework and item selection. Medical Care, 30, 473-483.

31.

Ware

J. E.

Kosinski

Turner-Bowker

D. M.

Gandek

(2002). How to score Version 2 of the SF-12 Health Survey. Lincoln, RI: Quality Metrics.

32.

Witte

Allen

(2000). A meta-analysis of fear appeals: Implications for effective public health campaigns. Health Education & Behavior, 27, 591-615.