357

Physical Activity Training forFunctional Mobility in Older Persons

Tom HickeyFredric M. Wolf

Lynne S. RobinsUniversity of Michigan

Marilyn B. WagnerCleveland State University

Wafa Harik

Case Western Reserve University

The effectiveness of low-intensity physical activity for improving functional ability and psycho-logical well-being in chronically impaired older individuals was demonstrated in a pilot study.Participants who completed 6 weeks of structured low-intensity exercise (N = 77) improved inthe time and number of steps required to walk a measured course, in self-assessments of mobilityand flexibility, and in three measures of well-being. Those who continued to exercise in a peer-ledprogram (n = 32) maintained improvements in mobility and optimism after 18 weeks.

Difficulties in balance and mobility and limitations in musculoskeletalstrength, range of motion, and flexibility are fairly common among olderpeople. For example, 19% of noninstitutionalized individuals age 65 yearsand over experience difficulty in walking, and approximately 8% havedifficulty in getting in and out of chairs and beds (National Center for HealthStatistics, 1987). Among individuals between the ages of 75 and 84 who werefollowed in the Framingham Disability Study, 39% were unable to stand fora 15-minute period, 23% were unable to walk one-half mile, 15% could notclimb stairs, 24% were unable to lift small items (up to 10 pounds), and morethan 50% were unable to lift heavier items or to stoop, crouch, or kneelwithout difficulty or assistance (Jette & Branch, 1981). Because these find-

AUTHORS’ NOTE: This research was supported by a grant from the Cleveland Foundationand conducted in collaboration with the Golden Age Centers of Greater Cleveland, Inc.,Cleveland, OH.

The Journal of Apphed Gerontology, Vol 14 No 4, December 1995 357-371 1@ 1995 The Southern Gerontological Society

358

ings were based on a relatively healthy older sample, the magnitude of suchdisability problems in the older population may be even greater.

These deficits in strength, range of motion, and flexibility contribute tothe loss of functional independence, placing frail older people at risk forinstitutionalization. In fact, limited mobility is a critical risk factor associatedwith health dependency among older people. Difficulties in walking andrelated physical problems are significant determinants of the need for insti-tutional care (Branch & Jette, 1982; Weissert & Cready, 1989).

There is evidence to suggest that losses in functional mobility can beprevented in many older persons through physical activity. Various forms ofexercise have been found to improve strength and mobility and to increasecardiac fitness necessary for active functioning (Holloszy, 1983; Lampman& Savage, 1988; Morey et al., 1991; Vallbona & Baker, 1984). Given theimportance of functional status for independence in the older population, thespecific role of different types of physical activity in maintaining or improv-ing the physical and functional dimensions of mobility requires close exami-nation in older adults at high risk of institutionalization.

This report summarizes a preliminary investigation of the effectiveness oflow-intensity exercise for improving the functional health of older adults withlimited mobility as a result of chronic health problems and prolongedphysical inactivity. The data reported in this article were collected in aresearch demonstration intended (a) to develop a low-intensity exerciseintervention that would appeal to chronically impaired older persons and berelatively easy for them to do and (b) to provide a basis for planning furtherresearch to determine whether light exercise could lead to measurable func-tional health benefits.

The exercise intervention was called SMILE, for So Much Improvementwith a Little Exercise. The SMILE program was suggested by an earlier studyof low-intensity exercise on people with diabetes, which demonstrated thatstructured physical activity could improve functional health and mobility inchronically impaired and sedentary adults (Howard et al., 1985). The SMILEprogram employs a similar nonaerobic, low-intensity exercise intervention.However, it was refocused to emphasize movements that address the specificlimitations in strength, range of motion, and mobility of physically inactiveolder persons for whom high-intensity exercise was not a realistic option.The intervention also was intended to minimize the risks associated with

undertaking a physical activity regimen for sedentary older persons withchronic impairments in functional mobility.

Although numerous studies have investigated various health-related ef-fects of exercise, most of what is known about the outcomes of physicalactivity on the health and functioning of older persons is based on relatively

359

healthy populations, and most interventions are of short duration and ofmoderate to high intensity, such as aerobic programs, where effects areevident after a short period of time. Only a few studies have investigated thefunctional health benefits of low-intensity activity (Gossard et al., 1986;Harkom, Lampman, Banwell, & Castor, 1985). With some exceptions (e.g.,Morey et al., 1991; Thompson, Crist, Marsh, & Rosenthal, 1988), little isknown about what beneficial outcomes might be obtained from a sustainedprogram of low-intensity physical activity for older people. The effects oflow-intensity exercise carried out over a longer period of time may be at leastas effective as high-intensity exercise of a shorter duration. In one study, forexample, differences between the effects of high- and low-intensity exerciseon aerobic capacity and blood lipids did not begin to disappear until 18 weeksof training (Gaesser & Rich, 1984). In other words, the low-intensity programmerely took longer to have essentially the same effects.

The impact of exercise on older persons has been measured largely interms of changes in heart rate or blood pressure or of overall health benefitssuch as weight reduction. Fewer studies have examined the specific effectsof structured physical activity on musculoskeletal movement, joint flexibil-ity, and balance and gait, which are essential for functional mobility andperformance of the basic activities of daily living (ADL). Participation inphysical activities designed to address these factors needs to be evaluated inmore functionally impaired older persons than have been studied previouslyto establish the potential efficacy of exercise interventions for older peopleat high risk of chronic disability.

Determining the functional health outcomes of low-intensity exercise onthe performance of basic ADL in a frail older population, however, representsa considerable challenge. In addition to the health and safety risks of involv-ing frail individuals in a physical activity intervention, initiating and main-taining an exercise program is especially difficult for physically inactivepersons who are limited by chronic impairments and who are typically lessmotivated to participate in any type of structured exercise program. Thepresent study was designed to address these challenges.

Method

Sample

Participants for this study were drawn from the client registry of anonprofit human services agency serving older people in senior centersaffiliated with public housing facilities in a large urban area. Through the use

360

of newsletter announcements, informational posters, and individual contactsby staff and peers, individuals from three senior centers were recruited toparticipate in the exercise program. Individuals were excluded on the basisof recent major surgery, myocardial infarction, or stroke. Persons with seriousmedical problems requiring major daily management or monitoring were alsoexcluded. The 90 participants who began the program were mostly female(94%), White (52%) or Black (48%), and living alone (61 %), with an averageage of 72.6 years. The dropout rate was consistent with similar exerciseprograms. At the end of 18 weeks, the sample size (N= 32) was slightly belowwhat is required for adequate power estimates (N= 34). This will be addressedin a follow-up study.

Most participants described their own health as only fair (44%) or poor(36%), and more than 66% expressed concern about their health and func-tioning. Participants reported an average of three chronic conditions thatcaused some impairment. All participants reported arthritis, hypertension,and related heart disease. Most of these physically inactive older people wereoverweight (average Body Mass Index > 28.6) and had not previouslyparticipated in an exercise program.

Measures

All assessments were conducted in random order at the senior center siteof the exercise intervention within 2 weeks of the beginning and end of theprogram, taking no more than 30 minutes for each participant. Functionalmobility was assessed in two ways. First, participants rated their abilities toperform basic ADL in five areas. Using the Functional Status Index (FSI),mobility (five items), personal care (four items), home chores (three items),hand activities (three items), and social roles (five items) were self-assessed(yes/no) according to whether they caused pain or difficulty or requiredassistance in carrying them out. Total scores for the five functional areas werederived from the number of yes responses for pain, difficulty, and assistanceon each item in that domain. Thus the possible scores ranged from 0 to 15for mobility, from 0 to 12 for personal care, from 0 to 9 for home chores,from 0 to 9 for hand activities, and from 0 to 15 for social roles. The FSI isa self-assessment of functioning that has demonstrated substantial constructvalidity and high reliability in studies of older persons with osteoarthritis andrelated functional mobility problems (Jette, 1980). Levels of internal consis-tency range from .66 to .91, and test-retest and interobserver reliabilities are.70 or better.A performance assessment of walk time and number of steps taken over

a measured course was also conducted as an additional measure of flexibility

361

and balance in mobility. Participants were asked to walk a 20-meter course,which entailed walking 10 meters in one direction, turning around, andwalking back to the starting point. Each individual was timed, and the numberof steps taken was counted by one of two different raters. Interrater reliabilitywas .94.

Assessment of psychological well-being was suggested by other studiesof older samples that have found both physical and mental health outcomesfrom different types of exercise (Emery & Blumenthal, 1990; Kaplan, Atkins,& Remsch, 1984; King, Taylor, & Haskell, 1989). Based on some of ourearlier work, we selected three measures for this study to assess the impactof a low-intensity exercise intervention on different dimensions of psycho-logical well-being as a basis for further research in this area. Participants wereasked to respond to 12 dichotomous items (agree/disagree) on the LifeOrientation Test (LOT), which measured future outlook and optimism as abasis for predicting behavioral outcomes (Scheier & Carver, 1985). Potentialscores could range between 0 and 12. A five-item Attitude Toward Agingfactor (agree/disagree) contained in the Morale Scale (Lawton, 1975) wasused to assess the impact of aging on participants’ daily lives. Thus the rangeof possible scores was 0 to 5.

Finally, based on other studies that have suggested the importance ofmeasuring expectations about the efficacy of participating in an exerciseprogram (e.g., Kaplan et al., 1984), an exercise-based self-efficacy scale wasdeveloped specifically for this study to assess participants’ beliefs about theirown abilities to perform five of the SMILE exercises successfully (i.e., footcircles, forward bends, arm flexes, shoulder rolls, and marching in place).This measure consisted of five items, each scored on a 3-point Likert-typescale, where 1 = not at all sure, 2 = somewhat sure, and 3 = very sure. Totalscores could range between 5 and 15. For all three measures, higher scoresindicated a more positive outlook. Evidence of the validity of our self-efficacy measure was supported by a pattern of predictable validity coefficientswith other measures in this study. For example, higher levels of baselineself-efficacy were associated with lower levels of ADL functioning (r = -.22,p <.05, two-tailed test) but with higher levels of morale (r =. 18) and optimism(r = .10). These latter correlations were in the predicted direction but werelow and suggest that self-efficacy as measured in the present study is a distinctconstruct and measures something different from morale or optimism.

Participants were asked two open-ended questions about the impact of theprogram on overall health and on their experience of muscle and joint pain.They were also asked whether they planned to continue to participate in theSMILE program with peer leaders.

362

Procedure

The SMILE program consisted of repeated stretching, flexibility, andrange of motion exercises based on their potential to strengthen muscle andjoint activity, to improve postural balance, and to enhance the performanceof basic activities essential for maintaining independent functioning (e.g.,standing, sitting, walking, reaching, stooping, and hand/finger flexibility).Specifically, the SMILE program combined 25 movements into a 30-minuteset routine that began and ended with diaphragmatic breathing for relaxation.The initial sequence was done in a seated position and involved 7 gentlemovements for the neck, shoulders, and hands, followed by 10 upper bodyexercises done in a standing position. These included various reaching andstretching exercises designed to promote shoulder flexibility and armstrength as well as spine twists, side stretches, and pelvic rotations for trunkflexibility and walking balance. This sequence was followed by 7 lower bodyexercises done in a seated position to strengthen stomach and pelvic musclesand to increase flexibility and range of motion in the hips (e.g., leg lifts, pelvictilts, hip hikes/rotations, and leg swings). Two ankle exercises were alsoperformed at this point to develop strength and coordination and to improvecirculation in the feet. The next segment of the SMILE routine involved 5exercises in a standing position to develop flexibility and strength in the lowerbody for such activities as climbing stairs or rising out of a chair (e.g., armcurls, heel lifts, leg swings, and side/back leg stretches). The first of twocool-down routines was done seated and at a slower pace. This routineinvolved gentle stretching movements for the upper body including shoulderstretches, forward bends to increase spine flexibility, posture exercises, andarm and finger movements for range of motion in arm, wrist, and finger joints.The final routine involved gentle stretching and flexing movements of thelegs and knees, arms, and shoulders as well as full body stretching in astanding position. The number of repetitions of each movement was basedon the level of difficulty of the movement (see Hickey, Wolf, Robins, Wagner,& Harik, 1992, and Hickey, Luchowski, & Wolf, 1994, for a completedescription of the SMILE program).

The SMILE intervention was conducted twice weekly in four groups of20 to 25 participants. After the first 6 weeks, the two professional instructorsbegan to encourage the participants to take turns leading short segments ineach exercise session as an initial step toward empowering the group mem-bers to exercise on their own. Individualized training was then provided for8 peer leaders who were selected by the participants to conduct the SMILEprogram following the departure of the professional instructors. During the

363

final 6 weeks, peer leaders gradually assumed leadership for the SMILEprogram, with the professionals providing technical assistance and advice.Participants were assessed prior to beginning the SMILE program, followingthe 6-week demonstration, and at the end of 18 weeks.

Results

Of the initial 90 participants, 77 (86%) were still in the program andavailable for assessment following 6 weeks. The number of participantsdeclined to 32 (35%) at the end of 18 weeks. Average weekly attendance wasjust under 70% for all groups and remained fairly stable. For purposes ofanalysis, individuals were treated as participants on the basis of attending atleast two thirds of the sessions. Although this rate is somewhat arbitrary, itwas suggested by earlier work of Morey et al. (1989). The participant groupwas easily identified: Those who missed two or three consecutive sessionstended to drop out of the program entirely; those who remained were fairlyconsistent in their participation, with only an occasional absence. Health-related problems were the primary reasons given for dropping out or missingsessions, and there were no differences in the number of chronic illnessesbetween those who continued to exercise and those who dropped out. No onewas injured or experienced any adverse health effects as a direct result oftheir participation in the SMILE program. After 6 weeks, more than 85% ofthe participants reported feeling better than when they had started, and 80%reported a decrease in the daily experience of muscle and joint pain. Almostall of the participants (97%) indicated their intent to continue exercisingregularly.

Exercise Effects

As indicated in Table 1, following 6 weeks of low-intensity exercise, therewas an improvement in participants’ assessments of mobility (p < .00), whichincluded walking, climbing stairs, rising from a chair, reaching overhead, andstooping or crouching. Participants also reported increased flexibility in handmovements (p < .03) related to writing, opening ajar or bottle, and dialing atelephone. Mean scores were based on the number of items with whichparticipants reported difficulty, pain, and/or the need for assistance. Thus themaximum score was 15 for both mobility and social role activities (each ofwhich had five items), 12 for personal care (four items), and 9 for both homechores and hand activities (three items each). Higher mean scores reflected

364

Table 1. Self-Assessments of Functional Performance and Observations of Se-lected Gait Parameters Between Baseline and 6-Week Follow-Up (N = 77)

Table 2. Analysis of Variance Results of Psychological Assessments BetweenBaseline and 6-Week Follow-Up (N = 77)

more pain/difficulty/need for assistance, and lower scores indicated greaterindependence in these functions. Performance assessments of mobilityshowed significant improvements in actual walk time (p < .0 ) and decreases

in the number of steps in normal gait used to cover a measured course (p < .02).Table 2 summarizes the outcomes of psychological assessments of opti-

mism, morale, and self-efficacy. As indicated in higher mean scores, partici-pants improved significantly across all three dimensions after 6 weeks of theSMILE intervention.

Maintenance Effects

Self-assessments and performance observations were conducted after 18weeks to determine whether initial improvements would be sustained in apeer-led maintenance program. As indicated in Table 3, those who continuedin the SMILE program (N = 32) maintained their initial improvements in

365

13

c

c<9«

ijMRmc

im

Sd

0

Ea.~m/5

ddoM

’5mo

i0

aOo

Se

o§

E

0roa

’50-S<7i:. N0 CO

II &dquo;

U.0&

cE - 0 0(A 03’2

Cboo

w35

Mt4*

a9

-dm_~51!triqva

i#§oo.s::

û)£ld

:s::Q)

.s::l#

N

8c:U

CD:s::&dquo;0’E$iE’cC)’5i<0

..v

Q)3l#Q)

EQ)

Easm

IJ2wen9Õ5oJ2*2lm

.25l

’i:~5E5i:6~3mfQ)

iiiI-OZ

366

self-assessments and in performance measures of mobility. Their self-assessments of pain, difficulty, and/or need for assistance in carrying outvarious social role activities (i.e., getting into and out of a vehicle, going toand from the senior center, and moving about the community includingshopping and visiting friends) were also more positive. The improvementsin well-being found at the end of the first 6 weeks were maintained in their18-week scores only on the optimism scale (see Table 4). Given the attritionrate, these findings may not be representative of the larger group assessed atbaseline.

Discussion

The results of this pilot study are indicative of the potential efficacy oflow-intensity physical activity for improving mobility in frail and function-ally impaired older persons in at least two ways. First, this project demon-strated that exercise interventions can be designed to address the limitationsand interests of physically inactive older people with considerable chronicimpairment. Second, as a result of participation in an intervention that wasdesigned to enhance the physical movement skills needed by older people tomaintain their independent functioning, there was improvement in self-reported assessments of functional ability in carrying out activities of dailyliving and in mobility as gauged by two simple evaluations. These prelimi-nary findings regarding mobility are encouraging for the development ofcontrolled interventions to evaluate more specific functional health effects ofthis type of intervention on the performance of basic ADL by frail olderpersons.

The continuing participation in a peer-led program at the conclusion ofthis research by more than one third of the initial group suggests thatpreviously sedentary individuals with considerable frailty can be encouragedto increase and sustain their levels of physical activity. This finding, whichis based on a group of older people more impaired than those studied by eitherThompson et al. (1988) or Morey et al. (1989), suggests the importance oftargeting exercise interventions and maintenance programs specifically forelderly patients with significant physical and functional impairments.

These findings are necessarily preliminary and descriptive, in that partici-pants remaining at the end of 18 weeks may have been a more select group.Also, a greater number of participants are necessary to provide sufficientstatistical power to demonstrate improvement over time. In addition, healthmaintenance effects are more difficult to evaluate in the absence of controlled

367

ÑCO11

z0OL:::I

i.S!au..19e0

310

c40’0Cate.5

m(I)0mc

s1

s

JScmE

i.aTEC)

.Cl)20.t=C)

0Do0o2’3.CD

&OElig;eO

cas

9-O19:11,0ca

.4a:is9

g

%zI~risv

C~.atu0

:20d

U

6Cl)

C6m%&reg;

mi5

Z8EctheTo

s8g

S

SEmElie£

°M

m0

rZ.°15

Sc

m

5l§TBs§mmS

tii

Oz

368

observations of the impact of the different peer leaders on participation andprogram adherence.

The study was further limited by the single-group design in that the initialgroup of participants may have been more motivated to exercise. The twobrief performance measures of walking behavior may have also been con-founded by practice effects and motivational factors. Despite these limita-tions, the results of this pilot investigation are an encouraging indication ofthe types of functional benefits that frail older individuals might derive fromparticipating in limited physical activity programs.

Careful clinical trials using additional performance measures along theline suggested by Tinetti and Ginter (1988) are now required to validate theseresults in older persons with different levels and types of impairments.Further research also will be important to determine the relationship ofspecific physical activity interventions to a broad range of functional mobilityoutcomes and to the performance of basic ADL. There is no reason to expect,for example, that improvements will be uniform across individuals withdiffering impairments or in all dimensions of mobility: For example, greatergains may be possible with some older individuals in hand, arm, or upperbody flexibility than in ambulation, balance, or gait.

However limited the results, performance assessments of basic mobilityfunctions appear to be sensitive measures of the effects of structured physicalactivity on functional mobility. For example, two simple observationalmeasures seemed to capture small improvements in walking behavior, andrelated observations of other components of balance and gait were alsopositive, although not significant. Additional clinical assessments of range-of-motion flexibility were also conducted with a subgroup of exercise par-ticipants to test the feasibility and validity of such measures for futureresearch. These participants generally sustained or improved their range-of-motion flexibility in neck, trunk, shoulder, and ankle movements, which areessential for successful balance and gait. Careful assessments at more fre-quent intervals are necessary to indicate the impact of the physical activityintervention on other dimensions of mobility than the limited gait parametersmeasured here. An important next step will be to observe and quantify thedimensions of change in the performance of specific physical movementsthat contribute to overall functional mobility.

Although participation in the SMILE program appeared to have a positiveimpact on overall well-being, this should be considered a very preliminaryfinding. Further research is needed to delineate more precisely the interactionof psychological factors with both participation and health outcomes. At thesame time, the positive effects shown here should not be ignored. Reports of

369

reduction in muscle and joint pain, which were an early and consistent resultof the intervention, may have been an important factor in the mental healthbenefits observed. The sense of mastery participants derived from doing theexercises may have contributed to the overall positive psychological out-comes as well. Participants were consistently enthusiastic about the SMILEsessions, as demonstrated by these typical comments: &dquo;Coming here givesme so much energy&dquo; and &dquo;This program makes me feel good.&dquo;

Conclusion

This pilot study has provided useful information for the development ofpractice interventions and of further applied research by demonstrating thepotential health effects of low-intensity physical activity in a sample ofrelatively frail older persons. It is consistent with related studies that haveshown that the initial physical and functional benefits derived from anexercise intervention with frail older persons can be maintained (Morey et al.,1991). However, it goes beyond other investigations by identifying whattypes of physical activity may lead to improvements in mobility and flexibil-ity in physically inactive and chronically impaired older people. This studysuggests that exercise interventions that both appeal to physically inactiveolder people who are limited in functional mobility and demonstrate positivehealth outcomes in a chronically impaired older population can be designed.Although these findings are based on a preliminary descriptive study, theysuggest that low-intensity physical activity may have longer term effects onvarious dimensions of functional mobility and overall health status.

References

Branch, L. G., & Jette, A. M. (1982). A prospective study of long-term care institutionalizationamong the aged. American Journal of Public Health, 24(4), 485-510.

Emery, C. E, & Blumenthal, J. A. (1990). Perceived change among participants in an exerciseprogram for older adults. The Gerontologist, 30(2), 184-188.

Gaesser, G. A., & Rich, R. G. (1984). Effects of high and low-intensity exercise training onaerobic capacity and blood lipids. Medicine and Science for Sports and Exercise, 16,269-274.

Gossard, D., Haskell, W. L., Taylor, C. B., Mueller, J. K., Rogers, F., Chandler, M., Ahn, D. K.,Miller, N. H., & DeBusk, R. F. (1986). Effects of low and high-intensity home-based exercisetraining on functional capacity in healthy middle-aged men. American Journal of Cardiology,57, 446-449.

370

Harkom, T. M., Lampman, R. M., Banwell, B. F., & Castor, C. W. (1985). Therapeutic value ofgraded aerobic exercise training in rheumatoid arthritis. Arthritis and Rheumatology, 28(1),32-38.

Hickey, T., Luchowski, A. T., & Wolf, F. M. (1994). Low-risk exercise for high-risk older women.The Female Patient, 19, 47-54.

Hickey, T., Wolf, F. M., Robins, L. S., Wagner, M. B., & Harik, W. (1992). So much improvementwith a little exercise [videotape with manuals]. Ann Arbor: University of Michigan Schoolof Public Health.

Holloszy, J. O. (1983). Exercise, health and aging: A need for more information. Medicine andScience for Sports and Exercise, 15, 1.

Howard, M., Funnell, M., Robins, L., Johnson, T., Lampman, R. M., & Wolf, F. (1985).Development and evaluation of a low-level exercise program for diabetics. Diabetes,34(Suppl. 1), 133A.

Jette, A. M. (1980). Functional Status Index: Reliability of a chronic disease evaluation instrument.Archives of Physical and Medical Rehabilitation, 61, 395-401.

Jette, A. M., & Branch, L. G. (1981). The Framingham disability study: Physical disability amongthe aging. American Journal of Public Health, 71(11), 1211-1216.

Kaplan, R. M., Atkins, C. J., & Remsch, S. (1984). Specific efficacy expectations mediateexercise compliance in patients with COPD. Health Psychology, 3, 223-242.

King, A. C., Taylor, C. B., & Haskell, W. L. (1989). Influence of regular exercise on psychologicalhealth: A randomized, controlled trial of healthy middle-aged adults. Health Psychology, 8,305-324.

Lampman, R. M., & Savage, P. J. (1988). Exercise and aging: A review of benefits and a planfor action. In J. R. Sowers & J. V. Felicetta (Eds.), The endocrinology of aging (pp. 307-335).New York: Raven.

Lawton, M. P. (1975). The Philadelphia Geriatric Center Morale Scale: A revision. Journal ofGerontology, 30, 85-89.

Morey, M. C., Cowper, P. A., Feussner, J. R., DiPasquale, R. C., Crowley, G. M., Kitzman,D. W., & Sullivan, R. J., Jr. (1989). Evaluation of a supervised exercise program in a geriatricpopulation. Journal of the American Geriatrics Society, 37, 348-354.

Morey, M. C., Cowper, P. A., Feussner, J. R., DiPasquale, R. C., Crowley, G. M., & Sullivan,R. J., Jr. (1991). Two-year trends in physical performance following supervised exerciseamong community-dwelling older veterans. Journal of the American Geriatrics Society, 39,549-554.

National Center for Health Statistics. (1987). Aging in the 80’s: Functional limitations ofindividuals age 65 and over. Washington, DC: U.S. Department of Health and HumanServices.

Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment and implica-tions of generalized outcome expectancies. Health Physiology, 4(3), 219-247.

Thompson, R. F., Crist, D. M., Marsh, M., & Rosenthal, M. (1988). Effects of physical exercisefor elderly patients with physical impairments. Journal of the American Geriatncs Society,36, 130-135.

Tinetti, M., & Ginter, S. F. (1988). Identifying mobility dysfunctions in elderly patients. Journalof the American Medical Association, 259, 1190-1193.

Vallbona, C., & Baker, S. B. (1984). Physical fitness prospects in the elderly. Archives of Physicaland Medical Rehabilitation, 65, 194-200.

Weissert, W. G., & Cready, C. M. (1989). Toward a model for improved targeting of aged at riskof institutionalization. Health Services Research, 24(4), 485-510.

371

Tom Hickey is a professor in the Department of Health Behavior and Health Educationat the University of Michigan School of Public Health and a faculty associate at theInstitute of Gerontology in Ann Arbor, MI. His current research is directed at improvingmobility in chronically impaired older adults. He is a coauthor of &dquo;The Attribution ofHealth Problems to Aging and Mortality Among Older Adults m a Longitudinal Study&dquo;

&dquo;

(American Journal of Public Health, 1992) and of &dquo;Adaptation of a General OptimismScale for Use With Older Women&dquo; (Psychological Reports, 1994).

Fredric M. Wolf is a professor of postgraduate medicine and health professions educationand the director of the Learning Resource Center and Laboratory for Computmg,Cognition and Clinical Skills at the University of Michigan Medical School. His researchinterests include the evaluation, dissemination, and adoption of new technology, decisionmaking and judgment, and the social-psychological aspects of chronic illness. He is theauthor of Meta-analysis: Quantitative Methods for Research Synthesis (Sage, 1986).

Lynne S. Robins is an assistant professor in the Department of Postgraduate Medicineand the director of curriculum evaluation at the University of Michigan Medical School,Her research interests include physician-patient interactions and the influence of genderand ethnicity on health care delivery and on medical education. She is a coauthor of&dquo;The Effect of Training on Medical Students’ Responses to Geriatric Patient Concerns:A Linguistic Analysis&dquo; (The Gerontologist,1989).

Marilyn B. Wagner is a physical therapist in the Department of Health Sciences atCleveland State University. Her teaching and research are focused on movement scienceand exercise with adult populations at special risk.

Wafa Harik is a nurse who was affiliated with the School of Nursing at Case WesternReserve University at the time this research was conducted. She directs exercise mter-ventions for functionally impaired older adults.