The manipulated content of each message was designed to be reflective of the key goal classifications outlined by socioemotional selectivity theory.10 The emotion and meaning message captured the emotionally meaningful outcomes that home exercise can afford older adults, such as spending time with loved ones and being there for family. The title of this message was also manipulated, stating “Let exercise take you there!” and contained the key phrases “live,” “laugh,” “love.” The images in this message also highlighted the emotion- and meaning-based outcomes of home exercise, picturing groups of older adults socializing, and sharing positive emotion. In accordance with socioemotional selectivity, this emotion and meaning message focused on ways in which adherence to home exercise can more immediately benefit a person socially and relationship-wise.
The facts and information message was reflective of the second class of goals outlined by socioemotional selectivity theory, knowledge-related goals. The benefits of exercise in this message highlighted the physiological outcomes that home exercise can afford older adults, such as stronger muscles and increased balance, which take time and persistence to realize. The title of this message, “Let exercise take your body there!”, as well as the key words “strength,” “balance,” “flexibility,” emphasized more of the facts about what exercise can do for your body, thus representing a more traditional informational appeal and the standard of care in communicating the importance of exercise adherence to older adults.
Baseline Intervention and Assessment
The baseline intervention and assessment took place any time between the last 2 weeks of the participants' therapy program and the day the participant was discharged from therapy. Following the randomization schedule, participants were assigned to either the facts and information message group or the emotion and meaning message group. They were given the flyer message that corresponded to their assigned message group and allowed unlimited time to read the flyer in a private room. Participants did not receive any special instructions regarding the flyer, except that it may be relevant to their current situation. Next, participants completed the baseline survey packet that included questions about their intentions to adhere to their prescribed home exercise program and basic demographic information. Upon completion of this session, participants were given a small honorarium, the flyer they read, and a smaller version of the flyer in the form of a refrigerator magnet in appreciation for their time. Participants were instructed to hang the magnet in a conspicuous location in their home.
The follow-up measures were completed by participants 2 weeks postdischarge from therapy. The follow-up session involved participants completing a paper survey at home and mailing it back to the principal investigator. When the follow-up survey was received by the principal investigator, the participants were mailed a small honorarium in appreciation for their time. The follow-up questionnaire measured participants' self-reported adherence to their home exercise program and self-reported exercise behavior outside of their home exercise program, as well as asking participants to report general information about their current health status. Participants were also asked about their use of the magnet they received (ie, Did they hang the magnet up? Did they find it motivating?). The questionnaire contained a total of 20 items that were created using Forkan and colleagues'9 home exercise program survey as a guide.
The main section asked specifically about participants' adherence to their home exercise program. Participants were first asked to indicate if they received a home exercise program, and if so, to answer a number of questions regarding that program. Participants were asked to list the names (or descriptions) of the exercises they did each time they performed their home exercise program. This list was cross-checked with the home exercise program on file for the participant and an “adherence score” was calculated (detailed in the “Statistical Analysis” section). Participants were also asked to indicate the number of times per week they performed the exercises they listed and what percentage of their fully prescribed home exercise program they felt they completed by performing these exercises.
All data were analyzed using SPSS version 17.0.1. Before testing the study hypotheses, descriptive statistics were conducted for all baseline and follow-up measures and tests of major statistical assumptions (ie, normality, homogeneity of variances) were performed. Appropriate statistical steps were taken when any assumptions were not met, such as the use of the nonparametric Mann-Whitney U Test in place of the independent-samples t test when the assumption of normality of the dependent variable was not met.15 In addition, preliminary analyses were conducted to determine if any of the sociodemographic variables (listed in Table 1) would be included as covariates in the analyses (bivariate correlations for continuous data; t tests and 1-way analyses of variances for categorical data). If an outcome variable was found to be significantly correlated with a single covariate or multiple potential covariates, an analysis of covariance was conducted to assess differences between the levels of the independent variable.
The different messages received by participants, either describing the benefits of home exercise as physiologically based (facts and information message) or emotion- and meaning-based (emotion and meaning message), represented the 2 manipulated levels of the independent variables. The key dependent outcome was participant adherence to their prescribed home exercise program. Participant adherence was calculated as an “adherence score” obtained by dividing the number of exercises listed by the participant in the follow-up survey that matched their prescribed home exercise program by the total number of exercises in the program (ie, a participant who listed 5 exercises of 10 prescribed received a score of 50%). The secondary dependent outcome, “extra exercise,” was calculated as the number of exercises listed by participants in the follow-up survey that were not in their prescribed home exercise program. The time at which participants completed the baseline session and received the study message, either before being discharged or on the day of discharge, was included in follow-up analyses as a potential moderating variable.
Most participants reported a high likelihood that they intended to be adherent to their home exercise program after discharge from therapy (median = 4.0 [48.3%]; min = 3.0 [3.3%]; max = 5.0 [48.3%]; range of scale 1 [extremely unlikely] to 5 [extremely likely]). All participants (100%) indicated having received a home exercise program. The majority of participants reported receiving 1 to 5 exercises (43.3%) or 6 to 10 exercises (46.7%) in their home exercise program and were asked to perform those exercises 7 days per week (51.7%). Regarding adherence to home exercise programs, participants averaged an adherence score of 57.5% (SD = 34.0%) with a range from 0% to 100%. Participants listed performing an average of 1.8 extra exercises outside of their home exercise program (SD = 2.2; range = 0-9). The majority of participants reported performing their home exercises 3 to 4 times per week (48.3%) and that it took them between 15 and 30 minutes each session (60.0%). Most felt that they completed at least 75% of their prescribed home exercise program each time they exercised (83.3%). Sixty-five percent of the participants reported hanging the study magnet in their home and half (53.8%) reported that it helped remind or motivate them to exercise.
To determine whether participants who received the emotion and meaning message were more adherent to their home exercise program than participants who received the facts and information message, the mean difference in the adherence score variable was analyzed between the groups. Based on preliminary analyses, gender, feelings of depression, and number of missed appointments were included as covariates because each was related to the adherence score variable. Thus, because of the inclusion of the 3 covariates, an analysis of covariance was conducted. Results indicated that, contrary to the hypothesis, participants who received the emotion and meaning message were not significantly more adherent to their home exercise program than participants who received the facts and information message (F1,55 = 1.4; p = .24; ηp2 = 0.03). Participants reading the emotion and meaning message had an average adherence score of 60.0% (SD = 34.4%), while participants reading the facts and information message had an average adherence score of 55.3% (SD = 34.0%). Adherence did not significantly differ between classes of diagnosis (p = .70), nor was diagnosis a significant moderator of the effect of the message on adherence (p = .17).
A 2×2 analysis of covariance revealed a significant interaction between message condition and time at which participants completed the baseline session and received the study message (F1,53 = 4.6; p = .04; ηp2 = 0.08) (Figure 3). Post hoc analyses separately examined the rates of adherence in participants who received the intervention message prior to discharge and participants who received the intervention message on the day of discharge. Participants who received the emotion and meaning message before being discharged and therefore continued supervised therapy after receiving the study message, reported somewhat higher rates of adherence to home exercise (M = 63.6%; SD = 34.3% than participants who received the facts and information message (M = 50.8%; SD = 39.0%) prior to discharge (F1,34 = 3.4; p = .07; ηp2 = 0.09). However, adherence scores were similar for participants who received either the emotion and meaning message (M = 60.5%; SD = 27.7%) or the facts and information message (M = 46.8%; SD = 34.7%) on the day of discharge (F1,16 = 1.4; p = .25; ηp2 = 0.08), meaning that the receipt of the study message coincided with the discontinuation of supervised therapy.
To determine whether participants who received the emotion and meaning message performed a greater number of extra exercises outside of their home exercise programs than participants who received the facts and information message, the mean difference in the extra exercises variable was analyzed between the groups using a Mann-Whitney U Test. The results were in the expected direction and approaching statistical significance [U(58) = 327.5; z = −1.9; p = .06; r = 0.24]. Participants reading the facts and information message reported an average of 1.3 extra exercises (SD = 1.8), while participants reading the emotion and meaning message reported an average of 2.4 extra exercises (SD = 2.5). This difference was most pronounced in the group of participants with a diagnosis of the low back [U(19) = 8.5; z = −3.3; p = .001; r = 0.72], where participants receiving the emotion and meaning message listed an average of 3.6 (SD = 2.2) extra exercises and participants receiving the facts and information message listed an average of 0.83 (SD = 1.0) extra exercises. There was no moderating effect of diagnosis (p = .08), nor was there was a moderating effect of the time at which the baseline session was completed (p = .12).
No significant difference was found in magnet usage between the message conditions (χ2(1, N = 60) = 4.2; p = .52). In comparing those who did and did not report hanging up the magnet, no significant difference was also reported on any of the dependent variables (all p ‘s > .255). Within the subgroup that hung the magnets, however, there was a significant difference (t37 = 2.35; p = .024) in adherence scores between those who did (M = 69.0%; SD = 24.9%) and did not (M = 46.5%; SD = 34.7%) report that the magnet helped to remind or motivate them to exercise.
The primary goal of this study was to examine the impact of a theoretically grounded message-based intervention on adherence to home exercise in older adults. Importantly, this study was the first to empirically test an adherence-enhancing intervention targeting post–physical therapy home exercise programs. Contrary to the study hypotheses, there was not a significant difference in participants' adherence to their home exercise program between the message groups. While the adherence score of those participants who received the emotion and meaning message was higher than that of those who received the facts and information message, the difference was not statistically significant and therefore definitive claims about the impact of the emotion and meaning message cannot be made for adherence specifically. There was, however, a much more reliable difference between the message groups in the number of extra exercises reported by participants that were not part of their prescribed home exercise program. When asked to list the specific exercises they did each time they completed their exercise program, participants who received the emotion and meaning message listed more exercises outside of their prescribed program than participants who received the facts and information message. Thus, while all participants may have been completing similar proportions of their prescribed home exercise programs, those who received the emotion and meaning message were somewhat more active in exercises that complemented their home exercise program exercises.
It is possible that the older adults who performed more extra exercises developed an adaptive strategy spurred on by an increased desire and motivation to be active as a result of viewing the emotion and meaning message. If participants in the emotion and meaning group were, in fact, more motivated to maintain the gains they made in physical therapy, it is likely that they adapted their prescribed home exercise program to best suit and accommodate their specific needs instead of giving up on exercising entirely when faced with challenges. The negative of a scenario such as this is that these extra exercises may not serve the same purpose as the prescribed exercises and thus may not aid in the maintenance of the progress made in physical therapy. In this case, an increased number of extra exercises may not actually be beneficial. Future research should examine to what extent older adults may utilize such a strategy in the context of home exercise adherence and the health costs and benefits associated with the utilization of this adaptive approach.
The time point at which participants completed the baseline session also appeared to play an important role in the present findings. Indeed, participants receiving the emotion and meaning message were more adherent to their home exercise program than participants receiving the facts and information message only when they received the message with time remaining in their therapy program (ie, while still in a supervised therapy program). Participants who received the study message on the last day of therapy (ie, the day of discharge) were equally adherent between message conditions. While these analyses and conclusions are certainly restricted by sample size issues, they do bring up some important questions for future intervention research to consider. For instance, at what point in therapy are older adults most receptive to information about home exercise? Is there a point in therapy when older adults are more receptive to physiological health information than emotion- and meaning-based information? Does the content of the message (physiological health vs emotion and meaning) need to be consistent throughout therapy? The answers to these questions would aid in the implementation of socioemotional selectivity theory–based health promotion practices into the geriatric health care setting.
A significant difference was not found between the message groups in reports of participants' perception of the percentage of their full home exercise program they completed each time they exercised—almost all of the participants felt that they performed 75% or more of their full home exercise program. In addition, the extra exercises variable used in the present analyses was not derived directly from participant self-report, but rather computed as the number of exercises participants listed that were not documented in their medical chart by their physical therapist. Thus, it is possible that participants had a difficult time distinguishing between the exercises that were listed in their home exercise program specifically and the exercises they performed in the clinic that they continued at home without the direction of the physical therapist.
Despite the fact that limited support was found for the specific study hypotheses related to home exercise adherence, this intervention was successful when considering the outcome in a larger scope. Across the entire sample, even with the conservative measure of adherence to home exercise (the match between participants' list of exercises and therapists' documented list of exercises), the average adherence score was nearly 60%. Considering the limited body of research that estimates a 44% rate of adherence to prescribed home exercise programs,3 and the fact that the greatest drop in home exercise adherence is the day patients are discharged from therapy,2 the 2-week postdischarge adherence rate found in this study is promising. Certainly, the results here speak to the value of future interventions to promote adherence to home exercise, even if this study failed to provide robust support for one type of message over another.
As the first study of its kind to propose and examine the effects of an intervention to enhance adherence to home exercise in older adults, the findings of this study have important implications for practice. It has been long recognized that adherence to home exercise after physical therapy is low,3 but until now a viable attempt has not been taken to establish a framework for addressing these low levels of adherence. Likewise, although theoretical variables have been examined as predictors of exercise in older adults,16,17 the application of theory into tangible practice in this area has thus far been sparse. The theoretical tenets of socioemotional selectivity theory are directly relevant to the older adult population, and the findings of this study reveal some new avenues for the integration of socioemotional selectivity theory into real-world applications.
For instance, the Academy of Geriatric Physical Therapy makes available for download a large section of patient exercise brochures.18 Physical therapists are encouraged to use these brochures in their clinics as patient education material. An examination of a sampling of these brochures reveals somewhat of a trend—all the brochures seem to resemble the style of the facts and information message used in this study. Each brochure is similar in that a section describes the benefits of exercise as it relates to physiological health. These physiological health benefits include increased muscle and bone strength and improved balance and flexibility, among others. Indeed, these common health benefits mirror the “strength, balance, flexibility” theme of the facts and information message. The findings from this study, despite being preliminary, can be used to better inform the design of educational materials, such as the exercise brochures from the American Physical Therapy Association highlighted here, that therapists may provide to their patients.
While it should not be forgotten that the materials therapists provide to their patients are meant to be educational in tone, the principle behind socioemotional selectivity theory is that older adults may respond better to education that has an emotional pitch. Instead of simply describing the many health benefits of exercise to older adults, as with the facts and information approach, it may be better to couple each health benefit with an associated outcome that provides emotion or meaning. For example, as opposed to just saying, “Exercise improves your balance,” one might say, “Exercise improves your balance, providing you with the mobility you need to stay active with your grandchildren.” If they do not already, physical therapists may want to incorporate this style of communication into their general interactions with older adult patients. Taking the time to understand the unique motivations that older adults may have for wanting to be healthy and integrating those emotionally salient goals into the patients' plan of care may help improve adherence rates in a “bottom-up” type manner. In other words, if patients are motivated to be in therapy for the right reasons from the beginning, attendance rates in the clinic might improve, spurring increased adherence to home exercise while in therapy, and potentially spilling over into adherence to home exercise once therapy has been completed.
Given that declines in cognitive function, such as memory, are well-documented as being a normal part of the aging process,19 it is likely that the single message exposure most participants received was not enough to power any significant observed differences in behavior. Indeed, significant differences in adherence were found when analyses were performed within the subgroup of participants who reported using the study magnets as suggested. Older adults who hung the magnet and thought it helped remind or motivate them to exercise had significantly higher rates of adhere to their prescribed home exercise program, averaging a 69.0% rate of adherence, than those who hung the magnet but did not think that it was helpful in motivating or reminding them to exercise, averaging only a 46.5% rate of adherence. Furthermore, participants who hung their magnets and found them helpful had a significantly higher rate of adherence than all other participants combined (who averaged only a 51.4% rate of adherence). What this finding suggests is that a message-based intervention can be successful in improving adherence to home exercise in older adults, but more work needs to be done to establish a framework that ensures that the messages are being used as intended and reaching participants appropriately.
The findings surrounding magnet usage in this study highlight an important point to consider for future interventions in this area and more broadly—that is, messages need to “reach” individuals in 2 distinct ways to be effective. As discussed earlier, the first “reach” is in a physical manner (ie, the message needs to be consistently accessible, such as on a magnet hung on a refrigerator). The second “reach” is in more of a psychological manner. Half of the older adults who reported hanging their magnet did not find it to be motivating and subsequently were less adherent to their home exercise program. In fact, post hoc analyses revealed that adherence rates of those participants who did not hang their magnets (55.5%) were actually higher than adherence rates of those who hung their magnets but did not find them to be helpful in motivating or reminding them to exercise (46.5%). Thus, while the physical reach may have been successful, the psychological reach was not and adherence seemed to suffer as a result.
Future studies would also benefit from the inclusion of additional measures of home exercise adherence (ie, exercise diaries), and in particular one that could provide a more objective measure of adherence rather than relying exclusively on participant self-report. Most participants in this study thought that they were being reasonably adherent to their prescribed home exercise program—the vast majority reporting that they completed 75% or more of their fully prescribed program—whereas their calculated adherence scores were far lower. Clearly, some disconnect was present between what participants thought they were supposed to be doing and what therapists prescribed for them to do. Future studies might consider contacting participants immediately after they have been discharged from therapy to have them provide an account of their home exercise program that can be verified with their therapist.
The American Physical Therapy Association endorses a program of physical therapy that aims to “restore, maintain, and promote not only optimal physical function but optimal wellness and fitness and optimal quality of life.”20(p21) The restoration and maintenance of optimal quality of life, in particular, is of paramount importance to the older adult population as declines in movement and physical functioning are natural components of the aging process. When performed as prescribed, older adults do benefit significantly from participation in supervised exercise programs, such as those typical of physical therapy.21 As adherence to the home exercise portion of physical therapy is considerably low in the older adult population,3 this study may reveal only the “tip of the iceberg” and be a small indication of the larger possibilities that this research can add to our understanding of how to best motivate older adults to live healthy lives and make healthy choices.
1. Collins J, Beissner KL, Krout JA. Home health physical therapy: practice patterns in Western New York. Phys Ther. 1998;78(2):170–179.
2. Sluijis EM, Knibbe JJ. Patient compliance with exercise: different theoretical approaches to short-term and long-term compliances. Patient Educ Couns. 1991;11(1):191–204.
3. Hardage J, Peel C, Morris D, et al. Adherence
to exercise scale for older patients (AESOP): a measure for predicting exercise adherence
in older adults
after discharge from home health physical therapy. J Geriatr Phys Ther. 2007;30(2):69–78.
4. Wolf B, Feys H, De Weerdt, et al. Effect of a physical therapeutic intervention for balance problems in the elderly: a single-blind, randomized, controlled multicentre trial. Clin Rehabil. 2001;15(6):624–636.
5. Campbell AJ, Robertson MC, Gardner MM, Norton RN, Buchner DM. Falls prevention over 2 years: a randomized controlled trial in women 80 years and older. Age Ageing. 1999;28(6):513–518.
6. Robertson MC, Devlin N, Gardner MM, Campbell AJ. Effectiveness and economic evaluation of a nurse delivered home exercise
programme to prevent falls. 1: Randomised controlled trial. BMJ. 2001;322(7288):697–701.
7. Hauer K, Specht N, Schuler M, Bartsch P, Oster P. Intensive physical training in geriatric patients after severe falls and hip surgery. Age Ageing. 2002;31(1):49–57.
8. Timonen L, Rantanen T, Ryynanen OP, Taimela S, Timonen TE, Sulkava R. A randomized controlled trial of rehabilitation after hospitalization in frail older women: effects on strength, balance and mobility. Scand J Med Sci Spor. 2002;12(3):186–192.
9. Forkan R, Pumper B, Smyth N, Wirkkala H, Ciol MA, Shunway-Cook A. Exercise adherence
following physical therapy intervention in older adults
with impaired balance. Phys Ther. 2006;86(3):401–410.
10. Carstensen LL, Isaacowitz DM, Charles ST. Taking time seriously: a theory of socioemotional selectivity. Am Psychol. 1999;54(3):165–181.
11. Robberson MR, Roger RW. Beyond fear appeals: negative and positive persuasive appeals to health and self-esteem. J Appl Soc Psychol. 1988;18(3):277–287.
12. Gallagher KM, Updegraff JA. Health message framing effects on attitudes, intentions, and behavior: a meta-analytic review. Ann Behav Med. 2012;43(1):101–116.
13. Conlon M, Anderson GC. Three methods of random assignment: comparison of balance achieved on potentially confounding variables. Nurs Res. 1990;39(6):376–379.
14. American Physical Therapy Association. Physical fitness for special populations. http://www.apta.org/pfsp/
. Accessed June 5, 2014.
15. Fay MP, Proschan MA. Wilcoxon-Mann-Whitney or t
-test? On assumptions for hypothesis tests and multiple interpretations of decision rules. Stat Surv. 2010;4(1):1–39.
16. King AC, Stokolos D, Talen E, Brassington GS, Killingsworth R. Theoretical approaches to the promotion of physical activity: forging a transdisciplinary paradigm. Am J Prev Med. 2002;23(2):15–23.
17. Resnick B, Jenkins L. Testing the reliability and validity of the self efficacy for exercise scale. Nurs Res. 2000;49(3):154–159.
19. Stuck AE, Walthert JM, Nikolaus T, Bula CJ, Hohmann C, Beck JC. Risk factors for functional status decline in community-living elderly people: a systematic literature review. Soc Sci Med. 1999;48(4):445–469.
20. American Physical Therapy Association. Guide to Physical Therapy Practice. 2nd ed. Alexandria, VA: American Physical Therapy Association; 2001.
21. Chang JT, Morton SC, Rubenstein LZ, et al. Interventions for the prevention of falls in older adults
: systematic review and meta-analysis of randomized control trials. BMJ. 2004;328(7441):680–687.
Keywords:Copyright © 2016 the Section on Geriatrics of the American Physical Therapy Association
adherence; home exercise; motivation; older adults; persuasive communications