Abstract
Occupational therapy is a client-centered health profession concerned with promoting health and well-being through occupation (World Federation of Occupational Therapists, 2010). Its primary goal is to enable people to participate in the activities of everyday life. Underlying occupational therapy theory and practice is the reported relationship between occupation and health and well-being (Townsend, 1997; Wilcock, 1993). A deeper understanding of how occupations pertain to health and well-being can enable the provision of more efficient occupational therapy services.
Client-centered practice, which embraces a philosophy of respect for, and partnership with, the people receiving services, is increasingly advocated to support the realization of meaningful occupation for clients (Falardeau & Durand, 2002; Law, Baptiste, & Mills, 1995; Palmadottir, 2003; Sumsion & Smyth, 2000; Taylor, 2003). Its success depends on the desire and ability of clients to take part in the decision-making process and the ability of occupational therapy practitioners to include clients in this process. However, inclusion of clients in decision making tends to depend on practitioners’ experience and values; Maitra and Erway (2006) reported a gap between occupational therapists and their clients in their perceptions of client-centered practice. Moreover, when clients do not appropriately convey a demand for a given occupation, the practitioner may not properly recognize clients’ subjective evaluation related to the occupation. Therefore, for practitioners to realize the occupation the client desires, it is necessary to facilitate sharing of the meaning of the occupation between practitioner and client. In other words, we think that practitioners are required to capture a subjective evaluation of their clients’ occupations.
One concept that captures the psychological state for an activity is flow, defined as “the state in which people are so involved in an activity that nothing else seems to matter; the experience itself is so enjoyable that people will do it even at great cost, for the sheer sake of doing it” (Csikszentmihalyi, 1990, p. 4). Flow can contribute to increased levels of happiness, self-esteem, work productivity, and joy in life (Csikszentmihalyi & Csikszentmihalyi, 1992; Haworth & Evans, 1995; Nielsen & Cleal, 2010). Studies have described the relationship between the experience of flow and occupation in everyday life (Larson & von Eye, 2010; Rebeiro & Polgar, 1999; Wright, Sadlo, & Stew, 2007).
Jackson and Csikszentmihalyi (1999) stated that the most important characteristic of flow is the challenge–skill balance: Flow is expected to occur when the perceived level of challenge provided by the activity and the person’s perceived level of skill are balanced. Activities in which the person’s skill is perceived to be high relative to the challenge provided would lead to boredom. Likewise, low-perceived skill and high-perceived challenge would produce anxiety, whereas low-perceived skill and low-perceived challenge would result in apathy. For example, Hirao, Kobayashi, Okishima, and Tomokuni (2012) reported that the physical health of older adults was significantly better in groups who experienced flow during important everyday activities than in a group who experienced apathy.
Therefore, we think that careful attention to the challenge–skill balance in client occupations, applying the concept of flow, can be useful in occupational therapy intervention. However, few studies have addressed challenge–skill balance for clients in clinical occupational therapy. The purpose of this study was to investigate whether occupational therapy that includes adjusting the challenge–skill balance significantly improves health-related quality of life (HRQOL) for older adults in comparison with standard occupational therapy.
Method
Participants
Participants were recruited between December 2013 and March 2014 at an adult day program in Japan. Inclusion criteria were age >60 yr and participation in occupational therapy at the center for >3 mo. Exclusion criteria were dementia and visual impairment. In reference to the results of a randomized controlled trial (RCT) in the field of occupational therapy, in which the effect size for QOL was 1.30 (Graff et al., 2007), we conducted a powerx analysis using G*Power (Version 3.1.7; Erdfelder, Faul, & Buchner, 1996) and determined that a total sample size of 54 clients (27 in each of two groups) would provide 80% power for detecting a difference with an effect size of 0.80 in HRQOL scores using a two-tailed test and an α level of .05. To compensate for possible attrition, we decided to enroll 63 clients. All provided written informed consent before participation.
Procedure
The study was designed as a single-blind RCT and reported in accordance with the CONSORT guidelines for reporting clinical trials (Schulz, Altman, & Moher, 2010). Clients were randomly assigned by blocked randomization (block size = 4) either to the experimental group, who received occupational therapy that included assessment and adjustment of the challenge–skill balance of the activities, or to the control group, who received occupational therapy as typically conducted at the adult day program.
Because the factors affecting the outcomes were homogeneous between the experimental and control groups, randomization was stratified by sex and EuroQol visual analog scale (EQ–VAS) scores for the self-assessment of general health in the EuroQol–5 Dimension scale (EQ–5D; EuroQol Group, 1990; high–low boundary = 50), resulting in four layers: (1) male and high EQ–VAS scores, (2) male and low EQ–VAS scores, (3) female and high EQ–VAS scores, and (4) female and low EQ–VAS scores. Order of blocks was randomly assigned by R software (Version 3.2.1; R Foundation, Vienna).
The clients were blinded to group allocation, but the therapists were aware of the treatment assigned. Occupational therapy for both groups consisted of 10 sessions of 20 min each held over about 10 wk (once a week); after this, the outcomes were reevaluated, and clients were asked whether they knew which group they were allocated to. Criteria for withdrawal from the study were admission to hospital, absenteeism, or death. The study protocol was approved by the ethics committee of Kibi International University, Okayama, Japan.
Experimental Group.
The intervention was implemented by occupational therapists with >5 yr experience in geriatric occupational therapy; the therapists had previously performed a trial study with several clients. The intervention program focused on the occupational performance of activities and was conducted individually. In both the intervention and control groups, the intervention was consistent with the Occupational Therapy Practice Framework: Domain and Process (3rd ed.; American Occupational Therapy Association [AOTA], 2014). The difference between the groups was in whether the evaluation and intervention were conducted on the basis of the challenge–skill balance.
In the first session of occupational therapy, the therapist assessed the client’s problems in daily living using the Canadian Occupational Performance Measure (COPM; Law et al., 1990). An activity that could be supported in the adult day program was selected from the problems identified.
In the second session, the client performed the selected activity and then evaluated the activity using the challenge and skill levels, which the therapist confirmed. Challenge level was defined as the “challenges of the activity” and was rated from 1 (very simple) to 7 (very difficult). Skill level was defined as “your skills in the activity” rated from 1 (not at all skillful) to 7 (very skillful; Csikszentmihalyi & Larson, 1984; Engeser & Rheinberg, 2008). At that time, the therapist explored the client’s reasons for his or her ratings of the challenge level and skill level.
On the basis of the client’s and therapist’s evaluation, the factors that made the occupational performance difficult (challenge components; e.g., environment, execution time, size of the place for the activity) and factors that improved the occupational performance (skill components) were determined. Adjustment of the challenge–skill balance in the activity was initiated on the basis of these components.
After the client performed the adjusted activity, the challenge and skill levels were reevaluated. If the challenge and skill levels were not balanced, the activity was readjusted in the next session. The intervention aimed at improving the skill level for the activity until the challenge and skill levels were balanced.
Control Group.
The first and second sessions were conducted in the same way as for the experimental group, except that the therapists did not address the client’s subjective perception of the challenge and skill levels for the activity. From the third session, the therapists simply assessed the client’s performance and conducted the therapy in the typical manner for the adult day program, following the general guidelines for occupational therapy practice.
Outcome Assessments
Clients were assessed before the intervention (pretest) and after the 10th session (posttest). The outcome measures, all self-reported by the clients, were HRQOL, flow experience, and self-perception of occupational performance.
HRQOL was assessed using the EQ–5D and the Japanese version of the eight-item Short-Form Health Survey (SF–8; Fukuhara & Suzukamo, 2004). The EQ–5D defines health in five dimensions: mobility, self-care, usual activities, pain and discomfort, and anxiety or depression. The EQ–5D includes the EQ–VAS, which clients use to self-assess their health on a scale from 0 (worst possible health) to 100 (best possible health). The SF–8 encompasses eight domains: General Health, Physical Functioning, Role Physical, Bodily Pain, Vitality, Social Functioning, Role Emotional, and Mental Health. The domains can be combined to yield a physical component summary score and a mental component summary score, whose reliability and validity have been confirmed (Fukuhara & Suzukamo, 2004).
Flow experience was assessed using the Flow State Scale for Occupational Tasks (Yoshida et al., 2013), developed for clinical situations, which consists of 14 items and three factors (scores range from 7 to 98). Its reliability and validity have been confirmed (Yoshida et al., 2013). The client’s self-perceived occupational performance was assessed using the COPM, which involves a semistructured interview format and structured scoring. Clients were rated on a scale from 1 to 10 for perceived performance capacity and for level of satisfaction with performance.
Information collected at pretest included the client’s age, sex, diagnosis, FIM® score (Granger, Hamilton, Keith, Zielezny, & Sherwin, 1986), and Frenchay Activities Index (FAI) score (Holbrook & Skilbeck, 1983). The FIM is an 18-item, 7-level scale that uniformly assesses the severity of the client’s disability and medical rehabilitation functional outcome (overall scores range from 7 to 126). The FAI is a 15-item self-report scale that provides an index of a client’s capacity to perform instrumental activities of daily living (overall scores range from 0 to 45).
Statistical Analysis
The primary outcome was the difference between pretest and posttest scores for HRQOL (EQ–5D and SF–8). The secondary outcomes were the differences between pretest and posttest scores on the Flow State Scale for Occupational Tasks and COPM. All data were analyzed on the basis of intention-to-treat analysis, applying the last-observation-carried-forward method for withdrawals. Using Graff et al.’s (2007) method, we evaluated pretest–posttest differences using analysis of covariance, with the pretest scores for the EQ–5D, the SF–8, and the Flow State Scale for Occupational Tasks as covariates. Moreover, 1,000 bootstrap replicates were performed.
The Mann−Whitney U test was used to evaluate changes in COPM scores (two-tailed, α = .05). We also reported the significance according to Benjamini and Hochberg’s (1995) method for adjusting the overall score controlling for the false discovery rate. The proportion of clients achieving a clinically relevant improvement, defined as an improvement of .05 points in the EQ–5D score (Robinson et al., 2013), and the numbers needed to treat with 95% confidence intervals (CIs) were calculated. Per protocol analysis was also carried out. Effect sizes (Cohen’s d) were calculated to assess the practical significance of the interventions relative to the control condition. Statistical analysis was performed with IBM SPSS Statistics (Version 22.0; IBM Corp., Armonk, NY).
Results
Sixty-three consecutive clients who received occupational therapy at the adult day program were evaluated for eligibility (see Figure 1). Of these, 5 were excluded and 2 did not consent to the study, leaving 56 clients for the analysis. Four clients left the study. In the experimental group, 1 died and 1 was admitted to the hospital before assessment. In the control group, 1 was admitted to the hospital immediately after randomization, and 1 was withdrawn from the study because he was not functioning well. Consequently, the per protocol analyses included 52 clients.

Flow diagram of participant allocation.
The pretest characteristics of the clients were well matched between the two groups (Table 1). The occupations determined using the COPM were classified as follows: in the experimental group, hobbies (6), housework (5), personal care (4), walking (6), basic movement (4), physical function (2), and learning (1), and in the control group, hobbies (5), housework (3), personal care (5), walking (8), basic movement (3), and physical function (3).
Participant Characteristics at Pretest
Note. EQ–VAS = EuroQol–5 Dimension visual analog scale; FAI = Frenchay Activities Index; SD = standard deviation.
The results analyzed by analysis of covariance using the bootstrap method and the changes pretest to posttest are shown in Tables 2 and 3. For the primary outcomes, statistically significant differences were observed between the two groups in EQ–5D scores and SF–8 scores for general health. However, no difference between groups was found in the SF–8 physical or mental component summary scores. For the secondary outcomes, significant differences were observed between the two groups in Flow State Scale for Occupational Tasks scores. Although the COPM Satisfaction and Performance scores both showed a trend toward improvement in favor of the experimental group, the difference was not significant (Tables 2 and 3).
Comparison of Primary Outcomes Postintervention, by Group
Note. CI = confidence interval; EQ–5D = EuroQol–5 Dimension; EQ–VAS = EQ–5D visual analog scale; M = mean; SD = standard deviation; SF–8 = 8-item Short-Form Health Survey.
Comparison of Secondary Outcomes Postintervention, by Group
Note. — = not applicable; CI = confidence interval; COPM = Canadian Occupational Performance Measure; M = mean; SD = standard deviation.
Mann−Whitney U test.
A clinically relevant improvement was achieved by 11 (39%) experimental participants and 5 (19%) control participants. The EQ–5D number needed to treat was 4.82, which was not statistically significant, 95% CI [−21.74, 3.06]. The effect sizes and 95% CIs are shown in Table 3. The per protocol analyses showed the following effect sizes and 95% CIs: EQ–5D, d = 0.78 [0.22, 1.34]; SF–8 General Health, d = 1.01 [0.45, 1.61]; SF–8 Physical Functioning, d = 0.66 [0.10, 1.22]; SF–8 Vitality, d = 0.69 [0.13, 1.25]; and Flow State Scale for Occupational Tasks, d = 0.80 [0.23, 1.37]. In addition, the blinding of the clients was checked; 13 (25.0%) of the clients knew their treatment allocation.
Discussion
In this study, older adults who received occupational therapy that focused on the adjustment of challenge–skill balance had significantly better EQ–5D and SF–8 QOL scores. This intervention was demonstrated to be highly effective for improving HRQOL, with a large effect size for the improvement in SF–8 General Health scores and medium effect sizes for the improvements in EQ–5D scores.
According to the Framework (AOTA, 2014), to help clients achieve their desired outcomes, occupational therapy practitioners facilitate interactions between the client and his or her environments and contexts and the occupations in which he or she engages. However, this process tends to depend on practitioners’ experiences and values (Maitra & Erway, 2006). In previous research (Yoshida, Mima, Nonaka, Kobayashi, & Hirao, 2016) it became clear that therapists could not properly evaluate the challenge and skill levels of activities for older adults unless the therapists checked with the clients; therefore, the client’s subjective evaluation using the perceived challenge level and skill level can facilitate the therapist’s understanding of the client’s perceptions of the occupation. Consequently, occupational therapy that includes adjustments to the challenge–skill balance can be one strategy to improve HRQOL and help ensure the quality of occupational therapy.
The results also showed improvement in flow experiences for occupational tasks with the adjusted challenge–skill balance. Csikszentmihalyi (1990) reported that flow occurs when a person perceives a balance between the challenge of the activity and his or her own skill. Thus, adjustment of the challenge–skill balance in the intervention group may have helped result in flow. In addition, because these participants’ flow state increased, the method of the present intervention appears to be appropriate. Further research could address issues regarding the relationship between occupational therapy and flow experiences.
No significant differences were found with respect to the COPM Satisfaction and Performance scores. We considered that improvement of the COPM score was small, because this study focused on partial activity that the client could achieve at that time regarding the occupation.
Limitations
A limitation of our study was the lack of follow-up data; we were unable to assess whether the observed changes in HRQOL were sustained after completion of the intervention. In addition, blinding of therapists regarding the intervention method was difficult because the therapists themselves performed the intervention. Nevertheless, only a quarter of the clients were able to identify their group assignment, suggesting that measurement bias on the client side did not strongly influence the outcomes. In the future, we want to do blinding of outcome assessors and implement a study of a multicenter RCT to reduce the potential for bias.
Implications for Occupational Therapy Practice
The results of this study suggest the following implications for occupational therapy practice:
Adjustment of the challenge–skill balance of activities in occupational therapy practice can positively influence older adults’ HRQOL.
Assessment and adjustment of challenge–skill balance may be a helpful method for understanding older adults’ perceptions of their occupations and can assist the occupational therapy process.
Conclusion
In examining the effects of an occupational therapy intervention using a focus on challenge–skill balance, we found that assessment and adjustment of the challenge–skill balance for activities of older adults positively influenced their HRQOL. This intervention approach may be a helpful method for understanding clients’ perceptions of their occupations. The present findings indicate that using a challenge–skill balance approach is a promising tool for occupational therapy with older adults.
Footnotes
Acknowledgments
This study is registered at
(No. NCT00232934). We acknowledge Harue Hospital, Kalmia Harue; Graduate School of Health Science Studies, Kibi International University; and Graduate School of Human Health Sciences, Tokyo Metropolitan University, as well as all the participants who took part in this study. We declare no conflict of interest.
