Abstract
Keywords
Occupational therapy advocates client-centered practice to promote client function and occupational performance (Law, Baum, & Dunn, 2005). However, research documenting therapy effectiveness in promoting actual functional outcome is sparse (Aldehag, Jonsson, & Ansved, 2005; Case-Smith, 2003; Jack & Estes, 2010; Rønningen & Kjeken, 2008; van der Giesen et al., 2007), with the measurement of treatment effectiveness often limited to impairment-focused measurement (Schoneveld, Wittink, & Takken, 2009). For example, although occupational therapists often treat patients with tendonitis who receive steroid injections and those who are not fit for steroid injections, no studies have systematically compared short-term functional improvement or long-term outcomes in these two groups. Likewise, no studies have compared treatment outcomes in patients who received steroid injections alone with those of patients who received a combination of injections and occupational therapy. As a result, whether and to what extent patients with tendonitis benefit from occupational therapy hand rehabilitation is not known.
This phenomenon is not unique to occupational therapy. In the field of hand surgery, studies have often reported pain and impairment reduction and seldom addressed short- or long-term functional outcomes. Hand surgeons providing steroid injection to treat patients with tendonitis have found that it is effective (for pain reduction) in only about 60% of patients; its long-term effect on function is unknown (Bailey, Kaskutas, Fox, Baum, & Mackinnon, 2009; McAuliffe, 2010). In a recent study, Swedish researchers reported that patients adopted different strategies using their hands after traumatic hand injuries, but again the researchers did not measure function; therefore, they did not know whether patients had improved function over time or not (Kingston, Tanner, & Gray, 2010).
Clinicians and researchers have agreed that patients’ ultimate goals for seeking treatment are to relieve symptoms and improve function, that is, to perform daily tasks with ease and comfort (Bailey et al., 2009; Davis et al., 1999). At this time, much information related to functional outcomes either is derived from outcome tools that assess component- or factor-level deficits (e.g., pain, sensation, range of motion, and strength are categorized as body function under client factors; see the Occupational Therapy Practice Framework: Domain and Process, 2nd Edition; American Occupational Therapy Association [AOTA], 2008) and performance skills (e.g., task performance that requires integrated skills such as dexterity or coordination) or comes from a patient’s self-report. Moreover, research has suggested that improvement in body function or component deficits does not always translate into improvement in activity participation or quality of life (Bindra et al., 2003; Rallon & Chen, 2008; Schoneveld et al., 2009).
Several functional assessments used in occupational therapy hand settings are self-reported (Disabilities of Arm, Shoulder and Hand Questionnaire [Hudak, Amadio, Bombardier, & the Upper Extremity Collaborative Group, 1996], Patient-Rated Wrist–Hand Evaluation Questionnaire [MacDermid & Tottenham, 2004], Michigan Hand Outcomes Questionnaire [Chung, Pilsbury, Walters, & Hayward, 1998]) and are supported in the literature as having reasonable construct validity and responsiveness (Hoang-Kim, Pegreffi, Moroni, & Ladd, 2011). However, a need does exist to correlate patient-reported outcome tools, which are subjective in nature, with those professionally administered by the clinician (Carlsson, Haak, Nygren, & Iwarsson, 2012) as an additional step toward expanding their use and thereby providing clinicians and researchers with a greater number of outcome measures that can be used to assess changes in participation.
The 36-item Manual Ability Measure–36 (MAM–36; Chen & Bode, 2010) is a newly developed assessment. Although the MAM–36 has gone through rigorous validation processes and demonstrated excellent internal validity, its responsiveness is unknown because it has not been used to investigate treatment effectiveness. Therefore, we sought to determine its responsiveness to functional change after treatment intervention compared with improvements in component deficits (i.e., grip strength) or performance skills (i.e., activities of daily living [ADL] task performance speed). This study posed two questions: (1) Is the MAM–36 sensitive enough to measure functional changes and, if so, (2) does the change detected by the MAM–36 correlate with changes measured by another performance-based, clinician-rated assessment?
Method
Research Design
We used a one-group pretest–posttest design to collect data. The study was approved by the institutional review boards of the participating institutions, and all patients provided consent before their participation in the study.
Participants
Patients who had various upper-extremity injuries or conditions and received occupational therapy outpatient services at an urban Israeli medical center were recruited to participate in the research. The exclusion criteria were as follows: (1) The patient was unable to understand or communicate with the therapist regarding items on the MAM–36 and (2) the patient’s injury or condition was too acute (or not safe) to perform the required performance-based assessments. The data consist of 2 cohorts of patients: Cohort A’s (n = 20) data were collected from late 2008 to early 2009 and Cohort B’s (n = 26) data were collected from mid-2009 to mid-2010.
Assessments
Manual Ability Measure–36.
The MAM–36 is a task-oriented, patient-reported functional assessment tool (Chen & Bode, 2010). It has 2 sections: The first section contains demographic and clinical information; the second section is a rating scale of 36 items. The 4-point rating scale (ranging from 1 = cannot do to 4 = easy) requires the patient to rate perceived ease or difficulty of performing 36 everyday tasks (e.g., cutting meat on a plate, taking things out of a wallet, opening a medicine bottle with a childproof cap or top) regardless of which hand is used and without the use of assistive devices. The psychometric properties of the MAM–36 were previously investigated with a diverse patient population using a Rasch measurement model (Chen & Bode, 2010), and it demonstrated adequate structural integrity and internal validity (i.e., .93 and .99 for person and item reliability, respectively).
The MAM–36 was first translated into Hebrew and back-translated into English by a therapist who was not involved in the study. The wording of the items was mostly retained, although some minimal modification was necessary as a result of inherent cultural or semantic differences; one such example was “squeezing a kitchen dish rag” instead of “wringing a towel,” as in the English version.
Smith Hand Function Test.
The Smith Hand Function Test (SHFT; Smith, 1973) is a performance-based hand function test in which the speeds of task performance and grip strength are used to measure component deficits and performance skills. The original test consists of 4 subtests: (1) unilateral grasp–release tasks (blocks, nails, coins, pegs), (2) bilateral simulated ADL tasks (safety pins, buckle, buttons, zipper, tying knot, tying bow, lacing shoes; each mounted on separate wooden board), (3) writing task, and (4) grip-strength measurement. Smith (1973) described the test in detail and published the norms for male and female participants comparing right- and left-hand performances regardless of hand dominance. However, no additional psychometric studies about the test were published. The bilateral tasks were mounted on wooden boards so that they could be administered to patients in a standardized manner. The writing task was only administered to patients whose symptoms were in their dominant hands and who could perform the task.
Procedures
Three occupational therapists from the same hospital volunteered to participate in the project and collected data. From late 2008 to early 2009, the therapists administered the MAM–36 to 20 patients (Cohort A) at intake and discharge to determine whether it was feasible to use in a clinical environment. The overall response from the patients and therapists was positive. Therefore, therapists decided to use it along with the SHFT and collected data (Cohort B) from mid-2009 to mid-2010 to examine the correlation between patient-reported function-based gain measured by the MAM–36 and clinician-assessed factor-level and skill-level gain measured by the SHFT.
All patients (Cohorts A and B) were assessed with the MAM–36 at the beginning of outpatient rehabilitation (i.e., during the 1st week of occupational therapy treatment) and again right before discharge (i.e., during the last week of occupational therapy treatment). Patients in Cohort B were also assessed with the SHFT. The MAM–36 and SHFT could be administered in one or two sessions, depending on the time needed, the patient’s tolerance for assessments, or both. The order of the two tests was randomized.
Occupational therapy treatments were provided 2 times/wk; the treatment period lasted 2–33 wk (Cohort B). Treatments incorporated the principles of tissue healing, edema reduction, scar tissue management, and promotion of range of motion, sensation tolerance, strength, endurance, coordination, dexterity, and so forth. Treatment techniques included the use of modalities (heat and cold), retrograde massage, manual scar mobilization, splinting, therapeutic exercises, occupation-based interventions, work hardening, and patient education. All treatments were provided by the three therapists.
Data Analysis
A series of preliminary analyses were conducted to examine the equivalence of the two cohorts. We found no significant difference in age, gender, hand dominance, and duration of treatment (Table 1); the MAM–36 data from both cohorts were therefore combined for subsequent analyses.
Demographic and Clinical Characteristics of the Participants
Note. M = mean; SD = standard deviation.
MAM–36 Variables: Mean Item Ratings and Rasch-Derived MAM.
The MAM–36 produces two levels of measures for each participant: mean item rating and an overall Rasch-derived manual ability measure (or the “MAM measure”; Bond & Fox, 2001; Chen & Bode, 2010). Mean item ratings at intake and discharge for each participant were calculated from their individual item ratings at Time 1 and Time 2. We use the terms intake and Time 1 and discharge and Time 2 interchangeably from this point on.
To calculate the scale-based overall MAM measures, we used Rasch analysis. First, intake and discharge MAM ratings from both cohorts were combined. Next, we used step and item anchoring procedures. Third, we converted MAM measures in log-odds units, or logits, into a 0–100 scale denoted as MAMT1 (intake MAM measure) and MAMT2 (discharge MAM measure). The step and item anchors were obtained from the earlier validation study based on a large sample of more than 300 patients (Chen & Bode, 2010). Anchoring was an important Rasch measurement technique, because the step and item anchors held the rating scale structure and the position of the items stable while allowing for the calibration of the MAM (person) measures. Higher MAM measures indicated greater manual ability.
Smith Hand Function Test Variables.
Because of a large amount of missing data in the writing subtest, we did not include it in the final data analyses. Therefore, the SHFT outcome variables included three speed variables (unilateral, bilateral, and total speed) and a strength variable. The speeds were measured by a stopwatch. Speedunilat1 and Speedunilat2 represent the average time needed to perform one-handed (or unilateral) tasks at intake (i.e., Speedunilat1) and discharge (i.e., Speedunilat2). (We summed up left-hand and right-hand speed separately and calculated the average time needed to perform the unilateral tasks at both intake and discharge.) Speedbilat1 and Speedbilat2 represent the time needed to perform the two-handed (or bilateral) tasks at intake and discharge, respectively. Speedtot1 and Speedtot2, respectively, represent the total time needed for all (unilateral and bilateral) tasks at intake and discharge. Strength was obtained by measuring patients’ affected hand using a Jamar dynamometer. For patients who had both hands affected, we obtained the average grip strength from both hands.
Parametric Analyses.
We conducted paired t tests to examine whether significant differences occurred from intake to discharge on (1) MAM measures (i.e., MAMT1 and MAMT2); (2) mean item ratings; (3) time needed to perform the SHFT (i.e., the three speed variables measuring unilateral, bilateral, and all (total) tasks (Cohort B only); and (4) grip strength in the affected hand (Cohort B only). We conducted Pearson correlation analyses to examine the associations between gains in MAM measures, strength, and speeds of performance among Cohort B patients. Last, we calculated the standard response mean (SRM), a measure of effect size, to estimate the responsiveness of the MAM. The SRM is computed by dividing the mean score change by the standard deviation of the change (Liang, Fossel, & Larson, 1990).
Results
Participants
A total of 46 patients participated in the study. Among them, 70% were female with an average age of 56.2 yr (standard deviation = 19.8). All but 3 persons were right handed; about half had injuries or pathology in their dominant hand. Twelve (26%) patients were working; another 12 were on leave from work because of their injuries; and 22 (48%) were not working, including those who were retired. Data on education and occupations were collected only from Cohort B patients: Eight had baccalaureate or graduate degrees, 10 had high school diplomas, and 5 had elementary or middle school education; 3 did not report education. Patients’ occupations varied and included manual laborers, managers, secretaries, teachers, homemakers, and other professionals. The patients’ diagnoses included a variety of upper-extremity injuries and conditions typically seen in an outpatient hand therapy setting: traumatic injuries to the fingers (n = 5), such as deep cuts, amputation, and crush injuries; various types of fractures (n = 27) of the finger, hand, wrist (i.e., distal radius, radius, ulnar), elbow, and shoulder; dislocated elbow (n = 1); status post–carpal tunnel release (n = 3); trigger finger release (n = 2); Dupuytren’s contracture release surgery (n = 1); status post–trapezectomy (n = 6); and neuropathy (n = 1).
Manual Ability Measure–36 (n = 46)
Change in Manual Ability.
We found a significant change in mean MAM measures between intake (MAMT1) and discharge (MAMT2), t(45)= 8.03, p < .0001, suggesting that manual ability improved over time. The means and standard deviations of the MAM measures and MAM gains are presented in Table 2. The gain in manual ability ranged from 1.9 to 76; 16 patients gained <10 points; 15 gained 11–20 points, 8 gained 21–30 points, 4 gained 31–40 points, and 3 gained >40 points.
MAM–36 Measures (N = 46)
Note. M = mean; MAM–36 = Manual Ability Measure–36; SD = standard deviation.
Perceived Change in Task Difficulty.
Paired t tests showed that the mean item ratings of all MAM items increased significantly (all ps < .01) from intake to discharge, suggesting that the patients rated the tasks easier at discharge than at intake. The mean changes ranged from 0.5 to 1.5 points. The items on which ratings changed the least were shuffling cards (changed only 0.52 points), turning pages (0.54), eating a sandwich (0.59), and washing hands (0.59). The items on which ratings changed the most were tying shoes with laces (1.5), peeling fruits and vegetables (1.3), cutting meat on a plate (1.39), and wringing a towel (1.41).
Smith Hand Function Test (n = 26)
Improvement in the Speed of Performance.
We used paired t tests to examine Time 1 (intake) and Time 2 (discharge) speeds of performance on one-handed, two-handed, and total SHFT tasks. All speed variables (i.e., Speedunilat, Speedbilat, and Speedtot) for the SHFT decreased significantly at discharge, indicating that less time was required for patients to perform the SHFT tasks. The means, standard deviations, and t values for the speed variables are presented in Table 3.
SHFT at Admission and Discharge (Cohort B, n = 26)
Note. M = mean; n/a = not applicable; SHFT = Smith Hand Function Test; SD = standard deviation.
All ps < .01. bAverage of the speeds of performing unilateral tasks by the right and left hands. cTime needed for performing unilateral tasks (right hand + left hand) and bilateral tasks.
Improvement in Grip Strength in the Affected Hand.
Grip strength of both hands was measured at intake and discharge as part of the SHFT among Cohort B patients; however, only strength of the affected hand was the focus of this study. Paired t tests showed that, for Cohort B patients as a group, grip strength of the affected hand increased significantly, t(25)= 4.44, p < .01. On inspection of individual patients, grip strength of the affected hand did not change in 2 patients, decreased in 2, and improved in 22. Among these 22 patients, the gain in strength ranged from 1.7 to 29.4 kg (mean = 9.89 kg, standard deviation = 7.75). With all patients included, however, the mean increase in grip strength of the affected hand was 7.79 (standard deviation = 8.93 kg).
Responsiveness of the Manual Ability Measure–36
We calculated the SRM of the MAM–36 (n = 46) by dividing the mean gain by the standard deviation of the gain. The SRM was 1.18, suggesting that the responsiveness is more than adequate (Liang et al., 1990).
Relationship Among Gain in Manual Ability, Strength, and Speed of Performance
On the basis of Cohort B’s data, gain in MAM measures correlated moderately with gain in grip strength of the affected hand (r = .69, p < .001) and gain in total speed of performing the SHFT tasks (r = .71, p < .001). Note that MAM measure gains were strongly and negatively correlated with MAM measures at intake (r = −.81, p < .001), suggesting that patients with low manual ability at intake tended to improve more. However, MAM gains were also strongly and positively correlated with MAMs at discharge (r = .73, p < .001), suggesting that patients’ improvement was strongly associated with discharge manual ability.
Discussion
The ultimate goal of occupational therapy upper-extremity treatment and rehabilitation is to improve patients’ function and their ability to assume their occupational roles and participate in everyday activities (Jack & Estes, 2010). However, few assessment tools currently in use provide evidence of the improvement in functional engagement and participation after intervention. The results of this study showed that patients’ scores on the MAM–36 improved—demonstrating enhanced self-reported or perceived hand ability—after occupational therapy treatment and that the changes were observed using tools that measure performance skills and factor-level abilities such as hand strength and control of movement. In all, patients made an average improvement of 15 points on the MAM measures, gained between 7.8 kg (all patients included) and 9.9 kg (among only those who improved) in grip strength in their affected hand, and performed the SHFT tasks faster.
The results of this study also demonstrate that the MAM–36 is capable of detecting changes over time. The SRM of the MAM measure gain was 1.18, which is considered large (Liang et al., 1990). Moreover, the improvement in manual ability was moderately correlated with the gain in grip strength (in the affected hand) and with the gain in performance speed of various unilateral and bilateral tasks. These results (i.e., correlations) are encouraging because they show that the patients’ perception (and report) of their manual ability improvement is coherent with their actual functional improvement: Not only did they gain strength in their affected hands, but they also performed the tasks faster.
The MAM–36 can be a useful tool in clinical settings in two different ways: (1) on the scale level and (2) on the item level. On the scale level, the MAM–36 produces a MAM person measure, which quantifies each person’s manual ability. The MAMs allow comparison of patients’ perception of their overall hand function over time. In our study, we noticed that all patients improved (i.e., the MAM measures improved at discharge when compared with intake). We were somewhat surprised to see the large range of MAM gains. However, close inspection of the cases showed that these results were reasonable.
The patient who improved the most (i.e., 75.9 points on MAM measure) was a high school boy who had a contusion of the dominant hand and wrist (with pain and swelling) and had great difficulty with most tasks at intake but showed substantial improvement after 3 wk of occupational therapy and rated almost all activities as easy. The patient who improved the least (i.e., 1.9 points on MAM) was a young man who was left handed and had a left arm (radius shaft) fracture. At intake, he rated most tasks as easy and rated only a few items as a little hard, very hard, or unable to do. Although the ratings of some items at discharge improved, the ratings of the most difficult items (e.g., opening medicine bottles, cutting nails, shuffling cards) did not change. We noticed, however, that the strength of his affected (and dominant) hand improved 6 kg. It is possible that as a left-handed individual, the patient had to be somewhat ambidextrous before the injury to survive in a right-handed world. As a result, although he had a fracture in his dominant arm, he was able to adapt from the beginning and therefore rated most MAM–36 items as easy or a little hard. Although he made some improvement, this improvement was not large enough to change the ratings on the most difficult items.
The MAM–36 can be examined on the item level. We noticed that item rating averages decreased at least 0.5 points, suggesting that the patients, as a group, perceived the tasks as easier to do at discharge than at intake. We can compare a patient’s item ratings at intake and discharge and assess his or her capabilities and difficulties with specific tasks, as the example given in the preceding paragraph shows. When examining item ratings, therapists can assess and document treatment outcomes in a very specific and evidence-based fashion. This level of understanding of the client’s performance capabilities will assist the clinician with demonstrating the effectiveness of interventions and with assisting the client to prepare for future functional challenges through education and adaptation if needed.
Limitations and Future Research
This study has several limitations. The sample size was small, and it was a convenience sample; no control group was used, and the treating therapists were also the assessors and were not blind to the purpose of the study. Therefore, the results should be interpreted with caution. Further research with a larger sample size and a more stringent research design is warranted.
Implications for Occupational Therapy Practice
The results of this study have the following implications for occupational therapy practice:
Provides evidence that function-based, patient-reported outcome assessments such as the MAM–36 may correlate well with traditional factor-level, clinician-rated assessments often used in hand rehabilitation settings
Provides evidence that patient-reported outcome tools have a place as part of a comprehensive assessment of upper-extremity functioning in a hand rehabilitation population
Provides support for occupational therapists interested in adopting a more functional outcome-based perspective with regard to hand rehabilitation outcomes
Expands the profession’s knowledge of the effectiveness of occupational therapy intervention as a means to enhance functional performance.
Conclusion
The results of this study demonstrate that the MAM–36 was responsive to changes in hand function in patients receiving occupational therapy outpatient rehabilitation services. These changes were identified on both the scale level and the average item level. In addition, the perceived manual ability gain correlated positively with improvements in task speed and grip strength assessed using the SHFT. These findings support the use of the MAM–36 as an outcome measure for occupational therapy hand rehabilitation patients when determination of activity- or participation-level functional outcome is desired.
Footnotes
Acknowledgments
We thank Hagit Harel, occupational therapist, for translating the MAM–36 into Hebrew; Sharon Werech and Hadas Pfizer, occupational therapists, for assisting in data collection; and the patients who participated in the study. An earlier version of the study was presented in the 2010 Joint Conference of American Congress of Rehabilitation Medicine and American Society for Neurorehabilitation.