Many individuals with stroke experience long-term upper-extremity impairments1 that cause restrictions in daily activities and participation.2 Inadequate training intensity and treatment adherence during subacute or chronic phases may further impede upper-extremity improvement,1,3–5 for this reason motivating tools for self-training are of value. The HandTutor glove,6 the MusicGlove,7 and the Piano keyboard8 are examples of tools developed specifically for rehabilitation that have been found to improve upper-extremity function and specifically hand dexterity (ie, the ability to use the fingers in a coordinative and fast manner) of individuals with stroke. However, these tools require the use of external devices worn or attached to the hand such as gloves or movement sensors that encumber the user, are costly, and not widely available for clinical use.
Recent developments in technology have opened new opportunities for intensive, goal-oriented, motivating training that might be suitable for home-based rehabilitation programs. Tablets computers (tablets) such as the iPad (Apple Inc, California; www.apple.com/ipad) and the associated apps are highly portable and relatively inexpensive. Many individuals own and use these devices for communicating, work, social, and leisure activities. Many of the game apps developed for these devices have the potential to encourage intensive, repetitive finger and hand movements (ie, dexterity). Apps can be motivating, enjoyable, and challenging, which are critical elements of motor learning,9 and therefore they might be useful for self-training of the upper extremity after stroke. In addition, apps can be used to train factors such as speed of motion and reaction time as well as to quantify the individual's performance for a precise evaluation of his or her progress. Individuals who used a tablet (ie, iPad) during subacute stroke rehabilitation perceived it helpful for enhancing engagement in therapeutic activities and socialization10; however, the tablet was not utilized for upper-extremity training.
Published studies that characterize the performance of people with disabilities performing tasks on touch screens are scarce. People with fine or gross upper-extremity disabilities (due to Parkinson disease or multiple sclerosis) performed worse on a tapping task11 or when entering digits12 compared with individuals without a disability. It is also known that hand function13 and manual dexterity14 decline with aging. When implementing new tools, and especially technology-based tools in rehabilitation, it is important for clinicians to have information to assist with the clinical reasoning. This information can guide recommendations related to apps for self-training, and for interpreting their clients' app performance (ie, the ability to perform activities on the tablet app).
Information regarding factors that affect app performance, such as hand dexterity and the use of technology, can be useful for clinicians when interpreting their clients' performance or deciding whether to use this technology. In the current study, we are targeting some of these factors, such as hand dexterity, which is known to change because of normal aging, the use of dominant versus nondominant hands, or experience (ie, practice effect). The client's health status is also likely to impede app performance. Moreover, it is unknown whether individuals with hand impairments because of stroke can use apps with their affected hand.
This study aimed to describe app performance of individuals without a disability (Experiment I) and individuals with stroke (Experiment II). Specifically, the objectives of each experiment were as follows:
- Experiment I: To compare hand performance when using a tablet app (percentage of success and time) in individuals from 3 age groups—young adults (18-35 years), middle-aged adults (46-64 years), and older adults (65-76 years)—and to assess differences in the performance of 2 trials performed with the dominant and 2 trials with the nondominant hand.
- Experiment II: To assess the feasibility of using tablet apps with individuals with stroke, and to characterize the experience (ie, user's feedback) and performance (eg, accuracy) of individuals with stroke using different tablet apps.
We hypothesized that dexterity would affect app performance, and therefore differences will be found between the 3 age groups to the participants with stroke. In addition, the feasibility testing is actually a proof of concept for future use of the tablet as a self-training tool to improve performance of the stroke-affected upper extremity poststroke. The methods and results of each study are presented separately, followed by a combined discussion of both experiments.
The study was approved by the Ethical Committees at the University of Haifa and Tel Aviv University (for the healthy participants; Experiment I) and Tel Aviv Sourasky Medical Center Institutional Helsinki Committee (for the participants with stroke; Experiment II). All participants provided written informed consent.
One hundred and seventy-two individuals without a disability aged 18 to 76 years from 3 age groups were included in the study: young adults (18-35 years old, middle-aged adults (46-64 years old), and older adults (65-76 years old). Study population was recruited using a convenience sample. All individuals were community-dwelling, and independent in basic and instrumental activities of daily living according to self-report.
The iPad app we used (Dexteria, BinaryLabs, Inc; http://goo.gl/BtoHqx) is commercially available, and was developed for training fine motor skills. The app includes 3 game-like tasks—“Write it,” Pinch it,” and “Tap it”—and provides a performance report. For this study the “Tap it” task was used wherein the user is required to continuously hold his/her thumb on an anchor shape while using different fingers to tap on colored shapes that appear and disappear on the screen (Figure 1). This task requires the ability to quickly and accurately isolate finger movements. Outcome measures included percentage of success (accuracy) and performance time (in seconds) recorded for each trial.
Assessment of Hand Dexterity
Dexterity was assessed by the Nine Hole Peg Test (NHPT),15 which measures the time (second) to insert and remove 9 small pegs into a wooden board. The NHPT is a reliable, valid, and responsive measure in rehabilitation and has been found to be sensitive to age-related decline in dexterity.16
Each participant performed 2 consecutive trials (trials 1 and 2) using “Tap it” (level 1) with the dominant hand, and 2 consecutive trials with the nondominant hand (trials 1 and 2). The NHPT was administered to both hands (dominant hand tested first) to describe the participants' level of dexterity using a reliable and valid assessment.
A convenience sample of 20 individuals recovering from stroke in the subacute (up to 4 months poststroke) and chronic (<12 months poststroke) stages was included in the study. Inclusion criteria were (1) at least 1 week poststroke, (2) ability to open and close their fingers (even partially; ie, 2 cm of movement), (3) full function of the hands before the stroke, (4) intact or corrected vision, and (5) cognitively intact or with a mild cognitive impairment as indicated by a score of 23 or above in the Mini-Mental State Examination.17 Exclusion criteria were (1) other neurological conditions (eg, Parkinson disease) and (2) acute orthopedic conditions of the upper extremities.
ScribbleKid (Mile 26 Studios; http://goo.gl/OdwDce)—participants were requested to draw shapes and then write their name on the touch screen with their index finger.
PegLight (Knert Consulting Inc; https://goo.gl/eng03F)—participants were requested to “insert” (tap) pegs (dots) on an enlarged “grid” on the tablet screen by using their index finger. The need for accuracy and speed is relatively low because the dots are large, can be inserted close to each other, and there is no time limit in the app itself. Practice of 5 dots was provided and then the time (second) to touch the screen 15 times to create 15 dots was recorded.
“Tap-it”—as described above for Experiment I. The “Tap it” accuracy of the weak hand was recorded.
Bowling game (Kronos Games; https://goo.gl/tTkE0j)—participants were requested to play 7 rounds of 5-pin bowling on the tablet. The game is played using only one finger while other fingers need to be flexed in order not to touch the screen. The user picks up a ball by touching it and rolls it down the alley by extending their finger. Overall performance was rated by the examiner on a 3-point scale (1 = not successful; 3 = successful).
Short Feedback Questionnaire.
The Short Feedback Questionnaire (SFQ)18 is a 5-point scale questionnaire that was developed to assess the participants' enjoyment, perception of success, and control while using any virtual environment or game. The SFQ has been used in various studies to query about participants' level of enjoyment and control while experiencing virtual environments or games.19,20 In the current study, participants filled in the SFQ regarding their experience after the Bowling App. Each question (eg, enjoyment) was rated on a scale of 1 (not at all) to 5 (very much).
Clinical Assessments to Characterize the Stroke-Affected Upper Extremity.
The following clinical assessments were used to characterize function of the stroke-affected upper extremity from various aspects: level of motor impairment, fine and gross dexterity, and grip strength. These clinical assessments are widely used during stroke rehabilitation and can provide comprehensive information regarding the participants' level of hand function for better understanding of the findings and to assist with the decision of which participants with stroke are suitable to for tablet use.
The Fugl-Meyer Motor Assessment21 (FMA), a common assessment of motor impairment, was used to quantify the ability to perform different movements. Each movement is rated 0 (inability to perform) to 2 (full movement). The total score ranges from 0 (no active movements) to 60 points (full movements) (not including the 6 points from the coordination section).
NHPT15—as described above.
The Box & Block Test22 (BBT) is a test of manual dexterity. Participants are required to pick up and transfer one block at a time over a divider of the box to the other side, within 60 seconds. The score is the number of blocks transferred in 60 seconds.
Grip strength was assessed using a dynamometer (Jamar, Sammons Preston Evaluation Equipment, Bolingbrook, Illinois), and the mean of 3 trials was calculated and recorded in kilograms. This test is valid and reliable for individuals with stroke.23
Participants were administered with the following clinical measures: the FMA, NHPT, BBT, and grip strength. Then participants were first introduced to ScribbleKid and PegLight apps to familiarize them with the tablet's touch screen. Thereafter they performed the activities with the “Tap it” app with their stroke-affected hand (trials 1 and 2), and finally they played the Bowling app, all with their stroke-affected hand. After the Bowling app they filled in the SFQ. An additional question regarding the relevance of the tablet for rehabilitation of their upper extremity was asked after their experience.
For Experiment I, statistical analyses were performed using The IBM SPSS Statistics Version 21.0 (IBM Corporation, Armonk, New York). Analysis of variance repeated-measures mixed design with one within-subjects factor (trial) and one between-subjects factor (age groups) was used to examine differences between the 3 age groups and the first and second trials of “Tap it.” Post-hoc Scheffe was used to examine differences between each pair of age groups. This was done for each hand for performance time and accuracy. One way analysis of variance followed by post-hoc Scheffe was used to examine the differences between the 3 age groups in performance of the NHPT.
For Experiment II, the feasibility, subjective experience, and tablet performance in participants with stroke were described using descriptive statistics. The participants were divided into 2 groups according to the median of the BBT scores, and description of their performance of app-based hand activities is provided as well.
One hundred and seventy-two individuals without a disability agreed to participate in the study: 79 young adults (63.3% women; mean [SD] age, 26.2 [3.9] years), 61 middle-aged adults (55.7% women; 55.9 [5.1] years), and 32 older adults (59.4% women; 68.7 [3.0] years). There were no significant differences (P > 0.01) between the younger, middle-aged adults and older groups in dexterity as assessed by the NHPT for the dominant hand (18.2 [2.1] seconds, 19.7 [3.6] seconds, 18.8 [6.0] seconds) and for the nondominant hand (19.5 [2.7] seconds, 20.7 [3.9] seconds, 19.5 [1.08] seconds), respectively (Table 1). However significant differences in “Tap it” performance for time and accuracy were found.
“Tap It” Performance Time
Significant main effects for trial (F(1,169) = 47.03; P = 0.0001) and age group (F(2,169) = 30.57; P = 0.0001) as well as interaction effect trial × age group (F(2,169) = 9.07; P = 0.0001) were found. Trial 1 was significantly longer than trial 2 (Figure 2 and Table 1). The younger group was significantly faster than the middle-aged adult and older adult groups for both trials (P < 0.01 for all comparisons), and the middle-aged adult group was significantly faster than the older adult group only for trial 1 (the dominant hand; P = 0.0001) (Figure 2 and Table 1).
Significant main effects for trial (F(1,169) = 19.13; P = 0.0001) and age group (F(2,169) = 35.09; P = 0.0001) were found. Trial 1 was significantly longer than trial 2. The younger group was significantly faster than the middle-aged adult and older adult groups for both trials (P < 0.001 for all comparisons), but no significant differences between the middle-aged adult and older adult groups (Figure 2 and Table 1) were found.
“Tap It” Accuracy
Significant main effects for trial (F(1,169) = 9.74; P = 0.002) and age group (F(2,169) = 25.20; P = 0.0001) were found. The accuracy score in trial 2 was significantly higher than in trial 1 (Figure 3 and Table 1). The younger group was significantly more accurate than the middle-aged adult and older adult groups for both trials (P < 0.01 for all comparisons), and the middle-aged adult group was significantly more accurate than the older adult group for both trials (P < 0.01 for all comparisons) (Figure 3 and Table 1).
A significant main effect was found only for age group (F(2,169) = 19.62; P = 0.0001). The younger group was significantly more accurate than the middle-aged adult and older adult groups for both trials 1 and 2 (P < 0.001 for all comparisons), with no significant differences between the other 2 groups (Figure 3 and Table 1).
Twenty individuals with stroke (13 men and 7 women), mean (SD) age 59.3 (13.7) years with moderate-mild upper-extremity motor impairment (as indicated by their score in the FMA) participated in this study. Fifteen participants were in inpatient rehabilitation and 5 participants were in the chronic stage poststroke, living at home, and not receiving formal rehabilitation services. Stroke-affected hemisphere and upper-extremity characteristics of each of the participants are given in Table 2.
Some of the participants were unable to use every app; therefore, we first report the number of participants who completed the task and then their scores (Table 3). PegLight was the only app that all participants were able to complete. From the 15 participants who were able to complete the 2 trials of “Tap it,” 11 (73.3%) improved their performance from trial 1 to trial 2, 3 participants showed a decline in their performance, and 1 participant showed no change in performance between trials 1 and 2. Participants with a higher score in the BBT had a higher score in trial 2 of “Tap it” as compared with those with a lower score in the BBT (Figure 4). In addition, they completed the 15 dots in the “Peg Light” app, faster (Figure 5). However, the clinical profile of participants 9, 12, and 16, who were not able to complete the “Tap it” or participants 4, 13, and 19, that their performance declined in trial 2, varied in terms of time since stroke and their scores in the clinical assessments (Tables 2 and 3).
Participants with stroke were positive regarding their experience with the tablet as reflected in their responses to the SFQ (3.6 [0.7] out of a maximum 5 points; median = 3.7) and they enjoyed the experience as seen from their responses to the first question of the SFQ (4.2 [1.2] out of a maximum 5 points; median = 5.0). All 20 participants thought that the tablet has potential to be a tool used for hand rehabilitation after stroke.
Existing tablet apps and specifically “Tap it” were used by individuals without a disability (Experiment I) and by individuals with mild-to-moderate upper-extremity impairment because of stroke (Experiment II). The findings emphasize that age-related differences and dexterity should be taken into consideration when using apps for rehabilitation poststroke as well as for developing apps for self-training poststroke.
Experiment I demonstrated that app performance is affected by aging and specifically by age-related dexterity decline. These results extend the knowledge that hand function and specifically dexterity16 declines with age,13,14 to app performance. Interestingly, we found significant differences in app performance despite the fact that these differences were not apparent in the NHPT, which is considered a valid and reliable measure of dexterity. As dexterity in older adults has previously been found to be a strong predictor of functional independence and disability in basic activities of daily living and instrumental activities of daily living24–26 and daily arm-hand use,27 it is important to maintain good dexterity. Apps could possibly be used for training of dexterity; however, this needs to be confirmed by future research. The short-term practice effect was demonstrated by an improvement in performance in trial 2 for each hand for all age groups.
Using the apps requires isolated finger movements, speed, and coordination, and therefore were difficult for participants with stroke. Although the participants with moderate-to-mild upper-extremity motor deficits were able to use various apps, their performance in “Tap it” trials 1 and 2 was inferior compared with the participants without disability. Because not all of the participants with stroke were able to use all apps, clinicians might need to try different existing apps to find a suitable and motivating app for their client. Despite this fact, similar to a previous report,10 the overall experience of the participants with stroke was positive and they thought the tablet has potential to be used as a tool for practicing movement of their stroke-affected hand. Because this study included only a small sample of participants with stroke, it is recommended to continue to characterize the motor and sensory components required to play apps.
The practice effect seen in the healthy participants was also seen in 11 of the 15 participants with stroke who were able to complete both trials with the app. This practice effect seems to be more apparent in participants with higher score in the BBT. This encouraging finding indicates the potential of using tablet apps for rehabilitation. Further studies would be of value to examine the practice and training effects on hand function poststroke. Interestingly, participants who were not able to complete the apps or participants with lower accuracy in trial 2 did not show a specific clinical profile; they varied in terms of the time since stroke, and most of them had good hand function (as indicated by high scores in the FMA, BBT, and NHPT assessments) and relatively strong grip strength. However, participants with a higher score in the BBT seem to be faster and more accurate in some of the apps. Nevertheless, participants who had more difficulty completing the apps were observed to have various motor difficulties such as isolating finger movements, extending the index to touch the screen, while flexing the remaining fingers to not touch the screen. In the future, when used as an intervention tool, wearing touch screen-compatible gloves might initially assist these participants, and might also be helpful for women with long fingernails who are not able to touch the screen with the finger pad.
Our inclusion criteria required the participants to have the ability to open and close their fingers. Despite this fact, and the fact that most of the participants had mild motor impairment of their upper extremity, not all of our participants were able to perform all of the apps. Therefore, it would be valuable to develop apps for hand rehabilitation to enable the grading of the app's level of difficulty to suit the individual's motor ability, and for generating a progress report that can be sent to the clinician. A protocol (iHOME) for a study using a newly developed app for the tablet computer (iPad) has been published and the effectiveness is now being tested.28 The goal of iHOME is to examine, relative to the standard of care, the feasibility and efficacy of an interactive software application, to enhance attention and fine motor performance. Another app (for android) called FINDEX has recently been developed for hand rehabilitation poststroke (including dragging, tapping, and stretching), and a pilot study of 3 people with stroke reported FINDEX to be feasible for hand rehabilitation poststroke.29 Development of more apps that provide a variety for self-training motor as well as cognitive tasks would be valuable, as would studies that evaluate the effectiveness of these apps, especially in comparison to other self-training programs.
Limitations of our study include the fact that only a few existing apps were utilized and we focused on the “Tap It” because it generated accuracy and time scores. The healthy participants were not randomly recruited and thus might not represent the population at large; however, their NHPT scores are comparable to the age and sex NHPT norms. In addition, the stroke group included a small convenience sample of individuals with stroke with varying motor deficits but with a relatively good return of active movements of the upper extremity. However, the proof of concept of the overall use of a tablet for individuals with stroke was established; all participants were able to complete the activities associated with at least one app and most of them were able to complete 2 or 3. Larger samples of individuals with stroke are needed to establish the tablet's usability and efficacy for stroke rehabilitation and to establish a training protocol and evidence for this type of practice.
Hand performance with tablet apps was affected by impaired hand dexterity in aging individuals without a disability and in individuals with stroke. The participants in our study indicated they enjoyed the experience of using the apps and thought that the tablet has a potential to be a tool for hand rehabilitation after stroke. Potentially, motivating apps may provide a way to facilitate self-training of repetitive, task-oriented, isolated finger and hand movements to improve hand dexterity and function after stroke.
The authors thank the following graduate students—P. Ben-Ami, S. Barzilay, N. Givon. and A. Keren—and undergraduate students from the departments of Occupational Therapy of the University of Haifa and Tel Aviv University, for their help in data collection.
2. Faria-Fortini I, Michaelsen SM, Cassiano JG, Teixeira-Salmela LF. Upper extremity function in stroke
subjects: relationships between the international classification of functioning, disability, and health domains. J Hand Ther. 2011;24:257–265.
3. Lang C, MacDonald J, Gnip C. Counting repetitions: an observational study of outpatient therapy for people with hemiparesis poststroke. J Neurol Phys Ther. 2007;31:3–10.
4. Lang C, MacDonald J, Reisman D, et al. Observation of amounts of movement practice provided during stroke
rehabilitation. Arch Phys Med Rehabil. 2009;90:1692–1698.
5. Kaur G, English C, Hillier S. How physically active are people with stroke
in therapy sessions aimed at improving motor function? A systematic review. Stroke
Research Treat. 2012; doi:10.1155/2012/820673.
6. Carmeli E, Peleg S, Bartur G, Elbo E, Vatine JJ. HandTutor™ enhanced hand rehabilitation after stroke
—a pilot study. Physiother Res Int. 2011;16(4):191–200.
7. Friedman N, Chan V, Reinkensmeyer AN, et al. Retraining and assessing hand movement after stroke
using the MusicGlove: comparison with conventional hand therapy and isometric grip training. J Neuroeng Rehabil. 2014;11:76.
8. Villeneuve M, Penhune V, Lamontagne A. A piano training program to improve manual dexterity
and upper extremity function in chronic stroke
survivors. Front Hum Neurosci. 2014;8:662.
9. Winstein C, Lewthwaite R, Blanton SR, Wolf LB, Wishart L. Infusing motor learning research into neurorehabilitation practice: a historical perspective with case exemplar from the accelerated skill acquisition program. J Neurol Phys Ther. 2014;38(3):190–200.
10. White J, Janssen H, Jordan L, Pollack M. Tablet technology during stroke
recovery: a survivor's perspective. Disabil Rehabil. 2015;37(13):1186–1192.
11. Irwin CB, Sesto ME. Performance and touch characteristics of disabled and non-disabled participants during a reciprocal tapping task using touch screen technology. Appl Ergonom. 2012;43(6):1038–1043.
12. Chen KB, Savage AB, Chourasia AO, Wiegmann DA, Sesto ME. Touch screen performance by individuals with and without motor control disabilities. Appl Ergonom. 2013;44(2):297–302.
13. Carmeli E, Patish H, Coleman R. The aging
hand. J Gerontol A Biol Sci Med Sci. 2003;58:146–152.
14. Desrosiers J, Hébert R, Bravo G, Rochette A. Age-related changes in upper extremity performance of elderly people: a longitudinal study. Exp Gerontol. 1999;34:393–405.
15. Kellor M, Frost F, Silberberg N, Iversen I, Cummings R. Hand strength and dexterity
. Am J Occup Ther. 1971;25:77–83.
16. Wang Y-C, Bohannon RW, Kapellusch J, Garg A, Gershon RC. Dexterity
as Measured with the 9-Hole Peg Test (9-HPT) across the Age Span. J Hand Ther. 2014. doi:10.1016/j.jht.2014.09.002.
17. Folstein M, Folstein S, McHugh P. Mini-mental state: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–198.
18. Kizony R, Katz N, Rand D, Weiss PL. A Short Feedback Questionnaire (SFQ) to enhance client-centered participation in virtual environments. 11th Annual Cyber Therapy Conference: Virtual Healing: Designing Reality, Gatineau, Canada, 2006.
19. Rand D, Kizony R, Weiss PTL. The Sony PlayStation II EyeToy: low-cost virtual reality for use in rehabilitation. J Neurolog Phys Ther. 2008;32:155–163.
20. Yang JC, Chen CH, Jeng MC. Integrating video-capture virtual reality technology into a physically interactive learning environment for English learning. Comp Edu. 2010;55:1346–1356.
21. Fugl-Meyer AR, Jääskö L, Leyman I, Olsson S, Steglind S. The post-stroke
hemiplegic patient. 1. A method for evaluation of physical performance. Scand J Rehabil Med. 1975;7:13–31.
22. Lyle RC. A performance test for assessment of upper limb function in physical rehabilitation treatment and research. Int J Rehabil Res. 1981;4:483–492.
23. Mathiowetz V, Weber K, Volland G, Kashman N. Reliability and validity of grip and pinch strength evaluations. J Hand Surg Am. 1984;9:222–226.
24. Van Heuvelen MJG, Kempen GIJM, Brouwer WH, de Greef MH. Physical fitness related to disability in older persons. Gerontology. 2000;46:333–341.
25. Williams ME, Hadler NM, Earp JA. Manual ability as a marker of dependency in geriatric women. J Chron Dis. 1982;35:115–122.
26. Ostwald SK, Snowdon DA, Rysavy DM. Manual dexterity
as a correlate of dependency in the elderly. J Am Geriatr Soc. 1989;37:963–969.
27. Rand D, Eng JJ. Arm-hand use in healthy older adults. Am J Occup Ther. 2010;64:877–885.
28. Saposnik G, Chow CM, Gladstone D, et al. iPad
technology for home rehabilitation after stroke
(iHOME): a proof-of-concept randomized trial. Int J Stroke
29. Carabeo CGG, Dalida CMM, Padilla EMZ, Rodrigo MMT. Stroke
patient rehabilitation: a pilot study of an android-based game. Simul Gam. 2014;45:151.