Original Article
Using a template to improve the accuracy and efficiency of the Jebsen–Taylor Hand Function Test: A comparative study
Hand Therapy 2014, Vol. 19(1) 11–16 ! The British Association of Hand Therapists Ltd 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1758998314522435 hth.sagepub.com
Daniel Harte, Denise Curran, Philip Hamill, Alison Porter-Armstrong and Lynn Wilson
Abstract Introduction: The Jebsen–Taylor Hand Function Test is a standardised assessment that relies upon precise test administration through the placement of a number of small items in each subset. This set up has been criticised in the literature as being time consuming and open to non-precision error in item placement. This study investigates whether application of a novel template board to the testing procedure of the Jebsen–Taylor Hand Function Test enhances accuracy and reduces the clinical time taken to administer the test when compared to non-template-based testing practices. Methods: The template board was marked to highlight where each test item should precisely be located during subtest administration. Additionally, three therapists completed 10 timed trials each in test preparation and setting up subtests 2, 3, 6 and 7 with and without the template to assess efficiency. Results: Results show that set up without using a template resulted in an average total of 10% accuracy in subtest 2, 0% accuracy in subtest 3 and 3.33% accuracy in subtests 6 and 7. The acceptable value on these tests to demonstrate accuracy is 100% (p < 0.05). The results also demonstrate that the total time to complete test set up was significantly less when using the template board (p < 0.05). Discussion: This study demonstrates the difficulty in achieving accuracy without a template board and the associated inefficiencies. The availability of standardised assessments that are easy to use in clinical practice and that have sound reliability, validity and responsiveness is necessary to objectively and accurately measure hand function.
Keywords Outcome measures, quantitative research, service evaluation, audit Accepted: 13 January 2014
Introduction Outcome measures can assist in diagnosis, goal setting and prognosis. They can demonstrate the benefit of prescribed therapy and showcase the value of services to stakeholders.1 Outcome measures must have evidence that its properties are reliable, valid and responsive.2 Standardised assessments are outcome measures with instructions that ensure that the assessment can be replicated in precisely the same manner each time.3 The Jebsen–Taylor Hand Function Test (JTT) is a standardised outcome measure commonly used in clinical practice. The test aims to assess the impact of hand dysfunction in activities of daily living by timing performance in writing, simulated page turning, lifting of a range of objects, feeding and stacking. The comparison of test results with published normative data is available for patients aged from six to 94 years old.4–6 There
has been reported use of the JTT as an outcome measure in more than 135 studies.7 In this paper, we introduce a template board to assist in administering the JTT to examine if it improves accuracy and efficiency in testing. Accuracy in test administration is an essential prerequisite to ensure reliability. The JTT is commercially available and is widely used in clinical practice and research as an outcome measure; yet, there have been Southern Health and Social Care Trust, Occupational Therapy Department, Craigavon Area Hospital, Portadown, UK Corresponding author: Daniel Harte, Southern Health and Social Care Trust, Occupational Therapy Department, Craigavon Area Hospital, Lurgan Road, Portadown BT63 5QQ, UK. Email:
[email protected]
Downloaded from hth.sagepub.com by guest on March 10, 2016
12
Hand Therapy 19(1)
questions about its accuracy and efficiency. There have been some attempts made to overcome these: for example, some studies describe modified versions of the JTT and the use of a board with markings to assist in the accurate positioning of test items.8,9 There have been descriptions of the JTT as a time consuming test to administer7 and demonstrating poor applicability when compared to other hand performance tests.10,11 Clinicians are cautioned that the JTT requires test administrators to be familiar with the test set up and instructions.12 The subtests of the JTT were designed ‘to be administered in precisely the same manner to each subject’.4 The aim of this study was to examine if use of a template board indicating the placement of test items enhanced accuracy of item placement when using the JTT. A secondary aim was to assess if use of the template would lead to a significant reduction in the time taken by clinicians to administer the JTT.
instructions and normal clinical practice. The outlining of each item on the table with a non-permanent marker subsequently measured the accuracy of item placement for each trial using the template by the researcher. At the completion of each set of subtests, the transparent template was in a central position on the tabletop and then levelled with the edge of the table closest to the patient. Once placed, the template highlighted the accuracy of test item placement. These were recorded by the researcher as either accurate (exact overlap with non-permanent marker outline with template)
Methods Ethics The Southern Health and Care Trust Research Governance Committee approved this study and it conformed to the ethical guidelines set out in the 1975 Declaration of Helsinki. All participants provided written, informed consent.
Materials A template board was made with 1365 mm 413 mm 2 mm clear acrylic glass (see Figures 1 to 3). Permanent pen marks denoted where each test item of the JTT should be located. These were colour-coded for each subtest. The test instruction manual provided directions on the precise locations of the markings. In the non-template-based assessments, the clinicians used a ruler so they could replicate normal clinical administration and measure out in real time where to place test items as per test manual guidelines.
Figure 1. Template board for the Jebsen–Taylor Hand Function Test (not to scale). Subtest 2: items placed on five horizontal rectangles. Subtest 3: items placed on large central circle with two small circles, two octagons and two oblongs to either side. Subtest 4: the partially occluded large central circle at bottom of template is for simulated feeding task. Subtest 5: items placed on four touching circles below dotted line. Subtest 6 and 7: items placed on five horizontal circles below dotted line. Area within dotted line illustrates where to place the wooden board on top of template to carry out subtests 4, 5, 6 and 7.
Procedure Strand 1 – Accuracy of item placement. Three right-handed occupational therapists, one senior therapist and two junior therapists with collectively over 15 years of experience, were involved in the study. Each therapist worked in the hand therapy clinic and was familiar with the JTT. They completed 10 consecutive trials each in attempting to accurately place test items on the table without the template for subtests 2, 3, 6 and 7. The provision of a ruler helped the therapists measure out where to place each test item on the table as per the test
Figure 2. The template assisting set up of subtest 6 (placement of wooden board and cans).
Downloaded from hth.sagepub.com by guest on March 10, 2016
Harte et al.
13
or inaccurate. There was blinding of each therapist on his or her performance on each trial. Strand 2 – Efficiency of test administration. The three therapists completed 10 consecutive timed trials each in test preparation and setting up of subtests 2, 3, 6 and 7 first with, and then without the template. The researcher was responsible for timing performance with a stopwatch in seconds. To ensure consistency, there was set up of a preparation table with all the test materials in the same position before each trial. The positioning of the preparation area was adjacent to the therapist on their right side with the assessment table in front. Each therapist sat on a swivel chair during the time trials. Placed on the preparation area were the items required to administer the test along with the template. In each trial, the therapist said, ‘Stop’ to indicate when they were finished. The first timed trial required the therapist to position masking tape at the table edge to mark the centre of the patient and a second strip of masking tape 12.7 cm from the table edge, as described in the test instructions. Preparation time using the template recorded the
length of time required to the lift the template off the preparation area and placing it central to the patient. Timed trials of subtest 2 and subtest 3 required the therapist to place the pages or objects, for therapists’ verbal instruction and for the removal of items from the test area. For timed trials using the template, the template was on the assessment table prior to the commencement of timing. Time trials of subtest 6 and subtest 7 required the therapist to place the wooden board and cans, for therapists’ verbal instruction and for removal of the cans from the test area.
Results Accuracy of item placement Results show from the raw data that set up without using a template resulted in an average total of 10% accuracy in subtest 2 (Simulated Page Turning), 0% accuracy in subtest 3 (Lifting Small Common Objects) and 3.33% accuracy in subtests 6/7 (Lifting Light/ Heavy Large Objects) (Table 1). A Chi-square test was carried out showing a significant difference between the observed performances and the benchmark (the acceptable value on these tests to demonstrate accuracy is 100% accuracy) (p < 0.001). The central card in subtest 2, the coffee tin in subtest 3 and the central can in subtest 6/7 are used as a reference point where other items must relate to in terms of distance so hence are not analysed.
Table 1. Accuracy of item placement. Therapist
Figure 3. The template assisting set up of subtest 3 (placement of coffee jar, coins, bottle tops and paperclips).
Number of accurate trials
Number of inaccurate trials
Subtest 2: Simulated page turning Therapist 1 0/10 10/10 Therapist 2 0/10 10/10 Therapist 3 3/10 7/10 Total 3/30 27/30 Subtest 3: Lifting small common objects Therapist 1 0/10 10/10 Therapist 2 0/10 10/10 Therapist 3 0/10 10/10 Total 0/30 30/30 Subtest 6/7: Lifting large objects Therapist 1 1/10 9/10 Therapist 2 0/10 10/10 Therapist 3 0/10 10/10 Total 1/30 29/30
Downloaded from hth.sagepub.com by guest on March 10, 2016
One-tailed t-test
p ¼ 0.001
p ¼ 0.001
p ¼ 0.001
14
Hand Therapy 19(1)
Efficiency of test administration The difference in the mean time taken to administer the test with and without the template was evaluated using a one-tailed t-test. Table 2 reports the descriptive statistics and tests for significant difference in the mean and median time taken to administer the JTT with and without the template. Consequently, as a robustness check, a test for significant difference in medians was completed using the Mann–Whitney test. The results demonstrate that time to complete test preparation (p ¼ 0.001), to administer subtest 2 (p ¼ 0.001), subtest 3 (p ¼ 0.001) and subtest 6 (p ¼ 0.001), and the total time of all four trialled components of the test (p ¼ 0.001) with the template are statistically significant. With p-values
