1038 ABILHAND: A Rasch-Built Measure of Manual Ability Massimo Penta, Zng, Jean-Louis Thonnard, PhD, Luigi Tesio, MD ABSTRACT. Penta M, Thonnard J-L, Tesio L. ABILHAND: a Rasch-built measure of manual ability. Arch Phys Med Rehabil 1998;79:1038-42. Objective: To apply the Rasch measurement model to the development of a clinical tool for measuring manual (dis)ability (ABILHAND). Design: Manual ability was evaluated in terms of the difficulty perceived by a hand-impaired patient on 57 represen- tative unimanual or bimanual activities. Setting: A clinical laboratory. Patients: Eighteen rheumatoid arthritis patients (14 women, 4 men) were interviewed after wrist arthrodesis (10 right, 4 left, and 4 both wrists). Their ages ranged from 38 to 77 years, time since diagnosis ranged from 7 to 41 years, and time since surgery ranged from 0.5 to 17 years. Main Outcome Measure: ABILHAND, administered at a mean duration of 7 years after arthrodesis. Results: Forty-six of the 57 items define a common, single manual ability continuum with widespread measurement range and regular item distribution. Items relating to feeding, groom- ing, and dressing upper body worked consistently with their counterparts in other disability scales. More difficult items extend the measurement range beyond that of most existing manual ability scales. Conclusion: Even in a small sample of patients, using the Rasch methodology enabled the investigators to produce a useful scale of manual (dis)ability and to define manual ability as a unique construct, at least in patients with rheumatoid arthritis. 0 1998 by the American Congress of Rehabilitation Medi- cine and the American Academy of Physical Medicine and Rehabilitation F UNCTIONAL INTEGRITY of the hand and upper limb is essential for many activities, yet it is lost as a result of a wide variety of impairments. Therefore, many tests have been developed to assess either the functional losses of mobility, grip strength, tactile sensitivity, and dexterity,ls2or a patient’s ability to execute manual activities. 3-11 The relation between impair- ment (dysfunction at the organ/segment level) and disability (dysfunction at the person level)12 is not straightforward. A patient may adopt either intrinsic or adaptative recovery mechanisms,13 the latter depending on the integrity of the From Rehabilitation and Physical Medicine Unit, Catholic University of Louvain, Brussels, Belgium (Mr. Penta, Dr. Thonnard); and the Department of Rehabilitation and Department of Research, Functional Assessment and Quality Assurance, “Salva- tore Maugeri” Foundation, Instituto di Ricoverdo e Cum a Carattere Scientifico, Pavia, Italy (Dr. T&o). Submitted for publication January 19.1998. Accepted February 20,1998. Supported by a grant from the Belgian Association Nation& d’Aide anx personnes HandicapCes. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated. Reprint requests to J.L. Thonnard, PhD, Rehabilitation and Physical Medicine Unit, Catholic University of Louvain, Tour Pasteur (5375), Avenue Mounier 53, 1200 Brussels, Belgium. 0 1998 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 0003-9993/98/7909-4825$3.00/O Arch Phys Med Rehabil Vol79, September 1998 unaffected organs/segments and on a complex interaction between psychosocial (eg, motivation), cognitive (eg, memory, attention, and spaceperception), and sensorimotor skills. Thus, disability must be measured per se and not merely inferred from the underlying impairments. Ability can be measured as the capacity of a person as a whole to execute activities, scaled along a continuous gradient of difficulty. But the question remains: which activities should be selected, and how should they be scored? For standardization purposes, many existing manual ability tests are built on artificial and/or mainly unimanual tasks.4s7,9 Moreover, the scoring criterion is generally restricted to the observation of a patient’s performance in different activities within an artificial, highly motivating clinical environment. These considerations limit a test’s adequacy to evaluate cor- rectly a patient’s performance in actual daily life activities. Other limitations arise from tests developed for specific levels of performance6 but lacking sensitivity at higher levels (eg, in the later stages of recovery). Also, in most existing manual ability tests, raw ordinal scoresare misused as measures,3,4,6,8,9J1 when in fact they are merely ranks unsuited to conventional arithmetic.14,15 Likewise, timed tests5,7,10 are only ordinal in nature, insofar as a patient being twice as fast as another is not necessarily twice as able in performing the task.16 The purpose of this study was to build a valid scale of manual ability. We selected a sample of patients with wrist arthrodesis, after rheumatoid arthritis @A). In fact, these patients exhibited homogeneous hand and upper limb impairments in the absence of more general movement-disturbing impairments (eg, spastic- ity, paresis, or ataxia). The Rasch mode117,1s appeared to be the ideal framework to achieve a valid measurement scale.15 The test was constructed as a self-estimation report to capture a patient’s average feeling of difficulty in natural tasks.t9 The selected items are centered on high ability levels where patients are functionally independent but still differ in their performance of manual activities. Therefore, the performance criterion, not the dependence criterion, was adopted for scoring.20 METHODS Sample The subjects were selected from among the 33 patients with RA who underwent arthrodesis of one or both wrists in the past 25 years at a university clinic. The same surgeon performed the surgery on all the patients. At the time of evaluation, one patient had died, six had changed their address and could not be located, and eight had difficulty moving and could not come to the laboratory. The 18 remaining patients (4 men and 14 women) took part in the study. Their ages ranged from 38 to 77 years (mean 59). The patients were evaluated between 6.8 and 40.9 years (mean 16.9) after diagnosis of RA. At the time of evaluation, none of the patients had an acute inflammatory disease. All patients were right-handed; 10 had undergone the arthrodesis on the right wrist, 4 on the left wrist, and 4 on both wrists, The surgical procedure, derived from the Mannerfelt and Malmsten operation for wrist fusionzl fixed the wrist in slight extension and ulnar deviation while maintaining pronation and supination of the forearm. The patients were tested between 0.5 and 16.8 years (mean 7.1) after the arthrodesis.