Validity of HAMIS: A Test of Hand Mobility in Scleroderma

Validity of HAMIS: A Test of Hand Mobility in Scleroderma Gunnel Sandqvist and Mona Eklund Objective. Hand Mobility in Scleroderma (HAMIS) is a hand function test for persons who have systemic sclerosis (scleroderma). The purpose of HAMIS is to obtain an estimation of the hand mobility that is precise enough to detect limitation of motion at the same time as it indicates the ability to use the hand in daily occupations. The aim of this study was to test psychometric properties of the HAMIS, and the following aspects of HAMIS were examined: 1) the applicability of HAMIS, 2) concurrent validity, and 3) discriminating ability. Methods. Forty-five patients with scleroderma were assessed for range of motion (ROM), HAMIS, and skin thickness. In addition, 15 healthy individuals completed HAMIS. Results. The applicability of HAMIS was good for items assessing finger and thumb mobility and moderate for items assessing mobility of the wrist and the forearm. The relationships of HAMIS to ROM and skin score were statistically significant for all items except for pronation and supination of the forearm. There were also statistically significant differences Supported by grants from the Swedish Rheumatism Association. Gunnel Sandqvist, MSc, OTR, Department of Rheumatology, Lund University Hospital, Lund, Sweden, and Department of Clinical Neuroscience, Division of Occupational Therapy, Lund University, Lund, Sweden; and Mona Eklund, Assistant Professor, PhD, OTR, Department of Clinical Neuroscience, Division of Occupational Therapy, Lund University, Lund, Sweden. Address correspondence to Gunnel Sandqvist, MSc, OTR, Department of Rheumatology, Lund University Hospital, S-221 85 Lund, Sweden. Submitted for publication February 9, 2000; accepted in revised form July 27, 2000. 2000 by the American College of Rheumatology. between the patients and the healthy individuals for all items except these two. Conclusion. HAMIS has a demonstrated concurrent validity compared with ROM and skin score, and it showed a good ability to discriminate between healthy individuals and persons with scleroderma, although a lack of variation in the items measuring pronation and supination inferred worse psychometric properties for these two items. INTRODUCTION Systemic sclerosis (SSc), or scleroderma, is a disorder of connective tissue characterized by induration and thickening of the skin, Raynaud s phenomenon, and a potential involvement of a wide range of internal organs (1). Based on the extent of skin involvement, two forms of SSc may be distinguished: limited systemic sclerosis (lssc) and diffuse systemic sclerosis (dssc). Skin induration and joint and muscle involvement often lead to a progressive reduction in range of motion that is the major cause of rehabilitative problems. Ninety percent of SSc patients report loss of hand grasp ability. Factors such as puffy fingers, calcium deposits, and wrist extension reduction have been identified as risk factors for later development of severe levels of hand disability (2). There are many aspects of hand function, such as anatomical integrity, mobility, muscle strength, sensation, grasp patterns, precision and accuracy, coordination and dexterity, unilateral and bilateral tasks, activities of daily living (ADL) tasks, and motivation (3). Several of these aspects, including mobility, are important to patients with SSc. Early hand deformities that occur in SSc are loss of flexion of the metacarpophalangeal (MCP) joints, loss of extension of 382 0893-7524/00/$5.00

Arthritis Care and Research Validity of HAMIS 383 the proximal interphalangeal (PIP) joints, loss of thumb abduction, opposition, and flexion, and finally loss of wrist motion in all planes. These typical hand deformities contribute to functional disability in varying degrees, and because of the nature of the disease the primary goal in treatment of the hand therefore is to maintain maximal range of motion (4). Mobility in the hand can be assessed in various ways, such as measurements of range of motion (ROM), using a goniometer and a ruler, or with a performance index. A performance index does not imply a testing of movements of isolated joints but gives an indication of the individual s ability to use the hand skillfully in the activities of daily living (5). Occupational therapists, framing function within the context of everyday activities, often find performance indexes more suitable in estimating function. Development and adoption of methods to quantify and longitudinally follow scleroderma s impact on functional capacity, self-assessment, quality of life, psychological well-being, and overall health would enhance both clinical trials and routine patient care (6). A new hand function test, Hand Mobility in Scleroderma (HAMIS), was devised in order to reflect the mobility of the scleroderma hand in an easy way and to reflect specific impairments, e.g., swelling (7). HAMIS is a performance index inspired by the hand function tests Signals of Functional Impairment (SOFI) (5) and Keitel Function Test (KFT) (8). The different performance areas of HAMIS are composed of different-sized grips and different movements, all related to tools and movements that are part of performance of daily occupations, and it includes all movements assessed in an ordinary ROM-measured hand mobility test. A performance index for evaluation of ROM is suitable for patients with SSc since it is mostly the skin tightness that has an effect on the mobility, and the ability of flexion and extension of the fingers is often the same for all fingers. HAMIS was developed with the purpose of obtaining an estimate of the function of the hand that is precise enough to detect limitation of motion at the same time as it indicates the ability to use the hand in daily occupations. HAMIS has turned out to be a reliable instrument (7), and the aim of the present study was to further test psychometric properties of HAMIS. The following aspects of HAMIS were examined: 1) the applicability of HAMIS, 2) concurrent validity, and 3) discriminating ability, as indicated by comparisons between patients with SSc and healthy individuals. PATIENTS AND METHODS Subjects. The test population comprised 45 consecutively selected patients (40 women and 5 men) who were hospitalized for 3 days for routine control at the department of rheumatology at a university hospital in southern Sweden. The sample included all patients in routine control during 7 months. Their age ranged from 22 to 74 years (median age 53 years), and the range of duration of illness was 1 to 17 years (median duration 4 years). Two patients had suspected SSc, 33 patients had lssc, and 9 patients had dssc. For one patient the specified diagnosis was not recorded. Fifteen healthy volunteers (14 women and 1 man) recruited from the hospital staff served as a comparison group. Their ages ranged from 36 to 62 years (median age 48 years). Forty-two individuals in the patient group were right dominant and 3 were left dominant. In the comparison group all individuals were right dominant. These data were collected during the same period as the patient data. Measurements. Three measures were employed: the ROM test using a goniometer and a ruler, HAMIS, and assessment of skin thickness by means of manual palpation (skin score). ROM test. The ROM test followed the procedure described by the American Academy of Orthopaedic Surgeons (9), but some modifications were made. Therefore, the procedure used in this study will be described in detail. Range of motion was measured as follows: Flexion deficits of digits 2 through 5 were defined in 2 ways: 1) distance in millimeters from the distal point of the digit to the flat of the hand, which denominates the deficit of flexion, and 2) distance in millimeters from the distal palmar crease to the point in the flat of the hand where the digit ended up, thus denominating the proximal flexion of the palmar crease. Extension deficits of digits 2 through 5 were measured with the forearm in a zero position and the ulnar side of the hand resting on the bottom of a box. The back surface of the hand flat was quite close to one side of the box, and the distance between the nail bed of the extended fingers and the box was recorded in millimeters. The volar abduction of the thumb was measured as the distance in millimeters between the fingertips of digits 1 and 2. Furthermore, extension deficits in PIP, flexion in MCP 1 and IP, volar flexion of the wrist, dorsal extension of the wrist, pronation of the forearm, and supination of the forearm followed the procedure described by the American Academy of Orthopaedic Surgeons (9).

384 Sandqvist and Eklund Vol. 13, No. 6, December 2000 HAMIS. HAMIS is a performance index that consists of 9 items (7) assessing the movements included in an ordinary range of motion test, i.e., finger flexion and extension, abduction of the thumb, dorsal extension and volar flexion of the wrist, and pronation and supination of the forearm. It also assesses the ability to make a thumb pincer grip and to make finger abduction. The different performance areas of HAMIS are composed of different-sized grips and different movements, all related to tools and movements that are part of daily occupations. For further description, see Sandqvist and Eklund (7). Each exercise is graded on a 0 3 scale, where 0 corresponds to normal function and 3 denotes that the individual is unable to perform the task. Skin score. The skin was assessed according to the modified Rodnan skin score technique, which has acceptable inter- and intra-observer reliability (10). The present study assessed only the skin score on the fingers, dorsally on the hand, and on the forearm. The scale step was 0 normal, 1 mild skin thickness, 2 moderate skin thickness, and 3 severe skin thickness with inability to pinch the skin into a fold. Procedure. The assessments with ROM and HAMIS were made in connection with the routine control by an occupational therapist (GS). First the ROM test was performed, followed by administration of HAMIS. A physician made the assessments of skin thickness. Analyses. All the measures were performed on both hands, but this report describes the evaluation on the patients dominant hand. To estimate whether HAMIS is a relevant test for patients with SSc and does not reach a floor or ceiling effect, an applicability analysis for HAMIS was performed by examining the frequency distribution of scores obtained for the different items. For the estimation of concurrent validity, two strategies were used. First, relationships between HAMIS and ROM, and between HAMIS and skin score, were calculated by means of Spearman s rank correlation coefficient. The finger abduction item of HAMIS was not included in this analysis since finger abduction was not measured in the ROM test. flexion according to the ROM test was assessed in two ways as mentioned above: flexion deficit and proximal flexion of the palmar crease. The HAMIS finger flexion item assesses an ability that reflects a combination of these 2 measurements, and this HAMIS item was therefore correlated with the summary of the 2 measurements of finger flexion in the ROM test. Second, in order to further test the Table 1. Range of motion among the patients (n 45) association between scores on HAMIS and ROM, the patients were grouped according to their HAMIS scores on each item. All patients who scored 0 on an item formed one subgroup, those who scored 1 formed another, and so on. Eight different groupings were done one for each HAMIS item where there was a corresponding ROM item. Subsequently, the Kruskal-Wallis H test was used to determine whether any differences were obtained between these HAMIS subgroups on the corresponding ROM scores. Discriminating ability for HAMIS was examined by testing for differences between the patient group and the comparison group of the 15 volunteers, using the Mann-Whitney U test. The ability of HAMIS to detect the patient group as well as the healthy individuals was further assessed by means of a forward stepwise logistic regression analysis. Data were analyzed by means of the SPSS software package, version 8.5 (SPSS, Chicago, IL). RESULTS Median (range) Normal Deficit of flexion, digits 2 5, mm 0.00 (0 45) 0.0* Proximal flexion of the palmar crease, digits 2 5, mm 20.00 (0 40) 0.2 Deficit of extension, digits 2 5, mm 1.25 (0 41) 0.3 Abduction, mm 130.00 (65 170) 156.0 Flexion of MCP 1 IP, degrees 90.00 (30 125) 140.0* Volar flexion, degrees 57.50 (0 80) 80.0* Dorsal flexion, degrees 60.00 (20 90) 70.0* Pronation, degrees 90.00 (60 92) 90.0* Supination, degrees 90.00 (70 90) 90.0* Skin score Skin score fingers 2.00 (0 3) 0 Skin score hand 1.00 (0 2) 0 Skin score arm 1.00 (0 3) 0 * Standards from American Academy of Orthopaedic Surgeons (ref. 9). Scores of a healthy group included in a previous study by Sandqvist (ref. 11). The normal skin constitutes the reference point and is set to 0. MCP metacarpophalangeal; IP interphalangeal. Hand function in the investigation group. Range of motion for the dominant hand is shown in Table 1. The patients pronation and supination of the forearm were almost normal, while the flexion abil-

Arthritis Care and Research Validity of HAMIS 385 Table 2. The distribution of scores (0 3) on HAMIS for the patient group (n 45) 0 1 2 3 Total Flexion 15 16 10 4 45 Extension 12 24 5 4 45 Abduction 19 21 5 45 Abduction 33 9 1 2 45 Pincer grip 29 12 4 45 Volar flexion 31 13 1 45 Dorsal extension 35 10 45 Pronation 44 1 45 Supination 27 18 45 ity of fingers and thumb, extension of the fingers, and volar flexion of the wrist were more affected. Applicability. HAMIS relevance and its ability to avoid the floor or ceiling effect for patients with SSc are demonstrated in Table 2, which shows the distribution of the patients scores on the different items of HAMIS. The most widely spread distribution of scores concerned the items finger flexion and finger extension. Scores above 0 were obtained for all items, and for the items finger flexion, finger extension, and thumb abduction all scores were represented, while there was just one score above 0 for the item pronation of the forearm. Three patients had totally normal HAMIS. Two of these also had normal ROM, while the third patient had abnormal ROM concerning finger flexion and extension. According to HAMIS, the patients range of motion was more limited in the fingers, and the patients mobility in pronation was almost normal. The healthy individuals in the comparison group obtained score 0 for all items, except for finger flexion (where one individual obtained score 1) and supination (where two individuals scored 1). Concurrent validity. Table 3 shows the correlations between the patients HAMIS scores and their ROM scores. The correlation was statistically significant for all items except pronation and supination of the forearm. As indicated above, a low HAMIS score denotes high functioning, and the same condition goes for finger flexion and finger extension in ROM. However, for the remaining ROM items the opposite is true, i.e., a high score indicates a high level of functioning. Thus, all the statistically significant correlations indicated associations between Table 3. Spearman s rank correlations between the patients HAMIS and ROM scores Flexion HAMIS ROM r s P value Extension Deficit of flexion proximal flexion of the palmar crease, mm Deficit of extension, digits 2 5, mm 0.79 0.000 0.80 0.000 Abduction Abduction, digit 1, mm 0.62 0.000 Pincer grip Flexion of MCP 1 and IP, 0.48 0.001 degrees* Volar flexion Volar flexion, degrees 0.50 0.001 Dorsal extension Dorsal flexion, degrees 0.40 0.008 Pronation Pronation, degrees 0.05 0.802 Supination Supination, degrees 0.34 0.059 * MCP metacarpophalangeal; IP interphalangeal. high functioning according to the respective measures. Table 4 shows correlations between HAMIS and skin score. There were no statistically significant correlations for pronation and supination of the forearm, but all the other HAMIS items were positively associated with skin score. Statistically significant differences were found between the HAMIS groupings on finger flexion (P 0.000), finger extension (P 0.000), finger abduction (P 0.008), thumb abduction (P 0.001), pincer grip of the thumb (P 0.006), volar flexion of the wrist (P 0.004), and dorsal extension of the wrist (P 0.009). This indicates that the variation in ROM Table 4. Spearman s rank correlations between the patients HAMIS and skin score HAMIS Skin score r s P value Flexion s 0.60 0.000 Extension s 0.52 0.000 Abduction s 0.47 0.001 Abduction s 0.31 0.038 Pincer grip s 0.36 0.016 Volar flexion Hand 0.33 0.029 Dorsal extension Hand 0.36 0.015 Pronation Hand 0.23 0.13 Supination Hand 0.01 0.93

386 Sandqvist and Eklund Vol. 13, No. 6, December 2000 scores within each grouping was less than the variation between the groups. No statistically significant differences were found on pronation of the forearm (P 0.80) or on supination of the forearm (P 0.06). Discriminating ability. Differences on HAMIS between healthy individuals and persons with scleroderma were used as indicators of the discriminating ability of HAMIS. There were statistically significant differences on finger flexion (P 0.000), finger extension (P 0.000), finger abduction (P 0.000), thumb abduction (P 0.027), pincer grip of the thumb (P 0.008), volar flexion of the wrist (P 0.015), and dorsal extension of the wrist (P 0.047). The differences were not statistically significant on pronation of the forearm (P 0.56) and supination of the forearm (P 0.06). The ability of HAMIS to distinguish patients with SSc from healthy individuals was further tested through a logistic regression analysis (P 0.0002), which resulted in the extension item as the only variable in the solution. This item predicted 100% of the healthy individuals and 73% of the patients in the correct groups. DISCUSSION This study demonstrated some good psychometric properties of HAMIS. It was found to be a valid test and to have good ability to discriminate patients with SSc from healthy individuals. There are times when HAMIS can replace the ROM test, e.g., to follow scleroderma s impact over time on the ability to move the hand, in surveying scleroderma patients ability to move the hand, and as a help when planning exercise programs. However, if there is need for a very careful and precise followup of treatment, an ROM test is more exact and ought to be used. HAMIS is a performance index that estimates mobility in the context of performance, not on the level of degrees or millimeters, and it is therefore not sensitive enough to record the very small variations detected in the ROM test. No correlations were found between HAMIS and ROM with respect to pronation and supination, which deviated from the result pattern of significant correlations found regarding the other items. The reason for this was probably that the patients range of motion in pronation and supination was good. The HAMIS items pronation and supination will still be kept in the instrument, because these items make the test complete for all movement lines, and they are important when using the hand in activities of daily living. Another argument for keeping these items is that they might separate patients with dssc from those with lssc, which was not investigated in this study because of the low number of patients with dssc. The pronation and supination should in the first place be related to skin score of the hand and arm, but this study could not demonstrate such a correlation (Table 4), possibly because there were too few patients with skin score 2 and 3. A disease-specific test should reflect the symptoms of the disease (12). HAMIS fulfills this demand since the correlations between HAMIS and skin score were statistically significant for all items except pronation and supination. The skin was more thickened on the fingers, the ROM was more limited on the fingers than on the wrist, and the HAMIS pointed to a substantial functional limitation of the fingers. This is in agreement with other studies, indicating that the most commonly affected functions are finger flexion, finger extension, and abduction of the thumb (4,13,14). The results from the logistic regression analysis, that finger extension as a single item could separate the patients from the healthy group, further underscored the fact that SSc is a disease that largely affects finger mobility. The patients ROM was very good in this study. This explains why some of the HAMIS items almost reached a floor effect, i.e., many of the patients reached score 0. An explanation for the patients good ROM could be that to a great extent they had received treatment early in the course of the disease and they were well informed about the importance of hand exercise. Melvin and colleagues recommended early treatment to prevent permanent limitation of range of motion (4). Because there was a tendency towards a floor effect on some of the items, the applicability of HAMIS has not been fully established, and the instrument needs to be tested on a sample with larger variation in hand function. HAMIS ability to discriminate between patients and healthy individuals was good despite the patients good range of motion in the hand. This indicates that HAMIS is sensitive even for relatively small limitations of ROM, but further studies comprising patients with greater variations of ROM are needed in order to estimate floor and ceiling effects. The aim of HAMIS is to reflect the mobility of the fingers, the wrist, and the forearm. There are of course other aspects of hand function that are of importance, e.g., dexterity. Many patients with SSc complain of fumbling, often caused by thickening of the skin, and this also has a great influence on the performance of daily occupations. A more complete estimation of the hand function of patients with SSc

Arthritis Care and Research Validity of HAMIS 387 should therefore consist of assessment of both mobility and dexterity. Therapists need simple and rapid tests that estimate function on the level of impairment but at the same time make it possible to estimate the ability to use the hand in the activities of daily living. HAMIS permits quantification based on a standardized evaluation procedure constructed on the basis of movements and objects used in daily living. It has a demonstrated concurrent validity set against ROM and skin score, and it discriminated between healthy individuals and persons with SSc. In an earlier study HAMIS demonstrated good interrater and intrarater reliability (8). Further studies are needed to estimate floor and ceiling effects and to test HAMIS sensitivity to outcome of intervention. HAMIS is mainly at the level of body functions and structure according to the ICIDH-2 (15), but it touches upon the level of activity, since it is a performance index composed of different-sized grips and different movements, all related to performance of daily occupations. There is therefore also a need for further research to investigate the relationships between HAMIS and tests of patients performance at the activity level. REFERENCES 1. Silman A, Åkesson A, Newman J, Henriksson H, Sandqvist G, Nihill M, et al. Assessment of functional ability in patients with scleroderma: a proposed new disability assessment instrument. J Rheumatol 1998; 25:79 83. 2. Poole JL, Watzlaf VJM, D Amico F. Hand risk factors for development of disability in scleroderma. Arthritis Rheum 1996;39:S312. 3. McPhee SD. Functional hand evaluations: a review. Am J Occup Ther 1987;41:158 63. 4. Melvin JL. Rheumatic disease in the adult and child: occupational therapy and rehabilitation. 3rd ed. Philadelphia: FA Davis; 1987. 5. Eberhard KB, Svensson B, Moritz U. Functional assessment of early rheumatoid arthritis. Br J Rheumatol 1988;27:364 71. 6. Merkel PA. Measurement of functional status, selfassessment, and psychological well-being in scleroderma. Curr Opin Rheumatol 1998;10:589 94. 7. Sandqvist G, Eklund M. Hand Mobility in Scleroderma (HAMIS): the reliability of a novel hand function test. Arthritis Care Res 2000;13:369 74. 8. Kalla AA, von Katze TJW, Meyers OL, Parkyn ND. Clinical assessment of disease activity in rheumatoid arthritis: evaluation of a functional test. Ann Rheum Dis 1988;47:773 9. 9. American Academy of Orthopaedic Surgeons. Joint motion: method of measuring and recording. London: Churchill Livingstone; 1966. 10. Clements PJ, Lachenbruch PA, Seibold JR, Zee B, Steen VD, Brennan P, et al. Skin thickness score in systemic sclerosis: an assessment of interobserver variability in 3 independent studies. J Rheumatol 1993;20:1892 6. 11. Sandqvist G. Kartläggning av greppförmågan hos patienter med sklerodermi. Hur och varför påverkas vardagsaktiviteter [Delineating grip ability among patients with scleroderma: how and why daily activities are effected]. Unpublished paper; 1995. 12. Wade DT. Measurement in neurological rehabilitation. Oxford: Oxford University Press; 1996. 13. Poole JL, Steen VD. The use of the Health Assessment Questionnaire (HAQ) to determine physical disability in systemic sclerosis. Arthritis Care Res 1991;5:27 31. 14. Poole JL. Grasp pattern variations seen in the scleroderma hand. Am J Occup Ther 1994;48:46 54. 15. ICIDH-2: International classification of functioning and disability. Beta-2 draft, short version. Geneva: World Health Organization; 1999.