Reliability and validity of the Duruöz Hand Index in persons with systemic sclerosis (scleroderma)

Authors


Abstract

Objective

To examine the test–retest reliability and concurrent validity of the Duruöz Hand Index (DHI) in persons with scleroderma.

Methods

Forty patients with scleroderma and no other major medical problems completed the DHI at 2 points in time to assess test–retest reliability. To assess validity, participants were administered the Arthritis Hand Function Test (AHFT), the Hand Mobility in Scleroderma Test (HAMIS), and the Keital Function Test (KFT), all performance-based tests. In addition, participants completed the Health Assessment Questionnaire (HAQ), a self report of functional ability.

Results

Test–retest reliability intraclass correlation coefficients for the DHI ranged from 0.81 to 0.97. Scores on the DHI did not correlate with HAMIS scores, but were significantly correlated with scores from the KFT (rs = 0.48, P < 0.01), HAQ (rs = 0.79, P < 0.01), and all sections of the AHFT (rs = 0.34–0.60, P < 0.05–0.01).

Conclusion

The results from this study show the DHI to be a reliable and valid test for persons with scleroderma.

INTRODUCTION

Systemic sclerosis (SSc; scleroderma) involves degenerative, inflammatory, and fibrotic changes in the skin, blood vessels, joints, tendons, skeletal muscles, and some internal organs. There are two classifications of SSc: limited cutaneous SSc (lSSc) and diffuse cutaneous SSc (dSSc). In lSSc or CREST syndrome (calcinosis, Raynaud's phenomenon, esophageal dysmotility, sclerodactyly, telangiectasias), slow and stable skin thickening occurs only in the distal extremities and face/neck. Visceral or internal organ involvement is observed in lSSc; however, this occurs later in the disease progression (1). In dSSc, rapid progression of skin thickening from distal to proximal is observed throughout the body; and unlike lSSc, severe visceral involvement occurs early in the course of the disease (1). Regardless of type of SSc, individuals experience hand involvement. There is range of motion limitation or loss in all planes of wrist motion; metacarpophalangeal (MCP) flexion; proximal interphalangeal (PIP) extension; and thumb abduction, opposition, and flexion. The distal interphalangeal joint may become fixed in midrange flexion (2, 3). The result is a claw type deformity with MCP extension, PIP flexion, thumb adduction, and the wrist in neutral position (2, 3).

Hand involvement in both lSSc and dSSC is a serious concern. Poole and Steen (4) suggested that hand involvement in SSc leads to functional disability based on correlations between disease status and grip strength and wrist and finger motion. However, their study did not use tasks of daily living to assess function, but relied on measures of strength and motion to imply functional limitations. In spite of the prevalence of hand involvement in SSc, there are only 2 assessments of hand function that have been found to be reliable and valid in persons with SSc: the Hand Mobility in Scleroderma Test (HAMIS) (5, 6) and the Arthritis Hand Function Test (AHFT) (7–9). Both are performance-based tests, require some training to administer, and require an extensive amount of easily obtainable equipment (5, 8). Because of these requirements, plus training in their use, they may be too complicated for routine use. Therefore, self-report questionnaires, which have been used extensively to assess general functional status in regards to abilities of daily living (10), may be appropriate to assess hand function. A recently developed self report of hand abilities, the Duruöz's Hand Index (DHI), shows promise as a reliable and valid assessment for hand function (11). The DHI (also called the Hand Functional Disability Scale and Cochin Scale) contains 18 items regarding hand ability in the kitchen, during dressing, while performing personal hygiene, while performing office tasks, and other general items. Persons rate their ability from 0 (no difficulty) to 5 (impossible to do). The questionnaire yields a score from 0 to 90 and takes about 3 minutes to complete. The DHI has been found to be reliable and valid in persons with 2 other rheumatic diseases: rheumatoid arthritis (11) and osteoarthritis (12). Thus, it may be useful to measure hand function in persons with SSc. Therefore, the purpose of this study was to examine the test–retest reliability and the concurrent validity of the DHI in persons with SSc (scleroderma).

PATIENTS AND METHODS

Patients.

This study included 40 participants who had been diagnosed with SSc according to the criteria for SSc subsets (13). Participants were excluded if they had >1 rheumatic disease or other chronic diseases, such as cardiac conditions, stroke, brain injury, or Parkinson's. The sample consisted of 34 women and 6 men, 35 of whom were right handed and 5 of whom were left handed. They ranged in age from 26 years to 74 years with the mean age of 53 years. Twenty-two of the participants in this study were classified as having lSSc, 15 had dSSc, and 3 had unclassified disease. Disease duration ranged from 3 months to 35 years with a mean duration of 11 years.

Measures.

Arthritis Hand Function Test.

The AHFT is a performance-based test examining the ability to use one's hands during daily life tasks. This standardized test consists of 11 items including grip and pinch strength, dexterity (pegboard), applied dexterity (buttoning, lacing a shoe and tying a bow, putting coins into a slot, opening and closing safety pins, and cutting with a knife and fork), and applied strength (lifting a tray of tin cans and pouring water) (8). The scores for the dexterity and applied dexterity items are based on the time it takes to complete each task. The strength items are measured using the following units: grip and pinch strength in pounds, number of cans, and milliliters of water lifted (8). The AHFT has been shown to be reliable and valid for use with persons with SSc (9).

Keital Functional Test (KFT).

The KFT was used to assess joint limitations. In this study, only the 11 upper-extremity items were used; these measure finger flexion, wrist extension and flexion, forearm pronation and supination, elbow flexion, and shoulder flexion (14). Scoring is done separately for each extremity, with the score ranging from 0 to 26. On the KFT, lower scores indicate less range of motion impairment. Literature has shown the KFT to be reliable and valid (14, 15).

Hand Mobility in Scleroderma test.

The HAMIS is one performance test that specifically examines the effects of SSc on hand function (5, 6). The HAMIS consists of 9 items: finger flexion and extension, abduction of the thumb, pincer grip, finger abduction, wrist flexion and extension, and forearm pronation and supination. Each item is scored according to prescribed criteria on a scale from 0 (no impairment) to 3 (cannot do), which yields a total possible score of 27 for each hand. The HAMIS was found to be a reliable and valid test to assess hand function in person with SSc (5, 6).

Health Assessment Questionnaire (HAQ).

The HAQ, a self-report measure, consists of 8 categories of daily living: dressing and grooming, arising, eating, walking, hygiene, reach, grip, and outside activity (16). Persons indicate how much difficulty they have with each item from 0 (no difficulty) to 3 (cannot do). A score for each category is the highest score for any question in the category. A disability index is calculated by adding the scores from each category and dividing by the number of categories answered. This yields a disability index from 0 (less disabled) to 3 (more disabled). Validity of the HAQ for use with clients with SSc has been established (17).

Skin scores.

The modified Rodnan skin score method was used to measure disease severity (18). A skin score is obtained by palpating the skin and rating these areas on a scale of 0 (normal skin thickness) to 3 (severe skin thickness). The higher the skin scores, the more severe the skin involvement in SSc (18). In the present study, only skin scores for the finger, hand, and forearm were obtained. Studies have shown this method of assessing skin involvement to be a valid (19).

Methods.

To determine test–retest reliability of the DHI, participants were asked to complete the DHI a second time within 1 week of the initial testing session and to return the completed form to the primary investigator. Upon return of the envelope to the primary investigator, the participants were paid $10 for their time and participation.

To determine the concurrent validity of the DHI, participants completed the 3 performance tests of hand function (AHFT, KFT, and HAMIS) and the HAQ. Disease severity was also assessed. The total testing time took ∼30–45 minutes. One rater (LMB), trained in the use of all these instruments, tested all the participants in the study. Intrarater agreement was 98%.

Data analysis.

Intraclass correlation coefficients (ICCs) type 3.1 were calculated to examine the test–retest reliability of the DHI (20). Spearman rho correlation coefficients were calculated to estimate the concurrent validity of the DHI with the AHFT, KFT, HAMIS, HAQ, and disease severity. Time 1 scores for the DHI were used because the other measures were also completed at time 1. A value P < 0.05 was considered statistically significant.

RESULTS

Table 1 shows the descriptive statistics for the performance of 40 participants on the DHI. The scores on the DHI suggest mild to moderate impairments in hand function in individuals with SSc. Because the 2 classifications of SSc display different rates of progression and disability (1), we thought it was important to compare the diffuse and limited SSc participant scores on the DHI (Table 2). The 3 individuals who were unclassified were not included in this analysis. As expected, the participants with diffuse SSc showed higher mean scores on all sections of the DHI than the limited SSc group, suggesting greater impairment of hand function in that group (Table 2). To compare the scores of the individuals with diffuse and limited SSc, t-tests were performed. No significance differences were found between the 2 groups on the DHI at time 1. At time 2, there were no differences between the 2 groups, except for the hygiene and office sections.

Table 1. Descriptive statistics for the DHI for time 1 (n = 40)*
DHI itemMean ± SD (range)
  • *

    DHI = Duruöz Hand Index.

Kitchen (0–40)9.95 ± 9.29006 (0–31)
Dress (0–10)2.25 ± 2.52932 (0–8)
Hygiene (0–10)1.68 ± 1.99213 (0–9)
Office (0–10)2.00 ± 2.26455 (0–8)
Other (0–20)4.98 ± 4.94839 (0–19)
Total (0–90)21.10 ± 19.25244 (0–66)
Table 2. Means, standard deviations, and significant differences for both trials of the DHI for diffuse and limited systemic sclerosis*
DHITime 1 (n = 40)Time 2 (n = 37)
lSScdSScPlSScdSScP
mean ± SDmean ± SDmean ± SDmean ± SD
  • *

    DHI = Duruöz Hand Index; lSSc = limited systemic sclerosis, dSSc = diffuse systemic sclerosis; NS = not significant.

Kitchen (0–40)8.68 ± 9.8812.00 ± 7.92NS6.10 ± 8.8212.23 ± 8.27NS
Dressing (0–10)2.05 ± 2.342.53 ± 2.83NS1.38 ± 1.912.53 ± 3.26NS
Hygiene (0–10)1.41 ± 2.111.93 ± 1.62NS0.57 ± 1.122.08 ± 1.750.028
Office (0–10)1.41 ± 2.042.93 ± 2.43NS0.95 ± 1.473.00 ± 2.000.007
Other (0–20)4.59 ± 5.435.53 ± 3.83NS3.24 ± 4.606.00 ± 4.14NS
Total (0–90)18.14 ± 20.2025.60 ± 16.22NS12.24 ± 17.1025.85 ± 17.85NS

The test–retest reliability of the DHI was calculated using 37 subjects because 3 subjects did not return the DHI for time 2. According to Portney and Watkins (20), the general guidelines for ICC values are as follows: values >0.75 represent good reliability and values <0.75 represent poor to moderate reliability. The ICCs for the test–retest reliability of the DHI between time 1 and time 2 were found to be excellent for all sections (Table 3). Each section showed ICC ≥ 0.92, except hygiene. The kitchen section had the highest ICCs at 0.97 and hygiene had the lowest at 0.81.

Table 3. Test–retest reliability of the DHI*
DHI items (n = 37)ICC
  • *

    DHI = Duruöz Hand Index; ICC = intraclass correlation coefficient.

Kitchen0.97
Dressing0.92
Hygiene0.81
Office0.96
Other0.92
Total0.97

Table 4 displays the descriptive statistics for the AHFT, HAQ, KFT, HAMIS, and skin scores. Because the AHFT has so many items, items were combined to create section totals for strength, dexterity, applied dexterity, and applied strength. The strength total was determined for each hand by summing the grip, 2-point pinch, and 3-point pinch items, which had been converted to pounds. The sum of all applied dexterity items was calculated to create an overall total in seconds for this section. To sum the 2 items in the applied strength section, the results of lifting cans was converted to milliliters. This was done by converting the number of ounces of a soup can to milliliters and multiplying by the number of cans lifted. The score for this item was added to the number of ml of water in the pitcher to yield a total applied strength score. The scores for the AHFT suggest moderate impairment of hand function when compared with scores obtained from adults who are nondisabled, as reported in the AHFT manual (7). Disease severity, assessed through skin scores (Table 4) of the fingers, back of hand, and forearm, also suggested mild to moderate disease severity. Scores on the KFT and HAMIS suggest minimal joint limitations. There were no significant differences between the scores for the lSSc and dSSc groups for the AHFT, HAQ, KFT, HAMIS, or skin scores. Therefore, the groups were not considered separately for the correlational analysis.

Table 4. Descriptive statistics for the performance of 40 subjects on the HAQ, KFT, HAMIS, Skin Scores, and the AHFT*
TestScore mean ± SD (range), n = 40
  • *

    HAQ = Health Assessment Questionnaire; KFT = Keital Functional Tests; HAMIS = Hand Mobility in Scleroderma test; AHFT = Arthritis Hand Function Test.

HAQ (0–3)1.03 ± 0.69 (0–2.75)
KFT total (0–52)18.15 ± 13.83 (4–50)
HAMIS total (0–54)10.93 ± 10.93 (0–40)
Skin score total (0–18)9.20 ± 5.50 (0–18)
AHFT sections 
 Strength total, pounds132.56 ± 46.50 (55.7–218.8)
 Dexterity total, seconds50.92 ± 12.39 (35–52)
 Applied dexterity total, seconds163.94 ± 62.63 (89–353)
 Applied strength total, milliliters5,012.53 ± 1,186.37 (2,271.51–5,814.53)

Table 5 shows the results from the correlation analysis. The DHI correlated significantly with the KFT, all sections of the AHFT, and the HAQ. Finally, the low correlations between the DHI and the HAMIS and skin scores were not statistically significant.

Table 5. Concurrent-validity of the DHI (n = 40)*
 DHI (Spearman rho)P
  • *

    DHI = Duruöz Hand Index, AHFT = Arthritis Hand Function Test; HAQ = Health Assessment Questionnaire; KFT = Keital Functional Test; HAMIS = Hand Mobility in Scleroderma Test; NS = not significant.

AHFT sections  
 Strength total, pounds−0.580.01
 Dexterity total, seconds0.390.05
 Applied dexterity, seconds0.430.01
 Applied strength, milliliters−0.580.01
HAQ0.790.01
KFT total0.480.01
HAMIS total0.28NS
Skin score total0.38NS

DISCUSSION

Due to time constraints, the need for equipment, and training, performance-based tests of hand function are not often used as outcomes in clinical trials or other studies of therapeutic interventions with SSc (21). Therefore, a self report of hand function that is reliable and valid, such as the DHI, may be useful as a screening tool to determine whether more detailed hand function assessment is necessary or as an outcome when examining the effectiveness of clinical and therapeutic interventions. Excellent test–retest reliability was found for 2 trials of the DHI. It is unclear why the ICC for the hygiene section was lower than the other sections. Previous studies of the DHI did not look at the reliability of individual sections.

Scores on the DHI showed no significant correlation with the HAMIS, 1 of 2 assessments of hand involvement previously reported to be reliable and valid for persons with SSc. This is not surprising because the HAMIS is a test of joint motion, not a test of one's ability to use their hands. However, what is interesting is the DHI did correlate with the KFT, another joint motion test, which has not been previously validated for persons with SSc. The DHI also correlated with the KFT in persons with rheumatoid arthritis (12). The DHI also did not correlate significantly with skin scores, agreeing with findings from other studies (4).

The DHI scale significantly correlated with all sections of the performance assessment of hand function, the AHFT. The correlations were good to moderate for the strength, applied dexterity, and applied strength sections of the AHFT. The dexterity section showed the weakest correlation with poor to moderate correlations. The dexterity item is the 9-hole peg test, which unlike the majority of the AHFT, is a nonfunctional and unfamiliar task. The rater noticed that the subjects put less effort into this item than other items, such as lacing a shoe or cutting with a knife and fork.

The strong correlation found between the DHI and the HAQ was expected because both questionnaires are self reports of functional ability. However, some of the items on the HAQ are not considered major problems by persons with SSc (4, 22). Thus, the benefit of the DHI is that it is hand-function specific.

The 2 classifications of SSc are very different in disease progression and effect; therefore, we felt it was important to look for significant differences between the 2 groups on the DHI. As was expected, individuals with dSSc reported more hand impairment on the DHI; however, there was no significant difference in scores for the first trial. Furthermore, there were no significant differences between the 2 groups for any of the other measures of hand strength, dexterity, joint motion, or disease severity. At time 2 for the DHI, the group with dSSc reported slightly higher hand impairment than at time 1. Participants with lSSc reported less hand impairment at time 2 of the DHI than at time 1. The DHI scores of participants in this study were higher than scores of person with rheumatoid arthritis (DHI = 17.17–17.29) (11) and osteoarthritis (DHI = 18.73) (12), indicating more hand disability. However, similar to these other studies on the DHI, impairment measures were generally not related to DHI scores. Assessment at the impairment level of the hand should be complimented by an evaluation at the activity level (International Classification of Functioning, Disability and Health) (23) because individuals may develop their own way of compensating.

A limitation of the present study is that the majority of the participants did not have severe joint limitations or severely impaired hand function. The geographic location of the sample in this study is also a limitation and may not be representative of the overall population with SSc.

Excellent interrater reliability has also been reported with the DHI when used with individuals with osteoarthritis and rheumatoid arthritis (11, 12). In addition to good to excellent psychometric properties, the DHI assesses at the level of activity rather than at the level of body function and structure (23). Because measures of body function and structure do not always reflect functional ability (24), the DHI may provide a more accurate measure of hand function. In conclusion, the DHI is a reliable and valid assessment of hand function at the activity level in persons with SSc. Future studies might want to examine the DHI's responsiveness and sensitivity to change in persons with SSc.

Ancillary