PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of handHand
 
Hand (N Y). 2009 December; 4(4): 362–367.
Published online 2009 April 21. doi:  10.1007/s11552-009-9191-x
PMCID: PMC2787220

The Validity, Reliability, Responsiveness, and Bias of the Manchester-Modified Disability of the Arm, Shoulder, and Hand Score in Hand Injuries

Abstract

The Manchester-Modified Disability of Arm, Shoulder and Hand questionnaire (M2 DASH) was developed by the authors as a modification to the original DASH questionnaire. In this study, we assessed the validity, reliability, responsiveness, and bias of the M2 DASH questionnaire for hand injuries using completed M2 DASH, Patient Evaluation Measure, and Michigan Hand Outcome questionnaires from 40 patients. The M2 DASH scores showed significant positive correlations with the Patient Evaluation Measure and Michigan Hand Outcome scores suggesting validity. There was also no evidence of a statistical difference in the M2 DASH scores when the condition had stabilized suggesting good test–retest reproducibility and reliability. The effect size and the standardized response mean for the M2 DASH score were greater than those for the Patient Evaluation Measure and Michigan Hand Outcome scores establishing that the M2 DASH is highly responsive. There was no gender, hand dominance, or dominant side injured bias for the M2 DASH score. There was, however, a relatively weak association between age and the M2 DASH score at presentation. We conclude that the M2 DASH questionnaire is a robust region-specific outcome measure. It is a valid and responsive questionnaire with test–retest reliability proven for hand injuries in this study. Gender, handedness, and side injured did not cause bias in the responses.

Keywords: M2 DASH, Validity, Reliability, Responsiveness, Hand injuries

Introduction

Outcome measures are important as they allow clinicians to assess the severity of injuries and disease and allow a comparison between different forms of management. A patient-completed questionnaire is efficient in terms of time and resources and allows the assessment of outcome without the need to attend an outpatient clinic.

The original DASH questionnaire was developed as a measure specific to the upper limb, but recent studies have shown that it also takes lower limb function into account [5, 6]. This implies that the original DASH questionnaire is not suited for patients with additional lower limb pathologies, or patients with polyarthropathies. The original DASH questionnaire is also limited by a high internal consistency suggesting redundancy in questions [4], difficulty with reproduction of the four pages it spans, time restraints for the patient and the clinician, and confusion over optional modules. These issues prompted the authors to create a modified shorter version, the Manchester-Modified Disability of Arm, Shoulder, and Hand questionnaire (M2 DASH) that is more sensitive and specific for the upper limb [1, 6]. The M2 DASH has a simple one-page layout and is easy to understand and complete for the patient. For the researcher, it is easy to enter into a database and to calculate and analyze the score.

It is important to establish the validity, reliability, responsiveness, and bias of any outcome measure. The validity establishes whether the outcome measure actually measures what it was designed to. An outcome tool is reliable if the same result is obtained when tested at different time points once the condition has stabilized. The responsiveness is the ability to detect clinically important changes in disability at various intervals. Bias can occur when assumed independent variables such as age and gender affect responses.

There are many patient-completed questionnaires available for the assessment of outcomes in the upper limb. The Patient Evaluation Measure (PEM) questionnaire was developed in the UK in 1995 [7], and the Michigan Hand Outcome (MHO) questionnaire was developed in the USA in 1998 [3]. Both questionnaires are region-specific outcome measures commonly used for hand injuries and, like the M2 DASH, are patient-completed questionnaires.

Previously, the only assessment of the validity of the M2 DASH questionnaire was by comparison with the original DASH questionnaire from which the M2 DASH scores had been extrapolated. It is important that a more thorough evaluation of this questionnaire is performed. In this study, we assessed the validity, reliability, responsiveness, and bias of the M2 DASH questionnaire for hand injuries using completed M2 DASH, PEM, and MHO questionnaires.

Patients and Methods

Between October 2006 and August 2007, we studied 59 patients who attended the fracture clinics of two local district general hospitals with hand injuries. Patients under 18 years of age, with preexisting disorders of the upper limb, with cognitive impairment preventing the completion of the questionnaires, or with language difficulties, were excluded. The M2 DASH (Fig. 1), PEM, and MHO questionnaires were completed by the patients at their first clinic visit, on discharge following their last visit, and at 6 months following discharge. The M2 DASH was also completed by the patients 12 months following discharge. Thirteen patients who started completing the questionnaires at presentation but failed to complete all sets of questionnaires at the various time points were excluded from the study. This included four patients where there were omissions from the questionnaires invalidating them. Six patients who started completing the questionnaires but sustained a further upper limb injury or had to seek further medical advice for their original injury following discharge were also excluded.

Figure 1
The M2 DASH questionnaire.

The completed questionnaires from 40 patients at the different time points were used to assess the validity, reliability, responsiveness, and bias of the M2 DASH questionnaire. Of the 40 patients for whom all sets of questionnaires were available, there were 24 men and 16 women. Their mean age was 43 years (standard deviation (SD) 18 years, range 18 to 85). Thirty-one were right-handed and nine were left-handed. Twenty-six patients had injured their dominant hand and 14 patients their nondominant hand. The M2 DASH, PEM, and MHO scores were then calculated as previously described [1, 3, 4, 7] giving a score out of 100.

The validity of the M2 DASH was assessed by determining how well the score at various time points correlated with the PME and MHO scores. The reliability was assessed by performing test–retest reproducibility analyses on the M2 DASH scores at 6 and 12 months after discharge; 6 months after discharge, as the patients had not sustained a further injury or sought further medical advice, it was assumed that that their condition had stabilized. A paired t test was used to check if the patients score had changed significantly over this time period.

Responsiveness was assessed by correlating the changes in the M2 DASH score with changes in the PEM and MHO scores and was measured by calculating the effect size and the standardized response mean. The effect size was calculated by considering the two periods between the three time points where there were changes in the scores: initial presentation, discharge, and 6 months following discharge. The effect sizes were calculated by dividing the mean of change in the score during the periods by the standard deviation of the baseline score. The standardized response mean was calculated by dividing the mean of change in the scores between initial presentation and 6 months following discharge by the standard deviation of the change in score.

The bias in the M2 DASH score was investigated by performing correlation studies for assumed independent variables including age, gender, hand dominance, and the side injured. Lastly, the values for the three questionnaires were plot against one another on scatter plots to further understand the relationships between them. All statistical analyses were performed on SPSS version 12.0 (SPSS Inc., Chicago, IL, USA) and a p value of <0.05 was taken as statistically significant.

Results

All the scores were normally distributed when assessed using histograms and Pearson correlations were used to compare them and assess validity of the M2 DASH score. The correlation between the M2 DASH scores and each of the PEM and MHO scores was calculated at presentation, at discharge and 6 months following discharge. The M2 DASH scores compared well with the PME and MHO scores. Table 1 shows that the M2 DASH scores showed highly significant positive correlations with both the PEM and MHO scores at all three time points (p < 0.001). The magnitude of the correlation coefficient was slightly lower with the MHO scores and at 6 months following discharge.

Table 1
The correlations of PEM and MHO scores with the M2 DASH scores.

Comparison of the M2 DASH scores 6 and 12 months after discharge shows no significant change over this period (p = 0.53). The mean change over the 6 months was −0.2 (−0.84, 0.44). This is less than 5% of the mean M2 DASH scores at 6 [4.83 (3.92, 5.73)] and 12 months [5.03 (4.19, 5.86)]. These results suggest good test–retest reproducibility and reliability.

Table 2 shows that the effect sizes for the M2 DASH were 1.45 and 1.07 for the time periods between presentation and discharge and between discharge and 6 months following discharge, respectively. These were greater than 0.8 indicating that they were valuable. The effect sizes for the two periods were greater than those for the PME (1.26 and 0.96) and MHO (1.23 and 0.60). The effect size for the MHO scores was not valuable for the second period, i.e., between discharge and 6 months following discharge. The standardized response mean for the M2 DASH score was 2.21. The standardized response mean for the PME and MHO scores were 1.80 and 1.50, respectively. These analyses establish that the M2 DASH is a highly responsive scale.

Table 2
The effect sizes and standardized response means for the M2 DASH, PME, and MHO scores.

There was evidence of a relatively weak association between age and the M2 DASH score at presentation (p = 0.02) but the correlation coefficient was not large (r = 0.37). At 12 months following discharge, there was no correlation between age and the M2 DASH score (r = −0.09, p = 0.57). Table 3 shows that there was no evidence of an association between the M2 DASH scores and gender, dominance, and injury to dominant side when we looked at the scores at presentation and 12 months following discharge.

Table 3
The association between the M2 DASH score and gender, dominance, and injury to dominant side.

When the scores were plot against each other (Fig. 2), the M2 DASH and PEM scores showed a fairly linear relationship. The MHO score, when compared to the other two scores, had a more sigmoid curve.

Figure 2
Scatter plots for M2 DASH and PEM scores (a), M2 DASH and MHO scores (b), and PEM and MHO scores (c).

Discussion

The M2 DASH is an upper limb outcome assessment tool that is not pathology or region specific within the upper limb. Previously, the only assessment of the validity of the M2 DASH questionnaire was by comparison with the original DASH questionnaire. In this study, we assessed the validity, reliability, responsiveness, and bias of the M2 DASH questionnaire for hand injuries using completed M2 DASH questionnaires and comparing it with the PEM and MHO questionnaires.

In this study, criterion validity testing showed strong correlations between the M2 DASH scores and the PME and MHO scores suggesting that the M2 DASH score measures what it is designed to and confirmed that the questionnaire is valid for injuries to the hand. The construct validity of the M2 DASH questionnaire appears sound as the questionnaire retains at least half of the questions from each of the six domains of the original DASH questionnaire [6]. Each individual question contributes to the complete evaluation of hand function. The M2 DASH includes some questions not routinely addressed in the objective assessments of the hand, suggesting good content or face validity of the M2 DASH questionnaire.

We have established that the M2 DASH is reliable with good test–retest reproducibility as there was no significant change in the scores at 6 and 12 months following discharge. Our study has also established that the M2 DASH questionnaire is responsive to change compared with the PME and MHO questionnaires over a period of time, from initial presentation following injury through discharge from the fracture clinic and to follow-up at 6 months.

Our study confirmed that no gender, hand dominance, or dominant side injured bias existed for the M2 DASH score at presentation or at 6-month follow-up. There was, however, a relatively weak association between age and the M2 DASH score at presentation. The positive correlation implies that older people tend to have high DASH scores when they present with hand injuries. This could be due to older patients sustaining more severe hand injuries or sustaining injuries of similar severity but suffering a greater disability. Although the latter one is more likely, further work is needed to explore this in more detail.

When the scores were plot against each other, the relationships of the MHO score with the other two scores was more sigmoid shaped than linear suggesting that at the extremes, the MHO scores were clustered and not spread out like the M2 DASH and PEM scores. This implies that the MHO score is less sensitive at both extremes.

The QuickDASH questionnaire is a shorter version of the DASH questionnaire developed using the concept-retention method [2]. The QuickDASH questionnaire remains limited in terms of sensitivity and specificity for the upper limb with only six of its 11 questions being present in the M2 DASH questionnaire [6]. Unlike the PEM and MHO questionnaires, the QuickDASH has not been validated for hand injuries and was not included in this study.

We conclude that the M2 DASH questionnaire is a robust region-specific outcome measure. It is a valid and responsive questionnaire with test–retest reliability proven for hand injuries in this study. Gender, handedness, and side injured did not cause bias in the responses. We acknowledge the need for further studies looking beyond hand injuries before a more complete picture on the validity of the M2 DASH for all upper limb pathologies could be drawn.

References

1. Beaton DE, Katz JN, Fossel AH. Measuring the whole or the parts? Validity, reliability and responsiveness of the Disabilities of the Arm, Shoulder, and Hand Outcome Measure in different regions of the upper extremity. J Hand Ther. 2001;14:128–46. [PubMed]
2. Beaton DE, Wright JG, Katz JN, Upper Extremity Collaborative Group. Development of the QuickDASH: comparison of three item-reduction approaches. J Bone Joint Surg Am. 2005;87:1038–46. doi:10.2106/JBJS.D.02060. [PubMed]
3. Chung KC, Pillsbury MS, Walters MR. Reliability and validity testing of the Michigan Hand Outcomes questionnaire. J Hand Surg. 1998;23A:575–87. [PubMed]
4. Dias JJ, Rajan RA, Thompson JR. Which questionnaire is best? The reliability, validity and ease of use of the patient evaluation measure, the disabilities of the arm, shoulder and hand and the Michigan hand outcome measure. J Hand Surg. 2008;33B:9–17. [PubMed]
5. Dowrick AS, Gabbe BJ, Williamson OD, Cameron PA. Does the disabilities of the arm, shoulder and hand (DASH) scoring system only measure disability due to injuries to the upper limb? J Bone Joint Surg. 2006;88B:524–7. [PubMed]
6. Khan WS, Jain R, Dillon B, et al. The ‘M(2) DASH’—Manchester-Modified Disabilities of Arm Shoulder and Hand Score. Hand. 2008;3:240–4. doi:10.1007/s11552-008-9090-6. [PMC free article] [PubMed]
7. Macey AC, Burke FD. Outcomes of hand surgery. J Hand Surg. 1995;20B:841–55. [PubMed]

Articles from Hand (New York, N.Y.) are provided here courtesy of American Association for Hand Surgery