Development and validation of the ankle fracture outcome of rehabilitation measure (A-FORM).

[

]

research report

Journal of Orthopaedic & Sports Physical Therapy® Downloaded from www.jospt.org at Otterbein University on July 1, 2014. For personal use only. No other uses without permission. Copyright © 2014 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

STEVEN M. MCPHAIL, PT, PhD1-3 • CYLIE M. WILLIAMS, Pod, PhD4,5 • MICHAEL SCHUETZ, FRACS, MD, PhD3,6 BRETT BAXTER, PT6 • PETER TONKS, PT6 • TERRY P. HAINES, PT, PhD4,5

Development and Validation of the Ankle Fracture Outcome of Rehabilitation Measure (A-FORM)

T

raumatic musculoskeletal injuries are a common problem and may result in short- or long-term disability.12,50,55,65,69 Ankle fractures are among the most common traumatic injuries requiring orthopaedic management.11,27,49 Despite their high

incidence, ankle fractures have been the subject of less empirical research in comparison to other fracture types.1,15,23,24,59,68 TTSTUDY DESIGN: Delphi panel and cohort study. TTOBJECTIVE: To develop and refine a condition-

specific, patient-reported outcome measure, the Ankle Fracture Outcome of Rehabilitation Measure (A-FORM), and to examine its psychometric properties, including factor structure, reliability, and validity, by assessing item fit with the Rasch model.

TTBACKGROUND: To our knowledge, there is no patient-reported outcome measure specific to ankle fracture with a robust content foundation.

TTMETHODS: A 2-stage research design was

implemented. First, a Delphi panel that included patients and health professionals developed the items and refined the item wording. Second, a cohort study (n = 45) with 2 assessment points was conducted to permit preliminary maximumlikelihood exploratory factor analysis and Rasch analysis.

TTRESULTS: The Delphi panel reached consensus on 53 potential items that were carried forward to the cohort phase. From the 2 time points, 81 questionnaires were completed and analyzed;

Prior investigations among patients following ankle fractures have reported substantial variability in the prevalence 38 potential items were eliminated on account of greater than 10% missing data, factor loadings, and uniqueness. The 15 unidimensional items retained in the scale demonstrated appropriate person and item reliability after (and before) removal of 1 item (anxious about footwear) that had a higher-than-ideal outfit statistic (1.75). The “anxious about footwear” item was retained in the instrument, but only the 14 items with acceptable infit and outfit statistics (range, 0.5-1.5) were included in the summary score.

TTCONCLUSION: This investigation developed and refined the A-FORM (Version 1.0). The A-FORM items demonstrated favorable psychometric properties and are suitable for conversion to a single summary score. Further studies utilizing the A-FORM instrument are warranted. J Orthop Sports Phys Ther 2014;44(7):488-499. Epub 22 May 2014. doi:10.2519/jospt.2014.4980 TTKEY WORDS: internal consistency, orthopaedics, patient-reported outcome, quality of life, Rasch

of negative patient outcomes. Some studies have reported that up to 87% of patients experience good to excellent clinical recovery following ankle fracture,5,14,25,27,36 whereas other investigations have indicated that fewer than half of patients experience complete recovery and report ongoing problems with physical symptoms,22 psychological health,59 and performing social or recreational activities.49,59 In addition to the uncertainty surrounding the frequency of poor outcomes for patients following ankle fracture, a Cochrane systematic review found limited available evidence to inform specific rehabilitation protocols for the management of ankle fracture.30 This review called for additional well-designed and appropriately powered research to document patient outcomes following ankle fracture and to evaluate interventions. Advancing this research agenda is important, given that insufficient or suboptimal rehabilitation has been cited as a potential cause of long-term disability following ankle fracture.46 A limiting factor when evaluating the effectiveness of ankle fracture rehabilitation interventions is the absence of a suitable ankle fracture–specific, patient-

1 Centre for Functioning and Health Research, Metro South Hospital and Health Service, Brisbane, Australia. 2School of Public Health and Social Work, Queensland University of Technology, Brisbane, Australia. 3Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia. 4Allied Health Research Unit, Monash Health, Cheltenham, Australia. 5Allied Health Research, School of Primary Health Care, Monash University, Frankston, Australia. 6Trauma (M.S.) and Physiotherapy Department (B.B. and P.T.), Princess Alexandra Hospital, Brisbane, Australia. Drs McPhail and Haines are supported by National Health and Medical Research Council (Australia) Fellowships. This project received funding from the Princess Alexandra Hospital Foundation. The study protocol was approved by the Princess Alexandra Hospital Human Research Ethics Committee. The authors certify that they have no affiliations with or financial involvement in any organization or entity with a direct financial interest in the subject matter or materials discussed in the article. Address correspondence to Dr Steven M. McPhail, c/o ABIOS / CFAHR, Princess Alexandra Hospital Campus, Corner of Ipswich Road and Cornwall Street, Woolongabba, QLD 4102. Australia. E-mail: [email protected] t Copyright ©2014 Journal of Orthopaedic & Sports Physical Therapy®

488 | july 2014 | volume 44 | number 7 | journal of orthopaedic & sports physical therapy

44-07 McPhail.indd 488

6/17/2014 7:35:44 PM


reported outcome measure with robust content foundation.42 The inclusion of patient-reported outcomes as primary measures has become increasingly common in investigations of people with musculoskeletal conditions. Most evident are measures of pain,29,54,60 physical function,2,18,48 and health-related quality of life.3,10,21,35,38,62 Patient-reported outcomes allow clinicians and researchers to better understand how a condition may impact various aspects of a patient’s life from the perspective of the patient.39,41,42,51 This information can enhance patient-centered care by targeting interventions to priority problem areas indicated by the patient, and by evaluating whether these interventions have a meaningful impact from the perspective of the patient.40,51 The aforementioned Cochrane review30 identified a range of patient-reported outcome measures that have been used in the context of ankle fracture rehabilitation.6,8,16,19,30,37,42,45,64 However, it has been suggested that these outcome measures lack a methodologically robust content foundation for use as ankle fracture–specific, patient-reported outcomes to evaluate quality-of-life impacts experienced by people recovering from this condition.13,26 It has also been found that existing measures lack empirical support for key elements of validity, reliability, and responsiveness, with the exception of the Lower Extremity Functional Scale (LEFS).16,17,19,26,37,42,64 Favorable clinimetric properties have been reported for the LEFS among individuals without ankle fracture6,56,63 and during the acute phase of ankle fracture recovery.31 However, a detrimental ceiling effect has been reported after the acute phase of ankle fracture recovery.31,42 Previous authors have recommended adding additional items to the LEFS to further improve the instrument.31 Perhaps the most important reason for the development of a patient-reported outcome to evaluate quality-of-life impacts following ankle fracture, despite the generally favorable performance of the LEFS, is the nature of the item content and pur-

pose of the LEFS. The item scoring and content of the LEFS focus on physical task performance.6 Though this targeted focus may not be considered a weakness in an instrument designed to assess patients’ ratings of their lower extremity physical function, it may not be suitable as an ankle fracture–specific outcome measure intended to capture broader quality-of-life impacts from the patient’s perspective.42,60 Perhaps the most widely used patientreported outcome among people with ankle fractures to date has been the Olerud-Molander Ankle Score.30,47 In summary, the Olerud-Molander Ankle Score includes 9 parameters: walking, stiffness, swelling, stair climbing, running, jumping, squatting, physical supports, and work capacity.47 While this scale is quite practical, it has been criticized for lacking a methodologically robust content foundation and focusing solely on physical symptoms and performance.26,42 Additionally, the assignment of scores to the various response options was arbitrarily described by the instrument’s authors, without justification for the values being evident.26 Without a condition-specific, patientreported outcome measure with content that is consistent with the life impacts reported by people who have experienced ankle fracture, the most pertinent aspects of the condition may not be evaluated (or addressed). A recent qualitative study has been published in preparation for the development of a condition-specific, patient-reported outcome measure known as the Ankle Fracture Outcome of Rehabilitation Measure (A-FORM) (APPENDIX, available online).42 This investigation reported life impacts following ankle fractures and outlined a thematic framework suitable for developing potential content for an ankle fracture condition-specific outcome measure. The thematic framework from that investigation has been reproduced in TABLE 1.42 This framework indicated that life impacts following ankle fractures commonly extend beyond short-term pain

and physical discomfort into several overarching areas of life.42 A condition-specific, patient-reported outcome measure for use among patients with ankle fracture during their rehabilitation should capture the effects of rehabilitation that patients (rather than health professionals) consider most important.13 These effects must also be evaluated in a way that is valid, reliable, and responsive to change over the entire rehabilitation period.43,44 The purpose of this study was to describe the process of developing the A-FORM, including the selection and refinement of items and response options, as well as to examine the psychometric properties of the A-FORM, including factor structure, reliability, and validity, by assessing item fit with the Rasch model. The A-FORM is intended to address a notable gap in the patient-reported outcome measures currently available for use in clinical and research settings with this clinical population. The A-FORM is designed to evaluate aspects of physical, social, and psychological recovery that are relevant to patients with ankle fractures, rather than purely measuring physical symptoms or other clinician-defined constructs. Appropriate development and validation of this measure will likely permit inclusion of the patient’s perspective in rigorous and efficient evaluations of rehabilitation approaches for individuals with ankle fractures.

METHODS Design

A

2-stage research design was implemented. First, a Delphi panel process was used for item development and wording refinement. Second, a cohort study was undertaken to permit a preliminary maximum-likelihood exploratory factor analysis, then Rasch analysis for items developed from stage 1.

Delphi Panel Process The Delphi panel included 8 members and was facilitated by a member of the research team (T.P.H.). The Delphi panel

journal of orthopaedic & sports physical therapy | volume 44 | number 7 | july 2014 | 489


6/17/2014 7:35:45 PM

[

]

Thematic Conceptual Framework of Life Impacts Following Ankle Fractures, Including Categories Represented Within Each Theme

TABLE 1


research report

Daily Living

Social

Occupational or Domestic

Physical

Psychological

1. P ain, ache, soreness, or discomfort

1. F eelings of anxiety 1. R educed partici1. N egative impact pation in preferred on relationship recreation or with spouse or leisure activities significant other

1. D ifficulty participating in usual work activity

Financial

Aesthetic

Medication Taking

1. R educed income

1. C hanged physical appearance due to weight gain

1. M edication usage (including associated side effects)

2. S welling

2. F eelings of depression

2. R educed 2. Increased dependence on others participation in in household health and fitness activities

2. D ifficulty com2. U se of savings pleting household tasks

3. D ecreased strength

3. F eelings of frustration

3. D ifficulty partici- 3. N egative impact pating in personal on personal care activities (inrelationships with cluding showering family or friends and dressing)

3. R educed discretionary spending

4. D ecreased range of movement/ stiffness

4. F eelings of tiredness or fatigue

4. D ifficulty sleeping

4. Increased cost of living (including health care costs)

2. N ow wear nonpreferred footwear

5. A ltered sensation 6. D ifficulty walking (including flat surfaces, slopes, and steps) Copyright © McPhail et al 2012.42 Reproduced with permission.

included representatives from a range of health professional backgrounds involved in the acute management and rehabilitation of people with ankle fractures, including orthopaedic surgery (1), physiotherapy (2), podiatry (1), occupational therapy (1), and nursing (1), as well as people who had sustained an ankle fracture (2). This cross-section of health professionals involved in the care of people with ankle fractures was selected to ensure ubiquitous understanding of question wording across multiple disciplines, as well as patients who had previously sustained an ankle fracture. Delphi panel members were identified and invited to participate through hospital orthopaedic department team meetings attended by 1 or more members of the research team. The 2 remaining Delphi panel members (patients who had previously sustained an ankle fracture) were already known to the research team from their involvement in the prior qualitative work in the lead-up to this investigation42 and had indicated interest in

participating in the Delphi panel process. Delphi panel members were provided with written stimulus material prior to the commencement of the first Delphi panel round. This stimulus material was drawn directly from the earlier qualitative patient reports of life impacts following ankle fractures. This stimulus material was intended to inform the content development and refinement to ensure that each category of life impacts across the 8 themes was represented in at least 1 stimulus item (TABLE 1). A member of the research team (T.P.H.) prepared preliminary items for inclusion in the stimulus material. Where possible, phrasing used by study participants in the prior qualitative research was used in the stimulus wording of each category item. For example, the phrase “My ankle feels stiff when...” was used by patients reporting the impacts of restricted ankle movement on activities in their daily lives. This specific phrasing was used in the stimulus material for 1 of the items that intended to

capture patient-perceived difficulty with ankle movement. In this way, a total of 50 stimulus items were provided to Delphi panel members, with item content covering the breadth of previously reported life impacts. Most of these stimulus items included content primarily related to 1 of the 8 previously reported themes. However, some stimulus items included content with relevance to more than 1 theme; these items typically involved the description of a psychological impact (eg, anxiety or frustration) in relation to an impact from another category (eg, social or occupational and domestic impacts). For example, 1 stimulus item asked about anxiety in relation to difficulty completing usual work or household tasks. This intertheme item content generally occurred as a result of using common patient phrasing in the stimulus content. Delphi panel participants received the list of potential items as stimulus via e-mail and were asked in each round to comment on the phrasing and scaling for



6/17/2014 7:35:45 PM


potential items that could be included in the A-FORM. Panel members discussed whether the content of the question was clear and unambiguous, suggested wording changes, commented on wording modifications suggested by others, and identified whether there were any other notable facets of ankle fracture life impacts that had not been adequately captured by the current item list. The Delphi panel took place via e-mails distributed to and from the central facilitator to ensure the anonymity of all participants. Three waves of Delphi panel feedback were discussed before consensus was attained on a total of 53 potential A-FORM items to be carried forward to the cohort phase of this investigation.

Patient Cohort Cohort participants included a convenience sample of patients who had sustained an ankle fracture and received treatment at a tertiary (university) hospital facility. For inclusion in the study, patients had to be willing to complete the questionnaire on at least 1 of 2 possible occasions, although completion on both occasions was requested. The 2 occasions were prior to the first follow-up outpatient appointment, within 1 week following cast removal (approximately 6 to 8 weeks postfracture), and prior to a 6- to 8-week post–cast removal reassessment (approximately 12 to 16 weeks postfracture). A total of 51 patients were invited to participate by a research assistant on the day of cast removal. Of these, 45 patients provided written informed consent, and their rights were protected throughout the investigation. The study protocol was approved by the Princess Alexandra Hospital Human Research Ethics Committee. The remainder indicated that they were not interested in taking part in the investigation (n = 4) or that they had already planned to receive subsequent care at another facility closer to their residence (n = 2). Consenting patients were sent a written copy of the preliminary question-

naire, which included all 53 items, with instructions to complete the questionnaire no more than 1 week prior to the first follow-up appointment after cast removal. Similarly, patients were sent a written copy of the same questionnaire with instructions to complete the questionnaire no more than 1 week prior to their 6- to 8-week assessment. Thus the participants were asked to complete the questionnaire on 2 occasions, but were not excluded from the study if they were only able to complete the questionnaire on 1 occasion. Participants who consented to completing the questionnaire but did not do so prior to their arrival at the clinic follow-up assessment were given an opportunity to complete the questionnaire in the clinic waiting area. Patients who completed the questionnaire but forgot to bring it to the clinic were provided with a stamped, addressed envelope by a member of the research team and were instructed to return the questionnaire via post. Patients who did not attend their clinic appointment at the scheduled time were mailed a return envelope with instructions to return the completed questionnaire via post.

Analysis Data from completed questionnaires at both assessments were included in the analyses to ensure data representation of severe limitations (immediately following removal of cast) and less severe limitations (6 to 8 weeks following removal of cast). The A-FORM was refined with the intent of developing a parsimonious instrument that could be used in both research and clinical contexts as a condition-specific measure of recovery. As a deviation from standard approaches to scale refinement, the investigators decided to be tolerant of items that appeared to have floor effects (indicating that participants had not recovered), anticipating that a longitudinal cohort study with a followup beyond 16 weeks postinjury would more readily detect changes in these items among high-functioning patients.

It has previously been suggested that if less than 5% of the data are missing it is inconsequential to the analysis, but if more than 10% of the data are missing it may introduce bias.4,52,53 Therefore, the starting pool of potential A-FORM items was initially screened for items that had a missing-response rate of more than 10%, indicating that the items might not be broadly applicable or might cause confusion during completion. An exploratory maximum-likelihood factor analysis was then pursued to examine whether a single-factor structure existed within the remaining items and to examine the relative contribution of each item. Items were sequentially removed on the basis of factor loading being less than 0.5. The internal consistency of items within the factors was then examined. Items were removed from factors on the basis of uniqueness values until a target Cronbach alpha of .90 was attained. This target value of .90 was considered ideal to ensure that the final instrument would likely be internally consistent, while minimizing the risk of item redundancy.57,58 Rasch analysis was then used to further examine and refine the prevailing single-factor structure, construct validity, and internal consistency of the measure. The Rasch model is suitable for the analysis of ordered-response category data.33 In the Rasch model of analysis, the relationships between the participant responses and the items are explored.66 Regarding the potential A-FORM items, this model first tests the theory that the person with a more functional ankle was less likely to report a negative impact on their quality of life, and, second, the theory that the “easier” the item on the A-FORM, the less likely the person was to report an impact. For example, respondents may be less likely to report difficulty walking on a flat surface than difficulty jumping. The Rasch model aligns a person’s perceived ability and item difficulty in logits, where the logit translates into a raw score. Higher logits represent increased item difficulty. Similarly, a person with



6/17/2014 7:35:46 PM

[ Person with most ability

3

Most able

research report

Most difficult

Sprinting

Most difficult

2 ‡

‡


Jump off both feet

1

†

Difficult

† There is swelling around my ankle My ankle feels stiff I feel pain in or around ankle

*

Average

* 0

†

Amount of work by others Anxious fitness Sleeping without waking; walk on flat surfaces Fatigued; getting back to sleep Relationship with extended family; restricted self-care † Depressed

Easy

Easiest

–1 ‡ ‡ –2

Person with least ability

–3

Least able

Least difficult

FIGURE. Item targeting (1 dot represents 1 participant). *Mean for participant ability and item difficulty. †1 standard deviation from the mean. ‡2 standard deviations from the mean.

a higher (perceived) ability would have a higher logit value. This was visually represented with an item map (FIGURE). This alignment means that a person with a logit of 1.0 would have a 50% likelihood of reporting difficulty on an item with the same logit value of 1.0, while being more likely to be impacted by “harder” items (logits greater than 1.0) and less likely to be impacted by “easier” items (logits less than 1.0). It has been reported that, for scale development with Rasch analysis, the use of a data set with between 64 and 100 responses indicates with 95% confidence that the items are stable at ±0.5 logits.67 It was anticipated that most participants in the study would provide responses at both assessments. Therefore, a target sample size of 40 participants for the patient cohort was considered sufficient to identify potential items for inclusion in

the A-FORM instrument. The A-FORM data were fitted using the WINSTEPS Version 3.74.0 software (John Linacre, Winsteps.com, Beaverton, OR).33 While the model function and software allow for response scale analysis, the preliminary maximum-likelihood factor analysis was conducted to ensure unidimensional data suitable for analysis within the Rasch model. Once unidimensionality was established, 3 main parameters were assessed: the fit statistics, separation indices, and item targeting. The 2 fit statistics commonly used within the Rasch model to assess the item fit included the infit and outfit. The infit is the weighted mean-square residual difference between the expected and observed responses, and the outfit statistic determines the unweighted mean-square residual and sensitivity to outliers. It has previously been suggested that mean-

] square infit and outfit statistics in the range of 0.5 to 1.5 are productive for measurement.34 Person or item responses that fall within this range can be retained within a questionnaire, having demonstrated favorable properties for inclusion in a unidimensional rating scale. For items that fell slightly outside this range (1.5-2.0), the authors used discretion as to whether the item would be retained or discarded. This use of discretion for items that performed well enough to not degrade the instrument but did not fall into the ideal range ensured that potentially useful items would not be discarded prematurely. The choice of whether to retain or discard items in the discretionary range was primarily based on whether an item with similar content was already in the questionnaire (a disincentive to retain the discretionary item), and whether it was plausible that inclusion of data from more assessment time points would contribute to the item falling into the ideal range (0.5-1.5). Any items with infit or outfit greater than 2.0 were automatically discarded. The person separation index gives an indication of the reliability of a questionnaire and its ability to discriminate between participants. The separation index must exceed 2 to achieve a reliability of at least 0.8.7 Differential item functioning (DIF) was also assessed to determine if participants responded differently to items over the 2 time points. The DIF was determined to be small or absent if the difference in logits was less than 0.50 logits, minimal (probably inconsequential) if the difference was 0.50 to 1.0 logits, and notable if the difference was greater than 1.0 logits. If notable DIF was found, it indicated that the scale performance differed between assessments, meaning that the scale would have been biased by the assessment time point. Another key factor of the Rasch analysis is the targeting of item difficulty compared to the ability of the participants.32 A well-formed questionnaire should give a range of scores across a broad range of abilities, and Rasch analysis produces



6/17/2014 7:35:47 PM


a hierarchy of item scores and person ability. Ideally, for a well-targeted questionnaire, the mean of the person scores should be similar to the mean of the item scores. Scaling was also conducted to allow for conversion of raw scores from items falling within the target infit and outfit range to a linear measure suitable for parametric analysis. To demonstrate this, raw scores from the A-FORM Version 1.0 were converted into a linear score between 1 and 100.

RESULTS

TABLE 2

Participant Demographics

Participants (n = 41) Age, y*

36.8 (26.1-53.8)

Gender (male), n (%)

30 (73)

Side of fracture (right), n (%)

22 (54)

Bone fractured, n (%) Tibia only

4 (9.8)

Fibula only

19 (46.3)

Both tibia and fibula

14 (34.1)

Talus only

2 (4.9)

Talus and fibula

1 (2.4)

Talus and calcaneus

1 (2.4)

Potential Items Following Delphi Panel

Open reduction internal fixation, n (%)

t the completion of the 3-round Delphi panel process, the 53 potential items carried forward to the cohort phase included content relating to the themes of physical impacts (16 items), psychological impacts (4 items), daily-living impacts (7 items), social impacts (3 items), occupational and domestic impacts (4 items), financial impacts (2 items), aesthetic impacts (2 items), medication-taking impacts (1 item), and content that addressed more than 1 of these themes within a single item (14 items). Items 1 through 11 included ordinal multiple-choice descriptive statements. The remainder of items had Likert-scale response options (5-point ordinal scale rating).

Weber classification of fibular fracture, n (% of fibular fractures)

A

Ankle Fracture Cohort Of the 45 consenting participants, 4 participants subsequently received followup orthopaedic care at another facility, did not return to the participating clinic, and did not complete the questionnaire at either assessment. The remaining 41 (91%) participants completed the questionnaire at 1 week post–cast removal and were included in the analyses. One of these participants was incarcerated at the time of the second assessment and did not attend the clinic or complete the questionnaire at the second assessment point. The remaining 40 (89%) participants completed the questionnaire at the second assessment. This resulted in

19 (46.3)

A

8 (23.5)

B

18 (52.9)

C

8 (23.5)

*Value is median (quartile 1-quartile 3).

a total of 81 completed questionnaires from the 2 time points for inclusion in the statistical analyses. Demographic details and fracture types present for the 41 participants who were included in the analyses are presented in TABLE 2. Examination of cohort study data revealed 5 items with more than 10% missing data, and these items were eliminated from further analyses. Exploratory maximum-likelihood factor analysis with the remaining 48 items indicated a distinctive single-factor solution (factor 1 eigenvalue, 24.2, where eigenvalues are representative of the amount of variance in all items accounted for by the factor20). Following this, there was a plateau of eigenvalues for subsequent factors. Items were further removed from the measure based on factor loadings and uniqueness values for the singlefactor solution, resulting in a preliminary instrument of 15 items. The final instrument included 3 symptom items (pain, swelling, stiffness), 5 functionallimitation items (walking on flat surface, jumping off both feet, sprinting, getting to sleep, returning to sleep if woken), 1 participation-restriction item (personal activities of daily living), 2 general

mental health symptoms (depressed, fatigued), 2 specific mental health symptoms (anxious about not being able to participate in your preferred health and fitness activities in the future, anxious about not being able to wear your preferred footwear), and 2 external impacts (the amount of extra work that others in your household have to do, relationships with your extended family and friends). The 15 unidimensional items retained in the scale were analyzed for fit and separation. TABLE 3 displays the fit statistics of the 15 items, with 70 participant responses. One item (“Are you anxious about not being able to wear your preferred footwear?”) showed high outfit (mean square, 1.75) and was removed from subsequent analysis to examine whether this improved the properties of the measure. The person and separation indices before and after removal of this item (anxious about footwear) and the mean infit and outfit responses (69 responses remaining) are shown in TABLE 4. These values were favorable, demonstrating appropriate person and item reliability after (and before) removal of the “anxious about footwear” item, and acceptable infit and outfit statistics for the



6/17/2014 7:35:47 PM

[

research report

]

TABLE 3

Total Score

Total Count

Measure

Model Standard Error

Infit Mean Square

Outfit Mean Square

1

238

81

–0.48

0.11

0.84

0.96

2

267

81

–0.80

0.11

0.78

0.94

There is swelling around my ankle

3

251

81

–0.62

0.11

1.28

1.28

My ankle feels stiff

Item Number


Ankle Fracture Outcome of Rehabilitation Measure Item Statistics

Item Code I feel pain in or around ankle

4

164

81

0.39

0.12

0.66

0.62

Walk on flat surface

5

326

81

–1.57

0.12

0.78

0.80

Jump off both feet

6

361

81

–2.22

0.15

1.00

1.17

Sprinting

7

167

81

0.35

0.12

1.01

0.94

Sleeping without waking

8

162

81

0.42

0.12

1.03

1.04

Getting back to sleep

9

143

81

0.70

0.13

0.64

0.55

Restricted self-care

10

135

81

0.83

0.13

1.16

1.01

Depressed

11

156

81

0.50

0.12

0.81

0.86

Fatigued

12

177

81

0.22

0.11

1.24

1.10

Anxious about fitness

13

101

81

1.56

0.16

1.99

1.75

Anxious about footwear

14

199

81

–0.04

0.11

1.50

1.34

Amount of work by others

15

140

81

0.75

0.13

0.81

0.66

Relationship with extended family

199.1  71.9

81.0  0.0

0.00  0.95

0.12  0.02

1.02  0.34

1.00  0.30

Mean  SD

other 14 items (range, 0.55-1.50). The person (and item) reliability and separation indices indicated that the measure was able to distinguish between persons (and items) in terms of their position on the latent variable. The high reliability of persons and items also indicated that the A-FORM was able to define each of the items (“difficulty”) and that there was an appropriate sample size and number of items within the A-FORM. Additionally, the small model error of measurements indicated that extreme scores were not common and was consistent with a favorable signal-to-noise ratio of true variance to observed variance. The DIF analysis indicated that notable DIF occurred between assessments for only the “anxious about footwear” item. In combination with the higher outfit for this item, this supported the removal of this question from subsequent analyses. The nonsignificant DIF for the other questions supported the use of the 2 time points to be used as a single data set. The FIGURE displays a person ability and item map. This map gives a visual representation of the range of person abilities across all the items, with a range

TABLE 4

The Person and Separation Indices Before and After Removal of 1 Item (Anxious About Footwear) and the Mean Infit and Outfit Responses Separation Index

Reliability

Average Infit Mean Square

Average Outfit Mean Square

Model Measurement Error ± SD

Person ability (a)

2.93

0.90

0.96

0.99

0.30  0.87

Item difficulty (a)

8.04

0.98

1.02

0.99

0.13  1.03

Person ability (b)

2.80

0.89

0.78

0.77

0.30  0.83

Item difficulty (b)

7.70

0.98

0.95

0.94

0.12  0.96

Parameter Before*

After†

*Responses, n = 81; items, n = 15. † Responses, n = 69; items, n = 14.

of difficulties. The person abilities were represented across the range of item difficulty, with no ceiling or floor effects noted. TABLE 5 displays a list of the raw scores (of the 14 items with acceptable infit and outfit statistics) converted to a scale of 1 to 100. This table demonstrates how, based on the analyses conducted, potential exists to convert the raw item scores to a single summary score (out of 100); however, the conversion process is not linear in nature.

DISCUSSION Main Findings, Strengths, and Limitations

T

his 2-phase research investigation included the development and refinement of potential items for inclusion in an ankle fracture conditionspecific outcome measure. The content of these potential items was founded on the previously described conceptual framework of life impacts that patients



6/17/2014 7:35:48 PM


TABLE 5

Raw Score to Rasch Score Conversion of the Ankle Fracture Outcome of Rehabilitation Measure Version 1.0

Raw Score

Logits

Converted Score

1

–7.80E

1

2

–6.48

1

3

–5.62

1

4

–4.63

1

5

–4.27

1

6

–3.95

1

7

–3.66

1

8

–3.39

1

9

–3.14

1

10

–2.90

1

11

–2.67

1

12

–2.45

1

13

–2.24

1

14

–2.05

1

15

–1.86

1

16

–1.69

2

17

–1.53

4

18

–1.37

5

19

–1.23

7

20

–1.09

9

21

–0.96

12

22

–0.84

13

23

–0.72

14

24

–0.61

18

25

–0.50

22

26

–0.40

26

27

–0.30

28

28

–0.20

28

29

–0.11

30

30

–0.02

33

31

0.07

37

32

0.15

40

33

0.23

41

34

0.31

45

35

0.39

49 Table continues on page 496.

reported as a result of an ankle fracture.42 Though all potential items included content that might have some relevance in clinical settings, only items that demonstrated favorable properties for inclusion in a patient-reported outcome measure were retained. The decision to retain items was informed by analyses of item response rates, factor structure, construct

validity, and internal consistency. At the end of these analyses, a decision was made as to whether to retain 14 or 15 items for inclusion in a parsimonious unidimensional outcome measure known as the A-FORM (Version 1.0). The item in question asked respondents to report whether they were anxious about not being able to wear their preferred footwear

due to their ankle fracture. Despite a high outfit value and notable DIF for this item, no similar content was covered by the other remaining 14 items, which prompted the investigators to retain the item in the instrument pending further empirical investigations. However, the summary score for the A-FORM (Version 1.0) was derived from only the 14 items meeting the required parameters from the Rasch analysis, on account of the high outfit value and poor DIF for the footwear item. Overall, data from this investigation were favorable, indicating that the AFORM (Version 1.0) had high levels of internal consistency (also known as internal consistency reliability) and construct validity. This instrument is intended for self-completion by patients in clinical and research contexts. Among this sample, all patients were able to self-complete all 15 questions without the need for interviewer administration. The findings from the A-FORM item data in this investigation are promising but must be considered with caveats, and further research is required. The nature of voluntary participation in the investigation of questionnaires may result in patients with lower levels of literacy or confidence in completing questionnaires self-electing not to participate in the study. Therefore, patients with the lowest levels of literacy may not be represented in this sample. This could have led to more favorable findings than those which might be observed among a complete cross-section of patients. Similarly, the sample included a higher percentage of males than females (TABLE 2), and only included participants up to 16 weeks postfixation of fracture, which might have impacted the external validity of these findings. However, this investigation had some protection against bias due to sampling on account of the high proportion of consenting participants in comparison to those invited to participate, and the high proportion of respondents who returned the questionnaire. The reasons provided for nonresponders were also rational (inability to return to



6/17/2014 7:35:49 PM

[


TABLE 5

research report

Raw Score to Rasch Score Conversion of the Ankle Fracture Outcome of Rehabilitation Measure Version 1.0 (continued)

Raw Score

Logits

36

0.46

Converted Score

37

0.53

57

38

0.61

62

52

39

0.67

64

40

0.74

68

41

0.81

73

42

0.88

76

43

0.94

78

44

1.01

80

45

1.00

86

46

1.14

91

47

1.20

93

48

1.26

94

49

1.33

95

50

1.39

96

51

1.46

96

52

1.52

96

53

1.59

98

54

1.66

99

55

1.73

99

56

1.80

99

57

1.88

99

58

1.96

99

59

2.04

99

60

2.13

99

61

2.23

99

62

2.34

99

63

2.46

100

64

2.59

100

65

2.75

100

66

2.95

100

67

3.21

100

68

3.58

100

69

4.24

100

70

5.41E

100

the participating facility for follow-up care or incarceration).

Clinical Implications There are several important clinical implications from the findings of this research. These analyses have provided evidence of a latent trait (a single factor) being evaluated by the items retained in the instrument. This is desirable for

a patient-reported outcome measure intended to generate a single summary score suitable for use in clinical practice and clinical trials. The authors are confident that the latent trait being evaluated by this instrument is patient-perceived impact (on the patient’s own quality of life) attributable to ankle fracture. The authors reached this conclusion after considering the content of the individual

] items and their derivation from qualitative reports of life impacts following ankle fracture. The retained items include representation of each of the overarching themes presented in TABLE 1, with the exception of financial impacts and medication taking. It is perhaps not a coincidence that financial impacts and medication-taking impacts were among the least frequently identified impacts from the previous qualitative reports. It is plausible that financial and medication-taking impacts did not consistently affect patients with ankle fractures, or that other life circumstances not related to ankle fracture had a greater influence on these aspects of life. It is also noteworthy that financial or medication-related impacts that are attributable to an ankle fracture may be indirectly represented through other items with related content, such as items pertaining to the physical impacts associated with taking pain medication or difficulty completing usual employmentrelated activities. The specific content in the retained items also has implications for use of the A-FORM in clinical practice (as well as clinical research). First, it is noteworthy that psychological constructs that have been associated with musculoskeletal pain conditions (eg, catastrophizing, pain-related fear, and fear avoidance28,61) were not identified in the previous qualitative investigation42 or in this study. It is possible that these psychological constructs may not occur as frequently among people who have sustained ankle fractures compared to other painful conditions, such as persistent low back pain.9 Second, the A-FORM includes a mix of item difficulty levels, including “easy” (eg, being able to care for oneself or fall asleep) and “difficult” (eg, sprinting or jumping). The authors consider the mix of item difficulty present in the retained items suitable for general clinical use to minimize any potential ceiling and floor effects. Many patients, particularly among increasingly sedentary and aging populations, may not have been



6/17/2014 7:35:49 PM


able to sprint or jump before their ankle fracture, but some patients may want to return to sports in which high-level functioning is required. On the other hand, some of the easier items are perhaps more important to a large proportion of this clinical population who may have severe consequences if recovery does not occur (eg, not being able to walk or care for oneself ). This mix of item difficulty is likely suitable for general clinical use, although it is plausible that it may have a ceiling effect if used among samples of elite or high-performing athletes recovering from ankle fractures. The decision to retain the footwear item, at least until further investigations are undertaken, was not straightforward. On one hand, the item outfit was slightly higher than the target range and a notable DIF value was evident, suggesting that it was undesirable to retain the item in the summary score of this unidimensional measure, based on data of this investigation. On the other hand, the content of the item aligned with previous qualitative reports and was not directly duplicating content in other retained items. Therefore, a decision to retain the item pending further testing but not to include it in the summary score was reached. It is plausible that subsequent testing may provide more definitive findings indicating that the item should be removed entirely or that it performs sufficiently to be included in the summary score derived from the A-FORM.

Comparison to Prior Research and Future Research Priorities The A-FORM (Version 1.0), developed by assessing item fit with the Rasch model, has a robust content foundation and demonstrated favorable single-factor structure, internal consistency reliability, and validity. This may be considered an advantage over the other instruments that have been previously used among people with ankle fractures, including the most widely used among these, the Olerud-Molander Ankle Score. There is currently a marked lack of empirical

evidence reporting favorable psychometric and clinimetric properties for that scale.17,26,43,47 However, further empirical examinations of the A-FORM and Olerud-Molander scales are warranted before firm conclusions can be drawn regarding the superiority in measurement properties of either instrument. Further investigations of the A-FORM are warranted, given the promising findings of this investigation. It would be useful for future studies to investigate other elements of reliability and validity of the A-FORM, as well as sensitivity to change and minimal clinically important difference values. This may include examining the association between A-FORM responses and objective measures of symptoms and function that are known to be impacted by ankle fractures. Investigation of the performance of the A-FORM among a sample with a longer time since ankle fracture would be valuable. It would also be useful for a future study to conduct a head-to-head comparison between the performance of the A-FORM instrument and the Olerud-Molander Ankle Score, or perhaps the LEFS.

CONCLUSION

T

his investigation developed and refined content for the A-FORM instrument. The A-FORM items demonstrated favorable psychometric properties in this study and are suitable for conversion to a single summary score. Given the promising findings from this investigation, further studies utilizing the A-FORM instrument among people with ankle fractures are warranted. Future research should examine the association between A-FORM responses and measures of functioning and symptoms relevant to people recovering from ankle fractures, as well as sensitivity to change and the minimal clinically important difference for this measure. t

KEY POINTS FINDINGS: This investigation reported

the development and validation of a

new ankle fracture condition-specific, patient-reported outcome known as the A-FORM. IMPLICATIONS: Findings from this research indicate that the 15-item A-FORM (Version 1.0) is suitable for use as a patientreported outcome measure among people recovering from ankle fractures. CAUTION: This investigation only included patients within the first 16 weeks postfixation; further research among patients in later stages of recovery is warranted.

REFERENCES 1. A hlborg HG, Rosengren BE, Järvinen TL, et al. Prevalence of osteoporosis and incidence of hip fracture in women – secular trends over 30 years. BMC Musculoskelet Disord. 2010;11:48. http://dx.doi.org/10.1186/1471-2474-11-48 2. Alexanderson H, Lundberg IE, Stenström CH. Development of the Myositis Activities Profile--validity and reliability of a self-administered questionnaire to assess activity limitations in patients with polymyositis/dermatomyositis. J Rheumatol. 2002;29:2386-2392. 3. Anderson RT, Aaronson NK, Wilkin D. Critical review of the international assessments of health-related quality of life. Qual Life Res. 1993;2:369-395. 4. Bennett DA. How can I deal with missing data in my study? Aust N Z J Public Health. 2001;25:464-469. 5. Beris AE, Kabbani KT, Xenakis TA, Mitsionis G, Soucacos PK, Soucacos PN. Surgical treatment of malleolar fractures. A review of 144 patients. Clin Orthop Relat Res. 1997:90-98. 6. Binkley JM, Stratford PW, Lott SA, Riddle DL. The Lower Extremity Functional Scale (LEFS): scale development, measurement properties, and clinical application. North American Orthopaedic Rehabilitation Research Network. Phys Ther. 1999;79:371-383. 7. Bond TG, Fox CM. Applying the Rasch Model: Fundamental Measurement in the Human Sciences. Mahwah, NJ: Lawrence Erlbaum Associates; 2007. 8. Brink O, Staunstrup H, Sommer J. Stable lateral malleolar fractures treated with aircast ankle brace and DonJoy R.O.M.-Walker brace: a prospective randomized study. Foot Ankle Int. 1996;17:679-684. 9. Buer N, Linton SJ. Fear-avoidance beliefs and catastrophizing: occurrence and risk factor in back pain and ADL in the general population. Pain. 2002;99:485-491. 10. Calvisi V, De Vincentiis B, Palumbo P, Padua R, Lupparelli S. Health-related quality of life in patients with anterior cruciate ligament insuf-



6/17/2014 7:35:50 PM

[

11.


12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

ficiency undergoing arthroscopic reconstruction: a practice-based Italian normative group in comorbid-free patients. J Orthop Traumatol. 2008;9:233-238. http://dx.doi.org/10.1007/ s10195-008-0034-2 Christey GR, Tomlinson M. Risk factors for ankle fracture requiring operative fixation. Aust N Z J Surg. 1999;69:220-223. Coyte PC, Asche CV, Croxford R, Chan B. The economic cost of musculoskeletal disorders in Canada. Arthritis Care Res. 1998;11:315-325. Dawson J, Doll H, Fitzpatrick R, Jenkinson C, Carr AJ. The routine use of patient reported outcome measures in healthcare settings. BMJ. 2010;340:c186. http://dx.doi.org/10.1136/bmj. c186 Day GA, Swanson CE, Hulcombe BG. Operative treatment of ankle fractures: a minimum tenyear follow-up. Foot Ankle Int. 2001;22:102-106. Den Hartog D, Van Lieshout EM, Tuinebreijer WE, et al. Primary hemiarthroplasty versus conservative treatment for comminuted fractures of the proximal humerus in the elderly (ProCon): a multicenter randomized controlled trial. BMC Musculoskelet Disord. 2010;11:97. http://dx.doi. org/10.1186/1471-2474-11-97 DiStasio AJ, 2nd, Jaggears FR, DePasquale LV, Frassica FJ, Turen CH. Protected early motion versus cast immobilization in postoperative management of ankle fractures. Contemp Orthop. 1994;29:273-277. Eechaute C, Vaes P, Van Aerschot L, Asman S, Duquet W. The clinimetric qualities of patientassessed instruments for measuring chronic ankle instability: a systematic review. BMC Musculoskelet Disord. 2007;8:6. http://dx.doi. org/10.1186/1471-2474-8-6 Engelberg R, Martin DP, Agel J, Obremsky W, Coronado G, Swiontkowski MF. Musculoskeletal Function Assessment instrument: criterion and construct validity. J Orthop Res. 1996;14:182192. http://dx.doi.org/10.1002/jor.1100140204 Finsen V, Saetermo R, Kibsgaard L, et al. Early postoperative weight-bearing and muscle activity in patients who have a fracture of the ankle. J Bone Joint Surg Am. 1989;71:23-27. Floyd FJ, Widaman KF. Factor analysis in the development and refinement of clinical assessment instruments. Psychol Assess. 1995;7:286-299. Ghislandi S, Apolone G, Garattini L, Ghislandi I. Is EQ-5D a valid measure of HRQoL in patients with movement disorders? A comparison with SF-36 and FIM questionnaires. Eur J Health Econ. 2002;3:125-130. http://dx.doi.org/10.1007/ s10198-002-0105-x Giannini S, Faldini C, Acri F, Leonetti D, Luciani D, Nanni M. Surgical treatment of posttraumatic malalignment of the ankle. Injury. 2010;41:1208-1211. http://dx.doi.org/10.1016/j. injury.2010.09.017 Goldhahn S, Kralinger F, Rikli D, Marent M, Goldhahn J. Does osteoporosis increase complication risk in surgical fracture treatment?

research report

24.

25.

26.

27.

28.

29.

30.

31.

32. 33. 34.

35.

36.

37.

38.

]

A protocol combining new endpoints for two prospective multicentre open cohort studies. BMC Musculoskelet Disord. 2010;11:256. http:// dx.doi.org/10.1186/1471-2474-11-256 Hallberg I, Bachrach-Lindström M, Hammerby S, Toss G, Ek AC. Health-related quality of life after vertebral or hip fracture: a seven-year follow-up study. BMC Musculoskelet Disord. 2009;10:135. http://dx.doi. org/10.1186/1471-2474-10-135 Johnson EE, Davlin LB. Open ankle fractures. The indications for immediate open reduction and internal fixation. Clin Orthop Relat Res. 1993:118-127. Kearney RS, Achten J, Lamb SE, Plant C, Costa ML. A systematic review of patient-reported outcome measures used to assess Achilles tendon rupture management: what’s being used and should we be using it? Br J Sports Med. 2012;46:1102-1109. http://dx.doi.org/10.1136/ bjsports-2011-090497 Lash N, Horne G, Fielden J, Devane P. Ankle fractures: functional and lifestyle outcomes at 2 years. ANZ J Surg. 2002;72:724-730. Leeuw M, Goossens ME, Linton SJ, Crombez G, Boersma K, Vlaeyen JW. The fear-avoidance model of musculoskeletal pain: current state of scientific evidence. J Behav Med. 2007;30:77-94. http://dx.doi.org/10.1007/ s10865-006-9085-0 Leung AS, Lam TH, Hedley AJ, Twomey LT. Use of a subjective health measure on Chinese low back pain patients in Hong Kong. Spine (Phila Pa 1976). 1999;24:961-966. Lin CW, Moseley AM, Refshauge KM. Rehabilitation for ankle fractures in adults. Cochrane Database Syst Rev. 2008:CD005595. http://dx.doi. org/10.1002/14651858.CD005595.pub2 Lin CW, Moseley AM, Refshauge KM, Bundy AC. The Lower Extremity Functional Scale has good clinimetric properties in people with ankle fracture. Phys Ther. 2009;89:580-588. http:// dx.doi.org/10.2522/ptj.20080290 Linacre JM. Rasch analysis of rank-ordered data. J Appl Meas. 2006;7:129-139. Linacre JM. A User’s Guide to WINSTEPS. Chicago, IL: MESA Press; 2005. Linacre JM. What do infit and outfit, meansquare and standardized mean? Rasch Meas Trans. 2002;16:878. Linde L, Sørensen J, Østergaard M, HørslevPetersen K, Hetland ML. Health-related quality of life: validity, reliability, and responsiveness of SF-36, 15D, EQ-5D [corrected] RAQoL, and HAQ in patients with rheumatoid arthritis. J Rheumatol. 2008;35:1528-1537. Lindsjö U. Operative treatment of ankle fracturedislocations. A follow-up study of 306/321 consecutive cases. Clin Orthop Relat Res. 1985:28-38. Mazur JM, Schwartz E, Simon SR. Ankle arthrodesis. Long-term follow-up with gait analysis. J Bone Joint Surg Am. 1979;61:964-975. McPhail S, Beller E, Haines T. Physical func-

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

tion and health-related quality of life of older adults undergoing hospital rehabilitation: how strong is the association? J Am Geriatr Soc. 2010;58:2435-2437. http://dx.doi. org/10.1111/j.1532-5415.2010.03163.x McPhail S, Comans T, Haines T. Evidence of disagreement between patient-perceived change and conventional longitudinal evaluation of change in health-related quality of life among older adults. Clin Rehabil. 2010;24:1036-1044. http://dx.doi.org/10.1177/0269215510371422 McPhail S, Haines T. Patients undergoing subacute rehabilitation have accurate expectations of their health-related quality of life at discharge. Health Qual Life Outcomes. 2012;10:94. http://dx.doi.org/10.1186/1477-7525-10-94 McPhail SM, Bagraith KS, Schippers M, Wells PJ, Hatton A. Use of condition-specific patientreported outcome measures in clinical trials among patients with wrist osteoarthritis: a systematic review. Adv Orthop. 2012;2012:273421. http://dx.doi.org/10.1155/2012/273421 McPhail SM, Dunstan J, Canning J, Haines TP. Life impact of ankle fractures: qualitative analysis of patient and clinician experiences. BMC Musculoskelet Disord. 2012;13:224. http:// dx.doi.org/10.1186/1471-2474-13-224 Mokkink LB, Terwee CB, Patrick DL, et al. The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study. Qual Life Res. 2010;19:539-549. http://dx.doi.org/10.1007/ s11136-010-9606-8 Mokkink LB, Terwee CB, Patrick DL, et al. The COSMIN study reached international consensus on taxonomy, terminology, and definitions of measurement properties for health-related patient-reported outcomes. J Clin Epidemiol. 2010;63:737-745. http://dx.doi.org/10.1016/j. jclinepi.2010.02.006 Moseley AM, Herbert RD, Nightingale EJ, et al. Passive stretching does not enhance outcomes in patients with plantarflexion contracture after cast immobilization for ankle fracture: a randomized controlled trial. Arch Phys Med Rehabil. 2005;86:1118-1126. http://dx.doi. org/10.1016/j.apmr.2004.11.017 Nilsson G, Nyberg P, Ekdahl C, Eneroth M. Performance after surgical treatment of patients with ankle fractures—14-month follow-up. Physiother Res Int. 2003;8:69-82. http://dx.doi. org/10.1002/pri.274 Olerud C, Molander H. A scoring scale for symptom evaluation after ankle fracture. Arch Orthop Trauma Surg. 1984;103:190-194. Ostelo RW, de Vet HC, Knol DL, van den Brandt PA. 24-item Roland-Morris Disability Questionnaire was preferred out of six functional status questionnaires for post-lumbar disc surgery. J Clin Epidemiol. 2004;57:268-276. http://dx.doi. org/10.1016/j.jclinepi.2003.09.005 Ponzer S, Nåsell H, Bergman B, Törnkvist H. Functional outcome and quality of life



6/17/2014 7:35:50 PM

50.


51.

52. 53.

54.

55.

56.

in patients with Type B ankle fractures: a two-year follow-up study. J Orthop Trauma. 1999;13:363-368. Praemer A, Furner S, Rice DP. Musculoskeletal Conditions in the United States. Park Ridge, IL: American Academy of Orthopaedic Surgeons; 1992. Sandman L, Munthe C. Shared decision making, paternalism and patient choice. Health Care Anal. 2010;18:60-84. http://dx.doi.org/10.1007/ s10728-008-0108-6 Schafer JL. Multiple imputation: a primer. Stat Methods Med Res. 1999;8:3-15. Schafer JL, Graham JW. Missing data: our view of the state of the art. Psychol Methods. 2002;7:147-177. Schoenfeld AJ, Bono CM. Measuring spine fracture outcomes: common scales and checklists. Injury. 2011;42:265-270. http://dx.doi. org/10.1016/j.injury.2010.11.040 Stewart WF, Ricci JA, Chee E, Morganstein D, Lipton R. Lost productive time and cost due to common pain conditions in the US workforce. JAMA. 2003;290:2443-2454. http://dx.doi. org/10.1001/jama.290.18.2443 Stratford PW, Kennedy DM, Hanna SE. Condition-specific Western Ontario McMaster Osteoarthritis Index was not superior to region-specific Lower Extremity Functional Scale at detecting change. J Clin Epidemiol.

57.

58.

59.

60.

61.

62.

63.

2004;57:1025-1032. http://dx.doi.org/10.1016/j. jclinepi.2004.03.008 Streiner DL. Starting at the beginning: an introduction to coefficient alpha and internal consistency. J Pers Assess. 2003;80:99-103. http:// dx.doi.org/10.1207/S15327752JPA8001_18 Tavakol M, Dennick R. Making sense of Cronbach’s alpha. Int J Med Educ. 2011;2:53-55. http://dx.doi.org/10.5116/ijme.4dfb.8dfd van der Sluis CK, Eisma WH, Groothoff JW, ten Duis HJ. Long-term physical, psychological and social consequences of a fracture of the ankle. Injury. 1998;29:277-280. Verra ML, Angst F, Staal JB, et al. Differences in pain, function and coping in Multidimensional Pain Inventory subgroups of chronic back pain: a one-group pretest-posttest study. BMC Musculoskelet Disord. 2011;12:145. http://dx.doi. org/10.1186/1471-2474-12-145 Vlaeyen JW, Linton SJ. Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art. Pain. 2000;85:317-332. Ware JE, Jr., Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care. 1992;30:473-483. Watson CJ, Propps M, Ratner J, Zeigler DL, Horton P, Smith SS. Reliability and responsiveness of the Lower Extremity Functional Scale and the Anterior Knee Pain Scale in patients with

64.

65.

66. 67. 68.

69.

anterior knee pain. J Orthop Sports Phys Ther. 2005;35:136-146. http://dx.doi.org/10.2519/ jospt.2005.35.3.136 Wetzler M, Whitelaw G, Lee P. The post-operative management of ankle fractures with pneumatic braces versus short leg cast [abstract]. Orthop Trans. 1991;15:719. Woolf AD, Pfleger B. Burden of major musculoskeletal conditions. Bull World Health Organ. 2003;81:646-656. Wright BD, Masters GN. Rating Scale Analysis. Chicago, IL: MESA Press; 1982. Wright BD, Stone MH. Best Test Design. Chicago, IL: MESA Press; 1979. Wu MH, Huang TJ, Cheng CC, Li YY, Hsu RW. Role of the supine lateral radiograph of the spine in vertebroplasty for osteoporotic vertebral compression fracture: a prospective study. BMC Musculoskelet Disord. 2010;11:164. http:// dx.doi.org/10.1186/1471-2474-11-164 Yelin E. Cost of musculoskeletal diseases: impact of work disability and functional decline. J Rheumatol Suppl. 2003;68:8-11.

@

MORE INFORMATION WWW.JOSPT.ORG

BROWSE Collections of Articles on JOSPT’s Website JOSPTs website (www.jospt.org) offers readers the opportunity to browse published articles by Previous Issues with accompanying volume and issue numbers, date of publication, and page range; the table of contents of the Upcoming Issue; a list of available accepted Ahead of Print articles; and a listing of Categories and their associated article collections by type of article (Research Report, Case Report, etc). Features further curates 3 primary JOSPT article collections: Musculoskeletal Imaging, Clinical Practice Guidelines, and Perspectives for Patients, and provides a directory of Special Reports published by JOSPT.



6/17/2014 7:35:52 PM


APPENDIX

journal of orthopaedic & sports physical therapy | volume 44 | number 7 | july 2014 | B1


6/18/2014 5:46:01 PM


[


research report

]

APPENDIX

B2 | july 2014 | volume 44 | number 7 | journal of orthopaedic & sports physical therapy

6/18/2014 5:46:01 PM

Development and Validation of the Keele Musculoskeletal Patient Reported Outcome Measure (MSK-PROM).

The Spine Functional Index: development and clinimetric validation of a new whole-spine functional outcome measure.

Development and validation of the Diabetic Peripheral Neuropathic Pain Impact (DPNPI) measure, a patient-reported outcome measure.

Development and validation of a patient-reported outcome measure for stroke patients.

Development and validation of the RAND Asthma Control Measure.

Development and Validation of a Measure of Self-Critical Rumination.

The Circumscribed Fear Measure: Development and initial validation of a trans-stimulus phobia measure.

Translation and validation of the German version of the foot and ankle outcome score.

Development and Initial Validation of a Brief Symptom Measure.

Metacognitive planning: development and validation of an online measure.

Validation of the Dutch language version of the Foot and Ankle Outcome Score.

The development and validation of a novel outcome measure to quantify mobility in the dysvascular lower extremity amputee: the amputee single item mobility measure.

Development and validation of the Surgical Outcome Risk Tool (SORT).

Treatment and outcome prognosis of patients with high-energy transsyndesmotic ankle fracture dislocation-the "Logsplitter" injury.

Fixing the Problem With Empathy: Development and Validation of the Affective and Cognitive Measure of Empathy.

Rehabilitation of the fracture patient.

The will and the ways: development and validation of an individual-differences measure of hope.

Development and validation of an ankle brachial index risk model for the prediction of cardiovascular events.

The development and validation of a measure of the psychiatrist's authoritative domain.

hip osteoarthritis outcome measure: process of development.

Current status of outcome measure development in vasculitis.

Development and initial psychometric evaluation of the Mobility Activities Measure for Inpatient Rehabilitation Settings (Mobam-in).

Validation of the Vitiligo Noticeability Scale: a patient-reported outcome measure of vitiligo treatment success.

Development and Validation of a Self-Report Measure of Mentalizing: The Reflective Functioning Questionnaire.