HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when Rapid Responses are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Roddey, T. S Articles by Hanten, W. Search for Related Content PubMed PubMed Citation Articles by Roddey, T. S Articles by Hanten, W.

Comparison of the University of California–Los Angeles Shoulder Scale and the Simple Shoulder Test With the Shoulder Pain and Disability Index: Single-Administration Reliability and Validity

TS Roddey, PT, PhD, OCS, FAAOMPT, is Associate Professor, School of Physical Therapy, Texas Woman's University, 1130 MD Anderson Blvd, Houston, TX 77030 (USA) (hg_roddey{at}twu.edu). She was a doctoral student, School of Physical Therapy, Texas Woman's University, and Research Fellow, Texas Orthopedic Hospital, Houston, Tex, at the time this research was conducted. Address all correspondence to Dr Roddey
SL Olson, PT, PhD, is Associate Professor, School of Physical Therapy, Texas Woman's University
KF Cook, PhD, is Psychometrician, Center for Healthy Aging (a VA Rehab R&D Center of Excellence), Houston, Tex, and Assistant Professor, Baylor College of Medicine, Houston, Tex
GM Gartsman, MD, is Orthopedic Surgeon, Fondren Orthopedic Group, Houston, Tex
W Hanten, PT, EdD, is Professor, School of Physical Therapy, Texas Woman's University
Dr Roddey, Dr Olson, and Dr Cook provided concept/research design, writing, and data analysis. Dr Roddey and Dr Gartsman provided data collection, Dr Roddey provided project management, and Dr Gartsman provided subjects. Dr Olson, Dr Cook, and Dr Hanten provided consultation (including review of manuscript before submission). The authors acknowledge the Rehabilitation Department of Texas Orthopedic Hospital for their support and the assistance of the employees of the Fondren Orthopedic Group with this project

Submitted July 19, 1999; Accepted April 24, 2000

	Abstract

 
Background and Purpose. Shoulder scales are often used to evaluate treatment efficacy, yet little is known about the psychometric properties of these scales. Only one scale has undergone psychometric scrutiny: the Shoulder Pain and Disability Index (SPADI). This study compared 2 shoulder measures—the University of California-Los Angeles (UCLA) Shoulder Scale and the Simple Shoulder Test (SST)—with the SPADI. Subjects. One hundred ninety-two patients with shoulder disorders were recruited from one physician's office to complete the self-report sections of the 3 scales. Methods. Cronbach alpha values and standard errors of measurement (SEM) were calculated for each of the multi-item subscales. Validity was examined through calculation of correlation coefficients among the 3 scales. Factor analysis was completed to assess the underlying constructs of the SPADI and the SST. Results. Cronbach alpha values ranged from .85 to .95. The SEM values for the multi-item scales ranged from 4.75 to 11.65. Evidence for validity to reflect function was indicated by the correlation between the SST and the SPADI disability subscale. The factor analysis of the SPADI revealed loading on 1 factor, whereas the SST loaded on 2 factors. Conclusion and Discussion. All scales demonstrated good internal consistency, suggesting that all items for each scale measure the same construct. However, the SEMs for all scales were high. Factor loading was inconsistent, suggesting that patients may not distinguish between pain and function.

Key Words: Outcome assessment (health care) • Psychometrics • Shoulder • Shoulder joint

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Conclusions Appendix References

 
The Institute of Medicine's (IOM) model of disability has 3 components: risk factors, the disablement process, and quality of life.¹ The disablement process consists of pathology (cellular/tissue level), impairment (organ or organ system level), functional limitation (organism level), and disability (societal level). In recent years, to better address the needs and concerns of patients, a shift has been advocated in how treatment interventions and strategies are evaluated. From the perspective of the IOM model, the proposed shift is from the organ level to the organism level, that is, from impairment-based measures to measures that assess patients' functional limitations.² The lack of association between changes in shoulder impairment and changes in functional status supports a movement away from a purely impairment-based approach. The promise of such research in shoulder function includes clinical studies for evaluating treatment effect and cost-effectiveness of interventions for shoulder dysfunction based on the needs and concerns of patients. The need for research has been established clearly by 2 systematic and recent reviews of the literature.^3,4 The success of such studies, however, will hinge on the existence of effective measures of shoulder outcomes.

A number of shoulder scales have been developed. Three of the most commonly used shoulder outcome scales are the University of California–Los Angeles (UCLA) Shoulder Scale,⁵ the Simple Shoulder Test (SST),⁶ and the Shoulder Pain and Disability Index (SPADI).⁷ There is little published evidence regarding the psychometric properties of these measures. Only one shoulder scale, the SPADI, has undergone psychometric scrutiny with more than one sample of subjects.^7–11 We found no studies in which reliability was measured for the SST, and in only one study was validity measured for the SST.⁹ Despite its frequent use in research, no studies have evaluated either the reliability or the validity of measurements obtained with the UCLA Shoulder Scale.

The test-retest reliability of measurements obtained with the SPADI has been assessed in 2 studies.^7,11 Both groups of investigators found the SPADI measurements to be what we would consider reliable (intraclass correlation coefficients [ICCs]=.66–.91), thus supporting its use for group-level comparisons. Shoulder scales are often used, however, not only in research but in clinical settings to document the status of patients. In comparison with the reliability required for group-level comparisons, we contend that much greater reliability is needed to justify a scale's use in individual assessment.¹² Only Beaton and Richards¹¹ have addressed the issue of the individual-level reliability of measurements obtained with the SPADI, and they reported that an ICC of .91 was adequate. The reliability of SPADI measurements was further evaluated by Roach et al⁷ through calculation of internal consistency values, with Cronbach alpha () values ranging from .86 to .95.

In several studies,^7–10 the SPADI measurements were evaluated for construct and criterion validity. Roach et al⁷ used factor analysis to evaluate the construct validity of SPADI measurements in a sample of 37 male veterans. They found minimal support for the division of the SPADI items into subscales measuring 2 dimensions (pain and disability). Because their sample was quite homogeneous (male veterans) and grossly undersized for a factor analysis,¹³ their results should be considered preliminary. In the same study,⁷ criterion validity was evaluated using active range of motion (AROM) of the shoulder as a "gold standard." As Heald et al¹⁰ have noted, there is no evidence to support the use of shoulder AROM to reflect function. Beaton and Richards⁹ found that AROM measurements obtained during shoulder elevation correlated poorly with scores on both the SPADI and the SST.

Researchers in 3 studies^8–10 examined the construct validity of SPADI measurements by comparing subjects' scores with those obtained from general health-related quality-of-life scales. The researchers in one of these studies⁹ also included the SST in their examination of construct validity. Williams et al⁸ compared the SPADI with the Medical Outcomes Study 20-Item Short-Form Health Survey (SF-20). The moderate correlation with the physical function and pain components of the SF-20 supported the construct validity of the SPADI measurements. Two groups of researchers^9,10 examined the construct validity of measurements obtained with shoulder outcome scales. Heald et al¹⁰ compared SPADI scores with scores on the Sickness Impact Profile (SIP). Beaton and Richards⁹ compared SPADI and SST scores with scores on the Medical Outcomes Study 36-Item Short-Form Health Survey (SF-36). The correlations found between these shoulder-specific measures and general health status measures demonstrated moderate associations.

In summary, of the 3 commonly used shoulder outcome scales—the SPADI, SST, and UCLA—only the SPADI has undergone psychometric scrutiny. Despite the methodological problems in some of these studies, we believe that, overall, they support a conclusion that there is acceptable reliability and validity of data obtained with the SPADI to support its use with group investigations. The UCLA Shoulder Scale and the SST, however, have little or no psychometric evidence of reliability at either the group or the individual level and little support for their use in either clinical evaluation or research regarding shoulder function. Therefore, we argue that the SPADI is the existing "gold standard" of shoulder outcome measures. The purposes of our study were (1) to add to the existing work on the reliability and validity of SPADI measurements and (2) to explore the psychometric properties of other 2 shoulder outcome measures, the UCLA and the SST. Specifically, the aims of our study were:

1. to assess the group-level reliability of SPADI and SST scores by calculating the internal consistency of total scores for the SPADI and SST and of scores for the multi-item subscales of the SPADI,
   
2. to assess the individual-level reliability of SPADI and SST scores by calculating the standard error of measurement (SEM) at the subscale and scale levels,
   
3. to evaluate the construct validity of SPADI and SST scores through factor analysis, and
   
4. to evaluate the convergent and discriminant validity of UCLA and SST scores by correlating the scores of these scales to those of the SPADI.
   

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Conclusions Appendix References

 
Subjects

The participants were 192 patients with shoulder disorders who were recruited from the private office of an orthopedic surgeon. While waiting for a scheduled appointment, the participants anonymously completed a questionnaire consisting of demographic questions and self-report sections of the shoulder scales examined in this study. The self-report sections of these 3 shoulder scales are shown in the Appendix. Patients who did not want to participate simply returned their blank forms to the office personnel. Because the data were collected anonymously, no additional data were collected from either the participants who completed the forms or those who returned the blank forms to the receptionist. All subjects approved for the physician's appointment had third-party payer authorization. After review, the shoulder questionnaire was approved by the Human Subject Review Committee at Texas Woman's University.

Demographics and surgical history for the sample are reported in Table 1. Fifty-eight percent of the subjects were male (112/192), and 41% (78/192) were female. Two subjects did not report their sex. The average age of the subjects was 47 years (SD=15, range=18–87). Forty-six percent of the subjects (88/192) reported that they had undergone surgery on their shoulder. The median time since surgery was 14 weeks (=30.1, SD=76.5, range=1.5–572). Fifty-three percent of the subjects (102/192) reported that they had not had surgery on their shoulder; 2 subjects did not answer that question.

SST.
The SST is a shoulder function scale consisting of 12 items that ask people about their ability to tolerate or perform 12 activities of daily living (ADL). The individual indicates that he or she is able or is not able to do the activity. The SST scores range from 0 to 100 and are reported as the percentage of answered items to which the person responds in the affirmative.⁶

SPADI.
The SPADI consists of 2 self-report subscales of pain and disability. The items of both subscales are visual analog scales (VASs). The 5-item pain subscale asks people about their pain during ADL, and each item is anchored by the descriptors "no pain" (left anchor) and "worst pain imaginable" (right anchor). The 8 disability items ask people about their difficulty in performing ADL. These items are anchored with the descriptors "no difficulty" (left anchor) and "so difficult it required help" (right anchor). Each item is scored by measuring the distance from the left anchor to the mark made by the person. Subscales are scored in a 3-part process. First, item scores within the subscale are summed. Second, this sum is divided by the summed distance possible across all items of the subscale to which the person responded. Third, this ratio is multiplied by 100 to obtain a percentage. Higher scores on the subscale indicate greater pain and greater disability. To obtain the SPADI total score, the pain and disability subscales scores are averaged.⁷

Data Analysis

Reliability.
The internal consistency of the multi-item subscales was assessed by calculating Cronbach alpha values for the subscales (SPADI pain subscale, SPADI disability subscale, and SST). The Cronbach alpha statistic is an estimate of the reliability of a scale's measurements calculated from a single administration of the scale.¹⁴ Therefore, in contrast to test-retest statistics, this method is particularly useful for evaluating scales measuring traits that can change over short periods of time. This statistic was not calculated for the UCLA pain or function subscales because these 2 subscales consist of single items.

The SEM is a statistic that is used to estimate how reliably a scale estimates an individual's "true score," that is, the score that would be obtained for the person if the scale measured perfectly, without error.^12,15 This statistic contrasts with other statistics, such as the Cronbach alpha, that evaluate how reliably a collection of scores can estimate the true mean score of a group of individuals, that is, the mean score that the group would have if the scale measured without error.

A useful property of the SEM statistic is that it can be used to establish a range of "confidence" around an individual score (eg, the score interval within which the examiner could be 95% confident that the individual's true score existed. The SEM estimates were calculated using the following equation:

where SD_x equals the standard deviation of the observed scores (x) and r equals the reliability estimate for the measure.¹⁴ Typically, with a test-retest format for reliability testing, correlation coefficients are used as the r value in calculating the SEM. However, in studies in which a single administration of the test is completed, some authors^16–18 argue that the Cronbach alpha statistic may be used as the reliability estimate (r) in calculating the SEM. For the purposes of this study, the Cronbach alpha statistic was used as the reliability estimate. The use of this statistic means that we did not examine whether there was a time-dependent effect on the measure's stability. However, a test-retest-based estimate would be confounded by actual changes in a person's status between administrations. Using the probabilities associated with a normal distribution, the SEMs were used to calculate the 95% confidence interval (CI) for scores on the SPADI pain and disability subscales, the SPADI total score, and the SST. The SEMs were not calculated for the UCLA pain or function subscales because Cronbach alpha estimates cannot be calculated for single-item subscales. The 95% CI for any observed score is defined as the observed score ± 1.96 SEMs.

Validity.
To assess the construct validity of the SPADI and SST scores, 2 principal-component factor analyses with varimax rotation were conducted. If the SPADI is measuring pain and disability as separate entities, we would expect a 2-factor solution, with disability items grouping on one factor and pain items grouping on the other factor. The SST is designed to measure a single dimension; therefore, its items are expected to load on a single dimension.

Two other kinds of validity were assessed: convergent validity and discriminant validity. Convergent validity estimates evaluate the degree to which 2 scales measure the same trait.¹⁹ Discriminant validity estimates evaluate the degree to which 2 scales measure different traits.¹⁹ In the present case, convergent and discriminant validity support the use of scale score to represent the status of patients with regard to their shoulder pain or shoulder function. Strong correlations between subscales purporting to measure the same construct are thought to be evidence of convergent validity. Weak correlations between subscales purporting to measure different constructs are, using the same logic, viewed as evidence of discriminant validity.

Convergent validity and discriminant validity were examined by calculating correlation coefficients between the scores of the SPADI and those of the SST and UCLA. To evaluate the convergent validity of the shoulder function and disability subscales, Spearman correlation coefficients (r_s) were calculated between the scores of the SPADI disability subscale and the scores of the UCLA function subscale and the SST. The Spearman correlation coefficient between the SPADI and UCLA pain subscales was also calculated. Discriminant validity was estimated by calculating the Spearman correlation coefficient between pairs of pain and function or disability subscales.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Conclusions Appendix References

 
Reliability

Internal consistency.
The calculated Cronbach alpha values were .89 for the SPADI pain subscale scores and .95 for the SPADI disability subscale scores. The Cronbach alpha values were .96 for the SPADI total scores and .85 for the SST scores.

Standard errors of measurement.
The calculated SEMs and 95% CIs for the SPADI pain and disability subscales and for the SPADI and SST scales as a whole are presented in Table 2. The SPADI scale demonstrated the most precision, with a 95% CI of ±9.3 points. The SST demonstrated the poorest precision, with a 95% CI of ±22.8 points. The SPADI pain and disability subscales had 95% CIs of ±15.3 and ±11.3 points, respectively.

Table 4 shows the loadings of each SST item on the first and second factors. After varimax rotation, the item with the strongest association with the first factor and weakest association with the second factor was the item "Can you wash the back of your opposite shoulder with the affected extremity?" (factor 1=.775, factor 2=–.037). The item "Does your shoulder allow you to sleep comfortably?" had the strongest association with the second factor and a very weak association with the first factor (factor 1=.023, factor 2=.764). The 2 items most strongly associated with the second factor (items 1 and 2) were those that might be expected to be strongly influenced by the amount of pain a person is experiencing. These items query respondents regarding their comfort in sleeping and in resting.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Conclusions Appendix References

The SST had lower internal consistency than either the SPADI pain and disability subscales or the SPADI total that combines items of the 2 subscales. It is somewhat curious that the largest Cronbach value was obtained for the SPADI total (=.96). When 2 subscales representing different constructs are combined, lower internal consistency would be expected.¹⁴ When we calculated in a single analysis the internal consistency of the pain and disability items of the SPADI, however, an even higher internal consistency value was obtained (=.96). One possible explanation for the higher value is that an increase in reliability is expected with a greater number of items. Another explanation is that the SPADI is essentially unidimensional. The latter possibility is consistent with 2 other findings in the study: (1) the principal-component factor analysis resulted in a single-factor solution, and (2) the correlational analysis revealed a high association between the SPADI subscales.

Standard errors of measurement.
Although the SEM values for the SST were the largest of the scales and subscales analyzed, the SEMs for the SPADI total and the SPADI disability and pain subscales also were quite large. These results demonstrate that a scale can meet generally accepted standards with regard to group-level reliability yet not have adequate precision for the measurement of individuals. Therefore, a scale that functions quite well in comparing group means in a research study may be inadequate for tracking changes in individual patients across the course of their rehabilitation. These results concur with those found by Yarnitsky and colleagues²⁰ when they evaluated the repeatability of VAS measurements of pain. They found that, although much research supports the reliability of VAS pain measurements, the reliability of measurements obtained with such scales for individuals warrants further investigation. In addition, with continued pressure by third-party payers to document treatment efficacy, utilization of a scale that has substantial measurement error can result in misleading information about a patient's progress.

Validity

Construct validity.
Roach et al⁷ conducted a principal-components factor analysis with varimax rotation of the SPADI pain and disability subscale items and found that they loaded on 2 factors. These 2 factors, however, did not delineate clearly between the pain and disability items. As already noted, the sample size (N=37) for the study by Roach and colleagues was insufficient for a principal-components factor analysis. In addition, their study sample was very homogeneous and failed to include nonveterans and women. In our larger and more heterogeneous sample, a single factor best accounted for the variance in the SPADI item scores. All of the SPADI items had at least moderately high correlations with the first factor (.613–.905), and there was little difference in magnitude between the factor loadings of the disability items and those of the pain items. These results suggest that, in their responses to the SPADI items, people do not distinguish between pain and disability. One possible explanation for this finding is the wording of the SPADI items. The disability items ask respondents to indicate the amount of difficulty they have with specified functions. It is possible that, when people report their difficulty in performing an activity, they consider pain to be part of what makes the activity difficult.

The results of the principal-components factor analysis of the SST provide an interesting contrast with those obtained for the SPADI. The SST purports to measure a single construct. Our results, however, support a 2-factor solution for the SST. Two items had a strong association with a second factor and a weak association with the first factor: "Is your shoulder comfortable with your arm at rest by your side?" and "Does your shoulder allow you to sleep comfortably?" In contrast to these 2 items, the other items of the SST, which loaded heavily on the first factor, query individuals regarding challenges to their strength or flexibility with the involved shoulder. Therefore, the first factor measures what a person can do with his or her shoulder; the second factor measures a person's comfort with the shoulder at rest.

Convergent validity.
To assess convergent validity, pain subscale scores were correlated with the SPADI pain subscale scores and function subscale scores were correlated with the SPADI disability subscale scores. The correlation between the SPADI and UCLA pain subscales was only moderate (r_s=–.63). This was also the case for pairs of function and disability scales. The strongest correlation was that between the SPADI disability subscale scores and the SST scores (r_s=–.80). Weaker correlations were found between the SPADI disability subscale scores and the UCLA function subscale scores (r_s=–.64). The correlation between the UCLA function subscale scores and the SST scores was .60. Generally, we believe there are 2 explanations for a lack of strong association between pairs of subscales developed to measure the same construct. These correlations will be low if (1) either or both subscales of a pair produce a large amount of measurement error or (2) the subscales actually measure different constructs. The relatively high internal consistency values obtained for the subscales in this study indicate that measurement error alone was not the sole cause of the moderate correlations.

Discriminant validity.
The strength of the obtained correlations between the SPADI pain and disability subscale scores and between the UCLA pain and function subscale scores does not support the discriminant validity of these scales. The correlation analyses suggest greater convergence than divergence between the subscales of these measures. This finding is exemplified by the fact that the association between the SPADI pain and disability subscale scores was stronger than that between the SPADI pain subscale scores and the UCLA pain subscale scores or between the SPADI disability subscale scores and the scores of either of the function subscales (SST or UCLA function subscale). Our results for the SPADI are similar to those obtained by Roach and Colleagues.⁷

The findings of our study support the following general assertions:

1. The SPADI appears to be a unidimensional scale. Therefore, it could be argued that the total score should combine and equally weight all of the items of both subscales.
   
2. Because of the weakness of the SPADI's construct validity, using it as a "gold standard" for comparing other scales and their respective subscales may result in misleading information.
   
3. Although the Cronbach alpha values of the SST and SPADI are acceptable for comparing groups of patients, their SEMs are quite large, suggesting that they are too imprecise for following an individual patient's change in status over time.
   
4. Because the UCLA Shoulder Scale depends on single-item scales, it was impossible to put it under the same psychometric scrutiny as the SPADI and SST. Therefore, our findings fail to either validate or invalidate its use with either group- or individual-level comparisons. However, when all other factor are equal, multiple-item scales yield much more reliable measurements than do single-item scales. It is doubtful, therefore, that the UCLA Shoulder Scale is precise enough to effectively follow the progress of individual patients in the clinic setting.
   

Limitations and Future Research

A limitation of this study was that the sample was obtained from the practice of a single orthopedic surgeon. Although the sample was similar to the general population of patients with shoulder disorders with respect to mean age and sex composition,²¹ it may contrast on other important variables. For example, information about ethnicity, shoulder diagnosis, and severity or acuteness of the impairment were not collected on the study population, nor did we account for socioeconomic status or vocation. It is possible, therefore, that our sample contrasted with the general population on these variables.

We used a classical approach for obtaining estimates of the SEMs.¹⁴ There are a number of drawbacks to such an approach.²² One drawback with substantial implications is that this is a "one-estimate-fits-all" method. It provides a single value to represent the precision of a scale on average across the whole range of possible scores. Scales vary, however, with regard to how precisely they measure people at different levels of the trait being measured.²³ As an example, it is possible, indeed likely, in our view, that a scale of shoulder function will measure with unequal precision individuals who have low, medium, and high levels of shoulder function. The classically obtained SEM estimate, because it is an average, underestimates how precisely a scale measures at some trait levels and overestimates its precision at other trait levels.

Another limitation of our study was the use of the SPADI as the "gold standard" to which the other 2 scales were compared in establishing validity. Although the SPADI has certainly undergone more psychometric testing than other shoulder scales, it has been evaluated in fairly homogeneous samples, primarily among male participants.^7,8 Future studies should examine the relationship between scores on shoulder outcome scales and external clinical criteria.

	Conclusions

Top Abstract Introduction Materials and Methods Results Discussion Conclusions Appendix References

 
Our results suggest that the multi-item scales evaluated in this study (SPADI pain subscale, SPADI disability subscale, and the SST) have good group-level reliability but questionable precision for documenting the status of individual patients. Our results also suggest that pain and function, although distinguishable conceptually, may not be distinguished by patients.

	Appendix

Top Abstract Introduction Materials and Methods Results Discussion Conclusions Appendix References

 

View larger version (130K): [in this window] [in a new window] Appendix.a Self-Report Sections of the University of California-Los Angeles (UCLA) Shoulder Scale,5 the Shoulder Pain and Disability Scale (SPADI),7 and the Simple Shoulder Test (SST)6

	Footnotes

 
This study was approved by the Human Subjects Review Committee at Texas Woman's University.

This study was supported, in part, by a doctoral research fellowship awarded to Ms Roddey by Texas Orthopedic Hospital and Columbia/HCA.

	References

Top Abstract Introduction Materials and Methods Results Discussion Conclusions Appendix References

 

1. Guide to Physical Therapist Practice. Rev ed. Alexandria Va: American Physical Therapy Association;1999 .
2. Jette AM. Outcomes research: shifting the dominant research paradigm in physical therapy. Phys Ther.1995; 75:965–970.[Abstract/Free Full Text]
3. Green S, Buchbinder R, Glazier R, Forbes A. Systematic review of randomised controlled trials of interventions for painful shoulder: selection criteria, outcome assessment, and efficacy. BMJ.1998; 316:354–360.[Abstract/Free Full Text]
4. van der Heijden GJ, van der Windt DA, de Winter AF. Physiotherapy for patients with soft tissue shoulder disorders: a systematic review of randomised clinical trials. BMJ.1997; 315:25–30.[Abstract/Free Full Text]
5. Ellman H, Hanker G, Bayer M. Repair of the rotator cuff: end-result study of factors influencing reconstruction. J Bone Joint Surg Am.1986; 68:1136–1144.[Abstract/Free Full Text]
6. Lippitt SB, Harryman DT, Matsen FA. A practical tool for evaluating function: the Simple Shoulder Test. In: Matsen FA, Fu FH, Hawkins RJ, eds. The Shoulder: A Balance of Mobility and Stability. Rosemont, Ill: American Academy of Orthopedic Surgeons,1993 :501–530.
7. Roach KE, Budiman-Mak E, Songsiridej N, Lertratanakul Y. Development of a Shoulder Pain and Disability Index. Arthritis Care Res.1991; 4:143–149.[Medline]
8. Williams JW Jr, Holleman DR Jr, Simel DL. Measuring shoulder function with the Shoulder Pain and Disability Index. J Rheumatol.1995; 22:727–732.[ISI][Medline]
9. Beaton DE, Richards RR. Measuring function of the shoulder: a cross-sectional comparison of five questionnaires. J Bone Joint Surg Am.1996; 78:882–890. '[Abstract/Free Full Text]
10. Heald SL, Riddle DL, Lamb RL. The Shoulder Pain and Disability Index: the construct validity and responsiveness of a region-specific disability measure. Phys Ther.1997; 77:1079–1089.[Abstract/Free Full Text]
11. Beaton DE, Richards RR. Assessing the reliability and responsiveness of 5 shoulder questionnaires. J Shoulder Elbow Surg.1998; 7:565–572.[ISI][Medline]
12. McHorney CA, Tarlov AR. Individual-patient monitoring in clinical practice: are available health status surveys adequate? Qual Life Res.1995; 4:293–307.[ISI][Medline]
13. McDonald RP. Factor Analysis and Related Methods. Hillsdale, NJ: Lawrence Erlbaum Associates Inc,1985 .
14. Anastasi A. Psychological Testing. New York, NY: Macmillan Publishing Co,1988 :122–127.
15. Diamond JJ. A practical application of reliability theory to family practice research. Fam Pract Res J.1991; 11:357–362.[Medline]
16. Miller DB, Briggs H, Corcoran K. Fear of AIDS and Homophobia Scales: additional estimates of reliability and validity. Psychol Rep.1997; 81(3 pt 1):783–786.
17. Sloan DA, Donnelly MB, Schwartz RW, et al. The use of the Objective Structured Clinical Examination (OSCE) for evaluation and instruction in graduate medical education. J Surg Res.1996; 63:225–230.[ISI][Medline]
18. Sheridan MJ. A psychometric assessment of the Children of Alcoholics Screening Test (CAST). J Stud Alcohol.1995; 56:156–160.[ISI][Medline]
19. Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. East Norwalk, Conn: Appleton & Lange Publishers,1993 .
20. Yarnitsky D, Sprecher E, Zaslansky R, Hemli JA. Multiple session experimental pain measurement. Pain.1997; 67:327–333.
21. Bergenudd H, Nilsson B. The prevalence of locomotor complaints in middle age and their relationship to health and socioeconomic factors. Clin Orthop.1994; 308:264–270.
22. Hamilton RK, Swamnathan H. Item Response Theory. Boston, Mass: Kluwer-Nijhoff Publishing,1985 .
23. McKinley R. An introduction to Item Response Theory. Measurement and Evaluation in Counseling and Development.1989; 22:37–57.

This article has been cited by other articles:

				J. R. Lynch, A. K. Franta, W. H. Montgomery Jr, T. R. Lenters, D. Mounce, and F. A. Matsen III Self-Assessed Outcome at Two to Four Years After Shoulder Hemiarthroplasty with Concentric Glenoid Reaming J. Bone Joint Surg. Am., June 1, 2007; 89(6): 1284 - 1292. [Abstract] [Full Text] [PDF]

				P. F Beattie, R. M Nelson, and A. Lis Spanish-Language Version of the MedRisk Instrument for Measuring Patient Satisfaction With Physical Therapy Care (MRPS): Preliminary Validation Physical Therapy, June 1, 2007; 87(6): 793 - 800. [Abstract] [Full Text] [PDF]

				F. Angst, J. Goldhahn, G. Pap, A. F. Mannion, K. E. Roach, D. Siebertz, S. Drerup, H. K. Schwyzer, and B. R. Simmen Cross-cultural adaptation, reliability and validity of the German Shoulder Pain and Disability Index (SPADI) Rheumatology, January 1, 2007; 46(1): 87 - 92. [Abstract] [Full Text] [PDF]

				A. Cacchio, M. Paoloni, A. Barile, R. Don, F. de Paulis, V. Calvisi, A. Ranavolo, M. Frascarelli, V. Santilli, and G. Spacca Effectiveness of Radial Shock-Wave Therapy for Calcific Tendinitis of the Shoulder: Single-Blind, Randomized Clinical Study Physical Therapy, May 1, 2006; 86(5): 672 - 682. [Abstract] [Full Text] [PDF]

				A. M. Smith, S. A. Barnes, J. W. Sperling, C. M. Farrell, J. D. Cummings, and R. H. Cofield Patient and Physician-Assessed Shoulder Function After Arthroplasty J. Bone Joint Surg. Am., March 1, 2006; 88(3): 508 - 513. [Abstract] [Full Text] [PDF]

				M. S. Kocher, M. P. Horan, K. K. Briggs, T. R. Richardson, J. O'Holleran, and R. J. Hawkins Reliability, Validity, and Responsiveness of the American Shoulder and Elbow Surgeons Subjective Shoulder Scale in Patients with Shoulder Instability, Rotator Cuff Disease, and Glenohumeral Arthritis J. Bone Joint Surg. Am., September 1, 2005; 87(9): 2006 - 2011. [Abstract] [Full Text] [PDF]

				A. M. Smith, J. W. Sperling, and R. H. Cofield Rotator Cuff Repair in Patients with Rheumatoid Arthritis J. Bone Joint Surg. Am., August 1, 2005; 87(8): 1782 - 1787. [Abstract] [Full Text] [PDF]

				A Paul, M Lewis, M F Shadforth, P R Croft, D A W M van der Windt, and E M Hay A comparison of four shoulder-specific questionnaires in primary care Ann Rheum Dis, October 1, 2004; 63(10): 1293 - 1299. [Abstract] [Full Text] [PDF]

				S D M Bot, C B Terwee, D A W M van der Windt, L M Bouter, J Dekker, and H C W de Vet Clinimetric evaluation of shoulder disability questionnaires: a systematic review of the literature Ann Rheum Dis, April 1, 2004; 63(4): 335 - 341. [Abstract] [Full Text] [PDF]

				E. V. Fehringer, B. Kopjar, R. S. Boorman, R. S. Churchill, K. L. Smith, and F. A. Matsen III Characterizing the Functional Improvement After Total Shoulder Arthroplasty for Osteoarthritis J. Bone Joint Surg. Am., August 12, 2002; 84(8): 1349 - 1353. [Abstract] [Full Text] [PDF]

				P. F Beattie, M. B. Pinto, M. K Nelson, and R. Nelson Patient Satisfaction With Outpatient Physical Therapy: Instrument Validation Physical Therapy, June 1, 2002; 82(6): 557 - 565. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when Rapid Responses are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Roddey, T. S Articles by Hanten, W. Search for Related Content PubMed PubMed Citation Articles by Roddey, T. S Articles by Hanten, W.