Development of a questionnaire to measure impact and outcomes of brachial plexus injury

Transcription

1 Washington University School of Meicine Digital Open Access Publications 2018 Development of a questionnaire to measure impact an outcomes of brachial plexus injury Carol A. Mancuso Weill Cornell Meical College Steve K. Lee Weill Cornell Meical College Eliana B. Saltzman Icahn School of Meicine at Mount Sinai Zina Moel Rutgers Robert Woo Johnson Meical School Zoe A. Laners Hospital for Special Surgery, New York, NY See next page for aitional authors Follow this an aitional works at: Recommene Citation Mancuso, Carol A.; Lee, Steve K.; Saltzman, Eliana B.; Moel, Zina; Laners, Zoe A.; Dy, Christopher J.; an Wolfe, Scott W.,,"Development of a questionnaire to measure impact an outcomes of brachial plexus injury." The Journal of Bone an Joint Surgery.100,3. e14. (2018). This Open Access Publication is brought to you for free an open access by Digital Commons@Becker. It has been accepte for inclusion in Open Access Publications by an authorize aministrator of Digital Commons@Becker. For more information, please contact engeszer@wustl.eu.

2 Authors Carol A. Mancuso, Steve K. Lee, Eliana B. Saltzman, Zina Moel, Zoe A. Laners, Christopher J. Dy, an Scott W. Wolfe This open access publication is available at Digital

3 e14(1) COPYRIGHT Ó 2018 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED Development of a Questionnaire to Measure Impact an Outcomes of Brachial Plexus Injury CarolA.Mancuso,MD,SteveK.Lee,MD,ElianaB.Saltzman,BS,ZinaMoel,BA,ZoeA.Laners,MSW, ChristopherJ.Dy,MD,MPH,anScottW.Wolfe,MD Investigation performe at the Hospital for Special Surgery, New York, NY Backgroun: The physical an psychological impact of brachial plexus injury (BPI) has not been comprehensively measure with BPI-specific scales. Our objective was to evelop an test a patient-erive questionnaire to measure the impact an outcomes of BPI. Methos: We evelope a questionnaire in 3 phases with preoperative an postoperative patients. Phase 1 inclue interviews of patients using open-ene questions aressing the impact of BPI an improvement expecte (preoperative patients) or receive (postoperative patients). Phase 2 involve assembling a raft questionnaire an aministering the questionnaire twice to establish test-retest reliability. Phase 3 involve selecting final items, eveloping a scoring system, an assessing valiity. Patient scores using the questionnaire were assesse in comparison with scores of the Disabilities of the Arm, Shouler an Han (DASH) an RAND-36 measures. Results: Patients with partial or complete plexopathy participate. In Phase 1 (23 patients), iscrete categories were iscerne from open-ene responses an became items for the preoperative an postoperative versions of the questionnaire. In Phase 2 (50 patients [14 from Phase 1]), test-retest reliability was establishe, with weighte kappa values of 0.50 for all items. In Phase 3, 43 items were retaine an groupe into 4 s: symptoms, limitations, emotion, an improvement expecte (preoperative) or improvement receive (postoperative). A score for each, ranging from 0 to 100, can be calculate, with higher scores inicating more symptoms, limitations, an emotional istress, an greater improvement expecte (or receive). Preoperative scores were worse than postoperative scores for the symptoms, limitations, an emotion s (composite score of 48 compare with 38; p = 0.05), an more improvement was expecte than was receive (69 compare with 53; p = 0.01). Correlations with the DASH (0.44 to 0.74) an RAND- 36 (0.23 to 0.80) for relate scales were consistent an moerate, inicating that the new questionnaire is vali an istinct. Conclusions: We evelope a patient-erive questionnaire that measures the physical an psychological impact of BPI on preoperative an postoperative patients an the amount of improvement expecte or receive from surgery. This BPIspecific questionnaire enhances the comprehensive assessment of this population. Brachial plexus injury (BPI) has a profoun impact on all aspects an quality of life 1,2. Patients are usually young aults who sustaine trauma, such as from a motor vehicle accient or sports injury 1,3,4. Severe motor an sensory ysfunction an variable egrees of neuropathic pain are hallmarks of the conition. Furthermore, the suen nature of the injury can lea to a realm of psychological challenges, such as emotional istress, loss of inepenence, an epression, that can overwhelm patients an potentially hamper recovery. Capturing these psychological elements within a BPI iseasespecific context is necessary in orer to thoroughly unerstan isability an recovery from this complex conition. For BPI, outcomes historically have been reporte on the basis of surgeon-grae muscle function; more recent efforts have incorporate functional an generic psychological questionnaires However, these outcomes o not simultaneously inclue broa perspectives of, an attention to, multiple physical an emotional aspects of life impacte by BPI over time. In aition, a BPI-specific questionnaire is neee to stanarize outcome reporting among clinicians an Disclosure: This work was supporte by the J.W. Kieckhefer Founation, the Howar J. Golen Family Research Fun, an a Richar Menschel institutional awar; these groups i not have roles in the investigation. The Disclosure of Potential Conflicts of Interest forms are provie with the online version of the article ( J Bone Joint Surg Am. 2018;100:e14(1-13)

4 e14(2) TABLE I Patient Demographic an Clinical Characteristics Phase 1 Phase 2 Characteristic Preop., N = 10 Postop., N = 13 Preop., N = 23 Postop., N = 27 Age* (yr) 38 ± 14 (24-63) 36 ± 15 (19-59) 41 ± 15 (23-71) 40 ± 16 (20-84) Male (no. [%]) 10 (100%) 9 (69%) 20 (87%) 24 (89%) Working (no. [%]) No, not because of injury 0 (0%) 1 (8%) 2 (9%) 1 (4%) No, because of injury 9 (90%) 8 (61%) 11 (48%) 13 (48%) Yes, with accommoations 1 (10%) 4 (31%) 8 (35%) 9 (33%) Yes, without accommoations 0 (0%) 0 (0%) 2 (9%) 4 (15%) Injury on ominant sie (no. [%]) 7 (70%) 6 (46%) 8 (35%) 18 (67%) How injury occurre (no. [%]) Motorcycle 6 (60%) 5 (39%) 9 (39%) 9 (33%) Motor vehicle 0 (0%) 6 (46%) 7 (30%) 5 (19%) Recreational vehicle 3 (30%) 0 (0%) 2 (9%) 6 (22%) Other 1 (10%) 2 (15%) 5 (22%) 7 (26%) Complete plexopathy (no. [%]) 4 (40%) 6 (46%) 13 (57%) 12 (44%) Time between: Injury an first interview (mo) 7 (3-14) 33 (14-153) 5 (3-7) 73 (13-195) First interview an surgery (ays) 12 (1-38) 21 (2-81) Surgery an first interview (mo) 14 (10-22) 58 (9-142) First an secon interviews (ays) 3 (1-9) 4 (2-11) DASH score 52 ± ± ± ± 20 RAND-36 physical health composite score# 33 ± 9 39 ± ± 9 41 ± 10 RAND-36 mental health composite score# 48 ± ± ± ± 12 If rest of life like past 24 hr (no. [%]) Delighte 0 (0%) 0 (0%) 0 (0%) 1 (4%) Please 0 (0%) 3 (23%) 0 (0%) 4 (15%) Mostly satisfie 0 (0%) 2 (15%) 2 (9%) 4 (15%) Mixe 2 (20%) 3 (23%) 0 (0%) 8 (30%) Mostly issatisfie 1 (10%) 2 (15%) 3 (13%) 5 (19%) Unhappy 3 (30%) 1 (8%) 7 (30%) 4 (15%) Terrible 4 (40%) 2 (15%) 11 (48%) 1 (4%) *The values are given as the mean an the stanar eviation, with the range in parentheses. Other surgery, raiation therapy, malposition while not conscious, sports injury, work injury, or knife woun. The values are given as the mean, with the range in parentheses. DASH = Disabilities of the Arm, Shouler an Han. Possible score ranges from 0 to 100, with a higher score inicating worse status. The values are given as the mean an the stanar eviation. #Possible score ranges from 0 to 100, with a higher score inicating better status. The values are given as the mean an the stanar eviation. researchers. We know of no uniform an wiely accepte patient-reporte questionnaires for ocumenting an comparing BPI outcomes accoring to clinical characteristics, such as nerve root levels involve, an results of novel surgical techniques, such as nerve transfers an grafting. In aition, there appear to be no existing questionnaires that simultaneously aress the unique collection of physical an psychological symptoms that affect patients with BPI. The objective of the current stuy was to evelop an test a BPI-specific questionnaire that aresses the physical an psychological impact of BPI an also aresses patients expectations for improvement from surgery an their assessment of actual improvement receive. We hypothesize that patients woul cite limitations an expectations with respect to multiple aspects of physical an mental well-being. Materials an Methos Patients with BPI who were unergoing, or who ha unergone, surgical reconstruction were enrolle in this multiple-phase stuy uring routine office visits an provie written informe consent. This stuy was approve by the institutional review boar at the Hospital for Special Surgery.

5 e14(3) TABLE II Weighte Kappa Values for Each Item of the BPI Questionnaire* Questionnaire Item Preop., N = 23 Postop., N = 27 Throbbing pain Stabbing pain Tingling Numbness Heaviness Bathing an hygiene Dressing Eating Depenence on opposite arm an han Activities with family an friens Depenent on others for tasks Depenent on others financially Depenent on others emotionally Decrease recreation or sports Self-conscious about appearance Self-conscious about isabilities Difficulty coping Deterioration in overall health Better if amputation Effect on employment/school Effect on future career plans Sa moo Stress Self-esteem Anger Guilt Frustration with limitations Frustration with time to heal Altere life priorities Paying for meical care Extra fatigue Relieve pain Relieve numbness an tingling Improve sleep Move arm, elbow, han Manage personal care Reuce nee for pain meicine Interact with family an friens Return to work Return to recreation or sports Restore emotional well-being Return to way was before injury Improvement expecte after 1 yr/as a result of surgery *Measuring agreement between first an secon aministrations of the questionnaire for preoperative an postoperative patients in Phase 2. BPI = brachial plexus injury.

6 e14(4) Fig. 1-A Figs. 1-A an 1-B Preoperative version of the Impact of Brachial Plexus Injury Questionnaire. (Reprouce with permission of Hospital for Special Surgery.) Phase 1: Ientifying Items for Draft Questionnaire Phase 1 was base on a previously reporte qualitative stuy of patients physical an psychosocial limitations ue to BPI an their expectations of improvement from surgery 11.Inbrief,patients were eligible if they were 18 years ol, spoke English, an were scheule for surgery (preoperative group) or ha unergone surgery within the previous 9 to 24 months (postoperative group) for partial or complete BPI. Patients were interviewe in person by a single investigator (C.A.M.) who was experience in qualitative research using stanar questions an techniques. Patients were aske open-ene questions about what bothere them most about their arm, what activities they ha curtaile, what accommoations they ha mae, an what their expectations were for improvement (preoperative group) or what improvement ha been achieve (postoperative group). Patients were encourage to volunteer as many comments as they wishe, an their responses were written own verbatim. Patients also complete the Disabilities of the Arm, Shouler an Han (DASH), a 21-item questionnaire measuring symptoms an limitations ue to upper-extremity ysfunction 12, an the RAND Health Survey (RAND-36), a 36-item questionnaire measuring general physical an mental health status 13.Patientsalso answere a single question as a global assessment of the conition of the arm aapte from a valiate measure of well-being, with 7 response options ranging from elighte to terrible (Table I) 14. Responses to the open-ene questions were assesse with stanar qualitative techniques using groune theory, a process by which responses are reviewe to ientify unique

7 e14(5) Fig. 1-B concepts, which are then groupe into larger categories through an iterative process 15,16. Phase 2: Assembling Draft Questionnaire an Establishing Test-Retest Reliability Categories from Phase 1 became the items for the raft questionnaire an were phrase using patients terminology. Response options also were wore accoring to patients terms. Using the same items, 2 versions of the questionnaire were create, to query patients about their conition before surgery (preoperative version) or after surgery (postoperative version). The raft versions were then teste among aitional patients who were 18 years ol, spoke English, an were either scheule for surgery (complete the preoperative version) or ha unergone surgery at least 9 months prior (complete the postoperative version). To establish testretest reliability, patients complete the same version of the questionnaire twice, several ays apart. In most cases, the first aministration occurre uring an in-person interview an the secon, uring a telephone interview. To aress external valiity, patients also complete the DASH, the RAND-36, an the global elighte-terrible question at thetimeofthefirst interview. Asamplesizeof50meetsrigorouscriteriaforrepeatability testing 17, an thus, we enrolle 50 patients in Phase 2. Phase 3: Selecting Final Items, Scoring, an Valiity The weighte kappa statistic was use to measure agreement between the first an secon aministrations for each item. The

8 e14(6) Fig. 2-A Figs. 2-A an 2-B Postoperative version of the Impact of Brachial Plexus Injury Questionnaire. (Reprouce with permission of Hospital for Special Surgery.) weighte kappa measures agreement above that ue to chance, an ranges from 0 to 1 (with 1 inicating perfect agreement) 18. An item was retaine for the final questionnaire if the kappa value was A system was evelope to generate scores for the s an omains (escribe below), an intraclass correlation coefficients (ICCs) were calculate to measure intrapatient agreement in scores. For both kappa an ICC, a value of <0.4 inicates slight/fair agreement; 0.4 to 0.6, moerate; >0.6 to 0.75, goo; an >0.75 to 1, excellent. External valiity was assesse by comparing scores on the evelope questionnaire with DASH an RAND-36 scores using Pearson correlations an with responses on the global assessment using Spearman correlations. Internal valiity was assesse with Cronbach alpha correlations. Results Phase 1: Ientifying Items for Draft Questionnaire Ten preoperative an 13 postoperative patients were enrolle from April 2013 to March The mean age (an stanar eviation) was 37 ± 14 years, 19 of the patients were male, all were working or were full-time stuents at the time of the BPI, but most were not working at enrollment because of the BPI (Table I). Most injuries were partial plexopathies an were ue to motorcycle or motor vehicle accients. Both preoperative an postoperative patients volunteere that the BPI ha ramatically impacte their lives in multiple ways 11, which inclue persistent pain, the inability to provie self-care, reliance on others for financial support,

9 e14(7) Fig. 2-B career moifications, an eterioration in general health. Psychological effects inclue being self-conscious about appearance, anger, an lower self-esteem. Expectations for improvement inclue improving the ability to move the arm, interact with others, return to work, an ecrease pain meications. Phase 2: Assembling Draft Questionnaire an Test-Retest Reliability The 43-item raft questionnaire aresse symptoms, limitations, emotions, an amount of improvement expecte (or receive). Likert response options were assigne for most items, except for the items regaring emotion,

10 e14(8) Fig. 3-A Figs. 3-A an 3-B Scoring instructions for the Impact of Brachial Plexus Injury Questionnaire. (Reprouce with permission of Hospital for Special Surgery.) which were assigne responses accoring to a numerical rating scale. Twenty-three preoperative an 27 postoperative patients were enrolle in Phase 2 (14 of these patients also participate in Phase 1) from August 2014 to February The mean age was 41 ± 15 years, 44 of the patients were male, most were working at the time of the BPI, but nearly half were not working because of the BPI at enrollment. Compare with the preoperative patients, the postoperative patients ha better DASH (p = 0.009) an RAND-36 physical health scores (p = 0.02) an were more likely to be satisfie if no further clinical improvement was anticipate (p < ). For both groups, the amount of time between the first an secon interviews was 3 to 4 ays. The meian time since surgery for the postoperative group was 3 years. Phase 3: Selecting Final Items, Scoring, an Valiity Weighte kappa values range from 0.50 to 0.92 for the preoperative version an from 0.52 to 0.96 for the postoperative version (Table II), an thus, all 43 items an their formats were retaine to form the final Impact of Brachial Plexus Injury Questionnaire (Figs. 1-A through 2-B). The questionnaire was assemble accoring to 4 thematic s that parallel the clinical scenario, namely, symptoms, limitations, emotion, an improvement. The symptoms has 5 items aressing the severity of pain, numbness, an tingling; the response options range from none to a lot, with numerical values on a 4-point Likert scale. The limitations has 16 items: 3 items aress ifficulty with personal care, with responses on a 5-point scale ranging

11 e14(9) Fig. 3-B from no ifficulty to someone ha to o it for me ; 11 items aress functional restrictions, with responses on a 5-point scale ranging from notatall to completely ; an 2 items aress work/school an career plans, with responses ranging from no change to not able to work/ atten school because of BPI. The emotion has 10 items aressing istress ue to the BPI; responses accoring to a 10-point numerical rating range from not at all to alot. The improvement expecte (preoperative) an the improvement receive (postoperative) s have 12 items aressing symptoms, movement, meications, employment, an emotions; responses range from complete improvement to no improvement. A score can be generate for each accoring to the scoring instructions (Figs. 3-A an 3-B); the numerical values assigne for each of the patient responses are summe, an the score is normalize on a scale of 0 to 100. Given that symptoms, limitations, an emotional istress reflect isability, higher scores for those s reflect more of that attribute an, therefore, worse status 19.An overall isability omain score also can be calculate as the mean of those s. This composite omain score is useful to gauge overall isability, while the scores provie information on which attributes are causing the most isability. Similarly, for the improvement, a higher score inicates more of that attribute, i.e., greater expectations for improvement (preoperative) or greater improvement receive (postoperative) 19. An improvement omain score can be reporte as the improvement expecte score for preoperative patients or the

12 e14(10) TABLE III Domain an Subscale Scores for Patients in Phase 2* Preop., N = 23 Postop., N = 27 Domain an Subscale Score No. with Min., Max. Score Cronbach Alpha Coefficient ICC Score No. with Min., Max. Score Cronbach Alpha Coefficient ICC Disability omain Symptoms Limitations Emotion 58 ± 25 (20-100) 41 ± 16 (11-77) 44 ± 23 (0-88) Overall 48 ± 18 (11-88) Improvement omain Expecte (preop.) Receive (postop.) 69 ± 20 (21-100) Overall 69 ± 20 (21-100) 0, ± 25 (13-100) 0, ± 14 (7-63) 1, ± 27 (0-97) 0, ± 18 (8-82) 0, , , , , ± 20 (23-88) 0, ± 20 (23-88) 0, , *Scores are from the first aministration of questionnaire. Possible scores range from 0 to 100. For the isability omain, a higher value inicates a worse conition, an for the improvement omain, a higher value inicates greater improvement expecte or receive. The ICC (intraclass correlation coefficient) measures agreement between the first an secon aministrations of the questionnaire. The values are given as the mean an the stanar eviation, with the range in parentheses. The overall score for the isability omain is a composite score, reflecting the mean of scores calculate for the symptoms, limitations, an emotions s. improvement receive score for postoperative patients. Subscale an omain scores were calculate an were assesse for mean values an ranges for patients in Phase 2. Results from the first aministration are summarize in Table III. For the preoperative version, scores for the s in the isability omain were normally istribute an spanne almost the entire possible range, with few patients having minimum (i.e., floor) or maximum (i.e., ceiling) scores. Cronbach alpha coefficients inicate goo to excellent internal valiity (0.61 to 0.91). Of the 3 isability s, the symptoms reflecte the greatest impact (ha the highest mean score, 58 compare with 41 for limitations an 44 for emotion, with an overall omain score of 48). Scores for the improvement-expecte also were normally istribute but were shifte towar higher values (mean of 69), inicating greater expectations. Results for the secon aministration of the questionnaire were similar, as reflecte by high ICC values (0.85 to 0.96). For the postoperative version, all scores from the first aministration also were normally istribute an spanne almost the entire possible range, with few patients having minimum or maximum scores (Table III). Cronbach alpha coefficients inicate goo to excellent internal valiity (0.64 to 0.94). The symptoms score reflecte the greatest impact (mean of 49 compare with 31 for emotion an 34 for limitations, with an overall omain score of 38), an scores for the improvement-receive also were normally istribute (mean of 53; interquartile range, 35 to 71). Results for the secon questionnaire aministration were similar, as reflecte by high ICC values (0.87 to 0.94). When mean scores were compare between the postoperative an preoperative groups, the overall isability omain score for the postoperative group was lower (38 compare with 48; p = 0.05), inicating less-severe symptoms, limitations, an emotional istress. The score for the amount of improvement also was lower in the postoperative group (53 compare with 69; p = 0.01); however, given the irection of scoring, this inicates that the actual improvement receive (postoperative group) was less than the expecte improvement (preoperative group). The final analyses assesse external valiity by comparing results from use of the questionnaire with outcomes using the stanar scales. There were multiple associations between the various s an stanar scales (Table IV). Worse symptoms an limitation scores

13 e14(11) TABLE IV Correlation Coefficients (an Corresponing P Values) from Comparisons Between Scores on the BPI Questionnaire an Stanar Scales for Patients in Phase 2* Preop., N = 23 Postop., N = 27 Domain an Subscale DASH RAND-36 Physical Health RAND-36 Mental Health Global Assessment DASH RAND-36 Physical Health RAND-36 Mental Health Global Assessment Disability omain Symptoms Limitations Emotion Overall 0.60 Improvement omain Expecte (preop.) Receive (postop.) 0.50 ( 0.01) Overall 0.50 ( 0.01) ( 0.01) ( 0.01) ( ) 0.74 ( ) ( ) ( 0.01) ( ) ( 0.01) ( ) *Base on first aministration of BPI (brachial plexus injury) questionnaire. The s of the isability omain, DASH (Disabilities of the Arm, Shouler an Han), an global assessment are score such that a higher score inicates worse status. The RAND-36 physical an mental health measures are score such that a higher score inicates better status, an the s of the improvement omain are score such that a higher score inicates more improvement expecte (preoperative) or receive (postoperative). were associate with worse DASH an RAND-36 physical health scores (Fig. 4). A worse emotion score was correlate with a worse RAND-36 mental health score. These associations existe for both preoperative an postoperative patients. There similarly were correlations between the improvement s an the DASH, RAND-36 physical health, an global assessment scores. However, the irection of the association was ifferent for preoperative an postoperative patients. Specifically, whereas a higher DASH score (worse status) preoperatively was associate with a higher score for expecte improvement, a higher DASH score postoperatively was associate with a lower score for receive improvement (Fig. 5). For most patients, our questionnaire took approximately 12 minutes to complete; the DASH, 10 minutes; the RAND-36, 6 minutes; an the transition from the isease-specific (DASH) to the more general (RAND-36) perspective, 5 minutes. There were no incomplete questionnaires, an no assistance in completing the questionnaires was require. Fig. 4 Disabilities of the Arm, Shouler an Han (DASH) versus Impact of Brachial Plexus Injury Questionnaire limitations scores preoperatively (Fig. 4-A) an postoperatively (Fig. 4-B).

14 e14(12) Fig. 5 Disabilities of the Arm, Shouler an Han (DASH) versus Impact of Brachial Plexus Injury Questionnaire preoperative improvement-expecte (Fig. 5-A) an postoperative improvement-receive (Fig. 5-B) scores. Discussion We evelope a questionnaire to measure physical an psychological isability from BPI an to assess improvement expecte an receive from surgery. The questionnaire items were erive from patients input, an preoperative an postoperative versions were assemble. Each item was teste for repeatability, an each was teste for repeatability an valiity. The questionnaire potentially can improve clinical care in several ways. First, the questionnaire provies a template from which patients can then iscuss the spectrum of physical an emotional effects of BPI with their surgical team. This, in turn, offers proviers the opportunity to comprehensively aress patients nees irectly or through referral. Secon, the questionnaire fosters iscussion of realistic outcome expectations, which are necessary to maintain motivation an ensure long-term participation in rehabilitation. Thir, the questionnaire provies a vali an stanarize metho for clinicians an researchers to ocument an compare patient-reporte outcomes from specific or novel surgical interventions. To facilitate the communication of results, we provie the option of calculating overall isability an improvement omain scores. The isability omain is compose of the symptoms, limitations, an emotion s. Although constitute by ifferent numbers of items, each is affore equal weight in the omain score to reflect its relevance to the clinical scenario. The improvement omain is compose of the improvement, tailore to the preoperative state (expecte benefit) or postoperative state (actual benefit). Our questionnaire contains 43 items with simple an varie response options base on patients terminology to maximize ease of completion an participant attentiveness. Despite current trens to use short surveys, our goal was to capture as much of this complex conition as possible using a necessary an sufficient number of items. Our approach is consistent with perspectives of other investigators who have avocate for comprehensive measures of impairment from BPI 2,8. Traitionally, BPI outcomes preominantly reflecte surgeon-measure physical parameters (range of motion, strength, sensation) 4,8,20.However,tofullyevaluateoutcomesof treatment, assessments from patients perspectives are essential, especially for complex injuries that affect every facet of life. The DASH is a wiely use questionnaire that measures general bilateral upper-extremity aily function 5,7-9,12, Several other upper-extremity questionnaires (none specific to BPI) an various general health questionnaires also have been use 1,3,8,9,20,22. Recently, a 31-item survey was evelope for BPI to measure mechanical function of the affecte limb for aily activities (e.g., put toothpaste on a toothbrush) an will be useful to track performance of these activities longituinally 6. Our questionnaire, in contrast, measures the impact of BPI on psychosocial life, activities of aily living, emotional aspects of recovery, an pain. A strength of our questionnaire is that it was base on patient input an therefore captures information that might otherwise not be inclue in a physician-erive questionnaire, such as amputation, altere life priorities, emotional well-being, an cost of BPI 7. Our methoology also permitte us to inclue a aressing preoperative expectations for improvement an then to capture the amount of improvement actually receive postoperatively. Measurement of expectations shoul be part of comprehensive assessments of BPI because patients usually o not know others with BPI, cannot witness outcomes in peers, an thus may not know realistic expectations 20. In our stuy, on the basis of group means, the preoperative patients ha high improvement-expecte scores (incluing expecting to be back to normal) an the postoperative patients ha lower improvement-receive scores. This fining provies evience that preoperative expectations may be unrealistic. Consequently, surgeons shoul consier counseling patients on appropriate goals that will better align expectations with realistic surgical outcomes. The issue of potentially unrealistic expectations also was uncovere through associations we foun between expecte improvement an preoperative function an between receive improvement an postoperative function. For example, patients who ha worse DASH scores preoperatively ha higher improvement-expecte scores. In contrast, patients who ha worse DASH scores postoperatively ha lower improvementreceive scores. Thus, while it is unerstanable that patients with the worst preoperative status woul have the most to expect, the salient question is whether such high expectations are realistic for these most-isable patients. For postoperative patients, it is logical that those with worse function woul report less improvement.

15 e14(13) This stuy ha several limitations. First, patients were enrolle from a tertiary center an may not represent patients in other settings. Secon, although aministere uring interviews, the first aministration of the questionnaire was inperson an the secon, by telephone. Thir, it woul have been ieal to have the same patients complete both preoperative an postoperative versions of the questionnaire. However, this was not feasible given that the time to recuperate from BPI is prolonge. Fourth, although our collective sample size is one of the largest reporte, it was not large enough for subanalyses to elineate the questionnaire s performance characteristics on the basis of emographic an clinical variables, such as the number of injure root levels. An ongoing longituinal stuy with a larger sample size will aress this an the responsiveness of the questionnaire. In summary, we evelope a questionnaire that measures the physical an psychological impact of BPI on preoperative an postoperative patients an that also captures the amount of improvement expecte an receive from surgery. Our questionnaire was constructe from patients perspectives an inclues items particularly important to them. Thus, our questionnaire fills a gap in the comprehensive assessment of patients with BPI by simultaneously aressing the spectrum of short-term an long-term physical an psychological consequences associate with this complex an life-altering conition an proviing a vali an stanarize metho for clinicians to report outcomes. n Carol A. Mancuso, MD 1,2 Steve K. Lee, MD 1,2 Eliana B. Saltzman, BS 3 Zina Moel, BA 4 Zoe A. Laners, MSW 1 Christopher J. Dy, MD, MPH 5 Scott W. Wolfe, MD 1,2 1 Hospital for Special Surgery, New York, NY 2 Weill Cornell Meical College, New York, NY 3 Icahn School of Meicine at Mount Sinai, New York, NY 4 Rutgers Robert Woo Johnson Meical School, New Brunswick, New Jersey 5 Washington University School of Meicine, St. Louis, Missouri aress for C.A. Mancuso: mancusoc@hss.eu ORCID id for C.A. Mancuso: X References 1. Hill BE, Williams G, Bialocerkowski AE. Clinimetric evaluation of questionnaires use to assess activity after traumatic brachial plexus injury in aults: a systematic review. Arch Phys Me Rehabil Dec;92(12): Choi PD, Novak CB, Mackinnon SE, Kline DG. Quality of life an functional outcome following brachial plexus injury. J Han Surg Am Jul;22(4): Ahme-Labib M, Golan JD, Jacques L. Functional outcome of brachial plexus reconstruction after trauma. Neurosurgery Nov;61(5): ; iscussion Dy CJ, Garg R, Lee SK, Tow P, Mancuso CA, Wolfe SW. A systematic review of outcomes reporting for brachial plexus reconstruction. J Han Surg Am Feb;40(2): Epub 2014 Dec Mancuso CA, Lee SK, Dy CJ, Laners ZA, Moel Z, Wolfe SW. Compensation by the uninjure arm after brachial plexus injury. Han (N Y) Dec;11(4): Epub 2016 Feb Hill B, Pallant J, Williams G, Olver J, Ferris S, Bialocerkowski AE. Evaluation of internal construct valiity an uniimensionality of the Brachial Assessment Tool, a patient-reporte outcome measure for brachial plexus injury. Arch Phys Me Rehabil Dec;97(12): Epub 2016 Jul Kretschmer T, Ihle S, Antoniais G, Seiel JA, Heinen C, Börm W, Richter HP, König R. Patient satisfaction an isability after brachial plexus surgery. Neurosurgery Oct;65(4)(Suppl):A Bengtson KA, Spinner RJ, Bishop AT, Kaufman KR, Coleman-Woo K, Kircher MF, Shin AY. Measuring outcomes in ault brachial plexus reconstruction. Han Clin Nov;24(4):401-15, vi. 9. Novak CB, Anastakis DJ, Beaton DE, Mackinnon SE, Katz J. Valiity of the Patient Specific Functional Scale in patients following upper extremity nerve injury. Han (N Y) Jun;8(2): Franzblau L, Chung KC. Psychosocial outcomes an coping after complete avulsion traumatic brachial plexus injury. Disabil Rehabil. 2015;37(2): Epub 2014 Apr Mancuso CA, Lee SK, Dy CJ, Laners ZA, Moel Z, Wolfe SW. Expectations an limitations ue to brachial plexus injury: a qualitative stuy. Han (N Y) Dec;10(4): Epub 2015 May Huak PL, Amaio PC, Bombarier C; The Upper Extremity Collaborative Group (UECG). Development of an upper extremity outcome measure: the DASH (Disabilities of the Arm, Shouler an Han) [correcte]. Am J In Me Jun;29(6): Erratum in: Am J In Me. 1996;30(3): Hays RD, Sherbourne CD, Mazel RM. The RAND 36-Item Health Survey 1.0. Health Econ Oct;2(3): Anrews FM, Withey SB. Social inicators of well-being: Americans perceptions of life quality. New York: Springer; Berkwits M, Inui TS. Making use of qualitative research techniques. J Gen Intern Me Mar;13(3): Strauss AL, Corbin JM. Basics of qualitative research: techniques an proceures for eveloping groune theory research. 2n e. Thousan Oaks: Sage Publications; Donner A, Eliasziw M. Sample size requirements for reliability stuies. Stat Me Jun;6(4): Kramer MS, Feinstein AR. Clinical biostatistics. LIV. The biostatistics of concorance. Clin Pharmacol Ther Jan;29(1): Health Measures. Patient reporte outcome measurement information system (PROMISÒ) scoring Accesse 2017 July Doakuni C, Doi K, Hattori Y, Sakamoto S, Fujihara Y, Takagi T, Fukua M. Outcome of surgical reconstruction after traumatic total brachial plexus palsy. J Bone Joint Surg Am Aug 21;95(16): Novak CB, Anastakis DJ, Beaton DE, Mackinnon SE, Katz J. Biomeical an psychosocial factors associate with isability after peripheral nerve injury. J Bone Joint Surg Am May 18;93(10): Franzblau LE, Shauver MJ, Chung KC. Patient satisfaction an self-reporte outcomes after complete brachial plexus avulsion injury. J Han Surg Am May;39(5): e4. Epub 2014 Mar 5.