Determinants of Return to Work After Carpal Tunnel Release

SCIENTIFIC ARTICLE

Determinants of Return toWork After Carpal

Tunnel Release

James Cowan, BA, Heeren Makanji, MS, Chaitanya Mudgal, MD, Jesse Jupiter, MD, David Ring, MD, PhD

Purpose The determinants of time to return to work—a common measure of treatmenteffectiveness—are incompletely defined. Our primary hypothesis was that employmentcircumstances are the strongest determinant of earlier return to work. Our secondaryhypothesis was that return to work in patients with desk-based jobs is predicted by patientexpectations and other psychosocial factors.

Methods We enrolled 65 employed patients with limited incision open carpal tunnel releasein a prospective cohort study. Patients completed validated measures of depression, copingstrategies, pain anxiety, and job burnout. Heavy lifting was not allowed for 1 month aftersurgery. Return to modified and full work duty was recorded in days. Although notspecifically an exclusion criterion, none of the patients had a workers’ compensation claimor other source of secondary gain.

Results Patients returned to modified duty an average of 11.8 days and full duty at an averageof 18.9 days after surgery. Predictors of earlier return to modified duty in multivariateanalyses included desk-based work and both the number of days patients expected to take offand the numbers of days they wanted to take off for the entire cohort, with an additionalinfluence from catastrophic thinking in desk-based workers. Predictors of earlier return tofull duty in multivariate analyses included desk-based work and number of days patientsexpected to take off before for the entire cohort, fewer days off desired in non–desk-basedworkers, fewer days off desired and change in work role in desk-based workers, and lowerpain anxiety in part-time workers.

Conclusions The most important determinant of return to full duty work after limited incision opencarpal tunnel release is job type, but psychological factors such as patient expectations, cata-strophic thinking, and anxiety in response to pain also have a role. (J Hand Surg 2012;37A:18–27. Copyright © 2012 by the American Society for Surgery of the Hand. All rights reserved.)

Type of study/level of evidence Prognostic II.

Key words Carpal tunnel release, disability, pain anxiety, patient expectations, work status.

TIME TO RETURN to work may not be an appro-priate measure of objective treatment effective-ness unless influences less related to pathophys-

iology (including the technical aspects of surgery) are

FromtheHandandUpperExtremityService,DepartmentofOrthopaedicSurgery,MassachusettsGeneralHospital, Boston, MA.

Received for publication January 9, 2008; accepted in revised form October 14, 2011.

D.R. received support from Wright Medical, the AO Foundation, Smith and Nephew, Small BoneInnovations, Joint Active Systems, and Biomet.

No benefits in any form have been received or will be received related directly or indirectly to the
subject of this article.
18 � © ASSH � Published by Elsevier, Inc. All rights reserved.

accounted for.1–3 For example, several studies haveindicated that the receipt of workers’ compensationmay delay a patient’s return to work after carpal tunnelrelease.3–6 In addition to the potential for secondary

Corresponding author: David Ring, MD, PhD, Hand and Upper Extremity Service, Department ofOrthopaedic Surgery, Harvard Medical School, Massachusetts General Hospital, Yawkey Center, Suite2100, 55 Fruit Street, Boston, MA 02114; e-mail: [email protected].

0363-5023/12/37A01-0005$36.00/0doi:10.1016/j.jhsa.2011.10.033

mailto:[email protected]

RETURN TO WORK AFTER CARPAL TUNNEL RELEASE 19

gain issues, Katz and colleagues5 found that return towork after carpal tunnel release was influenced by men-tal health status, including anxiety and depression. Fi-nally, preoperative factors such as grip strength, pinchstrength, ability to perform activities of daily living, andpain or numbness, including nocturnal paresthesias, arepoor predictors of return to work after carpal tunnelrelease.1

Carpal tunnel syndrome (CTS) is a useful conditionfor the study of factors affecting return to work becauseits clinical presentation and treatment options are rela-tively well understood,5 and when a specific operativetechnique is used, there is a relatively consistent phys-ical insult that should result in a narrow range of im-pairment and recovery among patients. In studies com-paring techniques with a slightly different physicalinsult (namely, endoscopic vs open carpal tunnel re-lease, or limited palmar incision vs traditional opentechnique), the observed differences between treatmentgroups have been in factors that are entirely subjective(pain), in objective factors that are effort- and pain-related (grip strength), or in time to return to work.4–10

It would be valuable to know whether return to work isalso related to attitude and circumstance and to whatdegree it is volitional. Symptoms and effort-related orvolitional measures are vulnerable to patient and sur-geon beliefs and may therefore be more subject tobias.11–13 For instance, patients who perceive that theyare treated with a minimally invasive technique or ad-vanced technology might be predisposed to perceivethemselves as recovering more rapidly.14 In the settingof total hip arthroplasty, patient preconditioning af-fected patient satisfaction, time to discharge, and walk-ing ability at discharge more than operative tech-nique.15

Moreover, it is possible that time to return to work isin part a foregone conclusion because forceful use ofthe hand is usually prohibited during the first monthafter open carpal tunnel release and also because returnto work may be largely determined by a patient’s psy-chosocial status (eg, secondary gain, job satisfaction,expectations regarding return to work, depression, painanxiety, coping skills). An instructive example of this isBond and colleagues’16 clinical trial of scaphoid frac-ture treatment with percutaneous screw fixation or castimmobilization. The only differences between treat-ments were in healing times (a notoriously unreliableoutcome measure,17 perhaps more so with implants inplace) and in return to work, which was determined atleast in part by the fact that these patients were in theNavy and by regulation could not be returned to active
duty in a cast.16 Differences in return to work may not
JHS �Vol A, Ja

represent objective evidence of treatment superiority ifreturn to work is determined largely by this type ofbureaucratic restriction or if it is strongly influenced bypsychosocial factors that can be influenced by the be-liefs and attitudes of the surgeon and the patient regard-ing different treatment arms.

Our impression is that return to work after opencarpal tunnel release is primarily related to work typeand secondarily to psychosocial factors. This studytested the primary hypothesis that patients employed indesk-based jobs return to work sooner than those work-ing in non–desk-based jobs after open carpal tunnelrelease. In secondary analyses, we investigated the in-fluence of psychological factors such as expectations,depression, job burnout, catastrophic thinking, and painanxiety on time to return to work.

MATERIALS AND METHODS

Enrollment and eligibility criteria

Between March 2005 and May 2010, we invited em-ployed adult patients with an isolated diagnosis of id-iopathic carpal tunnel syndrome electing operativetreatment to enroll in an investigation approved by thehuman research committee at our institution. The inter-val was long because we restarted the study to increasepower and we competed with other studies for enroll-ment of patients, and because a small percentage ofpatients requesting surgery for carpal tunnel syndromeare actively employed in our practices. The diagnosis ofCTS was based on characteristic reports of numbnesswith confirmatory physical examination and electrodi-agnostic testing. Because several of the questionnairesused were validated only for English-language use, weexcluded patients whose primary language was notEnglish. Although it was not an exclusion criterion, it isnotable that none of the enrolled subjects had filed aclaim for workers’ compensation.

Among the 92 adult patients who enrolled, 66 com-pleted the study, 13 declined to return and complete thestudy after suture removal, 8 did not return the initialstudy questionnaire before surgery, 4 elected not tohave surgery, and 1 withdrew from the study for un-specified reasons. We excluded from the final analysisdata for subjects who did not complete the study pro-tocol. Of these subjects, 18 completed the preoperativequestionnaires but did not complete the postoperativequestionnaires. We compared preoperative data forthese noncompleters with preoperative data of subjectswho completed the protocol, with no significant differ-
ences between the groups (Table 1).
nuary

20 RETURN TO WORK AFTER CARPAL TUNNEL RELEASE

Baseline measures

After enrolling and signing informed consent, each sub-ject completed a questionnaire regarding work classifi-cation (desk-based or light work, heavy labor, or inter-mediate work), employment role (leadership, midlevelmanagement, or employee), the amount of time thepatient would like to stay out of work, the amount oftime he or she expected to be out of work, relativedesire to return to work (as soon as possible, after taking

TABLE 1. Completers Versus Noncompleters

CompleProto

Age 49.

Gender

Male 1

Female 4

Work status

Full-time 5

Part-time 1

Work type

Desk-based 3

Non–desk-based 3

Work role

Employee 3

Midlevel 1

Leadership 2

Has CTS altered your role at work?

No 3

Somewhat 2

Yes

When do you wish to return to work?

As soon as possible 5

After 1 wk

Fully recovered

After how long do you expect to return to work? 10.

After how long do you want to return to work? 7.

Preoperative questionnaire scores

Preoperative DASH 28.

CES-D 11.

SMBM 3.

PCS 23.

PASS 49.

CES-D, Center for Epidemiologic Studies–Depression; SMBM, ShiromAnxiety Symptom Scale.

a week off, or when fully recovered), whether CTS had

JHS �Vol A, Ja

altered his or her role at work (yes, no, somewhat), andsymptoms (numbness, pain, sleep quality, or weak-ness). Individuals also completed the Disabilities of theArm, Shoulder, and Hand (DASH) questionnaire and abattery of validated psychological questionnaires mea-suring depression (Center for Epidemiologic Studies–Depression), job burnout (Shirom-Melamed BurnoutMeasure), catastrophic thinking (the Pain Catastrophiz-ing Scale), and anxiety in response to pain (Pain Anx-

ntire Studyn � 66)

Completed Only PreoperativeStudy Protocol (n � 18) P Value

11.3) 48.9 (SD 12.4) .80

%) 5 (28%) 1.00

%) 13 (72%)

%) 10 (56%) .10

%) 8 (44%)

%) 8 (44%) .70

%) 10 (56%)

%) 11 (61%) .21

%) 2 (11%)

%) 4 (23%)

%) 11 (61%) .81

%) 5 (28%)

%) 2 (11%)

%) 12 (67%) .29

) 3 (16%)

%) 3 (16%)

22.8) 19.7 (SD 28.3) .17

9.1) 17.2 (SD 28.5) .19

16.3) 30.0 (SD 20.4) .29

8.6) 14.0 (SD 9.8) .27

1.1) 3.6 (SD 1.0) .28

10.0) 22.8 (SD 9.7) .92

29.9) 48.5 (SD 26.5) .94

amed Burnout Measure; PCS, Pain Catastrophizing Scale; PASS, Pain

ted Ecol (

7 (SD

7 (26

9 (74

2 (80

3 (20

4 (52

2 (48

2 (48

2 (18

2 (34

4 (52

3 (35

9 (13

0 (76

6 (9%

9 (15

7 (SD

6 (SD

5 (SD

0 (SD

3 (SD

0 (SD

2 (SD

-Mel

iety Symptom Scale).

nuary


The DASH questionnaire is a 30-item, self-adminis-tered outcome measure focusing on an individual’supper extremity health status over the precedingweek.18 Twenty-one questions ask about the degree ofdifficulty of various tasks that caused arm, shoulder, orhand problems; 5 questions address the severity of pain,activity-induced pain, tingling, weakness, and stiffness;and 4 questions ask about the effect of arm, shoulder, orhand problems on social and work activities, sleep, andself-image. Each question is answered on a 5-pointLikert scale. The DASH scores are scaled to range from0 (no disability/symptoms) to 100 (severe disability/symptoms).

The Center for Epidemiologic Studies–Depressionscale is a self-reported measure of symptoms of depres-sion.19 The questionnaire lists and asks whether duringthe past week respondents had specific feelings rarelyor never, sometimes, moderately, or most of the time.Scores range from 0 (infrequent depressive symptoms)to 60 (frequent depressive symptoms).

The Shirom-Melamed Burnout Measure question-naire contains 3 scales measuring burnout, tension, andlistlessness.20 The burnout component consists of 8questions regarding physical fatigue and emotional ex-haustion. Items are scored from 1 (almost never) to 7(almost always). The tension and listlessness subscalesare measured by 4 items each, scored similarly to theburnout questions. Five items are negatively wordedand reverse scored. Scores are calculated by dividingthe sum of all scores by the total number of responses,with a higher score representing greater burnout.

The Pain Catastrophizing Scale assesses emotionaldistress and pain-related thoughts by measuring rumi-nation, magnification, and helplessness.21 Each ques-tion is answered on a 4-point Likert scale with totalscores ranging from 0 (severe catastrophic thinking) to100 (no catastrophic thinking).

The Pain Anxiety Symptom Scale is a 40-item mea-sure designed to measure cognitive, behavioral, andphysiological fear of pain.22 This scale asks respon-dents about how often (from never to always) they havecertain thoughts or engage in certain activities when inpain. Total scores range from 0 (no pain anxiety) to 200(severe pain anxiety).

Between 2 and 4 months after surgery, subjects com-pleted the DASH questionnaire and a 10-point ordinalscale measuring satisfaction with surgery and recordedthe number of days that passed between surgery andreturn to work to both modified duty and full duty.Modified duty was considered to be work that wasaltered as a result of disability owing to surgery. Full-
time research coordinators not involved in the care of
JHS �Vol A, Ja

patients administered and collected all study question-naires.

Intervention

All patients had an open carpal tunnel release using a2- to 3-cm longitudinal palmar incision and local anes-thesia. Activities requiring heavy lifting and forcefulgripping were prohibited for 1 month after surgery.Otherwise, patients were encouraged to return to nor-mal activities including work as soon as possible.

Sample size and statistical analysis

A power analysis determined that a cohort of 34 pa-tients would provide 80% power (� � 0.05, � � 0.2) todetect a significant difference (P � .05) of 7 days in thetime to return to work between desk-based workers andnon–desk-based workers using Student’s t-test and as-suming a standard deviation of 7 days in return to work(effect size � 1.0).

Bivariate comparisons of time to return to work usedindependent t-tests for dichotomous explanatory vari-ables (gender, work status, and work classification),Bonferroni 1-way analysis of variance for nominal vari-ables with more than 2 possible answers (employmentrole, desire to return to work, and whether CTS hasaltered role at work), and Pearson correlation to analyzecontinuous variables (days wanted and expected beforereturning to work, questionnaire scores, and days untilreturning to partial and full duty).

We performed multivariable analyses using back-ward, stepwise linear regression analyses starting withthe variables with P less than .08 in bivariate analysis.

RESULTSTables 1 and 2 list preoperative and postoperative de-scriptive statistics. In the cohort, there were 34 desk-based subjects, 32 non–desk-based subjects, 52 full-time subjects, and 13 part-time subjects. One subjectdid not specify being full-time or part-time.

Predictors of earlier return to modified work duty

In bivariate analysis, earlier return to modified workduty was associated with desk-based work, alteration ofwork role, fewer days patients expected to take off,fewer days patients wanted to take off, lower painanxiety, and less catastrophic thinking (Tables 3, 4).The best multivariate model included desk-based workand both the number of days patients expected to takeoff as well as the number of days patients wanted totake off, and accounted for 68% of the variability inreturn to modified work duty.

Among the subset of patients employed at desk-
based work, more rapid return to modified work duty
nuary


was associated with decreasing age, fewer days patientsexpected to take off, fewer days patients wanted to takeoff, work role, relative desire to return to work, cata-strophic thinking, and burnout in bivariate analysis. Thebest multivariable model included the number of dayspatients expected to take off and catastrophic thinking,and accounted for 61% of the variability in earlierreturn to modified work among desk workers.

In bivariate analysis of the subset of patients em-ployed in non-desk work, earlier return to modifiedwork duty was associated with fewer days patientsexpected to take off and fewer days patients wanted totake off. The best multivariate model included dayspatients both expected and wanted to take off, andaccounted for 70% of the variability in earlier return tomodified duty in non-desk workers.

For the subset of patients employed in part-timework, bivariate analysis demonstrated that earlier returnto modified duty was best predicted by relative desire toreturn to work, fewer days patients expected to take off,and fewer days patients wanted to take off. Only ex-pected time off remained in the multivariate model,which accounted for 57% of variability in days to return

TABLE 2. Postoperative Descriptive Statistics

AllSubjects

DesSu(n

After how long did you return tomodified work duty?

Days 11.8

Range 0–90 0–

SD 16.2 1

After how long did you return tonormal work duty?

Days 18.9

Range 0–90 0–

SD 21.9 1

Postoperative DASH Questionnaire

Score 9.9

Range 0–42.5 0–

SD 9.0

Postoperative Satisfaction

Score 9.0

Range 2–10 2–

SD 1.7

to part-time work duty.

JHS �Vol A, Ja

In bivariate analysis, earlier return to modified dutyin full-time workers was associated with fewer dayspatients expected to take off and fewer days patientswanted to take off. The multivariate model includeddays patients wanted to take off alone, and accountedfor 42% of the variability in days until return to mod-ified work duty.

Predictors of earlier return to full work duty

Earlier return to full work duty was associated withdecreasing age, desk-based work, altered work role,fewer days patients expected to take off, fewer dayspatients wanted to take off, less catastrophic thinking,and less anxiety in response to pain (Tables 5, 6). Thebest multivariate model included desk-based work,fewer days patients expected to take off, altered workrole, and lower pain anxiety, and accounted for 43% ofthe variability in days until return to full work duty.

Earlier return to full-duty desk work was associatedwith lower pain anxiety and lower postoperative DASHscores, and both were included in the best multivariatemodel, accounting for 29% of the variability. Earlierreturn to full-duty non-desk work was associated withexpected time off and wanted time off, and it accounted

eds)

Non–Desk-BasedSubjects (n � 32)

Full-TimeSubjects(n � 52)

Part-TimeSubjects(n � 13)

17.7 9.5 15.3

0–90 0–49 0–56

19.5 11.4 17.8

29.5 14.0 33.7

3–90 0–60 0–90

22.5 13.8 32.7

9.9 9.4 11.7

0–36.6 0–36.6 0–42.5

8.9 8.4 11.6

9.0 9.0 9.1

3–10 2–10 5–10

1.8 1.8 1.4

k-Basbject� 34

6.7

49

0.5

9.5

90

6.7

9.9

42.5

9.3

9.0

10

1.7

for 40% of the variability. Earlier return to full-duty

nuary


part-time work was associated with lower preoperativeDASH score, and lower pain anxiety accounted for47% of the variability.

Earlier return to unrestricted full-time work was as-sociated with decreasing age, desk-based work, alteredwork role, fewer days patients expected to take off, andfewer days patients wanted to take off. The best multi-variable model included desk-based work, altered work

TABLE 3. Bivariate Predictors (P Values) of Return

AllSubjects

Age .260

Gender

Male .180

Female

Work status

Full-time .160

Part-time

Work type

Desk-based .009

Non–desk-based

Work role

Employee vs leadership .620

Employee vs midlevel

Midlevel vs leadership


Yes vs no .026

Yes vs somewhat .320

Somewhat vs no .380


As soon as possible vs fully recovered .140

As soon as possible vs after 1 wk

After 1 wk vs fully recovered

After how long do you expect to return to work? .001

After how long do you want to return to work? �.001


Preoperative DASH .290

CES-D .480

SMBM .600

PCS .005

PASS .040

Postoperative DASH .660

Likert satisfaction .830

N/A, data not available; CES-D, Center for Epidemiologic Studies–Dephizing Scale; PASS, Pain Anxiety Symptom Scale.

role, and fewer days patients wanted to take off, and

JHS �Vol A, Ja

accounted for 47% of variability of days until returningto full work duty.

DISCUSSIONCertain aspects of our study are distinctive comparedwith other studies on return to work and carpal tunnelrelease. In our study, patients returned to normal workearlier (19 d) than in prior studies by Palmer and col-

Modified Work

esk-BasedSubjects(n � 34)

Non–Desk-BasedSubjects(n � 32)

Full-TimeSubjects(n � 52)

Part-TimeSubjects(n � 13)

.030 .130 .120 .320

.210 .270 .410 .400

.630 .180 N/A N/A

N/A N/A

N/A N/A .070 .130

N/A N/A

.030 .700 .120 .220

.240

.600

.880 .400 .430 .280

.040 .550 .110 .010

.990 N/A

.630 N/A

.001 �.001 �.001 .003

�.001 .019 �.001 .010

.940 .810 .220 .580

.270 .490 .460 .920

.007 .200 .120 .350

.012 .290 .070 .130

.840 .170 .540 .250

.670 .690 .190 .310

.320 .290 .640 .240

ion; SMBM, Shirom-Melamed Burnout Measure; PCS, Pain Catastro-

to

D

press

leagues9 (44 d), Trumble and colleagues10 (38 d), Agee

nuary

ABLE

4.M

ulti

vari

able

Pre

dict

ors

ofR

etur

nto

Mod

ified

Wor

ka

All

Subj

ects

Des

k-B

ased

Subj

ects

(n�

34)

Non

–Des

k-B

ased

Subj

ects

(n�

32)

Full-

Tim

eSu

bjec

ts(n

�52

)Pa

rt-T

ime

Subj

ects

(n�

13)

20.

680.

610.

700.

420.

57

43.8

26.0

35.0

33.9

15.8

�.0

01�

.001

�.0

01�

.001

.003

Exp

ecte

dtim

eun

tilre

turn

tow

ork

R2

�0.

36W

ante

dtim

eun

tilre

turn

tow

ork

R2

�0.

45E

xpec

ted

time

until

retu

rnto

wor

kR

2�

0.53

Wan

ted

time

until

retu

rnto

wor

kR

2�

0.42

Exp

ecte

dtim

eun

tilre

turn

tow

ork

R2

�0.

57

Wan

ted

time

until

retu

rnto

wor

kR

2�

0.06

Wor

kty

pe(d

esk

base

dvs

non–

desk

base

d)

R2

�0.

02PC

SR

2�

0.15

Wan

ted

time

until

retu

rnto

wor

kR

2�

0.05

CS,

Pain

Cat

astr

ophi

zing

Scal

e.a R

2re

port

edfo

rin

divi

dual

fact

ors

isth

epa

rtco

rrel

atio

nco

effic

ient

squa

red.


and colleagues4 (47 d), and Brown and colleagues (28d).7 This may reflect the practice setting and style of thetreating surgeons. For example, none of the patientswere receiving workers’ compensation, the diagnosiswas always verified by electrodiagnostic testing, thegoal of surgery was relief of numbness rather than pain,splints were never used after surgery, and nonforcefuluse of the hand was encouraged immediately after sur-gery. In our opinion, our data can be generalized to thispractice style.

There were several strengths of the design and exe-cution of our study. We employed a prospective designand used a consistent operative technique. Data wereself-reported and collected by an independent observernot involved with patient care. The use of validatedpsychological measures was more specific than previ-ous studies that investigated the influence of mentalhealth on return to work.5

Our study had some limitations. Overall enrollmentwas relatively low considering the number of patientstreated for CTS over the time period. The primaryreason for this was that most patients electing operativetreatment for CTS were not employed (many of themwere older and retired or homemakers). An additionallimitation of the study design is that multiple compar-isons may lead to spurious findings. The answer to ourprimary study question (that desk-based workers returnto work sooner) is therefore the most reliable, with theothers representing hypothesis-generating findings inwhich we have less confidence. Similarly, the signifi-cant factors for earlier return to work as determined bymultivariate analysis did not account for all of thevariability in return to work.

Another limitation was that the actual standard de-viation of return to full-duty work was 3 times ourestimate in the power analysis (22 d actual vs 7 destimated). To detect a difference of 7 days at 80%power and alpha less than or equal to 0.05 with astandard deviation of 22, we would need a sample sizeof 286 patients. On the other hand, a post hoc poweranalysis for the primary study question found that 80%power to detect the actual difference in return to workof 19 days with the actual the standard deviation of 22days is achieved with a sample of 46 patients. In otherwords, despite the error in our estimated standard de-viation, our sample of 65 patients provided adequatepower to address the primary study question.

A final limitation is that approximately 28% of theoriginal enrollment pool did not complete the study,although there were no differences between completersand noncompleters.

Our data supported the primary hypothesis of ourT R F P P

JHS �Vol A, January


study—that patients employed at desk-based workwould return to full-duty work more rapidly than pa-tients employed at non–desk-based work. In particular,return to full work duty was determined largely by workclassification (desk-based vs non–desk-based work).This finding is consistent with those of Katz and col-leagues,5 who found that working as a laborer or ma-

TABLE 5. Bivariate Predictors (P Values) of Return

AllSubjects

Age .050

Gender

Male .500

Female

Work status

Full-time .060

Part-time

Work type

Desk-based �.001

Non–desk-based

Work role

Employee vs midlevel .510

Employee vs leadership

Midlevel vs leadership


Yes vs no .110

Yes vs somewhat .990

Somewhat vs no .031


As soon as possible vs fully recovered .560

As soon as possible vs after 1 wk

After 1 wk vs fully recovered

After how long do you except to return to work? �.001

After how long do you want to return to work? .002


Preoperative DASH .690

CES-D .380

SMBM .290

PCS .028

PASS .005

Postoperative DASH .110

Likert satisfaction .500

CES-D, Center for Epidemiologic Studies–Depression; SMBM, ShiromAnxiety Symptom Scale.

chine operator correlated with work absence after carpal

JHS �Vol A, Ja

tunnel release. It is likely a result of the restrictionsplaced on forceful use of the hand for the first postop-erative month.

The amount of time subjects expected to take off andwanted to take off before returning to work was themost important determinant of earlier return to bothmodified work duty and full work duty for non–desk-

Full Work

esk-BasedSubjects(n�34)

Non-Desk-BasedSubjects(n�32)

Full-TimeSubjects(n�52)

Part-TimeSubjects(n�13)

.110 .530 .050 .170

.320 .590 .640 .400

.440 .230 N/A N/A

N/A N/A

N/A N/A �.001 .600

N/A N/A

.980 .240 .720 .720

.820 .430 .560 .860

.990

.007

.410 .850 .430 .110

.230 �.010 .110 .110

.350 .030 .010 .180

.760 .330 .350 .060

.310 .920 .230 .360

.230 .560 .250 .100

.120 .700 .670 .130

.003 .570 .350 .028

.004 .850 .920 .180

.140 .970 .740 .340

amed Burnout Measure; PCS, Pain Catastrophizing Scale; PASS, Pain

to

D

-Mel

based employees. This indicates that patient expecta-

nuary

ABLE

6.M

ulti

vari

ate

Pre

dict

ors

ofR

etur

nto

Ful

lW

orka

All

Subj

ects

Des

k-B

ased

Subj

ects

(n�

34)

Non

-Des

k-B

ased

Subj

ects

(n�

32)

Full-

Tim

eSu

bjec

ts(n

�52

)Pa

rt-T

ime

Subj

ects

(n�

13)

0.43

0.29

0.40

0.47

0.47

12.6

7.7

10.8

15.7

5.3

�.0

01.0

02�

.001

�.0

01.0

30

dels

Exp

ecte

dtim

eun

tilre

turn

tow

ork

R2

�0.

18PA

SSR

2�

0.09

Exp

ecte

dtim

eun

tilre

turn

tow

ork

R2

�0.

29W

ork

type

(des

kba

sed

vsno

n–de

skba

sed)

R2

�0.

23PA

SSR

2�

0.23

Wor

kty

pe(d

esk

base

dvs

non–

desk

base

d)

R2

�0.

06

Con

ditio

nso

mew

hat

alte

red

role

R2

�0.

03Po

stop

erat

ive

DA

SHR

2�

0.09

Wan

ted

time

until

retu

rnto

wor

kR

2�

0.06

Wan

ted

time

until

retu

rnto

wor

kR

2�

0.10

Preo

pera

tive

DA

SHR

2�

0.21

PASS

R2

�0.

03C

ondi

tion

alte

red

role

som

ewha

tR

2�

0.06

ASS

,Pa

inA

nxie

tySy

mpt

omSc

ale.

a R2

repo

rted

for

indi

vidu

alfa

ctor

sis

the

part

corr

elat

ion

coef

ficie

ntsq

uare

d.


tions, beliefs, and desires are important determinants ofreturn to work. This is important because the physi-cian’s preoperative counseling about postoperative re-covery, marketing of the procedure, and the media mayinfluence the patient’s perspective.15

Desk workers seem more susceptible to the influ-ences of coping strategies and psychological distressthan laborers. Catastrophic thinking was associatedwith delayed return to modified duty and anxiety inresponse to pain was associated with delayed return tofull duty work among desk workers. Katz and col-leagues5,23 also noted the importance of psychologicaland psychosocial factors in determining return to workin studies, and found that anxiety, depression, self-efficacy, social support, and psychological work de-mands were predictors of work absence 6 months aftersurgery.

According to our data, return to work after carpaltunnel release through a small open incision and withlocal anesthesia is determined by job type (desk workvs non–desk work), but also by patient beliefs andexpectations and psychological factors, particularly fordesk-based work. Therefore, when return to work isused to evaluate the efficacy or effectiveness of anintervention, the study design should attempt to controlfor these factors. We propose that preoperative educa-tion that stresses that return to desk work is safe andappropriate within a few days of surgery and that in-stilling a sense of confidence and well-being in thecontext of postoperative discomfort might result in ear-lier return to work.

REFERENCES1. Atroshi I, Johnsson R, Ornstein E. Patient satisfaction and return to

work after endoscopic carpal tunnel surgery. J Hand Surg 1998;23A:58–65.

2. Baldwin ML, Johnson WG, Butler RJ. The error of using returns-to-work to measure the outcomes of health care. Am J Ind Med1996;29:632–641.

3. Carmona L, Faucett J, Blanc PD, Yelin E. Predictors of rate of returnto work after surgery for carpal tunnel syndrome. Arthritis Care Res1998;11:298–305.

4. Agee JM, McCarroll HR Jr, Tortosa RD, Berry DA, Szabo RM,Peimer CA. Endoscopic release of the carpal tunnel: a randomizedprospective multicenter study. J Hand Surg 1992;17A:987–995.

5. Katz JN, Keller RB, Fossel AH, Punnett L, Bessette L, Simmons BP,et al. Predictors of return to work following carpal tunnel release.Am J Ind Med 1997;31:85–91.

6. Nathan PA, Meadows KD, Keniston RC. Rehabilitation of carpaltunnel surgery patients using a short surgical incision and an earlyprogram of physical therapy. J Hand Surg 1993;18A:1044–1050.

7. Brown RA, Gelberman RH, Seiler JG III, Abrahamsson SO,Weiland AJ, Urbaniak JR, et al. Carpal tunnel release. A prospective,randomized assessment of open and endoscopic methods. J BoneJoint Surg 1993;75A:1265–1275.

8. Jugovac I, Burgic N, Micovic V, Radolovic-Prenc L, Uravic M,
Golubovic V, et al. Carpal tunnel release by limited palmar incision
T

R2

F P Mo P

JHS �Vol A, January


vs traditional open technique: randomized controlled trial. CroatMed J 2002;43:33–36.

9. Palmer DH, Paulson JC, Lane-Larsen CL, Peulen VK, Olson JD.Endoscopic carpal tunnel release: a comparison of two techniqueswith open release. Arthroscopy 1993;9:498–508.

10. Trumble TE, Diao E, Abrams RA, Gilbert-Anderson MM. Single-portalendoscopic carpal tunnel release compared with open release: a pro-spective, randomized trial. J Bone Joint Surg 2002;84A:1107–1115.

11. Chaffin DB, Lee M, Freivalds A. Muscle strength assessment fromEMG analysis. Med Sci Sports Exerc 1980;12:205–211.

12. Gutierrez Z, Shechtman O. Effectiveness of the five-handle positiongrip strength test in detecting sincerity of effort in men and women.Am J Phys Med Rehabil 2003;82:847–855.

13. Smith GA, Nelson RC, Sadoff SJ, Sadoff AM. Assessing sincerity ofeffort in maximal grip strength tests. Am J Phys Med Rehabil1989;68:73–80.

14. Bozic KJ, Smith AR, Hariri S, Adeoye S, Gourville J, Maloney WJ,et al. The 2007 ABJS Marshall Urist Award: the impact of direct-to-consumer advertising in orthopaedics. Clin Orthop Relat Res2007;458:202–219.

15. Pour AE, Parvizi J, Sharkey PF, Hozack WJ, Rothman RH. Mini-mally invasive hip arthroplasty: what role does patient precondition-
ing play? J Bone Joint Surg 2007;89A:1920–1927.
JHS �Vol A, Ja

16. Bond CD, Shin AY, McBride MT, Dao KD. Percutaneous screwfixation or cast immobilization for nondisplaced scaphoid fractures.J Bone Joint Surg 2001;83A:483–488.

17. Dias JJ. Definition of union after acute fracture and surgery forfracture nonunion of the scaphoid. J Hand Surg 2001;26B:321–325.

18. Hudak PL, Amadio PC, Bombardier C. Development of an upperextremity outcome measure: the DASH (disabilities of the arm,shoulder and hand) [corrected]. The Upper Extremity CollaborativeGroup (UECG). Am J Ind Med 1996;29:602–608.

19. Radloff LS. The CES-D scale: a self-report depression scale forresearch in the general population. Appl Psychol Meas 1977;1:385–401.

20. Melamed S, Kushnir T, Shirom A. Burnout and risk factors forcardiovascular diseases. Behav Med 1992;18:53–60.

21. Sullivan MJL, Bishop SR, Pivik J. The Pain Catastrophizing Scale:development and validation. Psychol Assess 1995;7:524–532.

22. McCracken LM, Zayfert C, Gross RT. The Pain Anxiety SymptomsScale: development and validation of a scale to measure fear of pain.Pain 1992;50:67–73.

23. Katz JN, Amick BC, 3rd, Keller R, Fossel AH, Ossman J, Soucie V,et al. Determinants of work absence following surgery for carpal
tunnel syndrome. Am J Ind Med 2005;47:120–130.
nuary

Documents

Determinants of Return to Work After Carpal Tunnel Release