Abstract
Background: Patient-reported outcome measures vary more than expected based on underlying pathology, in part due to the substantial influence of mood and coping strategies. Methods: This study addressed the primary null hypothesis that the Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function score 1 month (range, 3-8 weeks) after hand surgery is not associated with PROMIS Depression and PROMIS Pain Interference scores prior to surgery, accounting for other factors. Using an institution-wide database of routinely collected patient-reported outcomes, we identified adult patients who underwent wrist ganglion excision, trapeziometacarpal arthroplasty, hand ganglion excision, trigger digit, De Quervain, and carpal tunnel release. Measures collected included the PROMIS Physical Function Computerized Adaptive Test (CAT), PROMIS Pain Interference CAT, and PROMIS Depression CAT. We sought factors associated with postsurgical PROMIS Physical Function scores and change between preoperative and postoperative score using multivariable linear regression, accounting for age, sex, surgery type, provider, and time from surgery to postsurgical measurement. Results: Higher postoperative PROMIS Physical Function score was independently associated with lower PROMIS Pain Interference scores, lower PROMIS Depression scores, younger age, and treatment by provider team 3. Greater change in PROMIS Physical Function score was independently associated with greater PROMIS Pain Interference scores, greater time from surgery, and treatment by provider team 3. Conclusions: Mood and effective coping strategies affect the level of symptoms and limitations during recovery from hand surgery and represent important treatment opportunities for enhancing recovery.
Keywords: coping strategies, depression, hand surgery, physical limitations, recovery
Introduction
Patient-reported outcome measures vary more than expected based on underlying pathology, and evidence suggests a substantial part of this variation is due to psychosocial factors.12 Symptom intensity and magnitude of limitations correlate with symptoms of depression and coping strategies.6 We are interested in how psychological factors affect the recovery process after common hand surgeries, measured with Patient-Reported Outcomes Measurement Information System (PROMIS). PROMIS was developed by the National Institutes of Health in an effort to create a reliable and reproducible outcome instrument, which can be administered as a Computerized Adaptive Test (CAT). This probability-based computer algorithm allows for a minimal number of adaptive questions, drawn from a pool of 124 potential questions, while still achieving high measurement precision.4 CAT PROMIS Physical Function (PF) has been demonstrated to be valid and reliable with high correlation to classic orthopedic patient-reported outcomes without fixed questionnaire limits.2 As we are moving toward a value-based health care system, patient-reported outcomes are becoming an increasingly important metric in assessing quality of care3 and patient outcomes may eventually determine financial reimbursement for health care providers.
A large database of routinely collected PROMIS questionnaires from patients visiting a hand surgeon allowed us to study the effect of psychological factors on functional outcomes after upper extremity surgery. The high volume of patients provides sufficient statistical power to detect even minor differences in patient-reported outcomes. This study measured factors associated with magnitude of limitations (measured with PROMIS PF CAT scores) during recovery from common hand surgeries, accounting for preoperative symptoms of depression (PROMIS Depression CAT) and the degree to which pain interferes with daily activities (PROMIS Pain Interference CAT). PROMIS Pain Interference was intended as a disability measure but is similar to and highly correlated with measures that assess coping strategies in response to pain (eg, the Pain Catastrophizing Scale and the Pain Self-Efficacy Questionnaire).8 We tested the primary null hypothesis that the PROMIS PF score 1 month (range, 3-8 weeks) after hand surgery is not associated with PROMIS Depression and PROMIS Pain Interference scores prior to surgery, accounting for other factors. We also tested the secondary null hypothesis that the change in PROMIS PF score in patients 1 month (range, 3-8 weeks) after a common hand surgery is not associated with initial PROMIS Depression score and PROMIS Pain Interference, accounting for other factors.
Materials and Methods
We used a database of patient-reported outcomes collected from all patients visiting an academic outpatient orthopedic clinic between 2015 and 2017. Adult (18 years or older) patients were eligible for this study if they had one of the following surgical treatments, identified by Current Procedural Terminology (CPT) codes: De Quervain release (25000), wrist ganglion excision (25111), trapeziometacarpal arthroplasty (25447), trigger finger release (26055), hand ganglion excision (26160), carpal tunnel release (open: 64721, endoscopic: 29848). Institutional review board approval was obtained for review of a hand division repository. PROMIS scores and CPT codes were linked, and the data set was deidentified prior to analysis.
Measures collected included the PROMIS PF CAT,14 PROMIS Pain Interference CAT,1 and PROMIS Depression CAT.15 These patient-reported outcome measures respectively measure physical limitations, effective coping strategies in response to nociception, and symptoms of depression. In total, 2668 patients (13 003 measurements) met the inclusion criteria. After excluding incomplete questionnaires (72% of patients who had incomplete questionnaires declined the survey, 11% quit the survey, 8.6% abandoned the survey, and 8.3% skipped one or more questions) and identifying patients with a preoperative measurement and a measurement between 21 days (3 weeks) and 56 days (8 weeks) after surgery (time 2), we were left with 831 patients (31%; Table 1) with complete data. At the time of the second measurement, most patients were still in the process of recovering from surgery, as the postoperative average PROMIS PF score was similar to the preoperative average score.
Table 1.
Patient Demographics.
| Variables | Patients without follow-up | Patients with follow-up | P value |
|---|---|---|---|
| N | 1164 | 831 | |
| Age | 57 ± 14 | 57 ± 13 | .57 |
| Male | 424 (36%) | 262 (32%) | .025 |
| Surgery | <.001 | ||
| De Quervain release | 39 (3.4%) | 45 (5.4%) | |
| Wrist ganglion excision | 75 (6.4%) | 47 (5.7%) | |
| Carpometacarpal arthroplasty | 64 (5.5%) | 159 (19%) | |
| Trigger finger release | 269 (23%) | 155 (19%) | |
| Hand ganglion excision | 141 (12%) | 36 (4.3%) | |
| Carpal tunnel release (endoscopic) | 115 (9.9%) | 112 (13%) | |
| Carpal tunnel release (open) | 461 (40%) | 277 (33%) | |
| Time to surgery (days) | 61 ± 75 | 50 ± 58 | <.001 |
| Provider | <.001 | ||
| 1 | 62 (5.3%) | 32 (3.9%) | |
| 2 | 268 (23%) | 152 (18%) | |
| 3 | 241 | 226 (27%) | |
| 4 | 110 | 128 (15%) | |
| 5 | 343 | 179 (22%) | |
| 6 | 76 | 53 (6.4%) | |
| 7 | 64 | 61 (7.3%) | |
| PROMIS Physical Function time 1 (t score) | 47 ± 9 | 45 ± 9 | <.001 |
| PROMIS Physical Function time 2 (t score) | 45 ± 9 | ||
| PROMIS Pain Interference (t score) | 57 ± 8 | 58 ± 8 | <.001 |
| PROMIS Depression (t-score) | 47 ± 10 | 48 ± 10 | <.001 |
Note. Bold indicates statistical significance, P < .05. Continuous variables as mean (±standard deviation); discrete variables as number (percentage). PROMIS = Patient-Reported Outcomes Measurement Information System.
Statistical Analysis
We conducted multivariable linear regression analysis of factors associated with PROMIS PF scores, accounting for age, sex, surgery type, provider (seen at first visit), and time from surgery to postsurgical measurement. We accounted for the surgeon-provider by coupling the surgeons and their advanced practice provider (physician assistant or nurse practitioner). The exception was one physician assistant who works with all the surgeons and who was considered as a specific provider. Only 1.1% of patients in our database switched providers. For the secondary null hypothesis, we subtracted the PROMIS PF score at time 1 from the score at time 2—a positive difference between these two numbers implies an increase in physical function score. The outcome in PROMIS PF score is reported in T-score: 50 is average for a large population, with each 10 points above or below 50 representing a standard deviation variation.12 Regression coefficients, standard errors, 95% confidence intervals, and P values are reported. All P values < .05 were deemed statistically significant.
A post hoc power analysis demonstrated that 831 patients provided > 99% power (α = 0.05) with a measured effect size of 0.15 and 8 predictor variables to determine factors associated with postoperative PROMIS PF scores.
Results
Three to 8 weeks after surgery, PROMIS PF scores were similar to the presurgical scores (Table 1). After controlling for confounding variables in multivariable analysis, the following variables were all independently associated with a higher postoperative PROMIS PF score: lower preoperative PROMIS Pain Interference score (regression coefficient [RC]: −0.44; 95% confidence interval [CI]: −0.52 to −0.36; partial R2: 0.13; P < .001), lower preoperative PROMIS Depression score (RC: −0.18; CI: −0.23 to −0.12; partial R2: 0.041; P < .001), younger age (RC: −0.10; CI: −0.14 to −0.063; partial R2: 0.030; P < .001), and treatment by provider team 3 (RC: 3.1; CI: 0.35 to 5.8; partial R2: 0.006; P = .027) (Table 2), but not type of surgery.
Table 2.
Multivariable Linear Regression Analysis of Factors Associated With PROMIS Physical Function Score.
| Variables | Regression coefficient |
Standard error | P value | Partial R² | Adjusted R2 |
|---|---|---|---|---|---|
| (95% confidence interval) | |||||
| PROMIS Pain Interference | −0.44 (−0.52 to −0.36) | 0.039 | <.001 | 0.13 | 0.32 |
| PROMIS Depression | −0.18 (−0.23 to −0.12) | 0.030 | <.001 | 0.041 | |
| Age | −0.10 (−0.14 to −0.063) | 0.021 | <.001 | 0.030 | |
| Sex | |||||
| Female | reference value | ||||
| Male | 0.64 (−0.46 to 1.7) | 0.56 | .25 | 0.002 | |
| Surgery | |||||
| De Quervain release | reference value | ||||
| Wrist ganglion excision | 2.2 (−0.77 to 5.2) | 1.5 | .15 | 0.003 | |
| Carpometacarpal arthroplasty | −1.0 (−3.5 to 1.5) | 1.3 | .42 | <0.001 | |
| Trigger finger release | 0.88 (−1.6 to 3.3) | 1.3 | .48 | <0.001 | |
| Hand ganglion excision | 2.0 (−1.2 to 5.3) | 1.7 | .22 | 0.002 | |
| Carpal tunnel release (endoscopic) | −0.78 (−3.6 to 2.0) | 1.4 | .59 | <0.001 | |
| Carpal tunnel release (open) | −1.0 (−3.4 to 1.3) | 1.2 | .40 | <0.001 | |
| Time from surgery (days) | 0.050 (−0.0033 to 0.10) | 0.027 | .066 | 0.004 | |
| Provider | |||||
| 1 | reference value | ||||
| 2 | 2.5 (−0.30 to 5.3) | 1.4 | .081 | 0.004 | |
| 3 | 3.1 (0.35 to 5.8) | 1.4 | .027 | 0.006 | |
| 4 | 1.3 (−1.6 to 4.1) | 1.5 | .38 | <0.001 | |
| 5 | 1.2 (−1.6 to 3.9) | 1.4 | .40 | <0.001 | |
| 6 | 2.4 (−0.84 to 5.6) | 1.6 | .15 | 0.003 | |
| 7 | 1.6 (−1.5 to 4.7) | 1.6 | .32 | 0.001 |
Note. Bold indicates statistical significance, P < .05. PROMIS = Patient-Reported Outcomes Measurement Information System.
In multivariable analysis, a greater increase in PROMIS PF score was independently associated with greater preoperative PROMIS Pain Interference (RC: 0.23; CI: 0.15 to 0.30; partial R2: 0.043; P < .001), greater time from surgery (RC: 0.086; CI: 0.035 to 0.14; partial R2: 0.013; P = .001), and treatment by provider team 3 (RC: 3.2; CI: 0.57 to 5.9; partial R2: 0.007; P = .017) (Table 3). About half of the patients (49%) had improved PROMIS PF scores, 6% had no change, and 45% had a decrease in score. In a secondary analysis, we found that patients who improved within 3 to 8 weeks of surgery had significantly lower presurgical PROMIS PF scores (42 ± 8.1 vs 48 ± 8.2, P < .001) and significantly higher presurgical PROMIS Pain Interference scores (60 ± 7.6 vs 57 ± 7.6, P < .001).
Table 3.
Multivariable Linear Regression Analysis of Factors Associated With Change in PROMIS Physical Function Score.
| Variables | Regression coefficient |
Standard error | P value | Partial R² | Adjusted R2 |
|---|---|---|---|---|---|
| (95% confidence interval) | |||||
| PROMIS Pain Interference | 0.23 (0.15 to 0.30) | 0.038 | <.001 | 0.043 | 0.090 |
| PROMIS Depression | −0.028 (−0.084 to 0.029) | 0.029 | .34 | 0.001 | |
| Age | −0.021 (−0.060 to 0.018) | 0.020 | .29 | 0.001 | |
| Sex | |||||
| Female | reference value | ||||
| Male | −0.52 (−1.6 to 0.54) | 0.54 | .33 | 0.001 | |
| Surgery | |||||
| De Quervain release | reference value | ||||
| Wrist ganglion excision | −0.011 (−2.9 to 2.9) | 1.5 | .99 | <0.001 | |
| Carpometacarpal arthroplasty | −2.1 (−4.5 to 0.23) | 1.2 | .077 | 0.0038 | |
| Trigger finger release | 0.73 (−1.6 to 3.1) | 1.2 | .55 | <0.001 | |
| Hand ganglion excision | 0.17 (−3.0 to 3.3) | 1.6 | .91 | <0.001 | |
| Carpal tunnel release (endoscopic) | −0.55 (−3.3 to 2.2) | 1.4 | .69 | <0.001 | |
| Carpal tunnel release (open) | 0.22 (−2.0 to 2.5) | 1.1 | .85 | <0.001 | |
| Time from surgery (days) | 0.086 (0.035 to 0.14) | 0.026 | .001 | 0.013 | |
| Provider | |||||
| 1 | reference value | ||||
| 2 | 2.4 (−0.30 to 5.1) | 1.4 | .082 | 0.004 | |
| 3 | 3.2 (0.57 to 5.9) | 1.3 | .017 | 0.007 | |
| 4 | 1.8 (−0.91 to 4.6) | 1.4 | .19 | 0.002 | |
| 5 | 0.14 (−2.5 to 2.8) | 1.3 | .92 | <0.001 | |
| 6 | 1.1 (−2.0 to 4.2) | 1.6 | .47 | <0.001 | |
| 7 | 1.7 (−1.3 to 4.7) | 1.5 | .26 | 0.002 |
Note. Bold indicates statistical significance, P < .05. PROMIS = Patient-Reported Outcomes Measurement Information System.
Discussion
There is greater variation in patient-reported outcome measures than accounted for based on differences in underlying pathology. Psychological factors such as symptoms of depression and pain interference account for a notable percentage of this variation6,13 and represent important opportunities for enhancing recovery. In this study, we tested what factors are associated with magnitude of limitations in patients recovering from common hand surgeries, accounting for preoperative symptoms of depression and pain interference.
This study should be read with the following limitations in mind. First, there was no routine measurement of race, education level, or household income. Second, the PROMIS PF CAT is not specific to the upper extremity, although there is evidence that PROMIS Upper Extremity (version 1) and PF CAT compare favorably with each other as well as with the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire.2,11 Third, there were statistically significant differences between the group with and without a follow-up examination within 3 to 8 weeks after surgery (Table 1). For example, there were fewer men, PROMIS PF scores were lower, and PROMIS Depression and Pain Interference scores were higher. There were many patients who did not complete the questionnaires in spite of our effort to have all patients complete them. Variations in socioeconomic status or literacy may affect their ability to use a tablet to complete the questionnaires. In addition, PROMIS is not translated into every language and there are patients who could not be assessed. Consequently, our findings might not apply to the average person undergoing hand surgery.
We found that older age, greater PROMIS Pain Interference, and PROMIS Depression scores were all independently associated with greater physical limitations between 3 and 8 weeks after surgery. It is not surprising that older patients have lower physical function after surgery, as physical function tends to decrease with age. The notable influence of resiliency (adaptivity to pain) and symptoms of depression on limitations is consistent with prior research among patients with upper extremity injury and illness.5,13 Treatment by one of the surgeon/physician assistant teams (provider team 3) was associated with better physical function after surgery, suggesting that patient selection, technical aspects of care, or communication/care strategies can influence limitations after surgery.
We found that a larger increase in PROMIS PF score was independently associated with a higher preoperative PROMIS Pain Interference score, greater time from surgery, and treatment by provider team 3. It is notable that a substantial percentage of patients had a decrease in physical function score, which is consistent with an early stage of recovery. Therefore, the average function score 3 to 8 weeks after surgical treatment is comparable to the average preoperative score. At this time, many patients were doing well enough that they wanted to return to the clinic as needed, so the ability for PROMIS PF CAT to detect a meaningful clinical change for an individual patient for a specific condition is unknown, but a change was present. Interestingly, the largest influence on change in physical function score was higher pain interference scores, which suggests that patients who are the most affected by pain in their daily lives benefited the most from surgery on average. This might reflect that patients with higher initial scores, due in part to less effective coping strategies, have the potential for greater improvement than more adaptive patients (by virtue of their scores being much lower to start compared with patients who are already near the top of their potential function in spite of their problem—a type of ceiling effect). Our secondary analysis supports this possibility. Another consideration is that interventions (particularly risky or painful interventions) are powerful stimulators of resiliency, which might have a greater effect on less resilient people who experience relatively greater hindrance of daily activities when in pain. Another point to consider is that PROMIS Pain Interference is to some degree a measure of symptoms and limitations and that there may be some confounding or collinearity with PROMIS PF. Several recent studies of response to treatment found that limitations persist longer, function is rated as lower, and pain intensity is higher with greater symptoms of depression and greater catastrophic thinking.7,10,16,17 It is not surprising that time passed since surgery is associated with better physical function, because the recovery is further along. This may be one explanation why some of our patients had lower PF scores postoperatively. Again, one specific provider was associated with a larger improvement in PF score, suggesting potential benefit in identifying, studying, and adopting best practices.
This study reinforces the influence of mood and coping strategies on patient-reported outcomes. It also reiterates that these effects occur on a continuum and are not all or none (eg, depressed or not depressed; catastrophizer or not catastrophizer). Patient mood and coping strategies represent important treatment opportunities to improve health by addressing these aspects of the illness. We might be able to reduce the stigma associated with considerations of coping strategies and mood by avoiding labels and categories, and by adjusting our communication and treatment strategies to ensure that they optimize capability, comfort, and calm.9
Acknowledgments
The authors thank Dave Mitten, MD, Allison McIntyre, MPH, CCCR, Chris DaSilva, BS, and the University of Rochester Health Lab for their assistance with collection and reporting of patient-reported outcomes through the UR VOICE platform.
Footnotes
Ethical Approval: Each author certifies that his or her institution approved the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.
Statement of Human and Animal Rights: All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008.
Statement of Informed Consent: Informed consent was obtained from all individual participants included in the study.
Declaration of Conflicting Interests: The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: One of the authors (D.R.) received royalties from Tornier (formerly Wright Medical; Bloomington, Minnesota) for elbow plates in the amount of less than USD 10 000 per year and Skeletal Dynamics Inc (Miami, Florida) for an internal joint stabilizer elbow in the amount of less than USD 10 000 per year; is a Deputy Editor for Hand and Wrist, Journal of Orthopaedic Trauma, and Clinical Orthopaedics and Related Research in the amount of USD 5000 per year; and received honoraria from meetings of AO North America, AO International, and various hospitals and universities.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iDs: DN. Bernstein 
https://orcid.org/0000-0002-1784-3288
References
- 1. Amtmann D, Cook KF, Jensen MP, et al. Development of a promis item bank to measure pain interference. Pain. 2010;150:173-182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Beckmann JT, Hung M, Voss MW, et al. Evaluation of the Patient-Reported Outcomes Measurement Information System Upper Extremity Computer Adaptive Test. J Hand Surg Am. 2016;41:739-744.e4. http://linkinghub.elsevier.com/retrieve/pii/S0363502316300922. Accessed May 7, 2018. [DOI] [PubMed] [Google Scholar]
- 3. Blackstone EA, Fuhr JP. Redefining health care: creating value-based competition on results. Atl Econ J. 2007;35:491-501. [Google Scholar]
- 4. Cella D, Riley W, Stone A, et al. The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005-2008. J Clin Epidemiol. 2010;63:1179-1194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Das De S, Vranceanu A-M, Ring DC. Contribution of kinesophobia and catastrophic thinking to upper-extremity-specific disability. J Bone Joint Surg Am. 2013;95:76-81. [DOI] [PubMed] [Google Scholar]
- 6. Hermanussen HH, Menendez ME, Chen NC, et al. Predictors of upper-extremity physical function in older adults. Arch bone Jt Surg. 2016;4:359-365. [PMC free article] [PubMed] [Google Scholar]
- 7. Karels CH, Bierma-Zeinstra SMA, Burdorf A, et al. Social and psychological factors influenced the course of arm, neck and shoulder complaints. J Clin Epidemiol. 2007;60:839-848. [DOI] [PubMed] [Google Scholar]
- 8. Kortlever JTP, Janssen SJ, van Berckel MMG, Ring D, Vranceanu AM. What is the most useful questionnaire for measurement of coping strategies in response to nociception? Clin Orthop Relat Res. 2015;473:3511-3518. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Liu TC, Bozic KJ, Teisberg EO. Value-based healthcare: person-centered measurement: focusing on the three C’s. Clin Orthop Relat Res. 2017;475:315-317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. London DA, Stepan JG, Boyer MI, et al. The impact of depression and pain catastrophization on initial presentation and treatment outcomes for atraumatic hand conditions. J Bone Joint Surg Am. 2014;96:806-814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Overbeek CL, Nota SPFT, Jayakumar P, et al. The PROMIS physical function correlates with the QuickDASH in patients with upper extremity illness. Clin Orthop Relat Res. 2015;473:311-317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Patient-Reported Outcomes Measurement Information System. A brief guide to the PROMIS Physical Function instruments: physical function. https://www.assessmentcenter.net/documents/PROMIS Physical Function Scoring Manual.pdf. Published 2015. Accessed January 1, 2017.
- 13. Ring D, Kadzielski J, Fabian L, et al. Self-reported upper extremity health status correlates with depression. J Bone Joint Surg Am. 2006;88:1983-1988. [DOI] [PubMed] [Google Scholar]
- 14. Schalet BD, Hays RD, Jensen SE, et al. Validity of PROMIS physical function measured in diverse clinical samples. J Clin Epidemiol. 2016;73:112-118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Schalet BD, Pilkonis PA, Yu L, et al. Clinical validity of PROMIS Depression, Anxiety, and Anger across diverse clinical samples. J Clin Epidemiol. 2017;73:119-127. doi: 10.1016/j.jclinepi.2015.08.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Vranceanu A-M, Barsky A, Ring D. Psychosocial aspects of disabling musculoskeletal pain. J Bone Joint Surg Am. 2009;91:2014-2018. [DOI] [PubMed] [Google Scholar]
- 17. Wegener ST, Castillo RC, Haythornthwaite J, et al. Psychological distress mediates the effect of pain on function. Pain. 2011;152:1349-1357. [DOI] [PubMed] [Google Scholar]