Algorithm Versus Expert: Machine Learning Versus Surgeon-Predicted Symptom Improvement After Carpal Tunnel Release.

BACKGROUND AND OBJECTIVES: Surgeons rely on clinical experience when making predictions about treatment effects. Incorporating algorithm-based predictions of symptom improvement after carpal tunnel release (CTR) could support medical decision-making. However, these algorithm-based predictions need to outperform predictions made by surgeons to add value. We compared predictions of a validated prediction model for symptom improvement after CTR with predictions made by surgeons. METHODS: This cohort study included 97 patients scheduled for CTR. Preoperatively, surgeons estimated each patient's probability of improvement 6 months after surgery, defined as reaching the minimally clinically important difference on the Boston Carpal Tunnel Syndrome Symptom Severity Score. We assessed model and surgeon performance using calibration (calibration belts), discrimination (area under the curve [AUC]), sensitivity, and specificity. In addition, we assessed the net benefit of decision-making based on the prediction model's estimates vs the surgeon's judgement. RESULTS: The surgeon predictions had poor calibration and suboptimal discrimination (AUC 0.62, 95%-CI 0.49-0.74), while the prediction model showed good calibration and appropriate discrimination (AUC 0.77, 95%-CI 0.66-0.89, P = .05). The accuracy of surgeon predictions was 0.65 (95%-CI 0.37-0.78) vs 0.78 (95%-CI 0.67-0.89) for the prediction model ( P = .03). The sensitivity of surgeon predictions and the prediction model was 0.72 (95%-CI 0.15-0.96) and 0.85 (95%-CI 0.62-0.97), respectively ( P = .04). The specificity of the surgeon predictions was similar to the model's specificity ( P = .25). The net benefit analysis showed better decision-making based on the prediction model compared with the surgeons' decision-making (ie, more correctly predicted improvements and/or fewer incorrectly predicted improvements). CONCLUSION: The prediction model outperformed surgeon predictions of improvement after CTR in terms of calibration, accuracy, and sensitivity. Furthermore, the net benefit analysis indicated that using the prediction model instead of relying solely on surgeon decision-making increases the number of patients who will improve after CTR, without increasing the number of unnecessary surgeries.

Explaining Personalized Activity Limitations in Patients With Hand and Wrist Disorders: Insights from Sociodemographic, Clinical, and Mindset Characteristics.

OBJECTIVES: To investigate the association of sociodemographic, clinical, and mindset characteristics on outcomes measured with a patient-specific patient-reported outcome measure (PROM); the Patient Specific Functional Scale (PSFS). Secondly, we examined whether these factors differ when a fixed-item PROM, the Michigan Hand Outcome Questionnaire (MHQ), is used as an outcome. DESIGN: Cohort study, using the aforementioned groups of factors in a hierarchical linear regression. SETTING: Twenty-six clinics for hand and wrist conditions in the Netherlands. PARTICIPANTS: Two samples of patients with various hand and wrist conditions and treatments: n=7111 (PSFS) and n=5872 (MHQ). INTERVENTIONS: NA. MAIN OUTCOME MEASURES: The PSFS and MHQ at 3 months. RESULTS: The PSFS exhibited greater between-subject variability in baseline, follow-up, and change scores than the MHQ. Better PSFS outcomes were associated with: no involvement in litigation (ß[95% confidence interval=-0.40[-0.54;-0.25]), better treatment expectations (0.09[0.06;0.13]), light workload (0.08[0.03;0.14]), not smoking (-0.07[-0.13;-0.01]), men sex (0.07[0.02;0.12]), better quality of life (0.07[0.05;0.10]), moderate workload (0.06[0.00;0.13]), better hand satisfaction (0.05[0.02; 0.07]), less concern (-0.05[-0.08;-0.02]), less pain at rest (-0.04[-0.08;-0.00]), younger age (-0.04[-0.07;-0.01]), better comprehensibility (0.03[0.01;0.06]), better timeline perception (-0.03[-0.06;-0.01]), and better control (-0.02[-0.04;-0.00]). The MHQ model was highly similar but showed a higher R2 than the PSFS model (0.41 vs 0.15), largely due to the R2 of the baseline scores (0.23 for MHQ vs 0.01 for PSFS). CONCLUSIONS: Health care professionals can improve personalized activity limitations by addressing treatment expectations and illness perceptions, which affect PSFS outcomes. Similar factors affect the MHQ, but the baseline MHQ score has a stronger association with the outcome score than the PSFS. While the PSFS is better for individual patient evaluation, we found that it is difficult to explain PSFS outcomes based on baseline characteristics compared with the MHQ. Using both patient-specific and fixed-item instruments helps health care professionals develop personalized treatment plans that meet individual needs and goals.

Treatment Invasiveness and Illness Perceptions Are Strongly Associated With Outcome Expectations in Patients Treated for Hand or Wrist Conditions: A Cross-sectional Study.

BACKGROUND: Multiple studies have shown that more-positive outcome expectations are associated with better treatment outcomes. Although this has not been shown to represent a causal relationship, there nonetheless is an interest in positively modifying outcome expectations to improve treatment outcomes. However, little is known about what is independently associated with outcome expectations in clinical practice. For example, it is unknown to what extent expectations are associated with treatment or patient characteristics such as sociodemographics or with patient-reported outcome measures (PROMs) on patient perceptions of physical or mental health or illness. Studying factors associated with outcome expectations may provide relevant information for clinicians and researchers aiming to improve outcome expectations. Improving expectations might, in turn, improve treatment outcomes. QUESTION/PURPOSE: Which factors (that is, sociodemographics, PROMs, illness perceptions, treatment, surgeon, and location) are independently associated with outcome expectations in patients with hand or wrist conditions? METHODS: This was a cross-sectional study. Between July 2018 and December 2021, we screened 21,327 patients with a diagnosed hand or wrist condition with complete baseline sociodemographic data such as age and workload. Sixty percent (12,765 of 21,327) of patients completed all relevant PROMs. We excluded patients receiving rare treatments, leaving 58% (12,345 of 21,327) for inclusion in the final sample. Those who participated were more often scheduled for surgical treatment and had higher expectations. We performed a multilevel analysis involving two steps. First, we evaluated whether patients receiving the same treatment, being counseled by the same surgeon, or being treated at the same location have more similar outcome expectations. We found that only patients receiving the same treatment had more similar outcome expectations. Therefore, we used a multilevel regression model to account for this correlation within treatments, and added treatment characteristics (such as nonsurgical versus minor or major surgery) to potential explanatory factors. Second, in the multilevel hierarchical regression analysis, we added sociodemographics (Model 1), PROMs for physical and mental health (Model 2), illness perceptions (Model 3), and treatment characteristics (most-definitive model) to assess the explained variance in outcome expectations per step and the relative association with outcome expectations. RESULTS: Sociodemographic factors such as age and workload explained 1% of the variance in outcome expectations. An additional 2% was explained by baseline PROMs for physical and mental health, 9% by illness perceptions, and 18% by treatment characteristics, resulting in an explained variance of 29% of the most-definitive model. A large number of patient and treatment characteristics were associated with outcome expectations. We used standardized betas to compare the magnitude of the effect of the different continuous and categorical variables. Among the associated variables, minor surgery (standardized beta [ß] = 0.56 [95% confidence interval 0.44 to 0.68]; p < 0.001) and major surgery (ß = 0.61 [95% CI 0.49 to 0.73]; p < 0.001) had the strongest positive association with outcome expectations (receiving surgery is associated with higher outcome expectations than nonsurgical treatment). A longer illness duration expected by the patient (-0.23 [95% CI -0.24 to -0.21]; p < 0.001) and being treated for the same condition as before (-0.08 [95% CI -0.14 to -0.03]; p = 0.003) had the strongest negative association with outcome expectations. CONCLUSION: Outcome expectations are mainly associated with the invasiveness of the treatment and by patients' illness perceptions; patients before surgical treatment have more positive expectations of the treatment outcome than patients before nonsurgical treatment, even after accounting for differences in clinical and psychosocial profiles. In addition, patients with a more-positive perception of their illness had more-positive expectations of their treatment. Our findings suggest expectation management should be tailored to the specific treatment (such as surgical versus nonsurgical) and the specific patient (including their perception of their illness). It may be more beneficial to test and implement expectation management strategies for nonsurgical treatments such as physical therapy than for surgical treatments, given that our findings indicate a greater need to do so. An additional advantage of such a strategy is that successful interventions may prevent converting to surgical interventions, which is a goal of the stepped-care principles of standard care. Future studies might investigate the causality of the association between pretreatment expectations and outcomes by performing an experimental study such as a randomized controlled trial, in which boosting expectations is compared with usual care in nonsurgical and surgical groups. LEVEL OF EVIDENCE: Level III, prognostic study.

Which Factors Are Associated With Satisfaction With Treatment Results in Patients With Hand and Wrist Conditions? A Large Cohort Analysis.

BACKGROUND: Satisfaction with treatment results is an important outcome domain in striving for patient-centered and value-based healthcare. Although numerous studies have investigated factors associated with satisfaction with treatment results, most studies used relatively small samples. Additionally, many studies have only investigated univariable associations instead of multivariable associations; to our knowledge, none have investigated the independent association of baseline sociodemographics, quality of life, improvement in pain and function, experiences with healthcare delivery, and baseline measures of mental health with satisfaction with treatment results. QUESTIONS/PURPOSES: (1) What factors are independently associated with satisfaction with treatment results at 3 months post-treatment in patients treated for common hand and wrist conditions? (2) What factors are independently associated with the willingness to undergo the treatment again at 3 months post-treatment in patients treated for common hand and wrist conditions? Among the factors under study were baseline sociodemographics, quality of life, improvement in pain and function, experiences with healthcare delivery, and baseline measures of mental health. METHODS: Between August 2018 and May 2020, we included patients who underwent carpal tunnel release, nonsurgical or surgical treatment for thumb-base osteoarthritis, trigger finger release, limited fasciectomy for Dupuytren contracture, or nonsurgical treatment for midcarpal laxity in one of the 28 centers of Xpert Clinics in the Netherlands. We screened 5859 patients with complete sociodemographics and data at baseline. Thirty-eight percent (2248 of 5859) of these patients had complete data at 3 months. Finally, participants were eligible for inclusion if they provided a relevant answer to the three patient-reported experience measure (PREM) items. A total of 424 patients did not do this because they answered "I don't know" or "not applicable" to a PREM item, leaving 31% (1824 of 5859) for inclusion in the final sample. A validated Satisfaction with Treatment Result Questionnaire was administered at 3 months, which identified the patients' level of satisfaction with treatment results so far on a 5-point Likert scale (research question 1, with answers of poor, moderate, fair, good, or excellent) and the patients' willingness to undergo the treatment again under similar circumstances (research question 2, with answers of yes or no). A hierarchical logistic regression model was used to identify whether baseline sociodemographics, quality of life, change in outcome (patient-reported outcome measures for hand function and pain), baseline measures of mental health (including treatment credibility [the extent to which a patient attributes credibility to a treatment] and expectations, illness perception, pain catastrophizing, anxiety, and depression), and PREMs were associated with each question of the Satisfaction with Treatment Result Questionnaire at 3 months post-treatment. We dichotomized responses to our first question as good and excellent, which were considered more satisfied, and poor, moderate, and fair, which were considered less satisfied. After dichotomization, 57% (1042 of 1824) of patients were classified as more satisfied with the treatment results. RESULTS: The following variables were independently associated with satisfaction with treatment results, with an area under the curve of 0.82 (95% confidence interval 0.80 to 0.84) (arranged from the largest to the smallest standardized odds ratio [SOR]): greater decrease in pain during physical load (standardized odds ratio 2.52 [95% CI 2.18 to 2.92]; p < 0.001), patient's positive experience with the explanation of the pros and cons of the treatment (determined with the question: "Have you been explained the pros and cons of the treatment or surgery?") (SOR 1.83 [95% CI 1.41 to 2.38]; p < 0.001), greater improvement in hand function (SOR 1.76 [95% CI 1.54 to 2.01]; p < 0.001), patients' positive experience with the advice for at-home care (determined with the question: "Were you advised by the healthcare providers on how to deal with your illness or complaints in your home situation?") (SOR 1.57 [95% CI 1.21 to 2.04]; p < 0.001), patient's better personal control (determined with the question: "How much control do you feel you have over your illness?") (SOR 1.24 [95% CI 1.1 to 1.40]; p < 0.001), patient's more positive treatment expectations (SOR 1.23 [95% CI 1.04 to 1.46]; p = 0.02), longer expected illness duration by the patient (SOR 1.20 [95% CI 1.04 to 1.37]; p = 0.01), a smaller number of symptoms the patient saw as part of the illness (SOR 0.84 [95% CI 0.72 to 0.97]; p = 0.02), and less concern about the illness the patient experiences (SOR 0.84 [95% CI 0.72 to 0.99]; p = 0.04). For willingness to undergo the treatment again, the following variables were independently associated with an AUC of 0.81 (95% CI 0.78 to 0.83) (arranged from the largest to the smallest standardized OR): patient's positive experience with the information about the pros and cons (determined with the question: "Have you been explained the pros and cons of the treatment or surgery?") (SOR 2.05 [95% CI 1.50 to 2.80]; p < 0.001), greater improvement in hand function (SOR 1.80 [95% CI 1.54 to 2.11]; p < 0.001), greater decrease in pain during physical load (SOR 1.74 [95% CI 1.48 to 2.07]; p < 0.001), patient's positive experience with the advice for at home (determined with the question: "Were you advised by the healthcare providers on how to deal with your illness or complaints in your home situation?") (SOR 1.52 [95% CI 1.11 to 2.07]; p = 0.01), patient's positive experience with shared decision-making (determined with the question: "Did you decide together with the care providers which care or treatment you will receive?") (SOR 1.45 [95% CI 1.06 to 1.99]; p = 0.02), higher credibility the patient attributes to the treatment (SOR 1.44 [95% CI 1.20 to 1.73]; p < 0.001), longer symptom duration (SOR 1.27 [95% CI 1.09 to 1.52]; p < 0.01), and patient's better understanding of the condition (SOR 1.17 [95% CI 1.01 to 1.34]; p = 0.03). CONCLUSION: Our findings suggest that to directly improve satisfaction with treatment results, clinicians might seek to: (1) improve the patient's experience with healthcare delivery, (2) try to influence illness perception, and (3) boost treatment expectations and credibility. Future research should confirm whether these suggestions are valid and perhaps also investigate whether satisfaction with treatment results can be predicted (instead of explained, as was done in this study). Such prediction models, as well as other decision support tools that investigate patient-specific needs, may influence experience with healthcare delivery, expectations, or illness perceptions, which in turn may improve satisfaction with treatment results. LEVEL OF EVIDENCE: Level III, therapeutic study.

What Are the Minimally Important Changes of Four Commonly Used Patient-reported Outcome Measures for 36 Hand and Wrist Condition-Treatment Combinations?

BACKGROUND: Patient-reported outcome measures (PROMs) are frequently used to assess treatment outcomes for hand and wrist conditions. To adequately interpret these outcomes, it is important to determine whether a statistically significant change is also clinically relevant. For this purpose, the minimally important change (MIC) was developed, representing the minimal within-person change in outcome that patients perceive as a beneficial treatment effect. Prior studies demonstrated substantial differences in MICs between condition-treatment combinations, suggesting that MICs are context-specific and cannot be reliably generalized. Hence, a study providing MICs for a wide diversity of condition-treatment combinations for hand and wrist conditions will contribute to more accurate treatment evaluations. QUESTIONS/PURPOSES: (1) What are the MICs of the most frequently used PROMs for common condition-treatment combinations of hand and wrist conditions? (2) Do MICs vary based on the invasiveness of the treatment (nonsurgical treatment or surgical treatment)? METHODS: This study is based on data from a longitudinally maintained database of patients with hand and wrist conditions treated in one of 26 outpatient clinics in the Netherlands between November 2013 and November 2020. Patients were invited to complete several validated PROMs before treatment and at final follow-up. All patients were invited to complete the VAS for pain and hand function. Depending on the condition, patients were also invited to complete the Michigan Hand outcomes Questionnaire (MHQ) (finger and thumb conditions), the Patient-rated Wrist/Hand Evaluation (PRWHE) (wrist conditions), or the Boston Carpal Tunnel Questionnaire (BCTQ) (nerve conditions). Additionally, patients completed the validated Satisfaction with Treatment Result Questionnaire at final follow-up. Final follow-up timepoints were 3 months for nonsurgical and minor surgical treatment (including trigger finger release) and 12 months for major surgical treatment (such as trapeziectomy). Our database included 55,651 patients, of whom we excluded 1528 who only required diagnostic management, 25,099 patients who did not complete the Satisfaction with Treatment Result Questionnaire, 3509 patients with missing data in the PROM of interest at baseline or follow-up, and 1766 patients who were part of condition-treatment combinations with less than 100 patients. The final sample represented 43% (23,749) of all patients and consisted of 36 condition-treatment combinations. In this final sample, 26% (6179) of patients were managed nonsurgically and 74% (17,570) were managed surgically. Patients had a mean ± SD age of 55 ± 14 years, and 66% (15,593) of patients were women. To estimate the MIC, we used two anchor-based methods (the anchor mean change and the MIC predict method), which were triangulated afterward to obtain a single MIC. Applying this method, we calculated the MIC for 36 condition-treatment combinations, comprising 22 different conditions, and calculated the MIC for combined nonsurgical and surgical treatment groups. To examine whether the MIC differs between nonsurgical and surgical treatments, we performed a Wilcoxon signed rank test to compare the MICs of all PROM scores between nonsurgical and surgical treatment. RESULTS: We found a large variation in triangulated MICs between the condition-treatment combinations. For example, for nonsurgical treatment of hand OA, the MICs of VAS pain during load clustered around 10 (interquartile range 8 to 11), for wrist osteotomy/carpectomy it was around 25 (IQR 24 to 27), and for nerve decompression it was 21. Additionally, the MICs of the MHQ total score ranged from 4 (nonsurgical treatment of CMC1 OA) to 15 (trapeziectomy with LRTI and bone tunnel), for the PRWHE total score it ranged from 2 (nonsurgical treatment of STT OA) to 29 (release of first extensor compartment), and for the BCTQ Symptom Severity Scale it ranged from 0.44 (nonsurgical treatment of carpal tunnel syndrome) to 0.87 (carpal tunnel release). An overview of all MIC values is available in a freely available online application at: https://analyse.equipezorgbedrijven.nl/shiny/mic-per-treatment/. In the combined treatment groups, the triangulated MIC values were lower for nonsurgical treatment than for surgical treatment (p < 0.001). The MICs for nonsurgical treatment can be approximated to be one-ninth (IQR 0.08 to 0.13) of the scale (approximately 11 on a 100-point instrument), and surgical treatment had MICs that were approximately one-fifth (IQR 0.14 to 0.24) of the scale (approximately 19 on a 100-point instrument). CONCLUSION: MICs vary between condition-treatment combinations and differ depending on the invasiveness of the intervention. Patients receiving a more invasive treatment have higher treatment expectations, may experience more discomfort from their treatment, or may feel that the investment of undergoing a more invasive treatment should yield greater improvement, leading to a different perception of what constitutes a beneficial treatment effect. CLINICAL RELEVANCE: Our findings indicate that the MIC is context-specific and may be misleading if applied inappropriately. Implementation of these condition-specific and treatment-specific MICs in clinical research allows for a better study design and to achieve more accurate treatment evaluations. Consequently, this could aid clinicians in better informing patients about the expected treatment results and facilitate shared decision-making in clinical practice. Future studies may focus on adaptive techniques to achieve individualized MICs, which may ultimately aid clinicians in selecting the optimal treatment for individual patients.