Study Design & Statistics

Principles of statistical integrity in orthopaedic research, focusing on the identification of spin and the application of the Fragility Index to assess outcome robustness.

Overview

Clinical translation of meniscal scaffolds remains constrained by methodological heterogeneity and small sample sizes, though short-to-mid-term outcomes show promise [1]. Definitive conclusions regarding robotic arm-assisted patellofemoral joint arthroplasty are currently precluded by a lack of randomized trials, small study numbers, and short follow-up periods [11].

Rigorous trial design is essential to mitigate bias and ensure clinical relevance. Trialists must employ stopping rules requiring large numbers of outcome events and adequate safety data to prevent biased decisions [2]. Fully randomized predictor-outcome pairings in large national databases frequently yield statistically significant results driven purely by sample size that lack clinical meaning [4]. Furthermore, changes to eligibility criteria and primary or secondary outcomes are often not reflected in registered trial data [5].

Future research must address specific methodological gaps to enhance validity. Multivariable analyses will remain the mainstay of observational orthopaedic research, provided they are used with appropriate study design to produce interpretable results [28]. Future matched studies require justification for included covariates and must properly account for matching in statistical analyses [13]. Understanding patient preferences can inform recruitment feasibility and future trial designs [12]. Orthopaedic surgeons prefer expertise-based randomized trials, which hold promise for reducing bias and enhancing feasibility, though this design is not a panacea [36]. Finally, the strengths and limitations of both registry cohort studies and randomized controlled trials must be understood to properly evaluate the literature [24].

Anatomy & Pathophysiology

Kinematics and Measurement

Currently available wearable devices assess spinal posture with good accuracy in the clinical setting [66], though a lack of current research exists to establish the psychometric properties of non-invasive 3D human posture-measuring instruments [97]. The proposed method for measuring the three-dimensional scoliosis angle from standard radiographs facilitates 3D-scoliosis assessment without sophisticated devices [98], while scoliotic curvatures in preoperative adolescent idiopathic scoliosis patients are largely represented by both biplanar low-dose stereoradiography and computed tomography despite differences in body positioning [102]. Variations in spinal growth velocity exert a more direct influence over changes in angle velocity than height velocity in peri-pubertal girls with idiopathic scoliosis [91]. Fatigue alters lower extremity kinematics during a single-leg stop-jump task without revealing gender differences in these kinematic changes [95], and weight-bearing status affects in vivo kinematics following mobile-bearing unicompartmental knee arthroplasty [107]. Surgical parameters influence paediatric knee kinematics and cartilage stresses in anterior cruciate ligament reconstruction, necessitating personalized surgical planning to restore native motion and tissue mechanics [111].

Osseous and Joint Mechanics

Whole-body vibration training may substantially increase bone mineral density in the L2–L4 lumbar spine segment, whereas it has a negligible effect on the L1–L4 lumbar spine segment and total hip region [96]. In in vitro knee rig investigations, no statistical differences were found in knee kinematics and retropatellar pressure distribution between rapid prototype and standard materials [99]. Both fixed- and mobile-bearing total knee arthroplasty prostheses show typical kinematics of an anterior instability, rendering them incapable of performing physiological movement [100]. Postoperative lateral laxity greater than 0.9 mm at 90-degree flexion was associated with physiological kinematic motion and fewer knee symptoms in patient-reported outcomes for posterior cruciate retaining total knee arthroplasty [110]. Patellofemoral pain may derive from a combination of physical activity in the context of pathological kinematics [73].

Spinal and Cervical Assessment

The Lyon brace is highly effective in correcting thoracic curves through its biomechanical action on vertebral modeling, particularly when SOSORT guidelines were adopted in addition to SRS criteria [93]. Method 2, measuring the angle from the inferior endplate of the vertebra above the degenerative disc to the inferior endplate of the vertebra below, showed the best intraobserver and interobserver reliability overall in assessing cervical sagittal rotation [103].

Soft Tissue and Implant Loading

Loading characteristics are more favorable when cement is placed along the entire back of the polyethylene glenoid component contacting the subchondral bone [106].

Classification

Foundational Epidemiology: Five studies published in AJSM over the past 50 years highlight the impact of epidemiology and statistics on injury prevention, scoring scales, and multicenter study design [19].

Osteochondritis Dissecans: A novel arthroscopic classification system for osteochondritis dissecans of the knee may be used clinically and to facilitate future research, including multicenter studies [30].

Proximal Radius: The Mason classification is the most reliable system for proximal radius fractures [33].

Rotator Cuff: Twenty-six different criteria described by multiple classification systems have been identified for the magnetic resonance assessment of rotator cuff after repair [45].

Knee Society Radiographic Evaluation: The Knee Society Radiographic Evaluation System is not predictive or prognostic but aims to accumulate data to potentially formulate future implant risk criteria [51].

Other Considerations: Reliable classification leads to more confidence that patients can be treated like similar patients in a study that had success, provided the reliability study environment is proper, the statistics are appropriate, and the magnitude of the ICC is acceptable [46]. Fundamental statistical concepts, terminology, and methodological principles such as randomization, blinding, and hypothesis testing are essential for the design, analysis, and interpretation of orthopaedic research [48]. Essential considerations for the analytic phase of a meta-analysis include the selection of effect estimators, interpretation of heterogeneity, and various sub-types of meta-analytic approaches to ensure data are appropriately handled and results are reliable [50]. Research should include comprehensive conceptual models that capture complete sets of important independent variables [52]. A two-stage approach using latent class analysis is preferable to a single-stage approach for identifying subgroups of patients with low back pain, as it yields more distinct and clinically meaningful subgroups [53]. Understanding valid statistical methods is essential for the correct design of research projects and for creating accurate predictive models to assist with evaluating surgeon performance and guiding clinical decisions [54]. A study's conclusions are only as strong as its methods [58]. Unsupervised machine learning methods including clustering techniques and dimensionality reduction tools carry great potential to elucidate structures and patterns in high-dimensional data and expand understanding of human physiology and disease [59]. Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true [40]. No difference in clinical outcome between a well-established total knee arthroplasty design and its successor was found, and the expected benefits of design modifications could not be observed in clinical outcome scores 2 years postoperatively [61].

Clinical Presentation

Evidence for clinical translation of meniscal scaffolds remains limited by methodological heterogeneity and small sample sizes, though short-to-mid-term outcomes appear promising [1]. In contrast, fully randomized predictor-outcome pairings in large national databases frequently yield statistically significant results lacking clinical relevance, where significance is driven purely by sample size rather than meaningful patient impact [4]. Clinicians must therefore evaluate treatment effects using the Minimum Clinically Important Difference (MCID) rather than relying solely on p-values, as statistically significant differences smaller than the MCID are unlikely to be perceived as important by patients [18]. Statistical significance is distinct from clinical significance, necessitating critical analysis of medical research literature to ensure proper application and interpretation of statistical analyses [23].

Readers must assess study validity, evaluate statistical significance, and interpret clinical relevance before altering practice [25]. Thresholds for WOMAC, EQ-5D, and Knee Society Scores facilitate the interpretation of patient-reported outcome measures in clinical practice [26]. However, current patient-reported outcome measures lack the discrimination and calibration necessary for clinical risk stratification [37]. Artificial intelligence analysis of biomedical large clinical registry data using machine learning requires tens of thousands of subjects and a focus on substantial clinical benefit, as the MCID is considered too low a bar for such analyses [31]. Consequently, registries should include validated patient-reported outcome scores and measures of clinical relevance and expectations [31].

Diagnostic Tools: The Nottingham Trauma Frailty Index is a clinically applicable tool using easily and routinely measured physiological and functional parameters to guide patient care and stratify the analysis of quality improvement and research projects [27]. For specific populations, representative regional reference standards are recommended for diagnosis when estimating T scores with Hologic using Native vs. Caucasian data in Indians [39]. The Italian version of the High-Activity Arthroplasty Score (HAAS-I) is recommended for clinical and research purposes [14].

Study Design & Reporting: The clinical course of low back pain symptoms follows a similar pattern in randomized clinical trials and cohort observational studies [3]. Changes to the final presentation of eligibility criteria and primary and secondary outcomes in sports medicine randomized controlled trials are frequently not reflected in registered trial data [5]. Several study characteristics influence the reporting of positive outcomes in distal radius fracture literature, though whether specific characteristics portend a greater chance of publication remains unclear [38]. The document regarding three-dimensional corrective exercise therapy for idiopathic scoliosis is a study protocol and does not report results or conclusions [6]. Similarly, the document titled "A Syllabus of Laboratory Examinations in Clinical Diagnosis" is a collection of book reviews and does not present original clinical research or specific conclusions regarding medical interventions [16].

Imaging & Outcomes: Extrusion, meniscal signal change, loss of shape, synovitis, and bone marrow oedema are reliable scoring parameters to assess MRI appearance post meniscal transplant [35]. Further research with larger meniscal allograft transplant cohort groups and patient reported outcome measures may be helpful to correlate clinical significance and guide further management [35]. No differences in clinical outcomes were detected between posterior-stabilised mobile-bearing and fixed-bearing total knee arthroplasty groups despite wider translations and rotations in the mobile-bearing group [32]. Multicenter prospective studies incorporating rigorous external validation protocols are needed to further establish the clinical utility of artificial intelligence in automated measurement of patellofemoral joint parameters [34].

Investigations

Plain radiography: EOS® standing full-leg radiographs provide more radiographic information than alternative methods while utilizing lower radiation doses [56]. For elbow anatomy, the most commonly utilized radiographic measures remain consistent between sexes, across the adolescent age group, and between adolescents and young adults [57]. In the context of osteoporotic thoracolumbar vertebral body fractures, the addition of CT offers limited value compared to conventional radiographs and MRI specifically for OF spine classification and the OF score [62]. Regarding knee pathology, baseline radiographic severity grade predicts future total knee arthroplasty risk only in the absence of a full-thickness defect, whereas full-thickness cartilage defects serve as important independent predictive factors for progression in older adults with minimal to moderate osteoarthritis [82].

MRI: Quantitative and semi-quantitative MR-imaging parameters, alongside protective clinical parameters, are associated with maintaining radiographically normal knee joints in an older population over 8 years [63]. MRI utilization by orthopaedic surgeons correlates with more appropriate interventions for patients whose symptoms and findings are most amenable to surgical intervention [68]. However, negative MRI results should not preclude diagnostic arthroscopy for grading knee chondral defects [69]. In children, selective magnetic resonance imaging does not provide enhanced diagnostic utility over clinical examination and should be used judiciously only when the clinical diagnosis is uncertain and MRI input will alter the treatment plan [70]. For meniscal extrusion, results support integrating dynamic factors and clinical outcomes into MRI-based classifications to inform treatment approaches [72]. In patellar instability, the sTTTG yields smaller values on MRI compared with CT, though the clinical significance of this difference remains undetermined [76]. Regarding ulnar collateral ligament injuries, a newly proposed 6-stage MRI-based classification utilizing grade and location demonstrates substantial to near perfect agreement among fellowship-trained observers [83]. Conversely, differences in angle of inclination findings for the native ACL may be attributed to variations in MRI and measurement techniques [65].

CT: CT offered the most accurate and precise assessment of acetabular component positioning in minimally invasive total hip arthroplasty, while other assessment methods remained within acceptable limits [78]. The fluoroscopically simulated Dunn view can measure femoral version with acceptable accuracy and obviates the need for repeat 3-dimensional imaging in patients who already possess an MRI scan without version analysis [80].

Other Considerations: Clinical translation of meniscal scaffolds is currently limited by differences in study methodology and small sample sizes, though it shows promise for improving clinical outcomes in the short to mid-term [1]. Specialist radiological imaging is specific for cartilage disease in the knee but demonstrates poorer sensitivity for determining therapeutic options [74]. Radiographs and MRI were not reliable for identifying the presence of a bone tunnel or evaluating bone tunnel widening after anterior cruciate ligament reconstruction [71]. The reproducibility of radiologic measurement of posterior femoral offset may be unreliable for informing total knee arthroplasty implantation [77]. Specific areas must be addressed in future studies to advance the meaningful and clinical use of AI for radiograph interpretation in acute scaphoid fracture detection [75].

Treatment

Non-Operative

An initial trial of nonoperative management may be considered in young active patients with isolated SLAP tears [79]. Clinicians and researchers must evaluate treatment effects in terms of the Minimum Clinically Important Difference (MCID) rather than relying solely on p values, as statistically significant differences smaller than the MCID are unlikely to be perceived as important by patients [18]. Low-intensity long-duration ultrasound may be used as a conservative non-pharmaceutical and non-invasive treatment option for patients with knee osteoarthritis [88]. Commencement of mobilisation within one week from injury for non-surgically managed proximal humerus fractures is safe and may confer short-term functional benefits compared to delayed mobilisation [92]. Twenty-five (24.0%) knees that underwent nonsurgical treatment for symptomatic intact discoid lateral meniscus met the overall failure criteria after a mean follow-up of 76.9 months, with older age and the presence of intrameniscal signs identified as risk factors for nonsurgical treatment failure [104]. No citations provided Level I or II evidence substantiating the effectiveness of 3 months of nonoperative treatment in patients who would otherwise be candidates for total joint arthroplasty [108].

Operative

Indications: Meniscectomy remains overused due to non-scientific factors like technical difficulty, cost, and patient preferences, despite evidence supporting meniscal repair and conservative treatment [67]. Definitive conclusions regarding robotic arm-assisted patellofemoral joint arthroplasty are precluded by the small number of available studies, short follow-up, and lack of randomized trials [11]. Understanding patient preferences can inform future trial designs and feasibility assessments regarding recruitment rates for placebo-controlled surgical trials of the knee [12].

Surgical Approach / Technique: The prognosis for successful repair of non-union of the shafts of the long bones is surprisingly similar whatever method is used, provided good judgment is exercised in the selection of the method [94]. Trialists should use stopping rules that require a large number of outcome events and ensure adequate safety data to prevent biased clinical decisions [2].

Implant Selection: The Italian version of the High-Activity Arthroplasty Score (HAAS-I) is recommended for clinical and research purposes following hip and knee total arthroplasty [14]. Classifications of good versus poor outcome following knee arthroplasty should not be defined using arbitrary cutoff scores, as this homogeneity impedes scientific progress [101].

Alignment / Balancing Strategy: Fully randomized predictor-outcome pairings in large national databases can yield frequent statistically significant results without clinical relevance, driven purely by sample size [4].

Pain Management: Longer follow-up data regarding the efficacy and safety of intra-articular steroid injections for hip osteoarthritis, along with potentially influential factors, remain unclear and require further confirmation [7].

Adjuncts: The APEX study aims to address the lack of conclusive evidence regarding acupuncture's clinical effectiveness and its superiority over sham interventions for older adults with knee pain [43].

Setting of Care: Non-responders to patient-reported outcomes in hip and knee arthroplasty had significantly increased mortality and significantly worse baseline scores despite similar demographics and revision risk [84].

Other Considerations: Clinical translation of meniscal scaffolds is limited by differences in study methodology and small sample sizes, though short to mid-term clinical outcomes are promising [1]. Observational data demonstrating the effectiveness of aspirin prophylaxis are strongly biased by the selection of lower-risk patients to receive aspirin and higher-risk patients to receive potent anticoagulants, meaning no specific prophylaxis regimen can be endorsed as 'best practice' without data from prospective randomized clinical trials [42]. Randomized controlled trials provide the most unbiased assessment of the risks and benefits of major new medical therapies and should be increasingly used in orthopaedic surgery to resolve controversies and ensure reliable information [49]. An intention-to-treat analysis should almost always be the primary analysis in superiority trials to maintain the integrity of randomization, whereas per-protocol analysis is usually better for equivalence or non-inferiority trials [44]. Examining the robustness of clinical trials requires an estimation of uncertainty rather than a misconstrued, dichotomous focus on statistical significance [47]. Adhering to simple statistical principles, such as ensuring congruence between research questions and data analysis, meeting parametric assumptions, correcting for multiple comparisons, selecting appropriate tests for data types, and justifying effect sizes in power analyses, can help authors avoid common statistical pitfalls [41]. Meta-analyses are valuable for synthesizing data to estimate treatment effects but must be performed with rigorous methodology to ensure validity, minimize bias, and avoid misleading conclusions [29]. Orthopaedic research basics include levels of evidence, tools for evaluating study quality, the use of clinical practice guidelines, grades of recommendations, ethical considerations, and the use and limitations of orthopaedic registries [15]. Limited data shows that high CONUT is associated with a non-significant increased risk of adverse outcomes after spinal surgeries [109]. The document for three-dimensional corrective exercise therapy for idiopathic scoliosis is a study protocol and does not report results or conclusions [6].

Complications

Other Considerations: Current evidence for clinical translation of meniscal scaffolds is limited by differences in study methodology and small sample sizes [1]. Trialists must implement stopping rules requiring a large number of outcome events and adequate safety data to prevent biased clinical decisions [2]. Frequent unreported changes to eligibility criteria and primary or secondary outcomes in registered trial data further compromise validity [5]. Propensity-score adjustments cannot substitute for adequately powered, well-controlled, prospective, randomized studies [81]. While propensity score matching is a popular technique to reduce bias in observational studies, researchers must carefully select the matching model and covariates [90]. This method carries the limitation of an inability to control for unobserved confounders [90]. Additionally, dependent observations are common in orthopaedic sports literature, yet most studies fail to adjust for this nonindependence [89]. Future matched studies require justification for included covariates and must properly account for matching in statistical analyses to enhance validity [13].

Long-term data regarding efficacy and safety for intra-articular steroid injection in hip osteoarthritis remain unclear and require further confirmation [7]. Longer follow-up is necessary to evaluate intermediate or long-term effects of open reduction and fixation for medium-sized posterior fragments in AO type B ankle fractures [8]. A long-term follow-up study has been designed to verify stable fixation of a new short uncemented, proximally fixed anatomic femoral implant [9]. Similarly, a large cohort long-term follow-up study is required to evaluate the intraoperative load sensor for posterior-stabilized TKA [10]. Longer follow-up will help determine the durability and long-term outcomes of surgeries for displaced femoral neck fractures in workers' compensation patients aged 45-65 years [17]. Future studies are warranted to capture longer follow-up of patient-reported outcomes following cemented versus cementless primary total knee arthroplasty [21]. Individualised total knee arthroplasty demonstrates 99.4% survival at 3 to 5 years [22]. Patients experiencing clinically relevant outcome improvement after osteochondral autograft transplantation of the knee in the short term continue to experience sustained benefits at longer-term follow-up [20]. Further research is needed to determine if improvements in sensor-guided robot-assisted TKA reach the minimal clinically important difference for long-term postoperative outcomes [64].

Recovery

Light activity (weeks): Evidence regarding specific timelines for light activity, such as desk work or driving, is not explicitly defined in the provided data. However, studies indicate that mobile and fixed-bearing all-polyethylene tibial component total knee arthroplasty designs functioned equivalently at the time of early follow-up in a low-to-moderate-demand patient group [113]. Additionally, time to surgery for delayed primary hip arthroplasty for geriatric low-energy femoral neck fracture was not an independent risk factor for mortality or functional outcomes when controlling for patient-specific factors [114].

Full activity (months): The clinical course of low back pain symptoms followed a pattern that was similar in randomised clinical trials and cohort observational studies [3]. Patients who experience clinically relevant outcome improvement after osteochondral autograft transplantation of the knee in the short term continue to experience sustained benefits at longer-term follow-up [20]. Clinical outcomes after lateral meniscal allograft transplantation improved compared with pre-surgery outcomes and were maintained throughout a long-term follow-up period averaging 10.3 years [60].

Complete recovery / outcome plateau (months): Longer follow-up is needed to evaluate intermediate or long-term effects of open reduction and fixation for medium-sized posterior fragments in AO type B ankle fractures [8]. Longer follow-up will help determine the durability and long-term outcomes of fixation versus joint replacement for displaced femoral neck fractures in workers' compensation patients aged 45-65 years [17]. Further studies with long-term follow-up are needed to determine whether the grafted area in autologous matrix-induced chondrogenesis for focal cartilage defects in the knee will maintain structural and functional integrity over time [55]. Future studies are warranted to capture longer follow-up of patient-reported outcomes following cemented versus cementless primary total knee arthroplasty [21]. The results of individualised total knee arthroplasty demonstrating 99.4% survival at 3 to 5 years warrant continued follow-up to evaluate long-term results [22]. Longer-term survival analysis for a short bone-conserving femoral stem still needs to be determined [115].

Rehabilitation protocol: Trialists should use stopping rules that require a large number of outcome events and ensure adequate safety data to prevent biased clinical decisions [2]. A long-term follow-up study has been designed to verify stable fixation and research clinical outcomes for a new short uncemented, proximally fixed anatomic femoral implant [9]. A long-term follow-up study with a large cohort is required to determine if an intraoperative load sensor improves early postoperative results of posterior-stabilized TKA for osteoarthritis with varus deformities [10].

Functional milestones: The longer follow-up data of efficacy and safety for intra-articular steroid injection in hip osteoarthritis are still unclear and need further confirmation [7]. Longitudinal studies are required to establish the potential role of intrinsic lumbar spine shape phenotypes as risk or prognostic factors for lumbar disc degeneration [112].

Key Evidence

• [L2] The evidence for clinical translation is limited by differences in study methodology and small sample sizes, but is promising in terms of improving clinical outcomes in the short to mid-term. (10.3390/ijms20030632)
• [L5] Trialists should use stopping rules that require a large number of outcome events and ensure adequate safety data to prevent biased clinical decisions. (10.2106/jbjs.k.01412)
• [L1] The clinical course of LBP symptoms followed a pattern that was similar in RCTs and cohort observational studies. (10.1186/1471-2474-15-68)
• [L4] Fully randomized predictor-outcome pairings in a large national database yielded frequent statistically significant results without clinical relevance, which were driven purely by sample size. (10.1016/j.arth.2025.08.053)
• [L2] Changes are also frequently made to the final presentation of eligibility criteria and primary and secondary outcomes that are not reflected in the registered trial data. (10.1177/0363546512448363)
• [L3] This document is a study protocol for a prospective non-randomized trial and does not report results or conclusions. (10.1186/s12891-022-05057-7)
• [L1] The longer follow-up data of efficacy and safety and potentially influential factors are still unclear and needed further confirmation. (10.1155/2020/6320154)
• [L1] Longer follow-up is needed to evaluate intermediate or long-term effects. (10.1302/0301-620x.107b4.bjj-2024-0521.r3)
• [L5] A long-term follow-up study has been designed to verify stable fixation and to research into the clinical outcome. (10.1186/1471-2474-9-147)
• [L2] A long-term followup study with a large cohort is required. (10.1007/s00167-018-5314-7)
• [L1] However, the small number of available studies, short follow-up and lack of randomised trials preclude definitive conclusions. (10.1002/ksa.70301)
• [L4] However, understanding their preferences can inform future trial designs and feasibility assessments with regard to recruitment rates. (10.1302/0301-620x.106b12.bjj-2023-1266.r2)
• [L1] Future matched studies should provide justification for included covariates and properly account for matching in their statistical analyses to enhance the validity of study findings. (10.1016/j.jse.2025.01.021)
• [L4] Its use is recommended for clinical and research purposes. (10.1186/s13018-018-0782-5)
• [L5] This document is a collection of book reviews and does not present original clinical research, patient data, or specific conclusions regarding medical interventions or outcomes. (10.2106/00004623-195133040-00044)
• [L3] Longer follow-up will help determine the durability and long-term outcomes of these surgeries. (10.1016/j.arth.2020.06.003)
• [Commentary] Clinicians and researchers must evaluate treatment effects in terms of the Minimum Clinically Important Difference (MCID) rather than relying solely on p values, as statistically significant differences smaller than the MCID are unlikely to be perceived as important by patients. (10.1007/s11999-017-5253-5)
• [L5] This editorial presents five foundational studies on epidemiology and statistics published in AJSM over the past 50 years, highlighting their impact on injury prevention, scoring scales, and multicenter study design. (10.1177/03635465221113347)
• [L3] Patients who experience clinically relevant outcome improvement after OATs in the short term continue to experience sustained benefits at longer-term follow-up. (10.1002/ksa.12422)
• [L3] Future studies are warranted to capture longer follow-up of PROs. (10.1186/s12891-022-05899-1)
• [L4] The results are promising and warrant continued follow-up to evaluate long-term results. (10.1002/ksa.70265)
• [L5] Statistical significance is different than clinical significance, and understanding clinical outcomes requires critical analysis of medical research literature to ensure that statistical analyses have been properly applied and interpreted. (10.1016/j.arthro.2017.03.013)
• [L3] Orthopaedic registry studies differ from randomized controlled trials in many ways and offer certain advantages; the strengths and limitations of registry cohort studies and RCTs must be understood to properly evaluate the literature. (10.2106/jbjs.n.01332)
• [L5] When looking back on findings of studies, the reader should assess the validity of the study, evaluate the statistical significance, and interpret the clinical relevance of the differences found; if all 3 conditions are met, the reader should consider changing practice to reflect the results of the study. (10.1177/0363546516651879)
• [L3] We have developed a clinically applicable tool using easily and routinely measured physiological and functional parameters, which clinicians and researchers can use to guide patient care and to stratify the analysis of quality improvement and research projects. (10.1302/0301-620x.106b4.bjj-2023-1058.r1)
• [L5] Multivariable analyses will continue to be the mainstay of observational orthopaedic research and are effective statistical tools that produce easily interpretable and useful results when used with appropriate study design. (10.1007/s00167-022-07215-9)
• [L5] Meta-analyses are valuable for synthesizing data to estimate treatment effects, but they must be performed with rigorous methodology to ensure validity, minimize bias, and avoid misleading conclusions. (10.1007/s00167-022-07304-9)
• [L3] In light of this reliability, this classification system may be used clinically and to facilitate future research, including multicenter studies for OCD. (10.1177/0363546516637175)
• [L5] Registries should include validated patient-reported outcome scores and measures of clinical relevance and expectations. (10.1016/j.arthro.2023.10.035)
• [L2] Despite this, no differences in clinical outcomes were detected between groups. (10.1007/s00167-023-07541-6)
• [L4] The Mason classification is the most reliable system. (10.1186/1471-2474-10-120)
• [L1] To further establish clinical utility, multicenter prospective studies incorporating rigorous external validation protocols are needed. (10.1186/s13018-025-06247-4)
• [L3] Further research with larger MAT cohort groups and patient reported outcome measures may be helpful to correlate its clinical significance and guide further management. (10.1007/s00167-021-06720-7)
• [Letter] The authors view the expertise-based design as an option that holds substantial promise to provide less biased results than the conventional design and to enhance feasibility of conducting randomized controlled trials in surgery, though they do not view it as a panacea. (10.1007/s11999-008-0576-x)
• [L5] Current patient-reported outcome measures (PROMs) lack the discrimination and calibration necessary for clinical risk stratification. (10.1016/j.arth.2025.10.039)
• [L2] Several study characteristics influence the reporting of positive outcomes, but whether the presence of these characteristics portends a greater chance of publication remains unclear. (10.1016/j.jhsa.2013.02.023)
• [L4] The authors recommend the adoption of representative regional reference standards for diagnosis. (10.1186/s12891-025-08599-8)
• [L5] Simulations show that for most study designs and settings, it is more likely for a research claim to be false than true. (10.1371/journal.pmed.0020124)
• [L5] Adhering to simple statistical principles, such as ensuring congruence between research questions and data analysis, meeting parametric assumptions, correcting for multiple comparisons, selecting appropriate tests for data types, and justifying effect sizes in power analyses, can help authors avoid common statistical pitfalls. (10.1016/j.jht.2013.09.004)
• [L5] Observational data demonstrating the effectiveness of aspirin prophylaxis are strongly biased by the selection of lower-risk patients to receive aspirin and higher-risk patients to receive potent anticoagulants; consequently, no specific prophylaxis regimen can be endorsed as 'best practice' without data from prospective randomized clinical trials. (10.2106/jbjs.19.01135)
• [L1] The study aims to address the lack of conclusive evidence regarding acupuncture's clinical effectiveness and its superiority over sham interventions. (10.1186/1471-2474-5-31)
• [L5] An intention-to-treat analysis should almost always be the primary analysis in superiority trials to maintain the integrity of randomisation, while per-protocol analysis is usually better for equivalence or non-inferiority trials. (10.1302/0301-620x.95b11.32419)
• [L4] Twenty-six different criteria described by multiple classification systems have been identified for the magnetic resonance assessment of rotator cuff after repair. (10.1007/s00167-014-3486-3)
• [L5] Reliable classification leads to more confidence that patients can be treated like similar patients in a study that had success, provided the reliability study environment is proper, the statistics are appropriate, and the magnitude of the ICC is acceptable. (10.1177/0363546517743761)
• [L5] Examining the robustness of clinical trials requires an estimation of uncertainty, rather than a misconstrued, dichotomous focus on statistical significance. (10.1016/j.arthro.2023.10.010)
• [L5] The article outlines essential considerations for the analytic phase of a meta-analysis, including the selection of effect estimators, interpretation of heterogeneity, and various sub-types of meta-analytic approaches to ensure data are appropriately handled and results are reliable. (10.1007/s00167-023-07328-9)
• [L2] Research should include comprehensive conceptual models that capture complete sets of important independent variables. (10.2106/jbjs.j.00296)
• [L4] The study concludes that a two-stage approach using latent class analysis is preferable to a single-stage approach for identifying subgroups of patients with low back pain, as it yields more distinct and clinically meaningful subgroups. (10.1186/s12891-017-1411-x)
• [L5] Understanding valid statistical methods is essential for the correct design of research projects and for creating accurate predictive models to assist with evaluating surgeon performance and guiding clinical decisions. (10.1016/j.arthro.2017.01.054)
• [L4] However, further studies with long-term follow-up are needed to determine whether the grafted area will maintain structural and functional integrity over time. (10.1007/s00167-010-1042-3)
• [L3] These images use lower radiation doses and contain more radiographic information. (10.1186/s12891-019-2746-2)
• [L4] Most commonly utilized radiographic measures were consistent between sexes, across the adolescent age group, and between adolescents and young adults. (10.1016/j.jse.2011.10.026)
• [L5] A study's conclusions are only as strong as its methods. (10.1016/j.arthro.2021.06.037)
• [L1] Unsupervised machine learning methods including clustering techniques and dimensionality reduction tools carry great potential to elucidate structures and patterns in high-dimensional data and expand understanding of human physiology and disease. (10.1007/s00167-022-07233-7)
• [L4] Clinical outcomes after surgery improved compared with those before surgery and were maintained throughout the long-term follow-up period. (10.1177/0363546519889046)
• [L1] No difference in clinical outcome between the two systems was found, and the expected benefits of design modifications could not be observed in clinical outcome scores 2 years postoperatively. (10.1007/s00167-020-06027-z)
• [L3] In terms of the OF classification and the OF score, the addition of CT adds limited value compared to conventional radiographs and MRI only. (10.1186/s12891-022-06056-4)
• [L2] Overall, this study provides protective clinical parameters as well as quantitative and semi-quantitative MR-imaging parameters associated with maintaining radiographically normal knee joints in an older population over 8 years. (10.1186/s12891-024-07590-z)
• [L3] Further research is needed to determine if these improvements reach the minimal clinically important difference for long‐term postoperative outcomes. (10.1002/ksa.70174)
• [L3] There are differences between the angle of inclination findings in this study and other studies, which could be due to MRI and measurement techniques. (10.1007/s00167-017-4419-8)
• [L1] Our findings suggest that currently available devices are capable of assessing spinal posture with good accuracy in the clinical setting. (10.1186/s12891-019-2430-6)
• [L5] The paper argues that while evidence supports meniscal repair and conservative treatment, meniscectomy remains overused due to non-scientific factors like technical difficulty, cost, and patient preferences; it calls for a critical analysis of literature to reduce unnecessary resections. (10.1007/s00167-014-3471-x)
• [L3] MRI utilization by orthopaedic surgeons results in more appropriate interventions for patients with symptoms and findings most amenable to surgical intervention. (10.2106/jbjs.n.00947)
• [L1] The negative results of MRI should not prevent a diagnostic arthroscopy. (10.1016/j.arthro.2012.04.138)
• [L3] Selective magnetic resonance imaging does not provide enhanced diagnostic utility over clinical examination, particularly in children, and should be used judiciously in cases where the clinical diagnosis is uncertain and magnetic resonance imaging input will alter the treatment plan. (10.1177/03635465010290030601)
• [L4] Radiographs and MRI were not reliable, even for simply identifying the presence of a bone tunnel. (10.1007/s00167-009-0952-4)
• [L4] Results also support integrating dynamic factors and clinical outcomes in MRI-based classifications to inform treatment approaches. (10.1002/ksa.12183)
• [L3] Rather, PFP may derive from a combination of physical activity in the context of pathological kinematics. (10.1177/0363546516679139)
• [L1] Specialist radiological imaging is specific for cartilage disease in the knee but has poorer sensitivity to determine the therapeutic options in this population. (10.1007/s00167-012-1905-x)
• [L4] Specific areas must be addressed in future studies to advance the meaningful and clinical use of AI for radiograph interpretation. (10.1177/17531934241312896)
• [L4] The sTTTG shows smaller values when measured on MRI compared with CT, but its clinical significance is yet to be determined. (10.1016/j.arthro.2025.01.045)
• [L3] The study concludes that the reproducibility of radiologic measurement to inform TKA implantation may be unreliable. (10.1007/s00167-015-3855-6)
• [L5] CT offered the most accurate and precise assessment of component positioning while other methods were within acceptable limits. (10.1016/j.arth.2008.11.016)
• [L3] An initial trial of nonoperative management may be considered in young active patients with isolated SLAP tear. (10.1016/j.jse.2015.09.008)
• [L2] With careful attention to technique, the fluoroscopically simulated Dunn view can be used to measure femoral version with acceptable accuracy and obviates the need for repeat 3-dimensional imaging for patients who already have an MRI scan without version analysis. (10.1016/j.arthro.2017.01.022)
• [L5] The results of the referenced studies make sense in theory if exogenous microbes are excluded and endogenous microbes are eliminated, but propensity-score adjustments are not a substitute for an adequately powered, well-controlled, prospective, randomized study. (10.2106/jbjs.19.00457)
• [L4] Baseline radiographic severity grade was only associated with future total knee arthroplasty risk in the absence of a full-thickness defect. (10.2106/jbjs.17.01657)
• [L2] The newly proposed 6-stage MRI-based classification utilizing grade and location of the injury had substantial to near perfect agreement among and within fellowship-trained observers. (10.1177/0363546518786970)
• [L3] Non-responders had significantly increased mortality and significantly worse baseline scores despite similar demographics and revision risk. (10.1302/0301-620x.108b1.bjj-2025-0683.r1)
• [L1] The clinical findings suggest that ultrasound may be used as a conservative non-pharmaceutical and non-invasive treatment option for patients with knee osteoarthritis. (10.1186/s13018-018-0965-0)
• [L4] The analysis of dependent observations is common in the orthopaedic sports literature, but most studies do not adjust for nonindependence in these situations. (10.1177/2325967118818410)
• [L5] Propensity score matching is a popular statistical technique used to reduce bias in observational studies, but researchers must carefully consider the matching model, covariates, and potential limitations such as the inability to control for unobserved confounders. (10.1016/j.arth.2025.04.048)
• [L3] Variations of spinal growth velocity exerted more direct influence over changes in angle velocity as compared with height velocity. (10.1186/s12891-016-1221-6)
• [L1] Our meta-analysis of RCTs showed that commencement of mobilisation within one week from injury for non-surgically managed PHFs is safe and may confer short-term functional benefits compared to delayed mobilisation. (10.1186/s12891-025-08371-y)
• [L4] The Lyon brace, through its biomechanical action on vertebral modeling, is highly effective in correcting thoracic curves, particularly when the SOSORT guidelines were adopted in addition to the SRS criteria. (10.1186/s12891-015-0782-0)
• [L4] The prognosis for successful repair of non-union is surprisingly similar whatever method is used, provided good judgment is exercised in the selection of the method. (10.2106/00004623-196143020-00001)
• [L3] The study did not reveal any gender differences in these kinematic changes. (10.1007/s00167-007-0432-7)
• [L1] The L2‒L4 lumbar spine segment may substantially increase, whereas the L1‒L4 lumbar spine segment and total hip region have negligible effect. (10.1186/s12891-026-09504-7)
• [L1] There is a lack of current research to establish the psychometric properties of non-invasive 3D human posture-measuring instruments. (10.1186/1471-2474-12-93)
• [L4] The proposed method facilitates 3D-scoliosis assessment without the use of sophisticated devices. (10.1186/s12891-020-03494-w)
• [L5] No statistical differences were found in knee kinematics and retropatellar pressure distribution between rapid prototype and standard materials. (10.1155/2015/185142)
• [L5] Both types of prostheses show typical kinematics of an anterior instability, hence they were incapable of performing physiological movement. (10.1007/s00167-012-1986-6)
• [L5] The authors argue that classifications of good versus poor outcome following knee arthroplasty should not be defined using arbitrary cutoff scores, as this homogeneity impedes scientific progress, and instead propose relying on non-biased statistical model-based approaches. (10.1186/s12891-020-03583-w)
• [L3] Findings suggested that scoliotic curvatures in preoperative AIS patients can be largely represented by both imaging modalities despite the difference in body positioning. (10.1186/s12891-020-03561-2)
• [L4] Method 2 (measuring the angle from the inferior endplate of the vertebra above the degenerative disc to the inferior endplate of the vertebra below) showed the best intraobserver and interobserver reliability overall in assessing cervical sagittal rotation. (10.1186/1471-2474-15-332)
• [L3] Twenty-five (24.0%) knees that underwent nonsurgical treatment met the overall failure criteria after a mean follow-up of 76.9 months. (10.1007/s00167-023-07586-7)
• [L5] Loading characteristics are more favorable when cement is placed along the entire back of the implant contacting the subchondral bone. (10.1186/s13018-015-0268-7)
• [L3] WB status affects the in vivo kinematics following mobile-bearing UKA. (10.1007/s00167-020-05893-x)
• [L1] Limited data shows that high CONUT is also associated with a non-significant increased risk of adverse outcomes. (10.1186/s13018-024-04771-3)
• [L3] Postoperative lateral laxity greater than 0.9 mm at 90-degree flexion was associated with physiological kinematic motion, leading to fewer knee symptoms in the PROMs. (10.1016/j.jisako.2024.100357)
• [L3] Study findings underscore subject-specific complexities in ACLR biomechanics, necessitating personalized surgical planning for effective restoration of native motion and tissue mechanics. (10.1002/ksa.12413)
• [L4] Longitudinal studies are required to establish the potential role of these risk or prognostic shape phenotypes. (10.1186/s12891-020-03346-7)
• [L1] The two designs functioned equivalently at the time of early follow-up in this low-to-moderate-demand patient group. (10.2106/jbjs.j.00157)
• [L3] The study found that time to surgery was not an independent risk factor for mortality or functional outcomes when controlling for patient-specific factors. (10.1016/j.arth.2025.05.099)
• [L1] However, longer-term survival analysis still needs to be determined. (10.1302/0301-620x.100b9.bjj-2017-1403.r1)

See Also

• Conservative Treatment
• Pain Management
• Deformities
• Disc Degeneration

References

[1] Are the Biological and Biomechanical Properties of Meniscal Scaffolds Reflected in Clinical Practice? A Systematic Review of the Literature. International Journal of Molecular Sciences. 2019. DOI: 10.3390/ijms20030632

[2] The Dangers of Stopping a Trial Too Early. Journal of Bone and Joint Surgery. 2012. DOI: 10.2106/jbjs.k.01412

[3] The clinical course of low back pain: a meta-analysis comparing outcomes in randomised clinical trials (RCTs) and observational studies. BMC Musculoskeletal Disorders. 2014. DOI: 10.1186/1471-2474-15-68

[4] Fully Randomized Predictor-Outcome Pairings Using a National Database Yield Frequent Statistical Significance Without Clinical Meaning. The Journal of Arthroplasty. 2025. DOI: 10.1016/j.arth.2025.08.053

[5] Publication of Sports Medicine–Related Randomized Controlled Trials Registered in ClinicalTrials.gov. The American Journal of Sports Medicine. 2012. DOI: 10.1177/0363546512448363

[6] Three-dimensional corrective exercise therapy for idiopathic scoliosis: study protocol for a prospective non-randomized trial. BMC Musculoskeletal Disorders. 2022. DOI: 10.1186/s12891-022-05057-7

[7] Intra‐Articular Steroid Injection for Patients with Hip Osteoarthritis: A Systematic Review and Meta‐Analysis. BioMed Research International. 2020. DOI: 10.1155/2020/6320154

[8] Open reduction and fixation does not improve short-term outcome of medium-sized posterior fragments in AO type B ankle fractures: one-year results of the POSTFIX randomized controlled trial. The Bone & Joint Journal. 2025. DOI: 10.1302/0301-620x.107b4.bjj-2024-0521.r3

[9] A new short uncemented, proximally fixed anatomic femoral implant with a prominent lateral flare: design rationals and study design of an international clinical trial. BMC Musculoskeletal Disorders. 2008. DOI: 10.1186/1471-2474-9-147

[10] An intraoperative load sensor did not improve the early postoperative results of posterior‐stabilized TKA for osteoarthritis with varus deformities. Knee Surgery, Sports Traumatology, Arthroscopy. 2018. DOI: 10.1007/s00167-018-5314-7

[11] Robotic arm‐assisted patellofemoral joint arthroplasty achieves good short‐term survival and functional outcomes: A systematic review and meta‐analysis. Knee Surgery, Sports Traumatology, Arthroscopy. 2026. DOI: 10.1002/ksa.70301

[12] Willingness to participate in placebo-controlled surgical trials of the knee. The Bone & Joint Journal. 2024. DOI: 10.1302/0301-620x.106b12.bjj-2023-1266.r2

[13] What are we matching on and why? A systematic review of matched study designs in shoulder arthroplasty. Journal of Shoulder and Elbow Surgery. 2025. DOI: 10.1016/j.jse.2025.01.021

[14] Development of the Italian version of the High-Activity Arthroplasty Score (HAAS-I) following hip and knee total arthroplasty: cross-cultural adaptation, reliability, validity and sensitivity to change. Journal of Orthopaedic Surgery and Research. 2018. DOI: 10.1186/s13018-018-0782-5

[15] Chapter 1 Orthopaedic Research. 2020.

[16] A Syllabus of Laboratory Examinations in Clinical Diagnosis. Critical Evaluation of Laboratory Procedures in the Study of the Patient. Thomas Hale Ham, B.S., M.D., Editor. Cambridge, Harvard University Press, 1950. $5.00. The Journal of Bone & Joint Surgery. 1951. DOI: 10.2106/00004623-195133040-00044

[17] Displaced Femoral Neck Fractures in Workers’ Compensation Patients Aged 45-65 Years: Is It Best to Fix the Fracture or Replace the Joint?. The Journal of Arthroplasty. 2020. DOI: 10.1016/j.arth.2020.06.003

[18] Editorial: The Minimum Clinically Important Difference—The Least We Can Do. Clinical Orthopaedics & Related Research. 2017. DOI: 10.1007/s11999-017-5253-5

[19] Top 5 AJSM Papers in Epidemiology and Statistics Over the Past 50 Years. The American Journal of Sports Medicine. 2022. DOI: 10.1177/03635465221113347

[20] Defining clinically relevant outcome thresholds for pain and function after osteochondral autograft transplantation of the knee. Knee Surgery, Sports Traumatology, Arthroscopy. 2024. DOI: 10.1002/ksa.12422

[21] Patient-reported outcomes following cemented versus cementless primary total knee arthroplasty: a comparative analysis based on propensity score matching. BMC Musculoskeletal Disorders. 2022. DOI: 10.1186/s12891-022-05899-1

[22] Individualised total knee arthroplasty demonstrates 99.4% survival at 3 to 5 years. Knee Surgery, Sports Traumatology, Arthroscopy. 2026. DOI: 10.1002/ksa.70265

[23] Research Pearls: The Significance of Statistics and Perils of Pooling. Arthroscopy. 2017. DOI: 10.1016/j.arthro.2017.03.013

[24] Understanding Orthopaedic Registry Studies. The Journal of Bone and Joint Surgery. 2016. DOI: 10.2106/jbjs.n.01332

[25] The Critical Reader—Magnitude and Strength of Findings. The American Journal of Sports Medicine. 2016. DOI: 10.1177/0363546516651879

[26] WOMAC,_EQ-5D_and_Knee_Society_Score_Thresholds_for_Treatment_Success_After_Total_S0883540315005240. n.d..

[27] Development and validation of a trauma frailty scale in severely injured patients: the Nottingham Trauma Frailty Index. The Bone & Joint Journal. 2024. DOI: 10.1302/0301-620x.106b4.bjj-2023-1058.r1

[28] Multivariable regression: understanding one of medicine’s most fundamental statistical tools. Knee Surgery, Sports Traumatology, Arthroscopy. 2022. DOI: 10.1007/s00167-022-07215-9

[29] A guide to appropriately planning and conducting meta‐analyses—Part 1: indications, assumptions and understanding risk of bias. Knee Surgery, Sports Traumatology, Arthroscopy. 2022. DOI: 10.1007/s00167-022-07304-9

[30] Novel Arthroscopic Classification of Osteochondritis Dissecans of the Knee. The American Journal of Sports Medicine. 2016. DOI: 10.1177/0363546516637175

[31] Editorial Commentary: Artificial Intelligence Analysis of Biomedical, Large, Clinical Registry Data Using Machine Learning Requires Tens of Thousands of Subjects and a Focus on Substantial Clinical Benefit: Minimal Clinically Important Difference Is too Low a Bar. Arthroscopy. 2024. DOI: 10.1016/j.arthro.2023.10.035

[32] Wider translations and rotations in posterior‐stabilised mobile‐bearing total knee arthroplasty compared to fixed‐bearing both implanted with mechanical alignment: a dynamic RSA study. Knee Surgery, Sports Traumatology, Arthroscopy. 2023. DOI: 10.1007/s00167-023-07541-6

[33] Are classifications of proximal radius fractures reproducible?. BMC Musculoskeletal Disorders. 2009. DOI: 10.1186/1471-2474-10-120

[34] Performance of artificial intelligence in automated measurement of patellofemoral joint parameters: a systematic review. Journal of Orthopaedic Surgery and Research. 2025. DOI: 10.1186/s13018-025-06247-4

[35] Extrusion, meniscal signal change, loss of shape, synovitis and bone marrow oedema are reliable scoring parameters to assess MRI appearance post meniscal transplant. Knee Surgery, Sports Traumatology, Arthroscopy. 2021. DOI: 10.1007/s00167-021-06720-7

[36] Reply to Letter to Editor: Orthopaedic Surgeons Prefer to Participate in Expertise-based Randomized Trials. Clinical Orthopaedics & Related Research. 2009. DOI: 10.1007/s11999-008-0576-x

[37] Reply to Letter to the Editor on “Can Preoperative Patient-Reported Outcome Measures Predict Clinical Outcomes Following Total Knee Arthroplasty?”. The Journal of Arthroplasty. 2026. DOI: 10.1016/j.arth.2025.10.039

[38] Analysis of Publication Bias in the Literature for Distal Radius Fracture. The Journal of Hand Surgery. 2013. DOI: 10.1016/j.jhsa.2013.02.023

[39] Estimation of T scores with Hologic using NatIve vs. Caucasian data in IndiAns (ETHNICA): a single center retrospective study. BMC Musculoskeletal Disorders. 2025. DOI: 10.1186/s12891-025-08599-8

[40] Why Most Published Research Findings Are False. PLoS Medicine. 2005. DOI: 10.1371/journal.pmed.0020124

[41] Avoiding common statistical pitfalls. Journal of Hand Therapy. 2014. DOI: 10.1016/j.jht.2013.09.004

[42] Selection Bias, Orthopaedic Style. Journal of Bone and Joint Surgery. 2019. DOI: 10.2106/jbjs.19.01135

[43] Is acupuncture a useful adjunct to physiotherapy for older adults with knee pain?: The "Acupuncture, Physiotherapy and Exercise" (APEX) study [ISRCTN88597683]. BMC Musculoskeletal Disorders. 2004. DOI: 10.1186/1471-2474-5-31

[44] Intention-to-treat analyses for interventional studies. The Bone & Joint Journal. 2013. DOI: 10.1302/0301-620x.95b11.32419

[45] Magnetic resonance imaging criteria for the assessment of the rotator cuff after repair: a systematic review. Knee Surgery, Sports Traumatology, Arthroscopy. 2015. DOI: 10.1007/s00167-014-3486-3

[46] The Critical Reader—Reliability. The American Journal of Sports Medicine. 2017. DOI: 10.1177/0363546517743761

[47] Editorial Comme ntary: The Statistical Fragility Index of Medical Trials Is Low By Design: Critical Evaluation of Confidence Intervals Is Required. Arthroscopy. 2024. DOI: 10.1016/j.arthro.2023.10.010

[48] Chapter 66 Statistics: Practical Applications for Orthopaedics. 2019.

[49] Randomized trials in orthopaedics: why, how, and when?.. The Journal of Bone and Joint Surgery. American Volume. 1989.

[50] A guide to appropriately planning and conducting meta‐analyses: part 2—effect size estimation, heterogeneity and analytic approaches. Knee Surgery, Sports Traumatology, Arthroscopy. 2023. DOI: 10.1007/s00167-023-07328-9

[51] Development_of_a_Modern_Knee_Society_Radiographic_Evaluation_System_and_Methodol_S0883540315004507. n.d..

[52] Evidence Summary. The Journal of Bone & Joint Surgery. 2011. DOI: 10.2106/jbjs.j.00296

[53] Identifying subgroups of patients using latent class analysis: should we use a single-stage or a two-stage approach? A methodological study using a cohort of patients with low back pain. BMC Musculoskeletal Disorders. 2017. DOI: 10.1186/s12891-017-1411-x

[54] Research Pearls: The Significance of Statistics and Perils of Pooling. Part 2: Predictive Modeling. Arthroscopy. 2017. DOI: 10.1016/j.arthro.2017.01.054

[55] Mid‐term results of Autologous Matrix‐Induced Chondrogenesis for treatment of focal cartilage defects in the knee. Knee Surgery, Sports Traumatology, Arthroscopy. 2010. DOI: 10.1007/s00167-010-1042-3

[56] The migration percentage measured on EOS® standing full-leg radiographs: equivalent and advantageous in ambulant children with cerebral palsy. BMC Musculoskeletal Disorders. 2019. DOI: 10.1186/s12891-019-2746-2

[57] Elbow radiographic anatomy: measurement techniques and normative data. Journal of Shoulder and Elbow Surgery. 2012. DOI: 10.1016/j.jse.2011.10.026

[58] Best Practice Guidelines for Propensity Score Methods in Medical Research: Consideration on Theory, Implementation, and Reporting. A Review. Arthroscopy. 2021. DOI: 10.1016/j.arthro.2021.06.037

[59] Unsupervised machine learning methods and emerging applications in healthcare. Knee Surgery, Sports Traumatology, Arthroscopy. 2022. DOI: 10.1007/s00167-022-07233-7

[60] Absolute Meniscal Extrusion After Lateral Meniscal Allograft Transplantation Does Not Progress During Long-term Follow-up: Average of 10.3 Years’ Follow-up Longitudinal Magnetic Resonance Imaging Study. The American Journal of Sports Medicine. 2019. DOI: 10.1177/0363546519889046

[61] Mid‐term results show no significant difference in postoperative clinical outcome, pain and range of motion between a well‐established total knee arthroplasty design and its successor: a prospective, randomized, controlled trial. Knee Surgery, Sports Traumatology, Arthroscopy. 2020. DOI: 10.1007/s00167-020-06027-z

[62] OF spine classification of osteoporotic thoracolumbar vertebral body fractures by MRI and conventional radiographs only leads to high inter-observer agreement rates-an additional CT adds limited information for the of classification and the OF score. BMC Musculoskeletal Disorders. 2022. DOI: 10.1186/s12891-022-06056-4

[63] Clinical and imaging findings associated with preservation of knee joint health over 8 years in individuals aged 65 and over: data from the OAI. BMC Musculoskeletal Disorders. 2024. DOI: 10.1186/s12891-024-07590-z

[64] Sensor‐guided robot‐assisted TKA more precisely achieves targeted functional knee phenotypes than conventional robotic‐assisted technique. Knee Surgery, Sports Traumatology, Arthroscopy. 2025. DOI: 10.1002/ksa.70174

[65] The angle of inclination of the native ACL in the coronal and sagittal planes. Knee Surgery, Sports Traumatology, Arthroscopy. 2017. DOI: 10.1007/s00167-017-4419-8

[66] The role of wearables in spinal posture analysis: a systematic review. BMC Musculoskeletal Disorders. 2019. DOI: 10.1186/s12891-019-2430-6

[67] Focusing on results after meniscus surgery. Knee Surgery, Sports Traumatology, Arthroscopy. 2014. DOI: 10.1007/s00167-014-3471-x

[68] MRI for the Evaluation of Knee Pain. The Journal of Bone and Joint Surgery-American Volume. 2015. DOI: 10.2106/jbjs.n.00947

[69] Accuracy of Magnetic Resonance Imaging in Grading Knee Chondral Defects. Arthroscopy. 2012. DOI: 10.1016/j.arthro.2012.04.138

[70] Diagnostic Performance of Clinical Examination and Selective Magnetic Resonance Imaging in the Evaluation of Intraarticular Knee Disorders in Children and Adolescents. The American Journal of Sports Medicine. 2001. DOI: 10.1177/03635465010290030601

[71] Comparison of plain radiography, computed tomography, and magnetic resonance imaging in the evaluation of bone tunnel widening after anterior cruciate ligament reconstruction. Knee Surgery, Sports Traumatology, Arthroscopy. 2009. DOI: 10.1007/s00167-009-0952-4

[72] Meniscal extrusion consensus statement: A collaborative survey within the Meniscus International Network (MenIN) Study Group. Knee Surgery, Sports Traumatology, Arthroscopy. 2024. DOI: 10.1002/ksa.12183

[73] Patellofemoral Kinematics and Tibial Tuberosity–Trochlear Groove Distances in Female Adolescents With Patellofemoral Pain. The American Journal of Sports Medicine. 2016. DOI: 10.1177/0363546516679139

[74] Accuracy of magnetic resonance imaging, magnetic resonance arthrography and computed tomography for the detection of chondral lesions of the knee. Knee Surgery, Sports Traumatology, Arthroscopy. 2012. DOI: 10.1007/s00167-012-1905-x

[75] Artificial intelligence and machine learning capabilities in the detection of acute scaphoid fracture: a critical review. Journal of Hand Surgery (European Volume). 2025. DOI: 10.1177/17531934241312896

[76] The Sagittal Tibial Tubercle Trochlear‐Groove Distance Is Not Equivalent on Magnetic Resonance Imaging and Computed Tomography Measurements in Patients With Patellar Instability. Arthroscopy. 2025. DOI: 10.1016/j.arthro.2025.01.045

[77] Radiographic measurement of the posterior femoral offset is not precise. Knee Surgery, Sports Traumatology, Arthroscopy. 2015. DOI: 10.1007/s00167-015-3855-6

[78] Six Sigma Analysis of Acetabular Component Positioning In MIS Total Hip Arthroplasty. The Journal of Arthroplasty. 2009. DOI: 10.1016/j.arth.2008.11.016

[79] Predictive factors associated with failure of nonoperative treatment of superior labrum anterior-posterior tears. Journal of Shoulder and Elbow Surgery. 2016. DOI: 10.1016/j.jse.2015.09.008

[80] Comparison of Intraoperative Fluoroscopic Dunn View With Magnetic Resonance Imaging to Determine Femoral Version. Arthroscopy. 2017. DOI: 10.1016/j.arthro.2017.01.022

[81] Are We All Still Listerians or Not?. Journal of Bone and Joint Surgery. 2019. DOI: 10.2106/jbjs.19.00457

[82] Full-Thickness Cartilage Defects Are Important Independent Predictive Factors for Progression to Total Knee Arthroplasty in Older Adults with Minimal to Moderate Osteoarthritis. Journal of Bone and Joint Surgery. 2019. DOI: 10.2106/jbjs.17.01657

[83] Interobserver and Intraobserver Reliability of an MRI-Based Classification System for Injuries to the Ulnar Collateral Ligament. The American Journal of Sports Medicine. 2018. DOI: 10.1177/0363546518786970

[84] Patient-reported outcomes may be ‘missing not at random’ in hip and knee arthroplasty. The Bone & Joint Journal. 2026. DOI: 10.1302/0301-620x.108b1.bjj-2025-0683.r1

[88] Effect of low-intensity long-duration ultrasound on the symptomatic relief of knee osteoarthritis: a randomized, placebo-controlled double-blind study. Journal of Orthopaedic Surgery and Research. 2018. DOI: 10.1186/s13018-018-0965-0

[89] Statistical Analysis of Dependent Observations in the Orthopaedic Sports Literature. Orthopaedic Journal of Sports Medicine. 2019. DOI: 10.1177/2325967118818410

[90] Propensity Score Matching in the Arthroplasty Literature. The Journal of Arthroplasty. 2025. DOI: 10.1016/j.arth.2025.04.048

[91] Spinal growth velocity versus height velocity in predicting curve progression in peri-pubertal girls with idiopathic scoliosis. BMC Musculoskeletal Disorders. 2016. DOI: 10.1186/s12891-016-1221-6

[92] Early versus delayed mobilisation for non-surgically treated proximal humerus fractures: a systematic review and meta-analysis of randomised trials. BMC Musculoskeletal Disorders. 2025. DOI: 10.1186/s12891-025-08371-y

[93] Lyon bracing in adolescent females with thoracic idiopathic scoliosis: a prospective study based on SRS and SOSORT criteria. BMC Musculoskeletal Disorders. 2015. DOI: 10.1186/s12891-015-0782-0

[94] Observations on Non-Union of the Shafts of the Long Bones, with a Statistical Analysis of 842 Patients. The Journal of Bone & Joint Surgery. 1961. DOI: 10.2106/00004623-196143020-00001

[95] Fatigue alters lower extremity kinematics during a single‐leg stop‐jump task. Knee Surgery, Sports Traumatology, Arthroscopy. 2007. DOI: 10.1007/s00167-007-0432-7

[96] Whole-body vibration training and bone mineral density in older adults: an updated systematic review and meta-analysis. BMC Musculoskeletal Disorders. 2026. DOI: 10.1186/s12891-026-09504-7

[97] The quality of evidence of psychometric properties of three-dimensional spinal posture-measuring instruments. BMC Musculoskeletal Disorders. 2011. DOI: 10.1186/1471-2474-12-93

[98] The method for measurement of the three-dimensional scoliosis angle from standard radiographs. BMC Musculoskeletal Disorders. 2020. DOI: 10.1186/s12891-020-03494-w

[99] Rapid Prototyping forIn VitroKnee Rig Investigations of Prosthetized Knee Biomechanics: Comparison with Cobalt-Chromium Alloy Implant Material. BioMed Research International. 2015. DOI: 10.1155/2015/185142

[100] Femoro‐tibial kinematics after TKA in fixed‐ and mobile‐bearing knees in the sagittal plane. Knee Surgery, Sports Traumatology, Arthroscopy. 2012. DOI: 10.1007/s00167-012-1986-6

[101] Classifications of good versus poor outcome following knee arthroplasty should not be defined using arbitrary criteria. BMC Musculoskeletal Disorders. 2020. DOI: 10.1186/s12891-020-03583-w

[102] Accuracy on the preoperative assessment of patients with adolescent idiopathic scoliosis using biplanar low-dose stereoradiography: a comparison with computed tomography. BMC Musculoskeletal Disorders. 2020. DOI: 10.1186/s12891-020-03561-2

[103] Intraobserver and interobserver reliability of measures of cervical sagittal rotation. BMC Musculoskeletal Disorders. 2014. DOI: 10.1186/1471-2474-15-332

[104] Older age and the presence of intrameniscal signs are risk factors for nonsurgical treatment failure of symptomatic intact discoid lateral meniscus. Knee Surgery, Sports Traumatology, Arthroscopy. 2023. DOI: 10.1007/s00167-023-07586-7

[106] An in vitro study comparing limited to full cementation of polyethylene glenoid components. Journal of Orthopaedic Surgery and Research. 2015. DOI: 10.1186/s13018-015-0268-7

[107] Weight‐bearing status affects in vivo kinematics following mobile‐bearing unicompartmental knee arthroplasty. Knee Surgery, Sports Traumatology, Arthroscopy. 2020. DOI: 10.1007/s00167-020-05893-x

[108] What_Is_the_Level_of_Evidence_Substantiating_the_Medicare_Local_Coverage_Determi_S0883540314007475. n.d..

[109] Do nutritional assessment tools (PNI, CONUT, GNRI) predict adverse events after spinal surgeries? A systematic review and meta-analysis. Journal of Orthopaedic Surgery and Research. 2024. DOI: 10.1186/s13018-024-04771-3

[110] Medial center of rotation and 90-degree lateral laxity improve patient-reported outcomes in posterior cruciate retaining total knee arthroplasty. Journal of ISAKOS. 2024. DOI: 10.1016/j.jisako.2024.100357

[111] Surgical parameters influence paediatric knee kinematics and cartilage stresses in anterior cruciate ligament reconstruction: Navigating subject‐specific variability using neuromusculoskeletal‐finite element modelling analysis. Knee Surgery, Sports Traumatology, Arthroscopy. 2024. DOI: 10.1002/ksa.12413

[112] Is intrinsic lumbar spine shape associated with lumbar disc degeneration? An exploratory study. BMC Musculoskeletal Disorders. 2020. DOI: 10.1186/s12891-020-03346-7

[113] Mobile and Fixed-Bearing (All-Polyethylene Tibial Component) Total Knee Arthroplasty Designs. Journal of Bone and Joint Surgery. 2010. DOI: 10.2106/jbjs.j.00157

[114] Delayed Primary Hip Arthroplasty for Geriatric Low-Energy Femoral Neck Fracture Is Not Associated With Worse Outcomes: A Study of the Arthroplasty for Hip Fracture Consortium. The Journal of Arthroplasty. 2025. DOI: 10.1016/j.arth.2025.05.099

[115] Primary stability of a short bone-conserving femoral stem. The Bone & Joint Journal. 2018. DOI: 10.1302/0301-620x.100b9.bjj-2017-1403.r1