Dataset Information

Can minimal clinically important differences in patient reported outcome measures be predicted by machine learning in patients with total knee or hip arthroplasty? A systematic review.

ABSTRACT:

Objectives

To systematically review studies using machine learning (ML) algorithms to predict whether patients undergoing total knee or total hip arthroplasty achieve an improvement as high or higher than the minimal clinically important differences (MCID) in patient reported outcome measures (PROMs) (classification problem).

Methods

Studies were eligible to be included in the review if they collected PROMs both pre- and postintervention, reported the method of MCID calculation and applied ML. ML was defined as a family of models which automatically learn from data when selecting features, identifying nonlinear relations or interactions. Predictive performance must have been assessed using common metrics. Studies were searched on MEDLINE, PubMed Central, Web of Science Core Collection, Google Scholar and Cochrane Library. Study selection and risk of bias assessment (ROB) was conducted by two independent researchers.

Results

517 studies were eligible for title and abstract screening. After screening title and abstract, 18 studies qualified for full-text screening. Finally, six studies were included. The most commonly applied ML algorithms were random forest and gradient boosting. Overall, eleven different ML algorithms have been applied in all papers. All studies reported at least fair predictive performance, with two reporting excellent performance. Sample size varied widely across studies, with 587 to 34,110 individuals observed. PROMs also varied widely across studies, with sixteen applied to TKA and six applied to THA. There was no single PROM utilized commonly in all studies. All studies calculated MCIDs for PROMs based on anchor-based or distribution-based methods or referred to literature which did so. Five studies reported variable importance for their models. Two studies were at high risk of bias.

Discussion

No ML model was identified to perform best at the problem stated, nor can any PROM said to be best predictable. Reporting standards must be improved to reduce risk of bias and improve comparability to other studies.

SUBMITTER: Langenberger B

PROVIDER: S-EPMC8772225 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:IntroductionThe purpose of this study is to evaluate the role of major psychiatric illness on patient outcomes after total joint arthroplasty.MethodsPatients with a diagnosis of a major psychiatric disorder undergoing total joint arthroplasty were retrospectively matched one-to-one with a cohort without such a diagnosis. Major psychiatric disorder in the registry was identified by diagnosis of anxiety, mood, or a psychotic disorder. Primary outcome of interest included perioperative Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). Secondary outcomes included EuroQol-5D, adverse events, length of stay, 30-day readmission, and discharge destination.ResultsTotal number of patients were 1828. The total hip arthroplasty (37.80 ± 17.91, p = 0.023) and the total knee arthroplasty psychiatric group (43.38 ± 18.41, p = 0.050) had significantly lower pre-operative WOMAC scores. At 3 months, the total hip arthroplasty (76.74 ± 16.94, p = 0.036) and total knee arthroplasty psychiatric group (71.09 ± 18.64, p < 0.01) again had significantly lower 3-month post-operative WOMAC score compared to the control groups. However, outcomes at 1 year were difficult to interpret, as patients with major psychiatric conditions had an extremely high loss to follow-up. Compared to the control groups, the total hip arthroplasty and total knee arthroplasty psychiatric group had an increased length of stay by 1.43 days (p < 0.01) and 0.77 days, respectively (p = 0.05). Similarly, the psychiatric groups were discharged directly home less often (total hip arthroplasty 86.9%, p = 0.024 and total knee arthroplasty 87.6%, p = 0.022) than the control groups.ConclusionPatients with the diagnosis of a major psychiatric illness have an increased length of stay and are more likely to require a rehabilitation facility, compared to the control groups. Arguably, of utmost importance, there is a very high rate of loss to follow-up within the psychiatric groups. As such, we recommend these patients should be treated for their diagnosis prior to total joint arthroplasty. Furthermore, importance of clinical follow-up should be emphasized carefully.

Project description:BackgroundThe number of hip, knee and shoulder arthroplasties continues to rise worldwide. The Organization for Economic Cooperation and Development has launched an initiative (called PaRIS Initiative) for the systematic collection of Patient Reported Outcome Measures (PROMs) in patients undergoing elective hip and knee arthroplasty. The Rizzoli Orthopedic Institute (IOR) was selected as a pilot center for the launch of the Initiative in Italy given that IOR hosts the Registry of Orthopedic Prosthetic Implants (RIPO), a region-wide registry which collects joint implant data from all the hospitals in the Emilia-Romagna Region. In this specific geographic area information related to PROMs after joint replacement is unknown. This paper describes the protocol of a study (PaRIS-IOR) that aims to implement the collection of a set of PROMs within an existing implant registry in Italy. The study will also investigate the temporal trend of PROMs in relation to the type of prosthesis and the type of surgical intervention.MethodsThe PaRIS-IOR study is a prospective, single site, cohort study that consists of the administration of PROMs questionnaires to patients on the list for elective arthroplasty. The questionnaires will be administered to the study population within 30 days before surgery, and then at 6 and 12 months following surgery. The study population will consist of consecutive adult patients undergoing either hip, knee or shoulder arthroplasty. The collected data will be linked with those routinely collected by the RIPO in order to assess the temporal trend of PROMs in relation to the type of prosthesis and the type of surgical intervention.DiscussionThe PaRIS-IOR study could have important implications in targeting the factors influencing functional outcomes and quality of life reported by patients after hip, knee and shoulder arthroplasty, and will also represent the first systematic collection of PROMs related to arthroplasty in Italy.Trial registrationProtocol version (1.0) and trial registration data are available on the platform www.clinicaltrial.gov with the identifier NCT03790267 , first posted on December 31, 2018.

Project description:Computer assisted surgery was pioneered in early 1990s. The first computer assisted surgery (CAS) total knee replacement with an imageless system was carried out in 1997. In the past 25 years, CAS has progressed from experimental in vitro studies to established in vivo surgical procedures.A comprehensive body of evidence establishing the advantages of computer assisted surgery in knee and hip arthroplasty is available. Established benefits have been demonstrated including its role as an excellent research tool. Its advantages include dynamic pre-operative and per-operative assessment, increased accuracy in correction of deformities, kinematics and mechanical axis, a better alignment of components, better survival rates of prostheses and a better functional outcome. Adoption of computer navigation in the hip arthroplasty is still at an early stage compared to knee arthroplasty, though the results are well documented. Evidence suggests improved accuracy in acetabular orientation, positioning, hip offset and leg length correction.Among the orthopaedic surgeons, navigated knee arthroplasty is gaining popularity though slowly. The uptake rates vary from country to country. The Australian joint registry data shows increased navigated knee arthroplasty from 2.4% in 2003 to 28.6% in 2015 and decreased revision rates with navigated knee arthroplasty in comparison with traditional instrumented knee arthroplasty in patient cohort under the age of 55 years.Any new technology has a learning curve and with practice the navigation assisted knee and hip arthroplasty becomes easy. We have actively followed the evidence of CAS in orthopaedics and have successfully adopted it in our routine practice over the last decades. Despite the cautious inertia of orthopaedic surgeons to embrace CAS more readily; we are certain that computer technology has a pivotal role in lower limb arthroplasty. It will evolve to become a standard practice in the future in various forms like navigation or robotics.