Tools for High-Value Orthopaedic Care Delivery
Elizabeth Duckworth, MD, MBA
Eugenia Lin, MD
Olivia Manickas-Hill, BA
Prakash Jayakumar, MD, PhD
None of the following authors or any immediate family member has received anything of value from or has stock or stock options held in a commercial company or institution related directly or indirectly to the subject of this chapter: Dr. Duckworth, Dr. Lin, Olivia Manickas-Hill, and Dr. Jayakumar.
INTRODUCTION
The shift toward value-based health care in the United States, since the enactment of the Patient Protection and Affordable Care Act (ACA) in 2010, has sparked a demand for tools and technologies to support stakeholders engaged in realizing this goal. Government agencies, regulatory bodies, professional groups, commercial insurers, pharmaceutical and biomedical device companies, and health care provider organizations are increasingly recognizing the need for such tools to better prepare them for delivering a range of value-oriented functions. This chapter defines concepts and unmet needs that provide the inspiration for tools and technologies enabling value-based health care delivery. A description of current state-of-the-art and future patient-facing, surgeon-facing, and systems-level tools contributing to an essential toolkit for high-value orthopaedic practices is also provided.
OVERVIEW OF HIGH-VALUE MODELS OF CARE DELIVERY
In orthopaedics, the largest transformation toward value-based reimbursement to date has been the Centers for Medicare & Medicaid Services (CMS) implementation of the Comprehensive Care for Joint Replacement (CJR) model in April 2016. Mandating hospitals across the country to accept bundled payment for a surgical episode of care around total hip and total knee arthroplasty, the CJR model places a greater emphasis on accountability for quality and costs of care by promoting coordination between hospitals, physicians, and postacute care providers. The CJR model and other value-oriented initiatives signal the requirement for tools and technologies, such as robust electronic health records and outcome measurement platforms, to deliver optimal care while also offering a lens on performance and benchmarking fit for value.1 Accountability for quality and costs of care for Medicare patient populations has also been realized in the engagement of groups of hospitals, practices, and clinicians within accountable care organizations.2
Value-based practice and payment reform in orthopaedics is now expanding beyond surgical procedures toward management of the condition over a full cycle of care, the original vision in the seminal work, Redefining Health Care.3 With this in mind, tools for high-value care can be key drivers for the development of alternative condition-based bundled payment models across orthopaedic care (Figure 1).
Value-based practice and payment reform in orthopaedics is now expanding beyond surgical procedures toward management of the condition over a full cycle of care, the original vision in the seminal work, Redefining Health Care.3 With this in mind, tools for high-value care can be key drivers for the development of alternative condition-based bundled payment models across orthopaedic care (Figure 1).
CONCEPTS AND UNMET NEEDS FOR DEVELOPING HIGH-VALUE TOOLS
Value-Based Health Care
Value-based health care is defined as care that achieves health outcomes benefiting patients relative to the costs of care.4 Importantly, this definition relies on the direct measurement of outcomes that matter to the individual over a full cycle of care. These outcomes should include a range of quality metrics from patients’ perspectives of their physical, emotional, and social health and well-being to clinically
effective outcomes including those related to hospitalization, complications, rehabilitation, and recurrences.5 The denominator includes the total cost of care for managing a patient’s condition and should incorporate direct and indirect costs spanning the gamut of inpatient, outpatient, rehabilitation, drugs and devices, physician services, equipment, and facility costs, as well societal costs through loss of productivity.5 Considering the costs of a full cycle of care may also support increased spending on higher value preventive services while disincentivizing the shift in costs from one type of service or provider of services to another.4
effective outcomes including those related to hospitalization, complications, rehabilitation, and recurrences.5 The denominator includes the total cost of care for managing a patient’s condition and should incorporate direct and indirect costs spanning the gamut of inpatient, outpatient, rehabilitation, drugs and devices, physician services, equipment, and facility costs, as well societal costs through loss of productivity.5 Considering the costs of a full cycle of care may also support increased spending on higher value preventive services while disincentivizing the shift in costs from one type of service or provider of services to another.4
FIGURE 1 Tools for high-value orthopaedic care delivery.
|
Patient-Centered Care
Patient-centered care is defined as care that respects and responds to individual patient preferences, values, and needs, while ensuring patients remain at the center of their care and clinical decision making.6 This concept includes the effect of care on patient-specific factors (physical, psychological, and social health and well-being) alongside the individual’s experience of care.7 Communication, trust, and empathy form some of the cornerstones of patient-centered care and enable patients to more effectively engage in their health care ecosystem. Shared decision making (SDM) is the expert communication of clinical information including management options to help clinicians and patients arrive at informed treatment decisions aligned with the patient’s preferences, values, and needs. This concept combines several patient-centered elements and is increasingly being adopted in orthopaedics.8 Patient-centered care has been embraced by the American Academy of Orthopaedic Surgeons via its appropriate use criteria, which utilize validated tools to extrapolate the appropriateness of an intervention within different patient and treatment combinations.9
Integrated Care
Integrated care is defined as care that is coordinated across professionals, facilities and support systems; continuous over time and between visits; tailored to patients’ needs and preferences; and based on shared responsibility between patients and caregivers while systematically measuring outcomes.10 In this regard, there has been growing interest in comprehensive, team-based, condition-focused approaches to orthopaedic care in the form of integrated practice units (IPUs).11 IPUs are structurally and functionally organized around conditions (rather than specific providers or procedures) over a full care cycle, involving a range of treatment strategies.11 The full range of treatments (including surgical and nonsurgical care) are delivered by a dedicated multidisciplinary team working within a common organizational unit, accountable for outcomes and costs of care under a bundled payment arrangement.12 Relatively few IPUs exist in current clinical practice; however, several institutions have adopted integrated care pathways (ICPs) that offer a specific, time-dependent regimen used to standardize care during a course of treatment.11 Vertically integrated institutions such as Geisinger Health and Kaiser Permanente have embraced ICPs for joint replacement surgery and seen reductions in length of hospital stay, complications, redundancy in ancillary services, and waste through enhanced interdisciplinary coordination.12,13 ICPs
have also been applied extensively in geriatric and nongeriatric fracture management and resulted in decreased complications, length of stay, time to surgery, and costs of care.14 Integrated care may also be considered comprehensive care by design because it aims to provide for the holistic needs of patients; one common example is the combined medical, surgical, and rehabilitation care provided by orthogeriatric teams in managing geriatric trauma patients.15 The benefits of a more complete integrated, biopsychosocial approach is increasingly being recognized in orthopaedics, especially in relation to the comprehensive management of osteoarthritis.16 The concepts of value-based health care, patient-centered care, and integrated care demand the development of practical, safe, and effective tools and technologies for delivering high-value orthopaedic care.
have also been applied extensively in geriatric and nongeriatric fracture management and resulted in decreased complications, length of stay, time to surgery, and costs of care.14 Integrated care may also be considered comprehensive care by design because it aims to provide for the holistic needs of patients; one common example is the combined medical, surgical, and rehabilitation care provided by orthogeriatric teams in managing geriatric trauma patients.15 The benefits of a more complete integrated, biopsychosocial approach is increasingly being recognized in orthopaedics, especially in relation to the comprehensive management of osteoarthritis.16 The concepts of value-based health care, patient-centered care, and integrated care demand the development of practical, safe, and effective tools and technologies for delivering high-value orthopaedic care.
PATIENT-FACING TOOLS DESIGNED FOR HIGH-VALUE ORTHOPAEDIC CARE
Tools enabling the capture of patient-generated health data are integral to valuebased health care. Patient-generated health data is defined as health-related data created, recorded, and gathered from patients (or family members or other caregivers) to help establish a patient’s health status.17 This type of information derived from patients is distinct from objectively reported data by clinicians or clinical systems.
Patient-Reported Outcome Measures
Patient-reported outcome measures (PROMs) are validated measures of physical, psychological, and social health and well-being reported by the patient that have revolutionized orthopaedic clinical research and are now being increasingly used in clinical practices across orthopaedic subspecialties.18 PROMs enable quantification of a patient’s perceptions of their health and responses to medical interventions with respect to function, symptoms, and quality of life.19 The introduction of national registries incorporating PROMs has had a meaningful effect in several countries through the evaluation of patient outcomes for different surgical techniques, analysis of positive predictors of these outcomes, and comparison across orthopaedic and non-orthopaedic conditions.20,21 The American Board of Orthopaedic Surgery collects PROMs as part of the Part II Board Certification process and the CMS incentivizes PROMs collection through quality reporting and bonus payment initiatives.22,23 Effective PROMs should demonstrate reliability (how well the tool repeatedly assesses the same item), validity (whether the tool measures the content it intends to measure), and responsiveness (the tool’s sensitivity to change over time) while also being user-friendly and easy to interpret (the degree to which qualitative and clinical meaning can be assigned to a PROMs quantitative scores or change in scores).24,25 Measuring PROMs before and after a treatment intervention provides an opportunity to objectively measure the effect of an intervention on the patient’s health from their perspective. This effect has traditionally been quantified using fixed-scale measures developed from classic test theory. In recent decades, there has been increased use of computer adaptive tests (CATs), developed using item response theory, which enables follow-up questions to be administered based on the patient’s response to a prior question.26 Although
fixed-scale PROMs require patients to answer most, if not all, questions to arrive at a valid score, CATs generally involve fewer but more tailored questions, and therefore result in more precise and efficient capture of patient outcomes. The most commonly used CATs are the Patient Reported Outcome Measurement Information System (PROMIS) measures developed by the National Institutes of Health. Development of PROMIS was prompted by the need for a valid, reliable, and generalizable set of measures that could be applied across clinical conditions, providing comprehensive coverage of different levels of a relevant health domain, while maintaining efficiency and reducing burden on responders.18,27 The PROMIS instruments score health domains using a common metric, normalized to the US general population. Over a 5-year period from 2014 to 2018, the volume of orthopaedic studies leveraging PROMIS increased sixfold, with most studies examining the domains of physical function, pain interference, and depression.28
fixed-scale PROMs require patients to answer most, if not all, questions to arrive at a valid score, CATs generally involve fewer but more tailored questions, and therefore result in more precise and efficient capture of patient outcomes. The most commonly used CATs are the Patient Reported Outcome Measurement Information System (PROMIS) measures developed by the National Institutes of Health. Development of PROMIS was prompted by the need for a valid, reliable, and generalizable set of measures that could be applied across clinical conditions, providing comprehensive coverage of different levels of a relevant health domain, while maintaining efficiency and reducing burden on responders.18,27 The PROMIS instruments score health domains using a common metric, normalized to the US general population. Over a 5-year period from 2014 to 2018, the volume of orthopaedic studies leveraging PROMIS increased sixfold, with most studies examining the domains of physical function, pain interference, and depression.28
PROMs provide a range of functions from tracking, screening, and segmentation of patient phenotypes, to enabling decision support and SDM.29 For instance, PROMs have been used in anterior cruciate ligament reconstruction and total joint arthroplasty (TJA) to aid surgeons in determining whether surgical intervention will benefit the patient by comparing a patient’s individual recovery to an expected recovery curve.30,31 Some institutions have committed to gathering PROMs for all patients across the spectrum of orthopaedic care, with the goal of longitudinally tracking outcomes of their population.26 Broadly, within orthopaedics, PROMs fall under the categories of health, health-related quality of life, and quality of life. Under health, PROMs range from disease-specific, region-specific, or conditionspecific PROMs, as well as psychosocial PROMs.32,33 The American Academy of Orthopaedic Surgeons has compiled a list of approved, open-access PROMs from a 2015 Quality Outcomes Data Work Group to evaluate PROMS for general health and condition or PROMs specific to an anatomic region.34
Patient-Reported Experience Measures
Patient-reported experience measures (PREMs) broadly capture patient perceptions of their experience with health care.35 PREMs range from measures of patient satisfaction with various structural and functional aspects of health care such as waiting times, access to facilities, and ability to navigate services, to (perhaps more importantly) the quality of communication and interpersonal interactions with clinicians and clinical teams.36 PREMs capturing satisfaction with clinician-patient communications and trust in providers, alongside confidence with the level of information received and involvement in care, are shown to have a positive association with PROMs in patients undergoing hip and knee arthroplasty.37 In the United States, the Consumer Assessment of Healthcare Providers and Systems (CAHPS) provide a suite of PREMs developed by the Agency for Healthcare Research and Quality, aiming to reflect key areas of overall patient experience related to structural, functional and interpersonal aspects of care. Since 2013, part of the value-based payment for hospitals initiated by CMS reflects the results of the hospital version of the Consumer Assessment of Healthcare Providers and Systems (known as H-CAPS).7
Patient Activation Measures
Patient activation measures (PAMs) are tools that measure an individual’s understanding, competence, and willingness to participate in care decisions and processes; rather, the knowledge, skills, and confidence a person has in engaging with their health and health care ecosystem.6 Active engagement of patients in their care has been demonstrated to improve health outcomes, patient experience, and lower health care costs.38 PAMs enable providers to understand both an individual patient’s level of activation and, more broadly, the levels of activation among various segments of their population. In particular, the Patient Activation Measure-13 (PAM-13) and the shorter PAM-10 have been used in orthopaedic care.39
Patient Decision Aids
SDM empowers patients to become active participants in their health and care by enabling both physician and patient to contribute to medical decision making. This approach depends on expert communication and education delivered by physicians that encourages patients to comfortably disclose their preferences, needs, and values to make informed treatment decisions.8,40 Studies have shown SDM leads to better patient satisfaction, improved decision quality, more appropriate use of health care resources by patients, and better outcomes.41,42,43 Based on a Cochrane review, SDM does not have a significant effect on the duration of the clinician-patient encounter, nor does it place additional burden on an already busy clinic schedule, adding only 2.55 minutes per encounter. Studies also suggest that these additional few minutes need not involve the physician but can be managed by midlevel health professionals.44
Patient decision aids (PDAs) are tools designed to facilitate SDM by helping patients understand relevant evidence-based information, empowering patients to better understand potential benefits and harms, and to aid communication between patients and clinicians.45 Importantly, PDAs are distinct from patient education materials (which are also useful tools) in that they more actively direct patients toward making an informed choice among multiple treatment options. Some PDAs also help align treatment options with patient preferences rather than providing information about specific treatments or treatment plans after they have already been set in place.46 In orthopaedic care, PDAs have been studied most frequently in patients with persistently painful preference-sensitive conditions (ie, where multiple valid treatment options exist) such as degenerative disease of the spine and osteoarthritis of the hip and knee.47 PDAs can take multiple forms, including written booklets, videos, and interactive online tools and can be provided to patients before, during, or after an initial encounter with an orthopaedic practitioner. PDAs may also be effective in empowering patients to make informed decisions at a given point in time as well as providing ongoing guidance along different phases of a care pathway. Evidence suggests that orthopaedic patients with hip osteoarthritis and degenerative disease of the spine recall only 38% and 45% of verbal information respectively, after outpatient clinic visits with a provider, and as little as 18% of information 6 weeks after surgery.48,49 PDAs could help address this knowledge and retention gap by providing patients with supportive resources on demand.
SDM-Related Outcome Metrics
SDM-related outcome measures (SROMs) can be used as part of an orthopaedic service’s SDM initiative alongside implementation of PDAs. Somewhat similar to PROMs, these tools provide a measure of effect from the patient’s perspective in relation to various elements of the decision-making process (eg, preparation for decisionmaking [Preparation for Decision-Making Scale], Decision Quality and level of SDM [Decision Quality Index, CollaboRATE survey], decisional conflict [Decision Conflict Scale, SURE 4-item screener], decisional regret [Decision Regret Scale], and decision support [Decision Support Analysis Tool, DSAT-10]), as well as numerical rating scales for various aspects of satisfaction with the consultation such as clinician-patient interaction.50 Further validated tools (eg, OPTION) have also been developed to independently observe, measure, and score the extent and quality of SDM delivered by clinicians. Tools and checklists for the development and application of PDAs are also provided by the International Patient Decision Aids Standards (IPDAS) initiative.51
CLINICAL APPLICATIONS FOR VALUE AND PERFORMANCE MEASUREMENT
A key component of these patient-facing tools, aside from their psychometric characteristics (validity, reliability, responsiveness), is the capability for users to interpret and apply them in real-world clinical settings and the potential for utilization as performance measures in gauging value. The ability to understand how the scores are generated by such tools is crucial to apply these metrics at the clinical and systems levels. The minimal clinically important difference (MCID) is the smallest change in a treatment outcome (eg, PROM scores) that a patient would identify as important and that would indicate a change in the patient’s management or health status.52 MCID thresholds of many commonly used PROMs are available in the orthopaedic literature and can be used to benchmark performance and assess clinically meaningful improvement.53 Differences smaller than the MCID are unlikely to matter to patients. MCID can be calculated using statistical methods (eg, the distribution method) based on typically selecting a threshold at 0.5 standard deviation within the distribution of outcome scores, or a subjective approach (eg, the anchor method), aligning scores to anchor questions reflecting patient satisfaction and patient perceptions around functional improvement.54 The anchor method is generally preferred, given that it more closely reflects patient perceptions of improvement, aligns with expectations, and helps to better define substantial clinical benefit (SCB) thresholds. SCB reflects an improvement in outcomes thought by the patient to be considerable. This may be particularly relevant in assessing the effect of interventions, such as TJA, which has a strong record in improving symptoms and function, and for which minimal clinical improvement should really be the standard outcome.55 SCB may also be useful in the treatment of sports injuries involving high-performing athletes and those with functional expectations on the higher end of the spectrum.56 Another tool for enabling clinically meaningful interpretation of patient-generated outcomes data is the patient acceptable symptom state, defined as an absolute threshold for symptoms experienced with a specific condition beyond which patients consider themselves well and therefore satisfied with treatment.57 Although MCID and SCB
can be defined using an existing PROMs dataset or with the simple addition of a single anchor question, these thresholds do not account for the fact that it is easier to achieve a larger magnitude of change in patients who have extreme baseline values, nor do they provide an indication of a patient’s successful return to activities of daily living, work, and recreation, or satisfaction with their level of improvement.58 Although the patient acceptable symptom state requires additional scoring, it can be used to determine success of an intervention in patients who do not have preoperative PROMs, is less sensitive to baseline symptom levels, and provides a useful tool for setting expectations around the potential outcomes of treatment.59 The use of these thresholds is gaining interest along with the development of PROMs as performance measures, defined as standardized tools, administered at designated time points, with an established risk-adjusted scoring methodology.60
can be defined using an existing PROMs dataset or with the simple addition of a single anchor question, these thresholds do not account for the fact that it is easier to achieve a larger magnitude of change in patients who have extreme baseline values, nor do they provide an indication of a patient’s successful return to activities of daily living, work, and recreation, or satisfaction with their level of improvement.58 Although the patient acceptable symptom state requires additional scoring, it can be used to determine success of an intervention in patients who do not have preoperative PROMs, is less sensitive to baseline symptom levels, and provides a useful tool for setting expectations around the potential outcomes of treatment.59 The use of these thresholds is gaining interest along with the development of PROMs as performance measures, defined as standardized tools, administered at designated time points, with an established risk-adjusted scoring methodology.60
Surgeon-Facing Tools
Clinical risk assessment tools (CRATs) provide validated metrics that can be utilized by surgeons to better understand the risks associated with a given treatment option. In orthopaedics, CRATs such as the Risk Assessment and Predictor Tool (RAPT), have commonly been used to predict discharge disposition home or to a rehabilitation facility following TJA.61 More recently, data have been gathered to power a similar risk assessment tool in geriatric patients with hip fractures.62 The largest drawback to CRATs is their varying levels of accuracy.63 Work continues to develop models involving higher levels of accuracy, with the goal of reducing costly postacute care.64 The Modified Frailty Index was introduced by the Canadian Study of Health and Aging and uses comorbidities (eg, hypertension, diabetes mellitus, vascular disease) to more comprehensively evaluate a patient’s health status to predict postoperative complications.65 This tool is shown to be predictive of increased mortality after femoral neck fractures as well as life-threatening complications and reoperation following total hip and total knee arthroplasty.66,67
Web-based orthopaedic personalized predictive tools provide personalized predictions of clinical outcomes based on the analysis of large volumes of data utilizing algorithmic mathematical modeling and predictive analytics.68 A 2019 study identified 31 discrete web-based orthopaedic personalized predictive tools designed to provide personalized prediction in various orthopaedic settings.69 These tools were designed for use in a variety of orthopaedic subspecialties including trauma, spine, TJA, and oncology; three-fourths of these tools had been developed since 2010.69
System-Level Tools: Core Platforms for Health Records and Patient Outcomes—Electronic Medical Records and Data Analytics Engines
A robust, interoperable, and user-friendly electronic health record is a core tool that can provide advanced management capabilities for value-oriented health care systems. These technologies can enable a more holistic longitudinal view of patients and improve care navigation (coordination and continuity of care) by multidisciplinary teams while reducing variation and waste. The functionalities of these platforms now extend beyond simple repositories of health information, toward the administration and collection of PROMs, PREMs, and other forms of
data including social determinants of unmet health and social needs. Electronic medical records (EMRs) provide access to this information at points of care and enable decision support. They may also provide functions beyond frontline applications at the patient level, enabling a more population-based health view with a lens on the health of their patient community. EMRs should also be interoperable and aligned with a strong data analytics engine, advanced enterprise data warehouse or cloud capabilities alongside dynamic real-time reporting and analytics.
data including social determinants of unmet health and social needs. Electronic medical records (EMRs) provide access to this information at points of care and enable decision support. They may also provide functions beyond frontline applications at the patient level, enabling a more population-based health view with a lens on the health of their patient community. EMRs should also be interoperable and aligned with a strong data analytics engine, advanced enterprise data warehouse or cloud capabilities alongside dynamic real-time reporting and analytics.
With the coronavirus disease 2019 (COVID-19) pandemic and a widening gap in access and affordability of care, structural and health-related social needs are found to be associated with health outcomes. In a move to address health needs upstream, there has been a recent call for systematic screening for social determinants of health, such as housing or food insecurity. Various effective screening tools exist, such as the Protocol for Responding to and Assessing Patients’ Assets, Risks, and Experiences, or PRAPARE, which is used and developed by the National Association of Community Health Centers.70 In an effort to align national outcome measurements, CMS compiled validated questions in an abbreviated screening tool known as the Accountable Health Communities Health-Related Social Needs Screening Tool.71 These population-based health tools equip organizations to address health-related social needs and, although not without its technical challenges, could be extremely effective if integrated into EMRs. This functionality could enable practices to identify and address the needs of patients through collaboration with community services, primary care networks, and other services upstream.
As behavioral health and health-related social needs become incorporated into discussions between clinicians and patients, EMRs and billing code development follows suit. A wider range of ICD-10 codes and “Z” codes, traditionally being used by behavioral health providers such as social workers and other allied health care professionals, are increasingly being implemented in primary and specialty care. EMR innovation must include access to data at population levels to improve the health of communities, whether through the incorporation of psychosocial PROMs or concordant billing codes.
PATIENT PORTALS
Patient portals enabling education, channels for communication, and continuity of care may be part of the EMR or stand alone, enhancing engagement of individuals in the care delivery process. Patient portals are secure websites that allow patients to access personal health information online using a secure username and password.72 Such portals facilitate longitudinal care by providing patients with easy access to records such as doctor’s visits, medication lists, and laboratory results, as well as a secure means of communication with health care providers to request prescription refills, ask questions, and schedule appointments. Although research into the use of patient portals is relatively new, it has been associated with improving adherence to medications, facilitating effective patient-clinician communication, and enabling the discovery of medical errors.73 Patient portals can also be used to send and record PROMs and PREMs, and several major EMRs have integrated patient portals with the capability to design, distribute, and
collect customizable questionnaires.74 The Partners Healthcare Patient Gateway at Massachusetts General Hospital/Brigham and Women’s Health in Boston, Massachusetts is a platform that includes PROMs such as the Knee injury and Osteoarthritis Outcome Score among other orthopaedic and nonorthopaedic PROMs that patients can complete before, during, or after their visit.75 In this system, physicians can help patients interpret their progress over time. The eventual goal is to build a robust database that can aid in clinical decision-making.
collect customizable questionnaires.74 The Partners Healthcare Patient Gateway at Massachusetts General Hospital/Brigham and Women’s Health in Boston, Massachusetts is a platform that includes PROMs such as the Knee injury and Osteoarthritis Outcome Score among other orthopaedic and nonorthopaedic PROMs that patients can complete before, during, or after their visit.75 In this system, physicians can help patients interpret their progress over time. The eventual goal is to build a robust database that can aid in clinical decision-making.
OUTCOME MEASUREMENT PLATFORM
A range of commercial and research-grade (eg, RecCAP) patient outcome measurement platforms exist, enabling electronic capture of PROMs and other forms of patient-generated health data. Digitization enables functionalities (eg, data visualization and analysis) and efficiencies (eg, time saving, automation) beyond manual paper entry. Most electronic PROM platforms support remote delivery, allowing patients to complete questionnaires by text, email, smartphone, tablet or their desktop computers at home.76 In addition, these electronic platforms offer integration into EMR systems, which enable real-time review with patients and other clinical team members during clinic appointments and tracking of results over time.77 Outcome measurement platforms have enabled the creation of large collections of PROMs data.
JOINT REPLACEMENT REGISTRIES
Data from PROMs are perhaps most valuable in aggregate and outcome measurement platforms and have enabled the creation of large collections of PROMs data. In the United States and across the world, these orthopaedic data are most typically aggregated in joint replacement registries, which provide insights into clinical outcomes at scale.78 The United States, with its multipayer infrastructure, has both regionally focused registries, such as the Michigan Arthroplasty Registry Collaborative Quality Initiative (MARCQI), and broader registries, such as the Function and Outcomes Research for Comparative Effectiveness in Total Joint Replacement (FORCE-TJR) registry and American Joint Replacement Registry (AJRR) developed in partnership with the American Association of Orthopaedic Surgeons in 2009.79,80 These registries also capture a range of clinical outcomes, including readmission rates, complications, and implant survival.77
ADVANCED COST ACCOUNTING PLATFORMS
Time-driven activity-based costing (TDABC), introduced by Kaplan and Anderson in 2004, provides a data-driven, targeted estimate of costs based on resources used by patients for their condition over a cycle of care.81 Using costs for supplied resources (generating a cost per minute of all aspects of a health care professional’s time), and practical capacity (actual productive time spent on each capacity-supplying resource), the method has enabled the definition of actual costs of care for orthopaedic conditions such as hip and knee osteoarthritis.82 TDABC can be considered a versatile tool for defining and increasing awareness of costs from the “bottom-up” compared to traditional cost accounting (eg, using ratio of costs to charges [RCC] and relative value units [RVU]), which takes a “top-down” approach and defines
costs based on charges and revenues. Although these legacy approaches are familiar and easier to perform, they risk an overestimation of total costs.81 In orthopaedic care, TDABC has estimated the costs of TJA and ankle fracture care to be 48% to 59% of traditional accounting estimates.83,84 Recent studies have used TDABC to identify implants and personnel to be the largest cost drivers in total hip, total knee, and total shoulder arthroplasty.82,85 However, TDABC has limitations: it often excludes substantial indirect overhead costs, although one proposed model has been to standardize the most frequently used indirect overhead costs (maintenance, information technology, hospital administration, and billing) in the model.86
costs based on charges and revenues. Although these legacy approaches are familiar and easier to perform, they risk an overestimation of total costs.81 In orthopaedic care, TDABC has estimated the costs of TJA and ankle fracture care to be 48% to 59% of traditional accounting estimates.83,84 Recent studies have used TDABC to identify implants and personnel to be the largest cost drivers in total hip, total knee, and total shoulder arthroplasty.82,85 However, TDABC has limitations: it often excludes substantial indirect overhead costs, although one proposed model has been to standardize the most frequently used indirect overhead costs (maintenance, information technology, hospital administration, and billing) in the model.86
CURRENT AND FUTURE TECHNOLOGIES FOR HIGH-VALUE ORTHOPAEDIC CARE
Digital Phenotyping and Passive Patient-Generated Health Data
Digital phenotyping is the moment-by-moment quantification of human phenotypes in situ using data related to activity, behavior, and communications obtained from personal digital devices, such as smartphones and wearable sensors. It offers a technology capable of passively capturing patient-generated health data.87 Personalized health information measured within an individual’s usual settings may provide metrics for tracking, decision support, and enhancing patientcentered care. A review of scope demonstrates an acceleration of work in this area over the past 5 years, particularly in the context of orthopaedic surgery.88 The measurement of and synthesis of activity, biometric, and communications data may provide suitable biomarkers of clinically meaningful health outcomes. Future work is geared to utilize this data alongside PROMs for the prediction of outcomes, risk profiling, SDM, and surgical optimization.
Artificial Intelligence, Machine Learning, Deep Learning
Artificial intelligence (AI) is broadly defined as a branch of computer science that simulates intelligent behavior in computers. In health care, AI offers the promise of enhanced predictive, diagnostic, and decision-making capabilities through three key domains: (1) advanced data discovery and extraction, (2) improved diagnostics and prediction, and (3) enhanced clinical and decision support.89 These tools aim to decrease overall spending by reducing time, resource utilization, manpower, and computational power.89 Data discovery and extraction has been successfully applied to work flows via natural language processing tools used for physician dictation as well as to clinical outcomes via stratification of patients by risk potential and prediction of adverse events.90 AI has improved diagnostics and prediction across several domains of orthopaedic care, including improved image acquisition and disease detection compared with conventional imaging for detecting long bone and fragility fractures, differentiating between benign and malignant bone tumors, determining prognosis for patients with cancer, determining the risk of mortality after arthroplasty, and mapping disease progression for developmental dysplasia of the hip and degenerative disease of the spine and lower extremities.91,92 In the domain of supporting clinical decision making, AI in combination with PROMs and demographic data have been applied to SDM in knee
osteoarthritis, anterior cruciate ligament deficiency, and complex degenerative spinal disease. More specifically, AI has been leveraged to provide patient-specific risk-benefit ratios and prediction of health outcomes following arthroplasty.93 Finally, AI is being applied to increase procedural efficiency in the operating room by guiding surgical teams through orthopaedic procedures with the goal of reducing variation in techniques in the operating room.
osteoarthritis, anterior cruciate ligament deficiency, and complex degenerative spinal disease. More specifically, AI has been leveraged to provide patient-specific risk-benefit ratios and prediction of health outcomes following arthroplasty.93 Finally, AI is being applied to increase procedural efficiency in the operating room by guiding surgical teams through orthopaedic procedures with the goal of reducing variation in techniques in the operating room.
AI has already had widespread effect across multiple domains of orthopaedic care; however, significant ethical, legal, policy, and practice considerations remain. The ethical challenges in using AI stem from the varying levels of risk to patient privacy and confidentiality related to data transfer and sharing, consent, and patient autonomy. In addition, there remain unknown factors over the potential for robots to autonomously adapt and alter the course of a procedure.94 Current regulatory frameworks and ethical guidelines need to keep pace with the progress in technology.95 Regulatory and professional bodies should provide guidance on interventions aligned with AI solutions, include meaningful endpoints (those driving changes beneficial to clinicians and patients), appropriate benchmarks (assuring these are dynamic based on evolving sets of real-world data and performance), and longitudinal audit mechanisms (incorporating postmarket surveillance of a solution evolving over time as AI-driven algorithms learn).95 On an individual level, surgeons should understand the population characteristics used to develop algorithms and work with programmers to calibrate their outputs to match the needs of the patient populations being treated.96 Future training of orthopaedic providers must focus on the interaction between clinicians, patients, and technology. There is also a need to demonstrate the effect of AI on value with high-quality cohort and randomized controlled trials using PROMs, costeffectiveness analyses, and rapid quality improvement studies. AI has been leveraged for orthopaedics in multiple studies97 (Table 1).
Machine learning (ML) is a process by which a computer improves its own performance (eg, via the analysis of image files) by continuously incorporating new data into an existing statistical model (Table 2). ML encompasses automatic procedures of incremental function optimization, which can be conducted under supervised or unsupervised learning. Supervised learning requires a human to manually input large volumes of data that are paired to a correct output function (eg, a correct diagnosis made using a radiologic image). In contrast, for unsupervised learning, the correct inputs are not known or given, and instead inferred by an algorithm on the basis of the relationship between the input data points.92 These techniques can be applied to individual-level and population-level data.98 Since 2010, the volume of orthopaedic literature citing ML has increased tenfold in areas ranging from disease detection (eg, spine pathology, osteoarthritis, fracture, anterior cruciate ligament/posterior cruciate ligament injury) and clinical assessment (eg, shoulder strength, skeletal bone age, gait analysis) to treatment efficacy (eg, optimal injection point, prosthesis control).99,100,101,102,103,104,105 These studies have utilized medical imaging, both biomechanical and patient generated, spanning a range of ML techniques.
Deep learning (DL) is as a form of AI that enables computers to learn from experience and understand systems in terms of information hierarchies.106 As the
computer gathers knowledge through cycles of learning and experience, requirements are minimal to none for ongoing manual input. This process allows the computer to learn complicated concepts.106 DL in orthopaedics to date has mostly been applied to imaging data with a focus on predicting pathology in adolescent idiopathic scoliosis, spondylolisthesis, spinal column and long bone fractures, hip osteoarthritis, and skeletal bone age.107,108,109,110,111,112 A major advantage of DL is efficiency and handling of large volumes of complex structured and unstructured data. It is critical for human users to understand and have access to the analytics behind the outputs, alongside controls to ensure safe handling and accurate interpretation of data.113,114
computer gathers knowledge through cycles of learning and experience, requirements are minimal to none for ongoing manual input. This process allows the computer to learn complicated concepts.106 DL in orthopaedics to date has mostly been applied to imaging data with a focus on predicting pathology in adolescent idiopathic scoliosis, spondylolisthesis, spinal column and long bone fractures, hip osteoarthritis, and skeletal bone age.107,108,109,110,111,112 A major advantage of DL is efficiency and handling of large volumes of complex structured and unstructured data. It is critical for human users to understand and have access to the analytics behind the outputs, alongside controls to ensure safe handling and accurate interpretation of data.113,114
|
TABLE 1 Artificial Intelligence Applied to Orthopaedics
|
||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
||||||||||||||||||||||||||||||||||||||||||||||||
Real-Time Location Systems
Real-time locating systems (or real-time location systems) (RTLS) are local systems that can identify and track assets in real (or near-real) time.115 This functionality
could support the development of orthopaedic multidisciplinary teams, enhance efficiencies in clinic workflow, and automating TDABC and other activity-based methods of cost accounting for increasing value. Such systems are composed of fixed readers that receive wireless signals from small unique tags attached to objects of interest. This network of readers can determine if the tagged object is located within a confined indoor or outdoor space. RTLS can provide location information to hospital information systems, admission/discharge/transfer systems, radiology information systems, and operating room systems via an open application programming interface.116 Both physical components of RTLS, the reader and the tag, can be outfitted with additional features including those for communication (eg, push/call buttons, indictor status buttons, voice-to-voice capability, buzzers, LED lights, LCD screens), additional data gathering (eg, temperature and motion sensors), and writeable memory to log and store data.117 RTLS systems can provide location at a variety of resolutions (Table 3) and be designed around a range of technologies (eg, infrared, ultrasonography, camera vision, Bluetooth, radiofrequency identification, GPS, WiFi, and cellular) that each have their benefits and drawbacks.118,119 Prior to selecting a given RTLS solution, a practice must consider
its own unique needs and physical and information technology infrastructure. For example, it will likely be challenging to rely on a cellular-based system in an older hospital complex with numerous concrete walls and subterranean floors. Finally, use of RTLS to monitor staff and/or patients requires clear disclosure and practices must be prepared to address privacy concerns prior to investing in RTLS.
could support the development of orthopaedic multidisciplinary teams, enhance efficiencies in clinic workflow, and automating TDABC and other activity-based methods of cost accounting for increasing value. Such systems are composed of fixed readers that receive wireless signals from small unique tags attached to objects of interest. This network of readers can determine if the tagged object is located within a confined indoor or outdoor space. RTLS can provide location information to hospital information systems, admission/discharge/transfer systems, radiology information systems, and operating room systems via an open application programming interface.116 Both physical components of RTLS, the reader and the tag, can be outfitted with additional features including those for communication (eg, push/call buttons, indictor status buttons, voice-to-voice capability, buzzers, LED lights, LCD screens), additional data gathering (eg, temperature and motion sensors), and writeable memory to log and store data.117 RTLS systems can provide location at a variety of resolutions (Table 3) and be designed around a range of technologies (eg, infrared, ultrasonography, camera vision, Bluetooth, radiofrequency identification, GPS, WiFi, and cellular) that each have their benefits and drawbacks.118,119 Prior to selecting a given RTLS solution, a practice must consider
its own unique needs and physical and information technology infrastructure. For example, it will likely be challenging to rely on a cellular-based system in an older hospital complex with numerous concrete walls and subterranean floors. Finally, use of RTLS to monitor staff and/or patients requires clear disclosure and practices must be prepared to address privacy concerns prior to investing in RTLS.
|
TABLE 2 Advantages and Disadvantages of Machine Learning
|
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
TABLE 3 Real-Time Locating Systems Resolution
|
||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
||||||||||||||||||||||||
SUMMARY
There are an increasing array of data-driven tools and technologies to support clinicians in practicing high-value orthopaedic care. Although these tools and capabilities are at various stages of development and adoption, the fundamental elements of value (outcomes that matter to patients and true costs of care) can start to be incorporated in orthopaedic practices using grassroots approaches, such as the use of paper-based or freely available research-grade software to administer PROMs. Advancement and availability of tools and technologies for driving value-based health care is inevitable and it will be just as important for culture change to adopt these capabilities to harness the full power of these solutions in highvalue orthopaedic practices.
References
1. Centers for Medicare & Medicaid Services: Comprehensive care for joint replacement model. Available at: https://innovation.cms.gov/innovation-models/cjr. Accessed January 14, 2021.
2. Centers for Medicare & Medicaid Services: Accountable care organizations (ACOs). Available at: https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/ACO. Accessed January 15, 2021.
3. Porter ME, Teisberg EO: Redefining Health Care: Creating Value-Based Competition on Results. Harvard Business School Press, 2006.
4. Porter ME: What is value in health care? N Engl J Med 2010;363(26):2477-2481.
5. Wei DH, Hawker GA, Jevsevar DS, Bozic KJ: Improving value in musculoskeletal care delivery: AOA critical issues. J Bone Joint Surg Am 2015;97(9):769-774.
6. Institute of Medicine (US) Committee on Quality of Health Care in America: Crossing The Quality Chasm: A New Health System For The 21st Century, 2001. Available at: https://www.med.unc.edu/neurosurgery/wp-content/uploads/sites/460/2018/10/Crossing-the-Quality-Chasm.pdf. Accessed May 3, 2023.
7. Harwood JL, Butler CA, Page AE: Patient-centered care and population health: Establishing their role in the orthopaedic practice. J Bone Joint Surg Am 2016;98(10):e40.
8. Elwyn G, Frosch D, Thomson R, et al: Shared decision making: A model for clinical practice. J Gen Intern Med 2012;27(10):1361-1367.
9. American Academy of Orthopaedic Surgeons. Preventing Venous Thromboembolic Disease In Patients Undergoing Elective Hip And Knee Arthroplasty: Evidence-Based Guideline And Evidence Report, 2011. Available at: https://www.aaos.org/quality/quality-programs/tumor-infection-and-military-medicine-programs/venous-thromboembolic-disease-in-elective-tka-and-tha-prevention/. Accessed May 3, 2023.
10. Singer SJ, Kerrissey M, Friedberg M, Phillips R: A comprehensive theory of integration. Med Care Res Rev 2020;77(2):196-207.
11. Manning BT, Callahan CD, Robinson BS, Adair D, Saleh KJ: Overcoming resistance to implementation of integrated care pathways in orthopaedics. J Bone Joint Surg Am 2013;95(14):e100, 1-6.
12. Jimenez Muñoz AB, Duran Garcia ME, Rodriguez Perez MP, Sanjurjo M, Vigil MD, Vaquero J: Clinical pathway for hip arthroplasty six years after introduction. Int J Health Care Qual Assur Inc Leadersh Health Serv 2006;19(2-3):237-245.
13. Pearson S, Moraw I, Maddern GJ: Clinical pathway management of total knee arthroplasty: A retrospective comparative study. Aust N Z J Surg 2000;70(5):351-354.
14. Kates SL, Mendelson DA, Friedman SM: The value of an organized fracture program for the elderly: Early results. J Orthop Trauma 2011;25(4):233-237.
15. Adunsky A, Arad M, Levi R, Blankstein A, Zeilig G, Mizrachi E: Five-year experience with the ‘Sheba’model of comprehensive orthogeriatric care for elderly hip fracture patients. Disabil Rehabil 2005;27(18-19):1123-1127.
16. Ring D: Editorial comment: Comprehensive orthopaedic care. Clin Orthop Relat Res 2018;476(4):694-695.
17. HealthIT.gov: Patient-Generated Health Data. Available at: https://www.healthit.gov/topic/scientific-initiatives/pcor/patient-generated-health-data-pghd. Accessed May 1, 2023.
18. Brodke DJ, Saltzman CL, Brodke DS: PROMIS for orthopaedic outcomes measurement. J Am Acad Orthop Surg 2016;24(11):744-749.
19. Patrick D, Guyatt GH, Acquadro C: Patient-reported outcomes, in Higgins JPT, Green S, eds: Cochrane Handbook for Systematic Reviews of Interventions Version 510 The Cochrane Collaboration, 2011. Available at: http:www.handbook.cochrane.org. Accessed May 3, 2023.
20. Lindgren JV, Wretenberg P, Karrholm J, Garellick G, Rolfson O: Patient-reported outcome is influenced by surgical approach in total hip replacement: A study of the Swedish hip arthroplasty register including 42,233 patients. Bone Joint J 2014;96-B(5):590-596.
21. Jameson SS, Mason J, Baker P, et al: A comparison of surgical approaches for primary hip arthroplasty: A cohort study of patient reported outcome measures (PROMs) and early revision using linked national databases. J Arthroplasty 2014;29(6):1248-1255.e1.
22. The American Board of Orthopaedic Surgery: Certification examinations part II: Patient reported outcomes. Available at: https://www.abos.org/certification-exams/part-ii/. Accessed May 2, 2023.
23. Stulberg J: The physician quality reporting initiative – A gateway to pay for performance: What every health care professional should know. Qual Manag Health Care 2008;17(1):2-8.
24. Ruzbarsky JJ, Marom N, Marx RG: Measuring quality and outcomes in sports medicine. Clin Sports Med 2018;37(3):463-482.
25. Dacombe PJ, Amirfeyz R, Davis T: Patient-reported outcome measures for hand and wrist trauma: Is there sufficient evidence of reliability, validity, and responsiveness? Hand (N Y) 2016;11(1):11-21.
26. Baumhauer JF, Bozic KJ: Value-based healthcare: Patient-reported outcomes in clinical decision making. Clin Orthop Relat Res 2016;474(6):1375-1378.
27. Horn ME, Reinke EK, Couce LJ, Reeve BB, Ledbetter L, George SZ: Reporting and utilization of Patient-Reported Outcomes Measurement Information System® (PROMIS®) measures in orthopedic research and practice: A systematic review. J Orthop Surg Res 2020;15(1):553.
28. Black N: Patient reported outcome measures could help transform health care. BMJ 2013;346:f167.
29. Andrawis J, Akhavan S, Chan V, Lehil M, Pong D, Bozic KJ: Higher preoperative patient activation, associated with better patient-reported outcomes after total joint arthroplasty. Clin Orthop Relat Res 2015;473(8):2688-2697.
30. Berliner JL, Brodke DJ, Chan V, SooHoo NF, Bozic KJ: John Charnley award: Preoperative patient-reported outcome measures predict clinically meaningful improvement in function after THA. Clin Orthop Relat Res 2016;474(2):321-329.
31. Papuga MO, Beck CA, Kates SL, Schwarz EM, Maloney MD: Validation of GAITRite and PROMIS as high-throughput physical function outcome measures following ACL reconstruction. J Orthop Res 2014;32(6):793-801.
32. Gagnier JJ: Patient reported outcomes in orthopaedics. J Orthop Res 2017;35(10):2098-2108.
33. MOTION Group: Patient-reported outcomes in orthopaedics. J Bone Joint Surg Am 2018;100(5):436-442.
34. American Academy of Orthopaedic Surgeons: Patient reported outcome measures. Available at: https://www.aaos.org/quality/research-resources/patient-reported-outcome-measures/. Accessed May 3, 2023.
35. NSW Agency for Clinical Innovation: About patient reported measures. Available at: https://aci.health.nsw.gov.au/statewide-programs/prms/about. Accessed May 3, 2023.
36. Weldering T, Smith SM: Patient-reported outcomes (PROs) and patient-reported outcome measures (PROMs). Health Serv Insights 2013;6:61-68.
37. Black N, Varaganum M, Hutchings A: Relationship between patient reported experience (PREMs) and patient reported outcomes (PROMs) in elective surgery. BMJ Qual Saf 2014;23(7):534-542.
38. Hibbard JH, Greene J: What the evidence shows about patient activation: Better health outcomes and care experiences; fewer data on costs. Health Aff (Millwood) 2013;32(2):207-214.
39. Grogan Moore ML, Jayakumar P, Laverty D, Hill AD, Koenig KM: Patient-reported outcome measures and patient activation: What are their roles in orthopedic trauma? J Orthop Trauma 2019;33(suppl 7):S38-S42.
40. Smith MA: The role of shared decision making in patient-centered care and orthopaedics. Orthop Nurs 2016;35(3):144-149.
41. Slover J, Shue J, Koenig K: Shared decision-making in orthopaedic surgery. Clin Orthop Relat Res 2012;470(4):1046-1053.
42. Bozic KJ, Belkora J, Chan V, et al: Shared decision making in patients with osteoarthritis of the hip and knee: Results of a randomized controlled trial. J Bone Joint Surg Am 2013;95(18):1633-1639.
43. Stacey D, Hawker G, Dervin G, et al: Decision aid for patients considering total knee arthroplasty with preference report for surgeons: A pilot randomized controlled trial. BMC Musculoskelet Disord 2014;15:54.
44. Stacey D, Légaré F, Col NF, et al: Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2014;2014(1):CD001431.
45. Adam JA, Khaw FM, Thomson RG, Gregg PJ, Llewellyn-Thomas HA: Patient decision aids in joint replacement surgery: A literature review and an opinion survey of consultant orthopaedic surgeons. Ann R Coll Surg Engl 2008;90(3):198-207.
46. Molenaar S, Sprangers M, Postma-Schuit F, et al: Feasibility and effects of decision aids. Med Decis Making 2000;20(1):112-127.
47. Ibrahim SA, Blum M, Lee GC, et al: Effect of a decision aid on access to total knee replacement for black patients with osteoarthritis of the knee: A randomized clinical trial. JAMA Surg 2017;152(1):e164225.
48. Langdon IJ, Hardin R, Learmonth ID: Informed consent for total hip arthroplasty: Does a written information sheet improve recall by patients? Ann R Coll Surg Engl 2002;84(6):404-408.
49. Saigal R, Clark AJ, Scheer JK, et al: Adult spinal deformity patients recall fewer than 50% of the risks discussed in the informed consent process preoperatively and the recall rate worsens significantly in the postoperative period. Spine (Phila Pa 1976) 2015;40(14):1079-1085.
50. Barr PJ, Elwyn G: Measurement challenges in shared decision making: Putting the ‘patient’ in patient-reported measures. Health Expect 2016;19(5):993-1001.
51. Elwyn G, O’Connor A, Stacey D, et al: Developing a quality criteria framework for patient decision aids: Online international Delphi consensus process. BMJ 2006;333(7565):417.
52. Crosby RD, Kolotkin RL, Williams GR: Defining clinically meaningful change in health-related quality of life. J Clin Epidemiol 2003;56(5):395-407.
53. Maltenfort M, Díaz-Ledezma C: Statistics in brief: Minimum clinically important difference – Availability of reliable estimates. Clin Orthop Relat Res 2017;475(4):933-946.
54. Glassman SD, Copay AG, Berven SH, Polly DW, Subach BR, Carreon LY: Defining substantial clinical benefit following lumbar spine arthrodesis. J Bone Joint Surg Am 2008;90(9):1839-1847.
55. Katz NP, Paillard FC, Ekman E: Determining the clinical importance of treatment benefits for interventions for painful orthopedic conditions. J Orthop Surg Res 2015;10:24.
56. Nwachukwu BU, Chang B, Fields K, et al: Defining the “substantial clinical benefit”’ after arthroscopic treatment of femoroacetabular impingement. Am J Sports Med 2017;45(6):1297-1303.
57. Muller B, Yabroudi MA, Lynch A, et al: Defining thresholds for the patient acceptable symptom state for the IKDC subjective knee form and KOOS for patients who underwent ACL reconstruction. Am J Sports Med 2016;44(11):2820-2826.
58. Carragee EJ: The rise and fall of the “minimum clinically important difference.” Spine J 2010;10(4):283-284.
59. Mannion AF, Loibl M, Bago J, et al: What level of symptoms are patients with adult spinal deformity prepared to live with? A cross-sectional analysis of the 12-month follow-up data from 1043 patients. Eur Spine J 2020;29(6):1340-1352.
60. Safran D: Feasibility and value of patient-reported outcome measures for value-based payment. Med Care 2019;57(3):177-179.
61. Gholson JJ, Pugely AJ, Bedard NA, Duchman KR, Anthony CA, Callaghan JJ: Can we predict discharge status after total joint arthroplasty? A calculator to predict home discharge. J Arthroplasty 2016;31(12):2705-2709.
62. Arshi A, Iglesias BC, Zambrana LE, et al: Postacute care utilization in postsurgical orthogeriatric hip fracture care. J Am Acad Orthop Surg 2020;28(18):743-749.
63. Menendez ME, Schumacher CS, Ring D, Freiberg AA, Rubash HE, Kwon YM: Does “6-clicks” day 1 postoperative mobility score predict discharge disposition after total hip and knee arthroplasties? J Arthroplasty 2016;31(9):1916-1920.
64. Goltz DE, Ryan SP, Attarian D, Jiranek WA, Bolognesi MP, Seyler TM: A preoperative risk prediction tool for discharge to a skilled nursing or rehabilitation facility after total joint arthroplasty. J Arthroplasty 2021;36(4):1212-1219.
65. Rockwood K, Andrew M, Mitnitski AB: A comparison of two approaches to measuring frailty in elderly people. J Gerontol A Biol Sci Med Sci 2007;62(7):738-743.
66. Patel KV, Brennan KL, Brennan ML, Jupiter DC, Shar A, Davis ML: Association of a modified frailty index with mortality after femoral neck fracture in patients aged 60 years and older. Clin Orthop Relat Res 2014;472(3):1010-1017.
67. Shin JI, Keswani A, Lovy AJ, Moucha CS: Simplified frailty index as a predictor of adverse outcomes in total hip and knee arthroplasty. J Arthroplasty 2016;31(11):2389-2394.
68. Bernstein DN, Fear K, Mesfin A, et al: Patient-reported outcomes use during orthopaedic surgery clinic visits improves the patient experience. Musculoskeletal Care 2019;17(1):120-125.
69. Curtin P, Conway A, Martin L, Lin E, Jayakumar P, Swart E: Compilation and analysis of web-based orthopedic personalized predictive tools: A scoping review. J Pers Med 2020;10(4):223.
70. National Association of Community Health Centers: PRAPARE implementation and action toolkit. Available at: http://www.nachc.org/research-and-data/prapare/tool-kit/. Accessed January 10, 2021.
71. Centers for Medicare and Medicaid Services: The accountable health communities health-related social needs screening tool. Available at: https://innovation.cms.gov/files/worksheets/ahcm-screeningtool.pdf. Accessed May 3, 2023.
72. Office for the National Coordinator for Health Information Technology: What is a patient portal? Available at: https://www.healthit.gov/faq/what-patient-portal. Accessed May 3, 2023.
73. Dendere R, Slade C, Burton-Jones A, Sullivan C, Staib A, Janda M: Patient portals facilitating engagement with inpatient electronic medical records: A systematic review. J Med Internet Res 2019;21(4):e12779.
74. Massachusetts Medical Society: Patient-reported outcome measures: current state and MMS principles. Available at: http://www.massmed.org/proms. Accessed May 3, 2023.
75. Partners Healthcare: About patient reported outcome measures (PROMs). Available at: http://caredecisions.partners.org/partners-care-decisions/patient-reported-outcome-measures-proms/. Accessed May 3, 2023.
76. Borowsky PA, Kadri OM, Meldau JE, Blanchett J, Makhni EC: The remote completion rate of electronic patient-reported outcome forms before scheduled clinic visits-a proof-of-concept study using patient-reported outcome measurement information system computer adaptive test questionnaires. J Am Acad Orthop Surg Glob Res Rev 2019;3(10):e19.00038.
77. Makhni EC: Meaningful clinical applications of patient-reported outcome measures in orthopaedics. J Bone Joint Surg Am 2021;103(1):84-91.
78. Rolfson O, Karrholm J, Dahlberg LE, Garellick G: Patient-reported outcomes in the Swedish hip arthroplasty register: Results of a nationwide prospective observational study. J Bone Joint Surg Br 2011;93(7):867-875.
79. Hallstrom B, Singal B, Cowen ME, Roberts KC, Hughes RE: The Michigan experience with safety and effectiveness of tranexamic acid use in hip and knee arthroplasty. J Bone Joint Surg Am 2016;98(19):1646-1655.
80. Ayers DC, Fehring TK, Odum SM, Franklin PD: Using joint registry data from FORCE-TJR to improve the accuracy of risk-adjustment prediction models for thirtyday readmission after total hip replacement and total knee replacement. J Bone Joint Surg Am 2015;97(8):668-671.
81. Kaplan R, Anderson S: Time-driven activity-based costing. Harv Bus Rev 2004;82(11):131-138, 150.
82. Menendez ME, Lawler SM, Shaker J, Bassoff NW, Warner JJP, Jawa A: Time-driven activity-based costing to identify patients, incurring high inpatient cost for total shoulder arthroplasty. J Bone Joint Surg Am 2018;100(23):2050-2056.
83. Palsis JA, Brehmer TS, Pellegrini VD, Drew JM, Sachs BL: The cost of joint replacement: Comparing two approaches to evaluating costs of total hip and knee arthroplasty. J Bone Joint Surg Am 2018;100(4):326-333.
84. McCreary DL, White M, Vang S, Plowman B, Cunningham BP: Time-driven activitybased costing in fracture care: Is this a more accurate way to prepare for alternative payment models? J Orthop Trauma 2018;32(7):344-348.
85. DiGioia AM, Greenhouse PK, Giarrusso ML, Kress JM: Determining the true cost to deliver total hip and knee arthroplasty over the full cycle of care: Preparing for bundling and reference-based pricing. J Arthroplasty 2016;31(1):1-6.
86. Pathak S, Snyder D, Kroshus T, et al: What are the uses and limitations of timedriven activity-based costing in total joint replacement? Clin Orthop Relat Res 2019;477(9):2071-2081.
87. Insel TR: Digital phenotyping: Technology for a new science of behavior. J Am Med Assoc 2017;318(13):1215-1216.
88. Jayakumar P, Lin E, Galea V, et al: Digital phenotyping and patient-generated health data for outcome measurement in surgical care: A scoping review. J Pers Med 2020;10(4):282.
89. Jayakumar P, Moore MLG, Bozic KJ: Value-based healthcare: Can artificial intelligence provide value in orthopaedic surgery? Clin Orthop Relat Res 2019;477(8):1777-1780.
90. Rajkomar A, Oren E, Chen K, et al: Scalable and accurate deep learning with electronic health records. NPJ Digit Med 2018;1:18.
91. Harris AHS, Kuo AC, Weng Y, Trickey AW, Bowe T, Giori NJ: Can machine learning methods produce accurate and easy-to-use prediction models of 30-day complications and mortality after knee or hip arthroplasty? Clin Orthop Relat Res 2019;477(2):452-460.
92. Cabitza F, Locoro A, Banfi G: Machine learning in orthopedics: A literature review. Front Bioeng Biotechnol 2018;6:75.
93. Sambare T, Uhler L, Bozic KJ: Shared decision making: Time to get personal. Available at: https://catalyst.nejm. Accessed May 1, 2023.
94. Makhni EC, Makhni S, Ramkumar PN: Artificial intelligence for the orthopaedic surgeon: An overview of potential benefits, limitations, and clinical applications. J Am Acad Orthop Surg 2021;29(6):235-243.
95. O’Sullivan S, Nevejans N, Allen C, et al: Legal, regulatory, and ethical frameworks for development of standards in artificial intelligence (AI) and autonomous robotic surgery. Int J Med Robot 2019;15(1):e1968.
96. Parikh RB, Obermeyer Z, Navathe AS: Regulation of predictive analytics in medicine. Science 2019;363(6429):810-812.
97. Myers TG, Ramkumar PN, Ricciardi BF, Urish KL, Kipper J, Ketonis C: Artificial intelligence and orthopaedics: An introduction for clinicians. J Bone Joint Surg Am 2020;102(9):830-840.
98. Bayliss L, Jones LD: The role of artificial intelligence and machine learning in predicting orthopaedic outcomes. Bone Joint J 2019;101-B(12):1476-1478.
99. Spampinato C, Palazzo S, Giordano D, Aldinucci M, Leonardi R: Deep learning for automated skeletal bone age assessment in X-ray images. Med Image Anal 2017;36(suppl):41-51.
100. Lemoyne R, Mastroianni T, Hessel A, Nishikawa K: Implementation of machine learning for classifying prosthesis type through conventional gait analysis. Annu Int Conf IEEE Eng Med Biol Soc 2015;2015:202-205.
101. Pogorelc B, Gams M: Diagnosing health problems from gait patterns of elderly. Annu Int Conf IEEE Eng Med Biol Soc 2010;2010:2238-2241.
102. Phinyomark A, Osis ST, Hettinga BA, Kobsar D, Ferber R: Gender differences in gait kinematics for patients with knee osteoarthritis. BMC Musculoskelet Disord 2016;17:157.
103. Atkinson EJ, Therneau TM, Melton LJ, et al: Assessing fracture risk using gradient boosting machine (GBM) models. J Bone Miner Res 2012;27(6):1397-1404.
104. Silver AE, Lungren MP, Johnson ME, O’Driscoll SW, An KN, Hughes RE: Using support vector machines to optimally classify rotator cuff strength data and quantify postoperative strength in rotator cuff tear patients. J Biomech 2006;39(5):973-979.
105. Jamaludin A, Lootus M, Kadir T, et al: ISSLS prize in bioengineering science 2017: Automation of reading of radiological features from magnetic resonance images (MRIs) of the lumbar spine without human intervention is comparable with an expert radiologist. Eur Spine J 2017;26(5):1374-1383.
106. Goodfellow I, Bengio Y, Courville A: Deep Learning. MIT University Press, 2016.
107. Thong W, Parent S, Wu J, Aubin CE, Labelle H, Kadoury S: Three-dimensional morphology study of surgical adolescent idiopathic scoliosis patient from encoded geometric models. Eur Spine J 2016;25(10):3104-3113.
108. Forsberg D, Sjöblom E, Sunshine JL: Detection and labeling of vertebrae in MR images using deep learning with clinical annotations as training data. J Digit Imaging 2017;30(4):406-412.
109. Olczak J, Fahlberg N, Maki A, et al: Artificial intelligence for analyzing orthopedic trauma radiographs. Acta Orthop 2017;88(6):581-586.
110. Al-Helo S, Alomari RS, Ghosh S, et al: Compression fracture diagnosis in lumbar: A clinical CAD system. Int J Comput Assist Radiol Surg 2013;8(3):461-469.
111. Xue Y, Zhang R, Deng Y, Chen K, Jiang T: A preliminary examination of the diagnostic value of deep learning in hip osteoarthritis. PLoS One 2017;12(6):e0178992.
112. Giordano D, Kavasidis I, Spampinato C: Modeling skeletal bone development with hidden Markov models. Comput Methods Programs Biomed 2016;124:138-147.
113. Cabitza F, Rasoini R, Gensini GF: Unintended consequences of machine learning in medicine. J Am Med Assoc 2017;318(6):517-518.
114. Caruana R, Lou Y, Gehrke J, Koch P, Sturm M, Elhadad N: Intelligible models for healthcare: Predicting pneumonia risk and hospital 30-day readmission, in Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. August 2015, pp 1721-1730. Accessed May 3, 2023.
115. Kamel Boulos MN, Berry G: Real-time locating systems (RTLS) in healthcare: A condensed primer. Int J Health Geogr 2012;11:25.
116. Dimond D: The optimal RTLS solution for hospitals. J Healthc Inf Manag 2009;23(2):4-5.
117. Malik A: RTLS For Dummies. Wiley Publishing, 2009.
118. Liu H, Darabi H, Banerjee P, Liu J: Survey of wireless indoor positioning techniques and systems. IEEE Trans Syst Man Cybern Part C Appl Rev 2007;37(6):1067-1080.
119. Curran K, Furey E, Lunney T, Santos J, Woods D, McCaughey A: An evaluation of indoor location determination technologies. J Location Based Serv 2011;5(2):61-78.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
