Proactive Polypharmacy Management Using Large Language Models: Opportunities to Enhance Geriatric Care.

Polypharmacy remains an important challenge for patients with extensive medical complexity. Given the primary care shortage and the increasing aging population, effective polypharmacy management is crucial to manage the increasing burden of care. The capacity of large language model (LLM)-based artificial intelligence to aid in polypharmacy management has yet to be evaluated. Here, we evaluate ChatGPT's performance in polypharmacy management via its deprescribing decisions in standardized clinical vignettes. We inputted several clinical vignettes originally from a study of general practicioners' deprescribing decisions into ChatGPT 3.5, a publicly available LLM, and evaluated its capacity for yes/no binary deprescribing decisions as well as list-based prompts in which the model was prompted to choose which of several medications to deprescribe. We recorded ChatGPT responses to yes/no binary deprescribing prompts and the number and types of medications deprescribed. In yes/no binary deprescribing decisions, ChatGPT universally recommended deprescribing medications regardless of ADL status in patients with no overlying CVD history; in patients with CVD history, ChatGPT's answers varied by technical replicate. Total number of medications deprescribed ranged from 2.67 to 3.67 (out of 7) and did not vary with CVD status, but increased linearly with severity of ADL impairment. Among medication types, ChatGPT preferentially deprescribed pain medications. ChatGPT's deprescribing decisions vary along the axes of ADL status, CVD history, and medication type, indicating some concordance of internal logic between general practitioners and the model. These results indicate that specifically trained LLMs may provide useful clinical support in polypharmacy management for primary care physicians.

Models and approaches for building knowledge translation capacity and capability in health services: a scoping review.

BACKGROUND: Building healthcare service and health professionals' capacity and capability to rapidly translate research evidence into health practice is critical to the effectiveness and sustainability of healthcare systems. This review scoped the literature describing programmes to build knowledge translation capacity and capability in health professionals and healthcare services, and the evidence supporting these. METHODS: This scoping review was undertaken using the Joanna Briggs Institute scoping review methodology. Four research databases (Ovid MEDLINE, CINAHL, Embase, and PsycInfo) were searched using a pre-determined strategy. Eligible studies described a programme implemented in healthcare settings to build health professional or healthcare service knowledge translation capacity and capability. Abstracts and full texts considered for inclusion were screened by two researchers. Data from included papers were extracted using a bespoke tool informed by the scoping review questions. RESULTS: Database searches yielded 10,509 unique citations, of which 136 full texts were reviewed. Thirty-four papers were included, with three additional papers identified on citation searching, resulting in 37 papers describing 34 knowledge translation capability building programmes. Programmes were often multifaceted, comprising a combination of two or more strategies including education, dedicated implementation support roles, strategic research-practice partnerships and collaborations, co-designed knowledge translation capability building programmes, and dedicated funding for knowledge translation. Many programmes utilised experiential and collaborative learning, and targeted either individual, team, organisational, or system levels of impact. Twenty-seven programmes were evaluated formally using one or more data collection methods. Outcomes measured varied significantly and included participant self-reported outcomes, perceived barriers and enablers of knowledge translation, milestone achievement and behaviour change. All papers reported that programme objectives were achieved to varying degrees. CONCLUSIONS: Knowledge translation capacity and capability building programmes in healthcare settings are multifaceted, often include education to facilitate experiential and collaborative learning, and target individual, team, organisational, or supra-organisational levels of impact. Although measured differently across the programmes, the outcomes were positive. The sustainability of programmes and outcomes may be undermined by the lack of long-term funding and inconsistent evaluation. Future research is required to develop evidence-informed frameworks to guide methods and outcome measures for short-, medium- and longer-term programme evaluation at the different structural levels.

Assessing the Utility of ChatGPT Throughout the Entire Clinical Workflow: Development and Usability Study.

BACKGROUND: Large language model (LLM)-based artificial intelligence chatbots direct the power of large training data sets toward successive, related tasks as opposed to single-ask tasks, for which artificial intelligence already achieves impressive performance. The capacity of LLMs to assist in the full scope of iterative clinical reasoning via successive prompting, in effect acting as artificial physicians, has not yet been evaluated. OBJECTIVE: This study aimed to evaluate ChatGPT's capacity for ongoing clinical decision support via its performance on standardized clinical vignettes. METHODS: We inputted all 36 published clinical vignettes from the Merck Sharpe & Dohme (MSD) Clinical Manual into ChatGPT and compared its accuracy on differential diagnoses, diagnostic testing, final diagnosis, and management based on patient age, gender, and case acuity. Accuracy was measured by the proportion of correct responses to the questions posed within the clinical vignettes tested, as calculated by human scorers. We further conducted linear regression to assess the contributing factors toward ChatGPT's performance on clinical tasks. RESULTS: ChatGPT achieved an overall accuracy of 71.7% (95% CI 69.3%-74.1%) across all 36 clinical vignettes. The LLM demonstrated the highest performance in making a final diagnosis with an accuracy of 76.9% (95% CI 67.8%-86.1%) and the lowest performance in generating an initial differential diagnosis with an accuracy of 60.3% (95% CI 54.2%-66.6%). Compared to answering questions about general medical knowledge, ChatGPT demonstrated inferior performance on differential diagnosis (ß=-15.8%; P<.001) and clinical management (ß=-7.4%; P=.02) question types. CONCLUSIONS: ChatGPT achieves impressive accuracy in clinical decision-making, with increasing strength as it gains more clinical information at its disposal. In particular, ChatGPT demonstrates the greatest accuracy in tasks of final diagnosis as compared to initial diagnosis. Limitations include possible model hallucinations and the unclear composition of ChatGPT's training data set.

Evaluating GPT as an Adjunct for Radiologic Decision Making: GPT-4 Versus GPT-3.5 in a Breast Imaging Pilot.

OBJECTIVE: Despite rising popularity and performance, studies evaluating the use of large language models for clinical decision support are lacking. Here, we evaluate ChatGPT (Generative Pre-trained Transformer)-3.5 and GPT-4's (OpenAI, San Francisco, California) capacity for clinical decision support in radiology via the identification of appropriate imaging services for two important clinical presentations: breast cancer screening and breast pain. METHODS: We compared ChatGPT's responses to the ACR Appropriateness Criteria for breast pain and breast cancer screening. Our prompt formats included an open-ended (OE) and a select all that apply (SATA) format. Scoring criteria evaluated whether proposed imaging modalities were in accordance with ACR guidelines. Three replicate entries were conducted for each prompt, and the average of these was used to determine final scores. RESULTS: Both ChatGPT-3.5 and ChatGPT-4 achieved an average OE score of 1.830 (out of 2) for breast cancer screening prompts. ChatGPT-3.5 achieved a SATA average percentage correct of 88.9%, compared with ChatGPT-4's average percentage correct of 98.4% for breast cancer screening prompts. For breast pain, ChatGPT-3.5 achieved an average OE score of 1.125 (out of 2) and a SATA average percentage correct of 58.3%, as compared with an average OE score of 1.666 (out of 2) and a SATA average percentage correct of 77.7%. DISCUSSION: Our results demonstrate the eventual feasibility of using large language models like ChatGPT for radiologic decision making, with the potential to improve clinical workflow and responsible use of radiology services. More use cases and greater accuracy are necessary to evaluate and implement such tools.

Hip resurfacing arthroplasty as an alternative to total hip arthroplasty in patients aged under 40 years.

Aims: The aims of the study were to report for a cohort aged younger than 40 years: 1) indications for HRA; 2) patient-reported outcomes in terms of the modified Harris Hip Score (HHS); 3) dislocation rate; and 4) revision rate. Methods: This retrospective analysis identified 267 hips from 224 patients who underwent an hip resurfacing arthroplasty (HRA) from a single fellowship-trained surgeon using the direct lateral approach between 2007 and 2019. Inclusion criteria was minimum two-year follow-up, and age younger than 40 years. Patients were followed using a prospectively maintained institutional database. Results: A total of 217 hips (81%) were included for follow-up analysis at a mean of 3.8 years. Of the 23 females who underwent HRA, none were revised, and the median head size was 46 mm (compared to 50 mm for males). The most common indication for HRA was femoroacetabular impingement syndrome (n = 133), and avascular necrosis ( (n = 53). Mean postoperative HHS was 100 at two and five years. No dislocations occurred. A total of four hips (1.8%) required reoperation for resection of heterotopic ossification, removal of components for infection, and subsidence with loosening. The overall revision rate was 0.9%. Conclusion: For younger patients with higher functional expectations and increased lifetime risk for revision, HRA is an excellent bone preserving intervention carrying low complication rates, revision rates, and excellent patient outcomes without lifetime restrictions allowing these patients to return to activity and sport. Thus, in younger male patients with end-stage hip disease and higher demands, referral to a high-volume HRA surgeon should be considered.

Massively parallel base editing to map variant effects in human hematopoiesis.

Systematic evaluation of the impact of genetic variants is critical for the study and treatment of human physiology and disease. While specific mutations can be introduced by genome engineering, we still lack scalable approaches that are applicable to the important setting of primary cells, such as blood and immune cells. Here, we describe the development of massively parallel base-editing screens in human hematopoietic stem and progenitor cells. Such approaches enable functional screens for variant effects across any hematopoietic differentiation state. Moreover, they allow for rich phenotyping through single-cell RNA sequencing readouts and separately for characterization of editing outcomes through pooled single-cell genotyping. We efficiently design improved leukemia immunotherapy approaches, comprehensively identify non-coding variants modulating fetal hemoglobin expression, define mechanisms regulating hematopoietic differentiation, and probe the pathogenicity of uncharacterized disease-associated variants. These strategies will advance effective and high-throughput variant-to-function mapping in human hematopoiesis to identify the causes of diverse diseases.

Assessing the Utility of ChatGPT Throughout the Entire Clinical Workflow.

IMPORTANCE: Large language model (LLM) artificial intelligence (AI) chatbots direct the power of large training datasets towards successive, related tasks, as opposed to single-ask tasks, for which AI already achieves impressive performance. The capacity of LLMs to assist in the full scope of iterative clinical reasoning via successive prompting, in effect acting as virtual physicians, has not yet been evaluated. OBJECTIVE: To evaluate ChatGPT's capacity for ongoing clinical decision support via its performance on standardized clinical vignettes. DESIGN: We inputted all 36 published clinical vignettes from the Merck Sharpe & Dohme (MSD) Clinical Manual into ChatGPT and compared accuracy on differential diagnoses, diagnostic testing, final diagnosis, and management based on patient age, gender, and case acuity. SETTING: ChatGPT, a publicly available LLM. PARTICIPANTS: Clinical vignettes featured hypothetical patients with a variety of age and gender identities, and a range of Emergency Severity Indices (ESIs) based on initial clinical presentation. EXPOSURES: MSD Clinical Manual vignettes. MAIN OUTCOMES AND MEASURES: We measured the proportion of correct responses to the questions posed within the clinical vignettes tested. RESULTS: ChatGPT achieved 71.7% (95% CI, 69.3% to 74.1%) accuracy overall across all 36 clinical vignettes. The LLM demonstrated the highest performance in making a final diagnosis with an accuracy of 76.9% (95% CI, 67.8% to 86.1%), and the lowest performance in generating an initial differential diagnosis with an accuracy of 60.3% (95% CI, 54.2% to 66.6%). Compared to answering questions about general medical knowledge, ChatGPT demonstrated inferior performance on differential diagnosis (ß=-15.8%, p<0.001) and clinical management (ß=-7.4%, p=0.02) type questions. CONCLUSIONS AND RELEVANCE: ChatGPT achieves impressive accuracy in clinical decision making, with particular strengths emerging as it has more clinical information at its disposal.

Patient-Specific Safe Zones for Acetabular Component Positioning in Total Hip Arthroplasty: Mathematically Accounting for Spinopelvic Biomechanics.

BACKGROUND: Despite a growing understanding of spinopelvic biomechanics in total hip arthroplasty (THA), there is no validated approach for executing patient-specific acetabular component positioning. The purpose of this study was to (1) validate quantitative, patient-specific acetabular "safe zone" component positioning from spinopelvic parameters and (2) characterize differences between quantitative patient-specific acetabular targets and qualitative hip-spine classification targets. METHODS: From 2,457 consecutive primary THA patients, 22 (0.88%) underwent revision for instability. Spinopelvic parameters were measured prior to index THA. Acetabular position was measured following index and revision arthroplasty. Using a mathematical proof, we developed an open-source tool translating a surgeon-selected, preoperative standing acetabular target to a patient-specific safe zone intraoperative acetabular target. Difference between the patient-specific safe zone and the actual component position was compared before and after revision. Hip-spine classification targets were compared to patient-specific safe zone targets. RESULTS: Of the 22 who underwent revision, none dislocated at follow-up (4.6 [range, 1 to 6.9]). Patient-specific safe zone targets differed from prerevision acetabular component position by 9.1 ± 4.2° inclination/13.3 ± 6.7° version; after revision, the mean difference was 3.2 ± 3.0° inclination/5.3 ± 2.7° version. Differences between patient-specific safe zones and the median and extremes of recommended hip-spine classification targets were 2.2 ± 1.9° inclination/5.6 ± 3.7° version and 3.0 ± 2.3° inclination/7.9 ± 3.5° version, respectively. CONCLUSION: A mathematically derived, patient-specific approach accommodating spinopelvic biomechanics for acetabular component positioning was validated by approximating revised, now-stable hips within 5° version and 3° inclination. These patient-specific safe zones augment the hip-spine classification with prescriptive quantitative targets for nuanced preoperative planning.

Evaluating ChatGPT as an Adjunct for Radiologic Decision-Making.

BACKGROUND: ChatGPT, a popular new large language model (LLM) built by OpenAI, has shown impressive performance in a number of specialized applications. Despite the rising popularity and performance of AI, studies evaluating the use of LLMs for clinical decision support are lacking. PURPOSE: To evaluate ChatGPT's capacity for clinical decision support in radiology via the identification of appropriate imaging services for two important clinical presentations: breast cancer screening and breast pain. MATERIALS AND METHODS: We compared ChatGPT's responses to the American College of Radiology (ACR) Appropriateness Criteria for breast pain and breast cancer screening. Our prompt formats included an open-ended (OE) format, where ChatGPT was asked to provide the single most appropriate imaging procedure, and a select all that apply (SATA) format, where ChatGPT was given a list of imaging modalities to assess. Scoring criteria evaluated whether proposed imaging modalities were in accordance with ACR guidelines. RESULTS: ChatGPT achieved an average OE score of 1.83 (out of 2) and a SATA average percentage correct of 88.9% for breast cancer screening prompts, and an average OE score of 1.125 (out of 2) and a SATA average percentage correct of 58.3% for breast pain prompts. CONCLUSION: Our results demonstrate the feasibility of using ChatGPT for radiologic decision making, with the potential to improve clinical workflow and responsible use of radiology services.

Concerns surrounding application of artificial intelligence in hip and knee arthroplasty : a review of literature and recommendations for meaningful adoption.

Literature surrounding artificial intelligence (AI)-related applications for hip and knee arthroplasty has proliferated. However, meaningful advances that fundamentally transform the practice and delivery of joint arthroplasty are yet to be realized, despite the broad range of applications as we continue to search for meaningful and appropriate use of AI. AI literature in hip and knee arthroplasty between 2018 and 2021 regarding image-based analyses, value-based care, remote patient monitoring, and augmented reality was reviewed. Concerns surrounding meaningful use and appropriate methodological approaches of AI in joint arthroplasty research are summarized. Of the 233 AI-related orthopaedics articles published, 178 (76%) constituted original research, while the rest consisted of editorials or reviews. A total of 52% of original AI-related research concerns hip and knee arthroplasty (n = 92), and a narrative review is described. Three studies were externally validated. Pitfalls surrounding present-day research include conflating vernacular ("AI/machine learning"), repackaging limited registry data, prematurely releasing internally validated prediction models, appraising model architecture instead of inputted data, withholding code, and evaluating studies using antiquated regression-based guidelines. While AI has been applied to a variety of hip and knee arthroplasty applications with limited clinical impact, the future remains promising if the question is meaningful, the methodology is rigorous and transparent, the data are rich, and the model is externally validated. Simple checkpoints for meaningful AI adoption include ensuring applications focus on: administrative support over clinical evaluation and management; necessity of the advanced model; and the novelty of the question being answered.Cite this article: Bone Joint J 2022;104-B(12):1292-1303.

Meaningless Applications and Misguided Methodologies in Artificial Intelligence-Related Orthopaedic Research Propagates Hype Over Hope.

There exists great hope and hype in the literature surrounding applications of artificial intelligence (AI) to orthopaedic surgery. Between 2018 and 2021, a total of 178 AI-related articles were published in orthopaedics. However, for every 2 original research papers that apply AI to orthopaedics, a commentary or review is published (30.3%). AI-related research in orthopaedics frequently fails to provide use cases that offer the uninitiated an opportunity to appraise the importance of AI by studying meaningful questions, evaluating unknown hypotheses, or analyzing quality data. The hype perpetuates a feed-forward cycle that relegates AI to a meaningless buzzword by rewarding those with nascent understanding and rudimentary technical knowhow into committing several basic errors: (1) inappropriately conflating vernacular ("AI/machine learning"), (2) repackaging registry data, (3) prematurely releasing internally validated algorithms, (4) overstating the "black box phenomenon" by failing to provide weighted analysis, (5) claiming to evaluate AI rather than the data itself, and (6) withholding full model architecture code. Relevant AI-specific guidelines are forthcoming, but forced application of the original Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis guidelines designed for regression analyses is irrelevant and misleading. To safeguard meaningful use, AI-related research efforts in orthopaedics should be (1) directed toward administrative support over clinical evaluation and management, (2) require the use of the advanced model, and (3) answer a question that was previously unknown, unanswered, or unquantifiable.

Clinical and pathological characterization of FLNC-related myofibrillar myopathy caused by founder variant c.8129G>A in Hong Kong Chinese.

FLNC-related myofibrillar myopathy could manifest as autosomal dominant late-onset slowly progressive proximal muscle weakness; involvements of cardiac and/or respiratory functions are common. We describe 34 patients in nine families of FLNC-related myofibrillar myopathy in Hong Kong ethnic Chinese diagnosed over the last 12 years, in whom the same pathogenic variant c.8129G>A (p.Trp2710*) was detected. Twenty-six patients were symptomatic when diagnosed; four patients died of pneumonia and/or respiratory failure. Abnormal amorphous material or granulofilamentous masses were detected in half of the cases, with mitochondrial abnormalities noted in two-thirds. We also show by haplotype analysis the founder effect associated with this Hong Kong variant, which might have occurred 42 to 71 generations ago or around Tang and Song dynasties, and underlain a higher incidence of myofibrillar myopathy among Hong Kong Chinese. The late-onset nature and slowly progressive course of the highly penetrant condition could have significant impact on the family members, and an early diagnosis could benefit the whole family. Considering another neighboring founder variant in FLNC in German patients, we advocate development of specific therapies such as chaperone-based or antisense oligonucleotide strategies for this particular type of myopathy.

Discrepancy in the epidemiology of nondiabetic chronic renal insufficiency and end-stage renal disease in black and white Americans: the third National Health and Nutrition Examination Survey and United States Renal Data System.

BACKGROUND: Epidemiologic data regarding the prevalence of chronic renal insufficiency (CRI) [from the third National Health and Nutrition Examination Survey (NHANES III)] and the incidence of end-stage renal disease (ESRD) [from the United States Renal Data System (USRDS)] are available. However, reconciliation of these separate particulars has not been performed objectively. The present work examines the epidemiology of CRI of nondiabetic etiology and ESRD in black and white Americans aged 20 years or greater. METHODS: Based on the incidence of ESRD in the study population (USRDS), the numbers of subjects with decreased Modification of Diet in Renal Disease (MDRD) glomerular filtration rate (GFR) <80 ml/min/1.73 m(2), <60 ml/min/1.73 m(2) and <30 ml/ min/1.73 m(2) in 1991 (on December 31 1991) were mathematically obtained based on a linear model of GFR decline. Similarly, the corresponding estimated prevalence figures of CRI were derived based on analyses of NHANES III data and the 1991 census counts of black and white Americans (aged 20 years or more). Unadjusted and adjusted (correcting for calibration differences between the NHANES III and MDRD laboratory) prevalences were calculated. Subsequently, the prevalence of different degrees of CRI based on the incidence of ESRD (USRDS) was compared to the corresponding figures of the estimated prevalence of CRI (NHANES III). RESULTS: By analyses of USRDS data, on December 31 1991, the prevalence of different degrees of reduced GFR in the study population was estimated to be as follows: 396,863 subjects with GFR <80 ml/min/1.73 m(2); 272,932 subjects with GFR <60 ml/min/1.73 m(2), and 115,065 subjects with GFR <30 ml/min/1.73 m(2). Using actual NHANES III creatinine values, the prevalence of different degrees of CRI in the study population was estimated as follows: 92,595,211 people with GFR <80 ml/min/1.73 m(2); 20,754,099 people with GFR <60 ml/min/1.73 m(2), and 415,082 people with GFR <30 ml/min/1.73 m(2). The data suggest that approximately 0.43% of subjects with GFR <80 ml/min/1.73 m(2), 1.32% of subjects with GFR <60 ml/min/1.73 m(2) and 27.72% of subjects with GFR <30 ml/ min/1.73 m(2) reached ESRD (USRDS). Using adjusted NHANES III creatinine values (downwardly correcting the NHANES III creatinine values to account for calibration differences with the MDRD measurements), the prevalence of different degrees of CRI in the study population was estimated as follows: 28,512,939 people with MDRD GFR <80 ml/min/1.73 m(2) (17.86%); 5,364,136 people with MDRD GFR <60 ml/min/1.73 m(2) (3.36%), and 255,435 people with MDRD GFR <30 ml/min/1.73 m(2) (0.16%). Of these, about 1.39% of the people with MDRD GFR <80 ml/min/1.73 m(2), 5.09% of the people with MDRD GFR <60 ml/min/1.73 m(2) and 45.07% of the people with MDRD GFR <30 ml/min/1.73 m(2) in 1991 reached ESRD. CONCLUSION: There is a major discrepancy in the epidemiology of nondiabetic CRI and ESRD amongst black and white Americans. The reasons for this need further study.

Slowing the progression of chronic renal failure: economic benefits and patients' perspectives.

Because of the predicted increase in end-stage renal disease (ESRD) incidence (projected increase from 1998 to 2010; 86,825 to 172,667), prevalence (projected increase from 1998 to 2010; 326,217 to 661,330), and cost (total cost based on 1998 ratio of Medicare versus non-Medicare cost; $16.74 billion in 1998 to $39.35 billion in 2010), a cohesive national effort is needed to develop strategies to slow the progression of chronic renal failure (CRF). The question arises to how much reduction in the progression of CRF would lead to a meaningful decrease in the prevalence and cost of ESRD. There are no objective data that show the economic impact of slowing the progression of CRF. We developed a mathematical model to assess the economic impact of decreasing the progression of CRF by 10%, 20%, and 30%. US Renal Data System (USRDS) projections were used to model the rate of increase in ESRD incidence and prevalence. Glomerular filtration rate (GFR) at the initiation of ESRD therapy and cost per patient-year were based on USRDS data. The average decline in GFR in subjects with CRF was estimated to be 7.56 mL/min/y. All dollar savings reflect 1998 costs, discounted for the future at 3% per annum. We also determined how much slowing of the progression of CRF is important from patients' perspectives by means of a written questionnaire (which inquired about willingness to go on a restricted diet, take six extra medications per day, and make six extra office visits per year) and calculation of the pre-ESRD time gained for different degrees of reduction in the progression of CRF. If the rate of decline in GFR decreased by 10%, 20%, and 30% after December 31, 1999, in all patients with GFRs of 60 mL/min or less, cumulative direct healthcare savings through 2010 would equal approximately $18.56, $39.02, and $60.61 billion, respectively. For a 10%, 20%, and 30% decrease in the rate of decline in GFR in all patients with a GFR of 30 mL/min or less, estimated cumulative savings through 2010 equal $9.06, $19.98, and $33.37 billion, respectively. Responses to the questionnaire showed that approximately 79% of subjects with CRF (n = 113) perceived a few weeks' dialysis-free period significant (P < or = 0.0001), a period corresponding to a 10% reduction in the rate of decline in GFR. Our data suggest that the cumulative economic impact of slowing the progression of CRF, even by as little as 10%, would be staggering. They provide strong support for the development and implementation of intensive reno-protective efforts beginning at the early stages of chronic renal disease and continued throughout its course.