1
|
Ryvlin J, Kim SW, De la Garza Ramos R, Hamad M, Stock A, Owolo E, Fourman MS, Eleswarapu A, Gelfand Y, Murthy S, Yassari R. External Validation of an Online Wound Infection and Wound Reoperation Risk Calculator After Metastatic Spinal Tumor Surgery. World Neurosurg 2024; 185:e351-e356. [PMID: 38342175 DOI: 10.1016/j.wneu.2024.02.005] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/15/2023] [Revised: 02/01/2024] [Accepted: 02/02/2024] [Indexed: 02/13/2024]
Abstract
STUDY DESIGN This was a single-institutional retrospective cohort study. OBJECTIVE Wound infections are common following spine metastasis surgery and can result in unplanned reoperations. A recent study published an online wound complication risk calculator but has not yet undergone external validation. Our aim was to evaluate the accuracy of this risk calculator in predicting 30-day wound infections and 30-day wound reoperations using our operative spine metastasis population. METHODS An internal operative database was used to identify patients between 2012 and 2022. The primary outcomes were 1) any surgical site infection and 2) wound-related revision surgery within 30 days following surgery. Patient details were manually collected from electronic medical records and entered into the calculator to determine predicted complication risk percentages. Predicted risks were compared to observed outcomes using receiver operator characteristic (ROC) curves with areas under the curve (AUC). RESULTS A total of 153 patients were included. The observed 30-day postoperative wound infection incidence was 5% while the predicted wound infection incidence was 6%. In ROC analysis, good discrimination was found for the wound infection model (AUC = 0.737; P = 0.024). The observed wound reoperation rate was 5% and the predicted wound reoperation rate was 6%. ROC analysis demonstrated poor discrimination for wound reoperations (AUC = 0.559; P = 0.597). CONCLUSIONS The online wound-related risk calculator was found to accurately predict wound infections but not wound reoperations within our metastatic spine surgery cohort. We suggest that the model may be clinically useful despite underlying population differences, but further work must be done to generate and validate accurate prediction tools.
Collapse
Affiliation(s)
- Jessica Ryvlin
- Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA.
| | - Seung Woo Kim
- Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA
| | - Rafael De la Garza Ramos
- Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA
| | - Mousa Hamad
- Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA
| | - Ariel Stock
- Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA
| | - Edwin Owolo
- Department of Orthopedic Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA
| | | | | | - Yaroslav Gelfand
- Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA
| | - Saikiran Murthy
- Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA
| | - Reza Yassari
- Department of Neurological Surgery, Montefiore Medical Center Albert Einstein College of Medicine, Bronx, New York, New York, USA
| |
Collapse
|
2
|
Tanner KT, Keogh RH, Coupland CAC, Hippisley-Cox J, Diaz-Ordaz K. Dynamic updating of clinical survival prediction models in a changing environment. Diagn Progn Res 2023; 7:24. [PMID: 38082429 PMCID: PMC10714456 DOI: 10.1186/s41512-023-00163-z] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/16/2023] [Accepted: 10/17/2023] [Indexed: 01/31/2024] Open
Abstract
BACKGROUND Over time, the performance of clinical prediction models may deteriorate due to changes in clinical management, data quality, disease risk and/or patient mix. Such prediction models must be updated in order to remain useful. In this study, we investigate dynamic model updating of clinical survival prediction models. In contrast to discrete or one-time updating, dynamic updating refers to a repeated process for updating a prediction model with new data. We aim to extend previous research which focused largely on binary outcome prediction models by concentrating on time-to-event outcomes. We were motivated by the rapidly changing environment seen during the COVID-19 pandemic where mortality rates changed over time and new treatments and vaccines were introduced. METHODS We illustrate three methods for dynamic model updating: Bayesian dynamic updating, recalibration, and full refitting. We use a simulation study to compare performance in a range of scenarios including changing mortality rates, predictors with low prevalence and the introduction of a new treatment. Next, the updating strategies were applied to a model for predicting 70-day COVID-19-related mortality using patient data from QResearch, an electronic health records database from general practices in the UK. RESULTS In simulated scenarios with mortality rates changing over time, all updating methods resulted in better calibration than not updating. Moreover, dynamic updating outperformed ad hoc updating. In the simulation scenario with a new predictor and a small updating dataset, Bayesian updating improved the C-index over not updating and refitting. In the motivating example with a rare outcome, no single updating method offered the best performance. CONCLUSIONS We found that a dynamic updating process outperformed one-time discrete updating in the simulations. Bayesian updating offered good performance overall, even in scenarios with new predictors and few events. Intercept recalibration was effective in scenarios with smaller sample size and changing baseline hazard. Refitting performance depended on sample size and produced abrupt changes in hazard ratio estimates between periods.
Collapse
Affiliation(s)
- Kamaryn T Tanner
- Department of Medical Statistics, London School of Hygiene and Tropical Medicine, London, WC1E 7HT, UK.
| | - Ruth H Keogh
- Department of Medical Statistics, London School of Hygiene and Tropical Medicine, London, WC1E 7HT, UK
| | - Carol A C Coupland
- Nuffield Department of Primary Health Care Sciences, University of Oxford, Oxford, OX2 6HT, UK
- Centre for Academic Primary Care, University of Nottingham, Nottingham, NG7 2UH, UK
| | - Julia Hippisley-Cox
- Nuffield Department of Primary Health Care Sciences, University of Oxford, Oxford, OX2 6HT, UK
| | - Karla Diaz-Ordaz
- Department of Statistical Science, University College London, London, WC1E 6BT, UK
| |
Collapse
|
3
|
van Dijk WB, Leeuwenberg AM, Grobbee DE, Siregar S, Houterman S, Daeter EJ, de Vries MC, Groenwold RHH, Schuit E. Dynamics in cardiac surgery: trends in population characteristics and the performance of the EuroSCORE II over time. Eur J Cardiothorac Surg 2023; 64:ezad301. [PMID: 37672025 PMCID: PMC10504469 DOI: 10.1093/ejcts/ezad301] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/27/2022] [Revised: 06/21/2023] [Accepted: 09/05/2023] [Indexed: 09/07/2023] Open
Abstract
OBJECTIVES The aim of this study was to investigate the performance of the EuroSCORE II over time and dynamics in values of predictors included in the model. METHODS A cohort study was performed using data from the Netherlands Heart Registration. All cardiothoracic surgical procedures performed between 1 January 2013 and 31 December 2019 were included for analysis. Performance of the EuroSCORE II was assessed across 3-month intervals in terms of calibration and discrimination. For subgroups of major surgical procedures, performance of the EuroSCORE II was assessed across 12-month time intervals. Changes in values of individual EuroSCORE II predictors over time were assessed graphically. RESULTS A total of 103 404 cardiothoracic surgical procedures were included. Observed mortality risk ranged between 1.9% [95% confidence interval (CI) 1.6-2.4] and 3.6% (95% CI 2.6-4.4) across 3-month intervals, while the mean predicted mortality risk ranged between 3.4% (95% CI 3.3-3.6) and 4.2% (95% CI 3.9-4.6). The corresponding observed:expected ratios ranged from 0.50 (95% CI 0.46-0.61) to 0.95 (95% CI 0.74-1.16). Discriminative performance in terms of the c-statistic ranged between 0.82 (95% CI 0.78-0.89) and 0.89 (95% CI 0.87-0.93). The EuroSCORE II consistently overestimated mortality compared to observed mortality. This finding was consistent across all major cardiothoracic surgical procedures. Distributions of values of individual predictors varied broadly across predictors over time. Most notable trends were a decrease in elective surgery from 75% to 54% and a rise in patients with no or New York Heart Association I class heart failure from 27% to 33%. CONCLUSIONS The EuroSCORE II shows good discriminative performance, but consistently overestimates mortality risks of all types of major cardiothoracic surgical procedures in the Netherlands.
Collapse
Affiliation(s)
- Wouter B van Dijk
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, Netherlands
| | - Artuur M Leeuwenberg
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, Netherlands
| | - Diederick E Grobbee
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, Netherlands
| | - Sabrina Siregar
- Department of Cardiothoracic Surgery, Erasmus Medical Center, Erasmus University, Rotterdam, Netherlands
| | | | - Edgar J Daeter
- Netherlands Heart Registration, Utrecht, Netherlands
- Department of Cardiothoracic Surgery, St. Antonius Hospital, Nieuwegein, Netherlands
| | - Martine C de Vries
- Department of Medical Ethics and Health Law, Leiden University Medical Center, Leiden University, Leiden, Netherlands
| | - Rolf H H Groenwold
- Department of Clinical Epidemiology, Leiden University Medical Center, Leiden University, Leiden, Netherlands
- Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden University, Leiden, Netherlands
| | - Ewoud Schuit
- Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, Netherlands
| |
Collapse
|
4
|
Yadava OP. Predicting the unpredictable in cardiothoracic surgery. Indian J Thorac Cardiovasc Surg 2023; 39:109-111. [PMID: 36785611 PMCID: PMC9918619 DOI: 10.1007/s12055-023-01478-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/27/2023] Open
|
5
|
Binuya MAE, Engelhardt EG, Schats W, Schmidt MK, Steyerberg EW. Methodological guidance for the evaluation and updating of clinical prediction models: a systematic review. BMC Med Res Methodol 2022; 22:316. [PMID: 36510134 PMCID: PMC9742671 DOI: 10.1186/s12874-022-01801-8] [Citation(s) in RCA: 27] [Impact Index Per Article: 13.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/02/2022] [Accepted: 11/22/2022] [Indexed: 12/14/2022] Open
Abstract
BACKGROUND Clinical prediction models are often not evaluated properly in specific settings or updated, for instance, with information from new markers. These key steps are needed such that models are fit for purpose and remain relevant in the long-term. We aimed to present an overview of methodological guidance for the evaluation (i.e., validation and impact assessment) and updating of clinical prediction models. METHODS We systematically searched nine databases from January 2000 to January 2022 for articles in English with methodological recommendations for the post-derivation stages of interest. Qualitative analysis was used to summarize the 70 selected guidance papers. RESULTS Key aspects for validation are the assessment of statistical performance using measures for discrimination (e.g., C-statistic) and calibration (e.g., calibration-in-the-large and calibration slope). For assessing impact or usefulness in clinical decision-making, recent papers advise using decision-analytic measures (e.g., the Net Benefit) over simplistic classification measures that ignore clinical consequences (e.g., accuracy, overall Net Reclassification Index). Commonly recommended methods for model updating are recalibration (i.e., adjustment of intercept or baseline hazard and/or slope), revision (i.e., re-estimation of individual predictor effects), and extension (i.e., addition of new markers). Additional methodological guidance is needed for newer types of updating (e.g., meta-model and dynamic updating) and machine learning-based models. CONCLUSION Substantial guidance was found for model evaluation and more conventional updating of regression-based models. An important development in model evaluation is the introduction of a decision-analytic framework for assessing clinical usefulness. Consensus is emerging on methods for model updating.
Collapse
Affiliation(s)
- M. A. E. Binuya
- grid.430814.a0000 0001 0674 1393Division of Molecular Pathology, the Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands ,grid.10419.3d0000000089452978Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden, The Netherlands ,grid.10419.3d0000000089452978Department of Clinical Genetics, Leiden University Medical Center, Leiden, The Netherlands
| | - E. G. Engelhardt
- grid.430814.a0000 0001 0674 1393Division of Molecular Pathology, the Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands ,grid.430814.a0000 0001 0674 1393Division of Psychosocial Research and Epidemiology, the Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands
| | - W. Schats
- grid.430814.a0000 0001 0674 1393Scientific Information Service, The Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands
| | - M. K. Schmidt
- grid.430814.a0000 0001 0674 1393Division of Molecular Pathology, the Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands ,grid.10419.3d0000000089452978Department of Clinical Genetics, Leiden University Medical Center, Leiden, The Netherlands
| | - E. W. Steyerberg
- grid.10419.3d0000000089452978Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden, The Netherlands
| |
Collapse
|
6
|
van Klaveren D, Zanos TP, Nelson J, Levy TJ, Park JG, Retel Helmrich IRA, Rietjens JAC, Basile MJ, Hajizadeh N, Lingsma HF, Kent DM. Prognostic models for COVID-19 needed updating to warrant transportability over time and space. BMC Med 2022; 20:456. [PMID: 36424619 PMCID: PMC9686462 DOI: 10.1186/s12916-022-02651-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/22/2022] [Accepted: 11/04/2022] [Indexed: 11/25/2022] Open
Abstract
BACKGROUND Supporting decisions for patients who present to the emergency department (ED) with COVID-19 requires accurate prognostication. We aimed to evaluate prognostic models for predicting outcomes in hospitalized patients with COVID-19, in different locations and across time. METHODS We included patients who presented to the ED with suspected COVID-19 and were admitted to 12 hospitals in the New York City (NYC) area and 4 large Dutch hospitals. We used second-wave patients who presented between September and December 2020 (2137 and 3252 in NYC and the Netherlands, respectively) to evaluate models that were developed on first-wave patients who presented between March and August 2020 (12,163 and 5831). We evaluated two prognostic models for in-hospital death: The Northwell COVID-19 Survival (NOCOS) model was developed on NYC data and the COVID Outcome Prediction in the Emergency Department (COPE) model was developed on Dutch data. These models were validated on subsequent second-wave data at the same site (temporal validation) and at the other site (geographic validation). We assessed model performance by the Area Under the receiver operating characteristic Curve (AUC), by the E-statistic, and by net benefit. RESULTS Twenty-eight-day mortality was considerably higher in the NYC first-wave data (21.0%), compared to the second-wave (10.1%) and the Dutch data (first wave 10.8%; second wave 10.0%). COPE discriminated well at temporal validation (AUC 0.82), with excellent calibration (E-statistic 0.8%). At geographic validation, discrimination was satisfactory (AUC 0.78), but with moderate over-prediction of mortality risk, particularly in higher-risk patients (E-statistic 2.9%). While discrimination was adequate when NOCOS was tested on second-wave NYC data (AUC 0.77), NOCOS systematically overestimated the mortality risk (E-statistic 5.1%). Discrimination in the Dutch data was good (AUC 0.81), but with over-prediction of risk, particularly in lower-risk patients (E-statistic 4.0%). Recalibration of COPE and NOCOS led to limited net benefit improvement in Dutch data, but to substantial net benefit improvement in NYC data. CONCLUSIONS NOCOS performed moderately worse than COPE, probably reflecting unique aspects of the early pandemic in NYC. Frequent updating of prognostic models is likely to be required for transportability over time and space during a dynamic pandemic.
Collapse
Affiliation(s)
- David van Klaveren
- Department of Public Health, Erasmus MC University Medical Center Rotterdam, Dr. Molewaterplein 50, 3015 GE, Rotterdam, The Netherlands. .,Predictive Analytics and Comparative Effectiveness Center, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, Boston, USA.
| | - Theodoros P Zanos
- Institute of Bioelectronic Medicine, Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY, USA
| | - Jason Nelson
- Predictive Analytics and Comparative Effectiveness Center, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, Boston, USA
| | - Todd J Levy
- Institute of Bioelectronic Medicine, Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY, USA
| | - Jinny G Park
- Predictive Analytics and Comparative Effectiveness Center, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, Boston, USA
| | - Isabel R A Retel Helmrich
- Department of Public Health, Erasmus MC University Medical Center Rotterdam, Dr. Molewaterplein 50, 3015 GE, Rotterdam, The Netherlands
| | - Judith A C Rietjens
- Department of Public Health, Erasmus MC University Medical Center Rotterdam, Dr. Molewaterplein 50, 3015 GE, Rotterdam, The Netherlands
| | - Melissa J Basile
- Division of Pulmonary Critical Care and Sleep Medicine, Department of Medicine, Donald and Barbara Zucker School of Medicine at Hofstra/Northwell Health, Hempstead, NY, USA
| | - Negin Hajizadeh
- Division of Pulmonary Critical Care and Sleep Medicine, Department of Medicine, Donald and Barbara Zucker School of Medicine at Hofstra/Northwell Health, Hempstead, NY, USA
| | - Hester F Lingsma
- Department of Public Health, Erasmus MC University Medical Center Rotterdam, Dr. Molewaterplein 50, 3015 GE, Rotterdam, The Netherlands
| | - David M Kent
- Predictive Analytics and Comparative Effectiveness Center, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, Boston, USA
| |
Collapse
|
7
|
Zhang X, Xue Y, Su X, Chen S, Liu K, Chen W, Liu M, Hu Y. A Transfer Learning Approach to Correct the Temporal Performance Drift of Clinical Prediction Models: Retrospective Cohort Study. JMIR Med Inform 2022; 10:e38053. [DOI: 10.2196/38053] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/17/2022] [Revised: 07/31/2022] [Accepted: 10/12/2022] [Indexed: 11/11/2022] Open
Abstract
Background
Clinical prediction models suffer from performance drift as the patient population shifts over time. There is a great need for model updating approaches or modeling frameworks that can effectively use the old and new data.
Objective
Based on the paradigm of transfer learning, we aimed to develop a novel modeling framework that transfers old knowledge to the new environment for prediction tasks, and contributes to performance drift correction.
Methods
The proposed predictive modeling framework maintains a logistic regression–based stacking ensemble of 2 gradient boosting machine (GBM) models representing old and new knowledge learned from old and new data, respectively (referred to as transfer learning gradient boosting machine [TransferGBM]). The ensemble learning procedure can dynamically balance the old and new knowledge. Using 2010-2017 electronic health record data on a retrospective cohort of 141,696 patients, we validated TransferGBM for hospital-acquired acute kidney injury prediction.
Results
The baseline models (ie, transported models) that were trained on 2010 and 2011 data showed significant performance drift in the temporal validation with 2012-2017 data. Refitting these models using updated samples resulted in performance gains in nearly all cases. The proposed TransferGBM model succeeded in achieving uniformly better performance than the refitted models.
Conclusions
Under the scenario of population shift, incorporating new knowledge while preserving old knowledge is essential for maintaining stable performance. Transfer learning combined with stacking ensemble learning can help achieve a balance of old and new knowledge in a flexible and adaptive way, even in the case of insufficient new data.
Collapse
|
8
|
Machine Learning Model Drift: Predicting Diagnostic Imaging Follow-Up as a Case Example. J Am Coll Radiol 2022; 19:1162-1169. [PMID: 35981636 DOI: 10.1016/j.jacr.2022.05.030] [Citation(s) in RCA: 9] [Impact Index Per Article: 4.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/28/2022] [Revised: 05/19/2022] [Accepted: 05/20/2022] [Indexed: 11/23/2022]
Abstract
OBJECTIVE Address model drift in a machine learning (ML) model for predicting diagnostic imaging follow-up using data augmentation with more recent data versus retraining new predictive models. METHODS This institutional review board-approved retrospective study was conducted January 1, 2016, to December 31, 2020, at a large academic institution. A previously trained ML model was trained on 1,000 radiology reports from 2016 (old data). An additional 1,385 randomly selected reports from 2019 to 2020 (new data) were annotated for follow-up recommendations and randomly divided into two sets: training (n = 900) and testing (n = 485). Support vector machine and random forest (RF) algorithms were constructed and trained using 900 new data reports plus old data (augmented data, new models) and using only new data (new data, new models). The 2016 baseline model was used as comparator as is and trained with augmented data. Recall was compared with baseline using McNemar's test. RESULTS Follow-up recommendations were contained in 11.3% of reports (157 or 1,385). The baseline model retrained with new data had precision = 0.83 and recall = 0.54; none significantly different from baseline. A new RF model trained with augmented data had significantly better recall versus the baseline model (0.80 versus 0.66, P = .04) and comparable precision (0.90 versus 0.86). DISCUSSION ML methods for monitoring follow-up recommendations in radiology reports suffer model drift over time. A newly developed RF model achieved better recall with comparable precision versus simply retraining a previously trained original model with augmented data. Thus, regularly assessing and updating these models is necessary using more recent historical data.
Collapse
|
9
|
Davis SE, Brown JR, Dorn C, Westerman D, Solomon RJ, Matheny ME. Maintaining a National Acute Kidney Injury Risk Prediction Model to Support Local Quality Benchmarking. Circ Cardiovasc Qual Outcomes 2022; 15:e008635. [PMID: 35959674 PMCID: PMC9388604 DOI: 10.1161/circoutcomes.121.008635] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
BACKGROUND The utility of quality dashboards to inform decision-making and improve clinical outcomes is tightly linked to the accuracy of the information they provide and, in turn, accuracy of underlying prediction models. Despite recognition of the need to update prediction models to maintain accuracy over time, there is limited guidance on updating strategies. We compare predefined and surveillance-based updating strategies applied to a model supporting quality evaluations among US veterans. METHODS We evaluated the performance of a US Department of Veterans Affairs-specific model for postcardiac catheterization acute kidney injury using routinely collected observational data over the 6 years following model development (n=90 295 procedures in 2013-2019). Predicted probabilities were generated from the original model, an annually retrained model, and a surveillance-based approach that monitored performance to inform the timing and method of updates. We evaluated how updating the national model impacted regional quality profiles. We compared observed-to-expected outcome ratios, where values above and below 1 indicated more and fewer adverse outcomes than expected, respectively. RESULTS The original model overpredicted risk at the national level (observed-to-expected outcome ratio, 0.75 [0.74-0.77]). Annual retraining updated the model 5×; surveillance-based updating retrained once and recalibrated twice. While both strategies improved performance, the surveillance-based approach provided superior calibration (observed-to-expected outcome ratio, 1.01 [0.99-1.03] versus 0.94 [0.92-0.96]). Overprediction by the original model led to optimistic quality assessments, incorrectly indicating most of the US Department of Veterans Affairs' 18 regions observed fewer acute kidney injury events than predicted. Both updating strategies revealed 16 regions performed as expected and 2 regions increasingly underperformed, having more acute kidney injury events than predicted. CONCLUSIONS Miscalibrated clinical prediction models provide inaccurate pictures of performance across clinical units, and degrading calibration further complicates our understanding of quality. Updating strategies tailored to health system needs and capacity should be incorporated into model implementation plans to promote the utility and longevity of quality reporting tools.
Collapse
Affiliation(s)
- Sharon E. Davis
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN
| | - Jeremiah R. Brown
- Departments of Epidemiology and Biomedical Data Science, Dartmouth Geisel School of Medicine, Hanover, NH
| | - Chad Dorn
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN
| | - Dax Westerman
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN
| | - Richard J. Solomon
- Department of Medicine, Larner College of Medicine, University of Vermont, Burlington, VT
| | - Michael E. Matheny
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN
- Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA
- Department of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA
- Tennessee Valley Healthcare System VA Medical Center, Veterans Health Administration, Nashville, TN, USA
| |
Collapse
|
10
|
Mehta Y, Kapoor PM, Maheswarappa HM, Saxena G. Noninvasive Bioreactance-Based Fluid Management Monitoring: A Review of Literature. JOURNAL OF CARDIAC CRITICAL CARE TSS 2022. [DOI: 10.1055/s-0041-1741491] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/19/2022] Open
Abstract
AbstractBody fluid balance is an independent predictor of mortality. For each liter of fluid over and above 5 L, risk-adjusted excess mortality is seen. Mortality increased by 2.3% for each 1 L of fluid and hospital costs increased by $999. Accordingly, most recent guidelines have endorsed dynamic modeling. Passive leg raising-induced increase of aortic blood flow ≥ 10% predicts fluid responsiveness with a sensitivity of 97% and a specificity of 94%. Thus, passive leg raising is often used as gold standard for validation of other procedures (though it's usefulness to assess respiratory variation in vena cava is not conclusive). STARLING, a device based on bioreactance, works on phase shift or time delay while bioimpedance works on the amplitude of the thoracic impedance. Unlike bioimpedance, bioreactance is not affected by the size of the patient, thoracic fluids, or position of sensors.STARLING is equipped with four sensor pads. Each pad contains two sensors, the outer sensor is a transmitting electrode and the inner sensor is a receiving electrode. The STARLING monitor induces a 75-KHz AC current. It then measures the time delay/phase shift.STARLING system, a bioreactance-based dynamic assessment system for fluid responsiveness, predicts it accurately, precisely, and noninvasively. It reduces invasive risks and is independently validated against pulmonary artery catheter. It is not affected by vasopressors or shock and has wide range of application.
Collapse
Affiliation(s)
- Yatin Mehta
- Medanta Institute of Critical Care and Anesthesiology, Medanta the Medicity, Gurugram, Haryana, India
| | - Poonam Malhotra Kapoor
- Department of Cardiac Anesthesiology, All India Institute of Medical Sciences, New Delhi, India
| | - Harish Mallapura Maheswarappa
- Division of Critical Care Medicine, Critical Care and Pain, Department of Anaesthesiology, Tata Memorial Hospital, Mumbai, Maharashtra, India
| | - Gaurav Saxena
- Medical Affairs Division, Baxter India Pvt Ltd, Gurugram, Haryana, India
| |
Collapse
|
11
|
Antoniou T, Mamdani M. Évaluation des solutions fondées sur l’apprentissage machine en santé. CMAJ 2021; 193:E1720-E1724. [PMID: 34750185 PMCID: PMC8584374 DOI: 10.1503/cmaj.210036-f] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/23/2022] Open
Affiliation(s)
- Tony Antoniou
- Centre de recherche et de formation en analytique des soins de santé Li Ka Shing (Antoniou, Mamdani), Réseau hospitalier Unity Health de Toronto; Institut du savoir Li Ka Shing (Antoniou, Mamdani), Réseau hospitalier Unity Health de Toronto; Département de médecine de famille et communautaire (Antoniou), Réseau hospitalier Unity Health deToronto et Université de Toronto; Faculté de médecine Temerty (Mamdani) et Faculté de pharmacie Leslie Dan (Mamdani), Université de Toronto; Institut des politiques, de la gestion et de l'évaluation de la santé (Mamdani), Université de Toronto, Toronto, Ont.
| | - Muhammad Mamdani
- Centre de recherche et de formation en analytique des soins de santé Li Ka Shing (Antoniou, Mamdani), Réseau hospitalier Unity Health de Toronto; Institut du savoir Li Ka Shing (Antoniou, Mamdani), Réseau hospitalier Unity Health de Toronto; Département de médecine de famille et communautaire (Antoniou), Réseau hospitalier Unity Health deToronto et Université de Toronto; Faculté de médecine Temerty (Mamdani) et Faculté de pharmacie Leslie Dan (Mamdani), Université de Toronto; Institut des politiques, de la gestion et de l'évaluation de la santé (Mamdani), Université de Toronto, Toronto, Ont
| |
Collapse
|
12
|
Reynard C, Martin GP, Kontopantelis E, Jenkins DA, Heagerty A, McMillan B, Jafar A, Garlapati R, Body R. Advanced cardiovascular risk prediction in the emergency department: updating a clinical prediction model - a large database study protocol. Diagn Progn Res 2021; 5:16. [PMID: 34620253 PMCID: PMC8499458 DOI: 10.1186/s41512-021-00105-7] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/27/2021] [Accepted: 09/27/2021] [Indexed: 12/23/2022] Open
Abstract
BACKGROUND Patients presenting with chest pain represent a large proportion of attendances to emergency departments. In these patients clinicians often consider the diagnosis of acute myocardial infarction (AMI), the timely recognition and treatment of which is clinically important. Clinical prediction models (CPMs) have been used to enhance early diagnosis of AMI. The Troponin-only Manchester Acute Coronary Syndromes (T-MACS) decision aid is currently in clinical use across Greater Manchester. CPMs have been shown to deteriorate over time through calibration drift. We aim to assess potential calibration drift with T-MACS and compare methods for updating the model. METHODS We will use routinely collected electronic data from patients who were treated using TMACS at two large NHS hospitals. This is estimated to include approximately 14,000 patient episodes spanning June 2016 to October 2020. The primary outcome of acute myocardial infarction will be sourced from NHS Digital's admitted patient care dataset. We will assess the calibration drift of the existing model and the benefit of updating the CPM by model recalibration, model extension and dynamic updating. These models will be validated by bootstrapping and one step ahead prequential testing. We will evaluate predictive performance using calibrations plots and c-statistics. We will also examine the reclassification of predicted probability with the updated TMACS model. DISCUSSION CPMs are widely used in modern medicine, but are vulnerable to deteriorating calibration over time. Ongoing refinement using routinely collected electronic data will inevitably be more efficient than deriving and validating new models. In this analysis we will seek to exemplify methods for updating CPMs to protect the initial investment of time and effort. If successful, the updating methods could be used to continually refine the algorithm used within TMACS, maintaining or even improving predictive performance over time. TRIAL REGISTRATION ISRCTN number: ISRCTN41008456.
Collapse
Affiliation(s)
- Charles Reynard
- grid.5379.80000000121662407Division of Cardiovascular Sciences, University of Manchester, Manchester, UK
- grid.498924.aEmergency Department, Manchester University NHS Foundation Trust, Manchester, UK
| | - Glen P. Martin
- grid.5379.80000000121662407Division of Informatics, Imaging and Data Science, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Manchester, UK
| | - Evangelos Kontopantelis
- grid.5379.80000000121662407Division of Informatics, Imaging and Data Science, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Manchester, UK
| | - David A. Jenkins
- grid.5379.80000000121662407Division of Informatics, Imaging and Data Science, Faculty of Biology, Medicine and Health, University of Manchester, Manchester Academic Health Science Centre, Manchester, UK
| | - Anthony Heagerty
- grid.5379.80000000121662407Division of Cardiovascular Sciences, University of Manchester, Manchester, UK
| | - Brian McMillan
- Centre for Primary Care and Health Services Research Division of Population Health, Health Services Research and Primary Care School of Health Sciences Faculty of Biology, Medicine and Health University of Manchestern, Manchester, UK
| | - Anisa Jafar
- grid.5379.80000000121662407Humanitarian and Conflict Response Institute, University of Manchester, Manchester, UK
| | - Rajendar Garlapati
- grid.439642.e0000 0004 0489 3782Emergency Department, Royal Blackburn Hospital, East Lancashire Hospitals NHS Trust, Burnley, UK
| | - Richard Body
- grid.5379.80000000121662407Division of Cardiovascular Sciences, University of Manchester, Manchester, UK
- grid.498924.aEmergency Department, Manchester University NHS Foundation Trust, Manchester, UK
| |
Collapse
|
13
|
Affiliation(s)
- Tony Antoniou
- Li Ka Shing Centre for Healthcare Analytics Research & Training (Antoniou, Mamdani), Unity Health Toronto; Li Ka Shing Knowledge Institute (Antoniou, Mamdani), Unity Health Toronto; Department of Family and Community Medicine (Antoniou), Unity Health Toronto and University of Toronto; Temerty Faculty of Medicine (Mamdani) and Leslie Dan Faculty of Pharmacy (Mamdani), University of Toronto; Institute of Health Policy, Management, and Evaluation (Mamdani), University of Toronto, Toronto, Ont.
| | - Muhammad Mamdani
- Li Ka Shing Centre for Healthcare Analytics Research & Training (Antoniou, Mamdani), Unity Health Toronto; Li Ka Shing Knowledge Institute (Antoniou, Mamdani), Unity Health Toronto; Department of Family and Community Medicine (Antoniou), Unity Health Toronto and University of Toronto; Temerty Faculty of Medicine (Mamdani) and Leslie Dan Faculty of Pharmacy (Mamdani), University of Toronto; Institute of Health Policy, Management, and Evaluation (Mamdani), University of Toronto, Toronto, Ont
| |
Collapse
|
14
|
Guo LL, Pfohl SR, Fries J, Posada J, Fleming SL, Aftandilian C, Shah N, Sung L. Systematic Review of Approaches to Preserve Machine Learning Performance in the Presence of Temporal Dataset Shift in Clinical Medicine. Appl Clin Inform 2021; 12:808-815. [PMID: 34470057 PMCID: PMC8410238 DOI: 10.1055/s-0041-1735184] [Citation(s) in RCA: 24] [Impact Index Per Article: 8.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/28/2021] [Accepted: 07/12/2021] [Indexed: 10/20/2022] Open
Abstract
OBJECTIVE The change in performance of machine learning models over time as a result of temporal dataset shift is a barrier to machine learning-derived models facilitating decision-making in clinical practice. Our aim was to describe technical procedures used to preserve the performance of machine learning models in the presence of temporal dataset shifts. METHODS Studies were included if they were fully published articles that used machine learning and implemented a procedure to mitigate the effects of temporal dataset shift in a clinical setting. We described how dataset shift was measured, the procedures used to preserve model performance, and their effects. RESULTS Of 4,457 potentially relevant publications identified, 15 were included. The impact of temporal dataset shift was primarily quantified using changes, usually deterioration, in calibration or discrimination. Calibration deterioration was more common (n = 11) than discrimination deterioration (n = 3). Mitigation strategies were categorized as model level or feature level. Model-level approaches (n = 15) were more common than feature-level approaches (n = 2), with the most common approaches being model refitting (n = 12), probability calibration (n = 7), model updating (n = 6), and model selection (n = 6). In general, all mitigation strategies were successful at preserving calibration but not uniformly successful in preserving discrimination. CONCLUSION There was limited research in preserving the performance of machine learning models in the presence of temporal dataset shift in clinical medicine. Future research could focus on the impact of dataset shift on clinical decision making, benchmark the mitigation strategies on a wider range of datasets and tasks, and identify optimal strategies for specific settings.
Collapse
Affiliation(s)
- Lin Lawrence Guo
- Program in Child Health Evaluative Sciences, The Hospital for Sick Children, Toronto, Canada
| | - Stephen R. Pfohl
- Biomedical Informatics Research, Stanford University, Palo Alto, California, United States
| | - Jason Fries
- Biomedical Informatics Research, Stanford University, Palo Alto, California, United States
| | - Jose Posada
- Biomedical Informatics Research, Stanford University, Palo Alto, California, United States
| | - Scott Lanyon Fleming
- Biomedical Informatics Research, Stanford University, Palo Alto, California, United States
| | - Catherine Aftandilian
- Division of Pediatric Hematology/Oncology, Stanford University, Palo Alto, United States
| | - Nigam Shah
- Biomedical Informatics Research, Stanford University, Palo Alto, California, United States
| | - Lillian Sung
- Program in Child Health Evaluative Sciences, The Hospital for Sick Children, Toronto, Canada
- Division of Haematology/Oncology, The Hospital for Sick Children, Toronto, Canada
| |
Collapse
|
15
|
Davis SE, Greevy RA, Lasko TA, Walsh CG, Matheny ME. Detection of calibration drift in clinical prediction models to inform model updating. J Biomed Inform 2020; 112:103611. [PMID: 33157313 PMCID: PMC8627243 DOI: 10.1016/j.jbi.2020.103611] [Citation(s) in RCA: 45] [Impact Index Per Article: 11.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/14/2020] [Revised: 10/21/2020] [Accepted: 10/29/2020] [Indexed: 10/23/2022]
Abstract
Model calibration, critical to the success and safety of clinical prediction models, deteriorates over time in response to the dynamic nature of clinical environments. To support informed, data-driven model updating strategies, we present and evaluate a calibration drift detection system. Methods are developed for maintaining dynamic calibration curves with optimized online stochastic gradient descent and for detecting increasing miscalibration with adaptive sliding windows. These methods are generalizable to support diverse prediction models developed using a variety of learning algorithms and customizable to address the unique needs of clinical use cases. In both simulation and case studies, our system accurately detected calibration drift. When drift is detected, our system further provides actionable alerts by including information on a window of recent data that may be appropriate for model updating. Simulations showed these windows were primarily composed of data accruing after drift onset, supporting the potential utility of the windows for model updating. By promoting model updating as calibration deteriorates rather than on pre-determined schedules, implementations of our drift detection system may minimize interim periods of insufficient model accuracy and focus analytic resources on those models most in need of attention.
Collapse
Affiliation(s)
- Sharon E Davis
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, USA.
| | - Robert A Greevy
- Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA.
| | - Thomas A Lasko
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, USA.
| | - Colin G Walsh
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Psychiatry, Vanderbilt University Medical Center, Nashville, TN, USA.
| | - Michael E Matheny
- Department of Biomedical Informatics, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Biostatistics, Vanderbilt University Medical Center, Nashville, TN, USA; Department of Medicine, Vanderbilt University Medical Center, Nashville, TN, USA; Geriatrics Research, Education, and Clinical Care, Tennessee Valley Healthcare System VA, Nashville, TN, USA.
| |
Collapse
|
16
|
Baxter RD, Fann JI, DiMaio JM, Lobdell K. Digital Health Primer for Cardiothoracic Surgeons. Ann Thorac Surg 2020; 110:364-372. [PMID: 32268139 DOI: 10.1016/j.athoracsur.2020.02.072] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/01/2019] [Revised: 01/03/2020] [Accepted: 02/23/2020] [Indexed: 12/12/2022]
Abstract
The burgeoning demands for quality, safety, and value in cardiothoracic surgery, in combination with the advancement and acceleration of digital health solutions and information technology, provide a unique opportunity to improve efficiency and effectiveness simultaneously in cardiothoracic surgery. This primer on digital health explores and reviews data integration, data processing, complex modeling, telehealth with remote monitoring, and cybersecurity as they shape the future of cardiothoracic surgery.
Collapse
Affiliation(s)
- Ronald D Baxter
- Department of Cardiothoracic Surgery, Baylor Scott and White, The Heart Hospital, Plano, Texas
| | - James I Fann
- Department of Cardiothoracic Surgery, Stanford University Medical Center, Stanford, California
| | - J Michael DiMaio
- Department of Cardiothoracic Surgery, Baylor Scott and White, The Heart Hospital, Plano, Texas
| | - Kevin Lobdell
- Sanger Heart and Vascular Institute, Atrium Health, Charlotte, North Carolina.
| |
Collapse
|
17
|
Siregar S, Nieboer D, Versteegh MIM, Steyerberg EW, Takkenberg JJM. Methods for updating a risk prediction model for cardiac surgery: a statistical primer. Interact Cardiovasc Thorac Surg 2019; 28:333-338. [DOI: 10.1093/icvts/ivy338] [Citation(s) in RCA: 19] [Impact Index Per Article: 3.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/18/2018] [Revised: 10/26/2018] [Accepted: 11/13/2018] [Indexed: 11/12/2022] Open
Affiliation(s)
- Sabrina Siregar
- Department of Cardio-thoracic Surgery, Leiden University Medical Center, Leiden, Netherlands
| | - Daan Nieboer
- Department of Public Health, Erasmus MC, University Medical Center Rotterdam, Rotterdam, Netherlands
| | - Michel I M Versteegh
- Department of Cardio-thoracic Surgery, Leiden University Medical Center, Leiden, Netherlands
- Board of the Netherlands Heart Registry, Utrecht, Netherlands
| | - Ewout W Steyerberg
- Department of Public Health, Erasmus MC, University Medical Center Rotterdam, Rotterdam, Netherlands
- Department of Statistics, Leiden University Medical Center, Leiden, Netherlands
| | - Johanna J M Takkenberg
- Department of Cardio-thoracic Surgery, Erasmus MC, University Medical Center Rotterdam, Rotterdam, Netherlands
| |
Collapse
|
18
|
Jenkins DA, Sperrin M, Martin GP, Peek N. Dynamic models to predict health outcomes: current status and methodological challenges. Diagn Progn Res 2018; 2:23. [PMID: 31093570 PMCID: PMC6460710 DOI: 10.1186/s41512-018-0045-2] [Citation(s) in RCA: 38] [Impact Index Per Article: 6.3] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/09/2018] [Accepted: 11/19/2018] [Indexed: 12/14/2022] Open
Abstract
BACKGROUND Disease populations, clinical practice, and healthcare systems are constantly evolving. This can result in clinical prediction models quickly becoming outdated and less accurate over time. A potential solution is to develop 'dynamic' prediction models capable of retaining accuracy by evolving over time in response to observed changes. Our aim was to review the literature in this area to understand the current state-of-the-art in dynamic prediction modelling and identify unresolved methodological challenges. METHODS MEDLINE, Embase and Web of Science were searched for papers which used or developed dynamic clinical prediction models. Information was extracted on methods for model updating, choice of update windows and decay factors and validation of models. We also extracted reported limitations of methods and recommendations for future research. RESULTS We identified eleven papers that discussed seven dynamic clinical prediction modelling methods which split into three categories. The first category uses frequentist methods to update models in discrete steps, the second uses Bayesian methods for continuous updating and the third, based on varying coefficients, explicitly describes the relationship between predictors and outcome variable as a function of calendar time. These methods have been applied to a limited number of healthcare problems, and few empirical comparisons between them have been made. CONCLUSION Dynamic prediction models are not well established but they overcome one of the major issues with static clinical prediction models, calibration drift. However, there are challenges in choosing decay factors and in dealing with sudden changes. The validation of dynamic prediction models is still largely unexplored terrain.
Collapse
Affiliation(s)
- David A. Jenkins
- 0000000121662407grid.5379.8Health e-Research Centre, Farr Institute, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UK
- 0000000121662407grid.5379.8NIHR Greater Manchester Patient Safety Translational Research Centre, The University of Manchester, Manchester, UK
- 0000000121662407grid.5379.8Faculty of Biology, Medicine and Health, University of Manchester, City Labs 1.0, Nelson Street, Manchester, M13 9NQ UK
| | - Matthew Sperrin
- 0000000121662407grid.5379.8Health e-Research Centre, Farr Institute, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UK
| | - Glen P. Martin
- 0000000121662407grid.5379.8Health e-Research Centre, Farr Institute, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UK
| | - Niels Peek
- 0000000121662407grid.5379.8Health e-Research Centre, Farr Institute, Faculty of Biology, Medicine and Health, Manchester Academic Health Science Centre, University of Manchester, Manchester, UK
- 0000000121662407grid.5379.8NIHR Greater Manchester Patient Safety Translational Research Centre, The University of Manchester, Manchester, UK
| |
Collapse
|