Development and validation of clinical prediction models for mortality, functional outcome and cognitive impairment after stroke: a study protocol

IntroductionStroke is a leading cause of adult disability and death worldwide. The neurological impairments associated with stroke prevent patients from performing basic daily activities and have enormous impact on families and caregivers. Practical and accurate tools to assist in predicting outcome after stroke at patient level can provide significant aid for patient management. Furthermore, prediction models of this kind can be useful for clinical research, health economics, policymaking and clinical decision support.Methods2869 patients with first-ever stroke from South London Stroke Register (SLSR) (1995–2004) will be included in the development cohort. We will use information captured after baseline to construct multilevel models and a Cox proportional hazard model to predict cognitive impairment, functional outcome and mortality up to 5 years after stroke. Repeated random subsampling validation (Monte Carlo cross-validation) will be evaluated in model development. Data from participants recruited to the stroke register (2005–2014) will be used for temporal validation of the models. Data from participants recruited to the Dijon Stroke Register (1985–2015) will be used for external validation. Discrimination, calibration and clinical utility of the models will be presented.EthicsPatients, or for patients who cannot consent their relatives, gave written informed consent to participate in stroke-related studies within the SLSR. The SLSR design was approved by the ethics committees of Guy’s and St Thomas’ NHS Foundation Trust, Kings College Hospital, Queens Square and Westminster Hospitals (London). The Dijon Stroke Registry was approved by the Comité National des Registres and the InVS and has authorisation of the Commission Nationale de l’Informatique et des Libertés.

Download Full-text

P–419 On prognosis after unexplained recurrent pregnancy losses (RPL); a systematic review and external validation of clinical prediction models

Human Reproduction ◽

10.1093/humrep/deab130.418 ◽

2021 ◽

Vol 36 (Supplement_1) ◽

Author(s):

A Youssef

Keyword(s):

Systematic Review ◽

Supportive Care ◽

Prediction Model ◽

Pregnancy Outcome ◽

Prediction Models ◽

External Validation ◽

Model Development ◽

Predictive Performance ◽

Pregnancy Rates ◽

Cochrane Library

Abstract Study question Which models that predict pregnancy outcome in couples with unexplained RPL exist and what is the performance of the most used model? Summary answer We identified seven prediction models; none followed the recommended prediction model development steps. Moreover, the most used model showed poor predictive performance. What is known already RPL remains unexplained in 50–75% of couples For these couples, there is no effective treatment option and clinical management rests on supportive care. Essential part of supportive care consists of counselling on the prognosis of subsequent pregnancies. Indeed, multiple prediction models exist, however the quality and validity of these models varies. In addition, the prediction model developed by Brigham et al is the most widely used model, but has never been externally validated. Study design, size, duration We performed a systematic review to identify prediction models for pregnancy outcome after unexplained RPL. In addition we performed an external validation of the Brigham model in a retrospective cohort, consisting of 668 couples with unexplained RPL that visited our RPL clinic between 2004 and 2019. Participants/materials, setting, methods A systematic search was performed in December 2020 in Pubmed, Embase, Web of Science and Cochrane library to identify relevant studies. Eligible studies were selected and assessed according to the TRIPOD) guidelines, covering topics on model performance and validation statement. The performance of predicting live birth in the Brigham model was evaluated through calibration and discrimination, in which the observed pregnancy rates were compared to the predicted pregnancy rates. Main results and the role of chance Seven models were compared and assessed according to the TRIPOD statement. This resulted in two studies of low, three of moderate and two of above average reporting quality. These studies did not follow the recommended steps for model development and did not calculate a sample size. Furthermore, the predictive performance of neither of these models was internally- or externally validated. We performed an external validation of Brigham model. Calibration showed overestimation of the model and too extreme predictions, with a negative calibration intercept of –0.52 (CI 95% –0.68 – –0.36), with a calibration slope of 0.39 (CI 95% 0.07 – 0.71). The discriminative ability of the model was very low with a concordance statistic of 0.55 (CI 95% 0.50 – 0.59). Limitations, reasons for caution None of the studies are specifically named prediction models, therefore models may have been missed in the selection process. The external validation cohort used a retrospective design, in which only the first pregnancy after intake was registered. Follow-up time was not limited, which is important in counselling unexplained RPL couples. Wider implications of the findings: Currently, there are no suitable models that predict on pregnancy outcome after RPL. Moreover, we are in need of a model with several variables such that prognosis is individualized, and factors from both the female as the male to enable a couple specific prognosis. Trial registration number Not applicable

Download Full-text

Evaluating Modeling and Validation Strategies for Tooth Loss

Journal of Dental Research ◽

10.1177/0022034519864889 ◽

2019 ◽

Vol 98 (10) ◽

pp. 1088-1095 ◽

Cited By ~ 2

Author(s):

J. Krois ◽

C. Graetz ◽

B. Holtfreter ◽

P. Brinkmann ◽

T. Kocher ◽

...

Keyword(s):

Tooth Loss ◽

Predictive Power ◽

Prediction Models ◽

Recursive Partitioning ◽

External Validation ◽

Model Development ◽

Gradient Boosting ◽

Extreme Gradient Boosting ◽

Complex Models ◽

Development And Validation

Prediction models learn patterns from available data (training) and are then validated on new data (testing). Prediction modeling is increasingly common in dental research. We aimed to evaluate how different model development and validation steps affect the predictive performance of tooth loss prediction models of patients with periodontitis. Two independent cohorts (627 patients, 11,651 teeth) were followed over a mean ± SD 18.2 ± 5.6 y (Kiel cohort) and 6.6 ± 2.9 y (Greifswald cohort). Tooth loss and 10 patient- and tooth-level predictors were recorded. The impact of different model development and validation steps was evaluated: 1) model complexity (logistic regression, recursive partitioning, random forest, extreme gradient boosting), 2) sample size (full data set or 10%, 25%, or 75% of cases dropped at random), 3) prediction periods (maximum 10, 15, or 20 y or uncensored), and 4) validation schemes (internal or external by centers/time). Tooth loss was generally a rare event (880 teeth were lost). All models showed limited sensitivity but high specificity. Patients’ age and tooth loss at baseline as well as probing pocket depths showed high variable importance. More complex models (random forest, extreme gradient boosting) had no consistent advantages over simpler ones (logistic regression, recursive partitioning). Internal validation (in sample) overestimated the predictive power (area under the curve up to 0.90), while external validation (out of sample) found lower areas under the curve (range 0.62 to 0.82). Reducing the sample size decreased the predictive power, particularly for more complex models. Censoring the prediction period had only limited impact. When the model was trained in one period and tested in another, model outcomes were similar to the base case, indicating temporal validation as a valid option. No model showed higher accuracy than the no-information rate. In conclusion, none of the developed models would be useful in a clinical setting, despite high accuracy. During modeling, rigorous development and external validation should be applied and reported accordingly.

Download Full-text

External Validation of the Totaled Health Risks in Vascular Events Score to Predict Functional Outcome and Mortality in Patients Entered into the China National Stroke Registry

Journal of Stroke and Cerebrovascular Diseases ◽

10.1016/j.jstrokecerebrovasdis.2016.03.021 ◽

2016 ◽

Vol 25 (10) ◽

pp. 2331-2337 ◽

Cited By ~ 2

Author(s):

Weiqi Chen ◽

Gaifen Liu ◽

Jiming Fang ◽

Yilong Wang ◽

Yanli Song ◽

...

Keyword(s):

Functional Outcome ◽

Health Risks ◽

External Validation ◽

Vascular Events ◽

Stroke Registry

Download Full-text

Machine Learning Approach Using Routine Immediate Postoperative Laboratory Values for Predicting Postoperative Mortality

Journal of Personalized Medicine ◽

10.3390/jpm11121271 ◽

2021 ◽

Vol 11 (12) ◽

pp. 1271

Author(s):

Jaehyeong Cho ◽

Jimyung Park ◽

Eugene Jeong ◽

Jihye Shin ◽

Sangjeong Ahn ◽

...

Keyword(s):

Machine Learning ◽

Random Forest ◽

Prediction Models ◽

External Validation ◽

Model Development ◽

Postoperative Mortality ◽

Random Forest Model ◽

Forest Model ◽

Laboratory Values ◽

Increased Risk

Background: Several prediction models have been proposed for preoperative risk stratification for mortality. However, few studies have investigated postoperative risk factors, which have a significant influence on survival after surgery. This study aimed to develop prediction models using routine immediate postoperative laboratory values for predicting postoperative mortality. Methods: Two tertiary hospital databases were used in this research: one for model development and another for external validation of the resulting models. The following algorithms were utilized for model development: LASSO logistic regression, random forest, deep neural network, and XGBoost. We built the models on the lab values from immediate postoperative blood tests and compared them with the SASA scoring system to demonstrate their efficacy. Results: There were 3817 patients who had immediate postoperative blood test values. All models trained on immediate postoperative lab values outperformed the SASA model. Furthermore, the developed random forest model had the best AUROC of 0.82 and AUPRC of 0.13, and the phosphorus level contributed the most to the random forest model. Conclusions: Machine learning models trained on routine immediate postoperative laboratory values outperformed previously published approaches in predicting 30-day postoperative mortality, indicating that they may be beneficial in identifying patients at increased risk of postoperative death.

Download Full-text

Prognostic Models for Predicting Remission of Diabetes Following Bariatric Surgery: A Systematic Review and Meta-analysis

10.2337/figshare.15173232 ◽

2021 ◽

Author(s):

Pushpa Singh ◽

Nicola J Adderley ◽

Jonathan Hazlehurst ◽

Malcolm Price ◽

Abd A Tahrani ◽

...

Keyword(s):

Systematic Review ◽

Bariatric Surgery ◽

Prediction Models ◽

Data Extraction ◽

External Validation ◽

Model Development ◽

Diabetes Remission ◽

Development Studies ◽

Cochrane Central Register

Background Remission of type 2 diabetes following bariatric surgery is well established but identifying patients who will go into remission is challenging. Purpose To perform a systematic review of currently available diabetes remission prediction models, compare their performance, and evaluate their applicability in clinical settings. Data sources A comprehensive systematic literature search of MEDLINE, MEDLINE In-Process & Other Non-Indexed Citations, EMBASE and Cochrane Central Register of Controlled Trials was undertaken. The search was restricted to studies published in the last 15 years and in the English language. Study selection and data extraction All studies developing or validating a prediction model for diabetes remission in adults after bariatric surgery were included. The search identified 4165 references of which 38 were included for data extraction. We identified 16 model development and 22 validation studies. Data synthesis Of the 16 model development studies, 11 developed scoring systems and 5 proposed logistic regression models. In model development studies, 10 models showed excellent discrimination with area under curve (AUC) ≥ 0.800. Two of these prediction models, ABCD and DiaRem, were widely externally validated in different populations, a variety of bariatric procedures, and for both short- and long-term diabetes remission. Newer prediction models showed excellent discrimination in test studies, but external validation was limited. Limitations and Conclusions Amongst the prediction models identified, the ABCD and DiaRem models were the most widely validated and showed acceptable to excellent discrimination. More studies validating newer models and focusing on long-term diabetes remission are needed.

Download Full-text

Development and external validation of a prognostic multivariable model on admission for hospitalized patients with COVID-19

10.1101/2020.03.28.20045997 ◽

2020 ◽

Cited By ~ 18

Author(s):

Jianfeng Xie ◽

Daniel Hungerford ◽

Hui Chen ◽

Simon T Abrams ◽

Shusheng Li ◽

...

Keyword(s):

Lactate Dehydrogenase ◽

Prediction Model ◽

Lymphocyte Count ◽

Prediction Models ◽

External Validation ◽

Model Development ◽

Added Value ◽

Clinical Prediction ◽

Healthcare Resources ◽

D Dimer

SummaryBackgroundCOVID-19 pandemic has developed rapidly and the ability to stratify the most vulnerable patients is vital. However, routinely used severity scoring systems are often low on diagnosis, even in non-survivors. Therefore, clinical prediction models for mortality are urgently required.MethodsWe developed and internally validated a multivariable logistic regression model to predict inpatient mortality in COVID-19 positive patients using data collected retrospectively from Tongji Hospital, Wuhan (299 patients). External validation was conducted using a retrospective cohort from Jinyintan Hospital, Wuhan (145 patients). Nine variables commonly measured in these acute settings were considered for model development, including age, biomarkers and comorbidities. Backwards stepwise selection and bootstrap resampling were used for model development and internal validation. We assessed discrimination via the C statistic, and calibration using calibration-in-the-large, calibration slopes and plots.FindingsThe final model included age, lymphocyte count, lactate dehydrogenase and SpO2 as independent predictors of mortality. Discrimination of the model was excellent in both internal (c=0·89) and external (c=0·98) validation. Internal calibration was excellent (calibration slope=1). External validation showed some over-prediction of risk in low-risk individuals and under-prediction of risk in high-risk individuals prior to recalibration. Recalibration of the intercept and slope led to excellent performance of the model in independent data.InterpretationCOVID-19 is a new disease and behaves differently from common critical illnesses. This study provides a new prediction model to identify patients with lethal COVID-19. Its practical reliance on commonly available parameters should improve usage of limited healthcare resources and patient survival rate.FundingThis study was supported by following funding: Key Research and Development Plan of Jiangsu Province (BE2018743 and BE2019749), National Institute for Health Research (NIHR) (PDF-2018-11-ST2-006), British Heart Foundation (BHF) (PG/16/65/32313) and Liverpool University Hospitals NHS Foundation Trust in UK.Research in contextEvidence before this studySince the outbreak of COVID-19, there has been a pressing need for development of a prognostic tool that is easy for clinicians to use. Recently, a Lancet publication showed that in a cohort of 191 patients with COVID-19, age, SOFA score and D-dimer measurements were associated with mortality. No other publication involving prognostic factors or models has been identified to date.Added value of this studyIn our cohorts of 444 patients from two hospitals, SOFA scores were low in the majority of patients on admission. The relevance of D-dimer could not be verified, as it is not included in routine laboratory tests. In this study, we have established a multivariable clinical prediction model using a development cohort of 299 patients from one hospital. After backwards selection, four variables, including age, lymphocyte count, lactate dehydrogenase and SpO2 remained in the model to predict mortality. This has been validated internally and externally with a cohort of 145 patients from a different hospital. Discrimination of the model was excellent in both internal (c=0·89) and external (c=0·98) validation. Calibration plots showed excellent agreement between predicted and observed probabilities of mortality after recalibration of the model to account for underlying differences in the risk profile of the datasets. This demonstrated that the model is able to make reliable predictions in patients from different hospitals. In addition, these variables agree with pathological mechanisms and the model is easy to use in all types of clinical settings.Implication of all the available evidenceAfter further external validation in different countries the model will enable better risk stratification and more targeted management of patients with COVID-19. With the nomogram, this model that is based on readily available parameters can help clinicians to stratify COVID-19 patients on diagnosis to use limited healthcare resources effectively and improve patient outcome.

Download Full-text

Evaluating Methodological Quality of Prognostic Prediction Models on Patient Reported Outcome Measurements after Total Hip Replacement and Total Knee Replacement Surgery: a systematic review protocol

10.21203/rs.3.rs-903247/v1 ◽

2021 ◽

Author(s):

Wei-Ju Chang ◽

Justine Naylor ◽

Pragadesh Natarajan ◽

Spiro Menounos ◽

Masiath Monuja ◽

...

Keyword(s):

Systematic Review ◽

Knee Osteoarthritis ◽

Prediction Model ◽

Methodological Quality ◽

Prediction Models ◽

External Validation ◽

Model Development ◽

Patient Reported ◽

Modelling Studies

Abstract Background Prediction models for poor patient-reported surgical outcomes after total hip replacement (THR) and total knee replacement (TKR) may provide a method for improving appropriate surgical care for hip and knee osteoarthritis. There are concerns about methodological issues and the risk of bias of studies producing prediction models. A critical evaluation of the methodological quality of prediction modelling studies in THR and TKR is needed to ensure their clinical usefulness. This systematic review aims to: 1) evaluate and report the quality of risk stratification and prediction modelling studies that predict patient-reported outcomes after THR and TKR; 2) identify areas of methodological deficit and provide recommendations for future research; and 3) synthesise the evidence on prediction models associated with post-operative patient-reported outcomes after THR and TKR surgeries. Methods MEDLINE, EMBASE and CINAHL electronic databases will be searched to identify relevant studies. Title and abstract and full-text screening will be performed by two independent reviewers. We will include: 1) prediction model development studies without external validation; 2) prediction model development studies with external validation of independent data; 3) external model validation studies; and 4) studies updating a previously developed prediction model. Data extraction spreadsheets will be developed based on the CHARMS checklist and TRIPOD statement and piloted on two relevant studies. Study quality and risk of bias will be assessed using the PROBAST tool. Prediction models will be summarised qualitatively. Meta-analyses on the predictive performance of included models will be conducted if appropriate. Discussion This systematic review will evaluate the methodological quality and usefulness of prediction models for poor outcomes after THR or TKR. This information is essential to provide evidence-based healthcare for end-stage hip and knee osteoarthritis. Findings of this review will contribute to the identification of key areas for improvement in conducting prognostic research in this field and facilitate the progress in evidence-based tailored treatments for hip and knee osteoarthritis. Systematic review registration: Submitted to PROSPERO on 30 August 2021.

Download Full-text

Development and validation of Prediction models for Risks of complications in Early-onset Pre-eclampsia (PREP): a prospective cohort study

Health Technology Assessment ◽

10.3310/hta21180 ◽

2017 ◽

Vol 21 (18) ◽

pp. 1-100 ◽

Cited By ~ 11

Author(s):

Shakila Thangaratinam ◽

John Allotey ◽

Nadine Marlin ◽

Ben W Mol ◽

Peter Von Dadelszen ◽

...

Keyword(s):

Early Onset ◽

Prediction Models ◽

Antihypertensive Drugs ◽

External Validation ◽

Model Development ◽

Maternal Complications ◽

Data Set ◽

Internal Validation ◽

Development Data ◽

S Model

BackgroundThe prognosis of early-onset pre-eclampsia (before 34 weeks’ gestation) is variable. Accurate prediction of complications is required to plan appropriate management in high-risk women.ObjectiveTo develop and validate prediction models for outcomes in early-onset pre-eclampsia.DesignProspective cohort for model development, with validation in two external data sets.SettingModel development: 53 obstetric units in the UK. Model transportability: PIERS (Pre-eclampsia Integrated Estimate of RiSk for mothers) and PETRA (Pre-Eclampsia TRial Amsterdam) studies.ParticipantsPregnant women with early-onset pre-eclampsia.Sample sizeNine hundred and forty-six women in the model development data set and 850 women (634 in PIERS, 216 in PETRA) in the transportability (external validation) data sets.PredictorsThe predictors were identified from systematic reviews of tests to predict complications in pre-eclampsia and were prioritised by Delphi survey.Main outcome measuresThe primary outcome was the composite of adverse maternal outcomes established using Delphi surveys. The secondary outcome was the composite of fetal and neonatal complications.AnalysisWe developed two prediction models: a logistic regression model (PREP-L) to assess the overall risk of any maternal outcome until postnatal discharge and a survival analysis model (PREP-S) to obtain individual risk estimates at daily intervals from diagnosis until 34 weeks. Shrinkage was used to adjust for overoptimism of predictor effects. For internal validation (of the full models in the development data) and external validation (of the reduced models in the transportability data), we computed the ability of the models to discriminate between those with and without poor outcomes (c-statistic), and the agreement between predicted and observed risk (calibration slope).ResultsThe PREP-L model included maternal age, gestational age at diagnosis, medical history, systolic blood pressure, urine protein-to-creatinine ratio, platelet count, serum urea concentration, oxygen saturation, baseline treatment with antihypertensive drugs and administration of magnesium sulphate. The PREP-S model additionally included exaggerated tendon reflexes and serum alanine aminotransaminase and creatinine concentration. Both models showed good discrimination for maternal complications, with anoptimism-adjustedc-statistic of 0.82 [95% confidence interval (CI) 0.80 to 0.84] for PREP-L and 0.75 (95% CI 0.73 to 0.78) for the PREP-S model in the internal validation. External validation of the reduced PREP-L model showed good performance with ac-statistic of 0.81 (95% CI 0.77 to 0.85) in PIERS and 0.75 (95% CI 0.64 to 0.86) in PETRA cohorts for maternal complications, and calibrated well with slopes of 0.93 (95% CI 0.72 to 1.10) and 0.90 (95% CI 0.48 to 1.32), respectively. In the PIERS data set, the reduced PREP-S model had ac-statistic of 0.71 (95% CI 0.67 to 0.75) and a calibration slope of 0.67 (95% CI 0.56 to 0.79). Low gestational age at diagnosis, high urine protein-to-creatinine ratio, increased serum urea concentration, treatment with antihypertensive drugs, magnesium sulphate, abnormal uterine artery Doppler scan findings and estimated fetal weight below the 10th centile were associated with fetal complications.ConclusionsThe PREP-L model provided individualised risk estimates in early-onset pre-eclampsia to plan management of high- or low-risk individuals. The PREP-S model has the potential to be used as a triage tool for risk assessment. The impacts of the model use on outcomes need further evaluation.Trial registrationCurrent Controlled Trials ISRCTN40384046.FundingThe National Institute for Health Research Health Technology Assessment programme.

Download Full-text

Statistical Development and Validation of Clinical Prediction Models

Anesthesiology ◽

10.1097/aln.0000000000003871 ◽

2021 ◽

Author(s):

Steven J. Staffa ◽

David Zurakowski

Keyword(s):

Prediction Model ◽

Prediction Models ◽

External Validation ◽

Model Development ◽

Predictive Ability ◽

Binary Outcome ◽

Clinical Prediction ◽

Clinical Prediction Models ◽

Surgery Research ◽

Development And Validation

Summary Clinical prediction models in anesthesia and surgery research have many clinical applications including preoperative risk stratification with implications for clinical utility in decision-making, resource utilization, and costs. It is imperative that predictive algorithms and multivariable models are validated in a suitable and comprehensive way in order to establish the robustness of the model in terms of accuracy, predictive ability, reliability, and generalizability. The purpose of this article is to educate anesthesia researchers at an introductory level on important statistical concepts involved with development and validation of multivariable prediction models for a binary outcome. Methods covered include assessments of discrimination and calibration through internal and external validation. An anesthesia research publication is examined to illustrate the process and presentation of multivariable prediction model development and validation for a binary outcome. Properly assessing the statistical and clinical validity of a multivariable prediction model is essential for reassuring the generalizability and reproducibility of the published tool.

Download Full-text