Correction: Candidemia Risk Prediction (CanDETEC) Model for Patients With Malignancy: Model Development and Validation in a Single-Center Retrospective Study (Preprint)

The rising incidence of cutaneous melanoma over the past few decades has prompted substantial efforts to develop risk prediction models identifying people at high risk of developing melanoma to facilitate targeted screening programs. We review these models, regarding study characteristics, differences in risk factor selection and assessment, evaluation, and validation methods. Our systematic literature search revealed 40 studies comprising 46 different risk prediction models eligible for the review. Altogether, 35 different risk factors were part of the models with nevi being the most common one (n = 35, 78%); little consistency in other risk factors was observed. Results of an internal validation were reported for less than half of the studies (n = 18, 45%), and only 6 performed external validation. In terms of model performance, 29 studies assessed the discriminative ability of their models; other performance measures, e.g., regarding calibration or clinical usefulness, were rarely reported. Due to the substantial heterogeneity in risk factor selection and assessment as well as methodologic aspects of model development, direct comparisons between models are hardly possible. Uniform methodologic standards for the development and validation of risk prediction models for melanoma and reporting standards for the accompanying publications are necessary and need to be obligatory for that reason.

Download Full-text

Developing a spatial risk prediction model for child maltreatment

International Journal for Population Data Science ◽

10.23889/ijpds.v3i5.1084 ◽

2018 ◽

Vol 3 (5) ◽

Author(s):

Ken Steif ◽

Matthew Harris ◽

Dyann Daley

Keyword(s):

Machine Learning ◽

Child Maltreatment ◽

Risk Prediction ◽

Model Development ◽

Data Governance ◽

Household Level ◽

Spatial Risk ◽

Child Maltreatment Risk ◽

Integration Data ◽

Development And Validation

While predicting child maltreatment risk at the household level is useful for allocating limited child welfare resources, significant privacy, data integration, data governance and legal hurdles make such an algorithm economically and politically difficult to put into production. In this project, we take a different approach to child maltreatment risk prediction, developing machine learning models that predict, not for a household but for a small spatial areal unit, such as the block. The only private health data required for this use case are geocoded maltreatment events. We present the results of a machine learning analysis in Richmond Virginia, including exploratory analysis, feature engineering, model development and validation. We then interpret our models in a resource allocation context.

Download Full-text

Predicting risk of hospital admission in patients with suspected COVID-19 in a community setting : protocol for development and validation of a multivariate risk prediction tool (RECAP) (Preprint)

10.2196/preprints.29072 ◽

2021 ◽

Author(s):

Ana B Espinosa-Gonzalez ◽

Ana Luisa Neves ◽

Francesca Fiorentino ◽

Denys Prociuk ◽

Laiba Husain ◽

...

Keyword(s):

Primary Care ◽

Risk Prediction ◽

Patient Outcomes ◽

Model Development ◽

Community Setting ◽

Signs And Symptoms ◽

Multivariate Risk ◽

Primary Care Clinicians ◽

Development And Validation ◽

Remote Consultations

BACKGROUND During the pandemic, remote consultations have become the norm for the assessment of patients with signs and symptoms of COVID-19 in order to decrease the risk of transmission. This has added to the already existing challenges experienced by primary care clinicians when assessing suspected COVID-19 patients due to the uncertainty around disease progression (e.g., risk of deterioration around the 8th day of disease) and has prompted the use of risk prediction scores, such as NEWS2, to assess severity and guide treatment. However, the risk prediction tools available have not been validated in a community setting and have not been designed to capture the idiosyncrasy of COVID-19 infection. OBJECTIVE The objective of this study is to produce a multivariate risk prediction tool (RECAP–V1) to support primary care clinicians in the identification of those COVID-19 patients that are at higher risk of deterioration and facilitate the early escalation of their treatment with the aim of improving patient outcomes. METHODS The study follows a prospective cohort observational design, whereby patients presenting in primary or community care with signs and symptoms suggestive of COVID-19 will be followed and their data linked with hospital outcomes (hospital admission, intensive care unit admission and death). The collection of the primary data for the model will be carried out by primary care clinicians in four arms, i.e., North West London Clinical Commissioning Groups (NWL CCG), Royal College of General Practitioners Research and Surveillance Centre (RCGP RSC), Covid Clinical Assessment Service (CCAS) and South East London CCGs (Doctaly platform), and will involve the use of an electronic template that incorporates a list of items (known as RECAP-V0) thought to be associated with worse disease outcome according to previous qualitative work.. This data will be linked to patient outcomes in highly secure environments (iCARE and ORCHID secure environments). We will then use multivariate logistic regression analyses for model development and validation. RESULTS Recruitment of participants started in October 2021. Initially, only NWL CCGs and RCGP RSC arms were active. As of 24th of March 2021, we have recruited a combined sample of 3,827 participants in these two arms. CCAS and Doctaly joined the study in February 2021, with CCAS starting recruitment process on the 15th of March 2021. The first part of the analysis (RECAP-V1 model development) is planned to start in April 2021 using the first half of the NWL CCG and RCGP RSC combined datasets. Posteriorly, the model will be validated with the rest of NWL CCG and RCGP RSC data as well as CCAS and Doctaly datasets. The study was approved by the Research Ethics Committee on the 27th of May 2020 (IRAS number 283024, REC reference number: 20/NW/0266) and badged as NIHR Urgent Public Health Study on 14th of October 2020. CONCLUSIONS We believe the validated RECAP-V1 early warning score will be a valuable tool for the assessment of suspected COVID-19 patients’ severity in the community, either in face-to-face or remote consultations, and will facilitate the timely escalation of treatment with the potential to improve patient outcomes. CLINICALTRIAL ISRCTN registry (ISRCTN13953727)

Download Full-text