Development and validation of a pancreatic cancer risk model for the general population using electronic health records: An observational study

2021 ◽  
Vol 143 ◽  
pp. 19-30
Author(s):  
Limor Appelbaum ◽  
José P. Cambronero ◽  
Jennifer P. Stevens ◽  
Steven Horng ◽  
Karla Pollick ◽  
...  
Keyword(s):  
Pancreatic Cancer ◽  
General Population ◽  
Cancer Risk ◽  
Observational Study ◽  
Electronic Health Records ◽  
Risk Model ◽  
Health Records ◽  
Pancreatic Cancer Risk ◽  
Electronic Health ◽  
Development And Validation

Development and validation of various phenotyping algorithms for Diabetes Mellitus using data from electronic health records

2017 ◽  
Vol 152 ◽  
pp. 53-70 ◽  
Author(s):  
Santiago Esteban ◽  
Manuel Rodríguez Tablado ◽  
Francisco E. Peper ◽  
Yamila S. Mahumud ◽  
Ricardo I. Ricci ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Electronic Health Records ◽  
Health Records ◽  
Electronic Health ◽  
Using Data ◽  
Development And Validation

scholarly journals What characterizes cancer family history collection tools? A critical literature review

2018 ◽  
Vol 25 (4) ◽  
Author(s):  
J.E. Cleophat ◽  
H. Nabi ◽  
S. Pelletier ◽  
K. Bouchard ◽  
M. Dorval
Keyword(s):  
Risk Assessment ◽  
Family History ◽  
Cancer Risk ◽  
Electronic Health Records ◽  
Health Professionals ◽  
Cancer Risk Assessment ◽  
Health Records ◽  
Cancer Family ◽  
Cancer Family History ◽  
Electronic Health

Background Many tools have been developed for the standardized collection of cancer family history (fh). However, it remains unclear which tools have the potential to help health professionals overcome traditional barriers to collecting such histories. In this review, we describe the characteristics, validation process, and performance of existing tools and appraise the extent to which those tools can support health professionals in identifying and managing at-risk individuals.Methods Studies were identified through searches of the medline, embase, and Cochrane central databases from October 2015 to September 2016. Articles were included if they described a cancer fh collection tool, its use, and its validation process.Results Based on seventy-nine articles published between February 1978 and September 2016, 62 tools were identified. Most of the tools were paper-based and designed to be self-administered by lay individuals. One quarter of the tools could automatically produce pedigrees, provide cancer-risk assessment, and deliver evidence-based recommendations. One third of the tools were validated against a standard reference for collected fh quality and cancer-risk assessment. Only 3 tools were integrated into an electronic health records system.Conclusions In the present review, we found no tool with characteristics that might make it an efficient clinical support for health care providers in cancer-risk identification and management. Adequately validated tools that are connected to electronic health records are needed to encourage the systematic identification of individuals at increased risk of cancer.

scholarly journals Development and validation of an algorithm to estimate the risk of severe complications of COVID-19 to prioritise vaccination

2021 ◽  
Author(s):  
Ron MC Herings ◽  
Karin MA Swart ◽  
Bernard AM Van der Zeijst ◽  
Amber A van der Heijden ◽  
Koos van der Velden ◽  
...  
Keyword(s):  
General Population ◽  
Electronic Health Records ◽  
Community Dwelling ◽  
Prediction Algorithm ◽  
List Type ◽  
Health Records ◽  
C Statistic ◽  
Morbidity Score ◽  
Confirmation Test ◽  
Electronic Health

AbstractObjectiveTo develop an algorithm (sCOVID) to predict the risk of severe complications of COVID- 19 in a community-dwelling population to optimise vaccination scenarios.DesignPopulation based cohort studySetting264 Dutch general practices contributing to the NL-COVID databaseParticipants6074 people aged 0-99 diagnosed with COVID-19Main outcome measuresSevere complications (hospitalisation, institutionalisation, death). The algorithm was developed from a training dataset comprising 70% of the patients and validated in the remaining 30%. Potential predictor variables included age, sex, a chronic co-morbidity score (CCS) based on risk factors for COVID-19 complications as defined by the National Institute of Public Health and the Environment (RIVM), obesity, neighborhood deprivation score (NDS), first or second COVID wave, and confirmation test. Six different population vaccination scenarios were explored: 1) random (naive), 2) random for persons above 60 years (60plus), 3) oldest patients first in age bands of five years (oldest first), 4) target population of the annual influenza vaccination program (influenza) and 5) those 25-65 years of age first (worker), and 6) risk-based using the prediction algorithm (sCOVID). For each vaccination strategy the amount of vaccinations needed to reach a 50% reduction of severe complications was calculated.ResultsSevere complications were reported in 243 (4.8%) people with 59 (20.3%) nursing home admissions, 181 (62.2%) hospitalisations and 51 (17.5%) deaths. The algorithm included age, sex, CCS, NDS, wave, and confirmation test with a c statistic of 0.91 (95% CI 0.88-0.94) in the validation set. Applied to different vaccination scenarios, the proportion of people needed to be vaccinated to reach a 50% reduction of severe complications was 67.5%, 50.0%, 26.1%, 16.0%, 10.0%, and 8.4% for the worker, naive, infuenza, 60plus, oldest first, and sCOVID scenarios respectively.ConclusionCOVID-19 related severe complications will be reduced most efficiently when vaccinations are risk-based, prioritizing the highest risk group using the sCOVID algorithm. The vaccination scenario, prioritising oldest people in age bands of 5 years down to 60 years of age, performed second best. The sCOVID algorithm can readily be applied to identify persons with highest risks from data in the electronic health records of GPs.What is already known on this topic?Severe COVID-19 complications may be reduced when persons at the highest risk will be vaccinated first.To identify persons at a high risk for hospitalization or death in the general population, a limited number of prediction algorithms have been developed.Most of these algorithms were based on data from the first wave of infections (spring 2020) when widespread testing was not always possible, limiting the usefulness of these algorithms.What this study addsIncluding data up to January 2021, we developed and validated a prediction algorithm (sCOVID) with a c-statistic of 0.91 (95% CI 0.88-0.94) based on age, sex, chronic comorbidity score, economic status, wave, and a confirmation test to identify patients in the general population that are at risk of severe COVID-19 complication.Using the algorithm, a 50% reduction of patients with severe complications could be obtained with a vaccination coverage of only 8%. This vaccination scenario based on this algorithm was superior to other calculated vaccination scenarios.The sCOVID algorithm can readily be implemented in the electronic health records of general practitioners.

scholarly journals Development and Validation of an Electronic Frailty Index Using Routine Electronic Health Records: An Observational Study From a General Hospital in China

2021 ◽  
Vol 8 ◽  
Author(s):  
Yao-Dan Liang ◽  
Yi-Bo Xie ◽  
Ming-Hui Du ◽  
Jing Shi ◽  
Jie-Fu Yang ◽  
...  
Keyword(s):  
Electronic Health Records ◽  
Hospital Stay ◽  
General Hospital ◽  
Older Adult ◽  
Validation Cohort ◽  
Frailty Index ◽  
Health Records ◽  
Electronic Health ◽  
Independent Association ◽  
Development And Validation

Background: This study aimed to develop and validate an electronic frailty index (eFI) based on routine electronic health records (EHR) for older adult inpatients and to analyze the correlations between frailty and hospitalized events and costs.Methods: We created an eFI from routine EHR and validated the effectiveness by the consistency of the comprehensive geriatric assessment-frailty index (CGA-FI) with an independent prospective cohort. Then, we analyzed the correlations between frailty and hospitalized events and costs by regressions.Results: During the study period, 49,226 inpatients were included in the analysis, 42,821 (87.0%) of which had enough data to calculate an eFI. A strong correlation between the CGA-FI and eFI was shown in the validation cohort of 685 subjects (Pearson's r = 0.716, P < 0.001). The sensitivity and specificity for an eFI≥0.15, the upper tertile, to identify frailty, defined as a CGA-FI≥0.25, were 64.8 and 88.7%, respectively. After adjusting for age, sex, and operation, an eFI≥0.15 showed an independent association with long hospital stay (odds ratio [OR] = 2.889, P < 0.001) and death in hospital (OR = 19.97, P < 0.001). Moreover, eFI values (per 0.1) were positively associated with total costs (β = 0.453, P < 0.001), examination costs (β = 0.269, P < 0.001), treatment costs (β = 0.414, P < 0.001), nursing costs (β = 0.381, P < 0.001), pharmacy costs (β = 0.524, P < 0.001), and material costs (β = 0.578, P < 0.001) after adjusting aforementioned factors.Conclusions: We successfully developed an effective eFI from routine EHR from a general hospital in China. Frailty is an independent risk factor for long hospital stay and death in hospital. As the degree of frailty increases, the hospitalized costs increase accordingly.

Sign in / Sign up

Export Citation Format

Share Document