Related Experiment Video
Updated: Jun 5, 2025
'Boden Food Plate': Novel Interactive Web-based Method for the Assessment of Dietary Intake
Published on: September 18, 2018
Predicting food insecurity in a pediatric population using the electronic health record
Joseph Rigdon1, Kimberly Montez2,3, Deepak Palakshappa2,3,4,5,6
1Department of Biostatistics and Data Science, Wake Forest School of Medicine, Winston-Salem, USA.
Insights
A new prediction model can identify children facing food insecurity (FI) using electronic health records. This tool aids healthcare providers in addressing FI among pediatric patients.
Area of Science:
- Pediatric Healthcare
- Public Health
- Health Informatics
Background:
- Over 5 million US children experience food insecurity (FI), a critical unmet social need.
- Limited guidance exists for healthcare systems to effectively screen for pediatric FI.
- Electronic health records (EHRs) contain valuable data for predicting FI.
Purpose of the Study:
- To develop and validate a prediction model for identifying pediatric food insecurity.
- To utilize demographic, geographic, medical, and historical unmet health-related social needs data from EHRs.
- To enhance healthcare system capabilities in addressing childhood FI.
Main Methods:
- Retrospective longitudinal cohort study of children in an academic pediatric primary care clinic (2017-2021).
- Incorporated American Community Survey data for neighborhood-level socioeconomic factors.
- Compared logistic regression, random forest, and gradient-boosted machine models for FI prediction.
Main Results:
- Logistic regression with a 12-month look-back and neighborhood variables achieved the best performance (C-statistic 0.70, PPV 0.92).
- This model outperformed random forest (C=0.65) and gradient boosted machine (C=0.68) models.
- Model performance remained consistent even when handling missing data.
Conclusions:
- The developed models demonstrate predictive capability for pediatric food insecurity.
- Further research is necessary to refine and enhance the robustness of pediatric FI prediction models.
- Improved prediction tools can support targeted interventions for at-risk children.
Introduction:
More than 5 million children in the United States experience food insecurity (FI), yet little guidance exists regarding screening for FI. A prediction model of FI could be useful for healthcare systems and practices working to identify and address children with FI. Our objective was to predict FI using demographic, geographic, medical, and historic unmet health-related social needs data available within most electronic health records.
Methods:
This was a retrospective longitudinal cohort study of children evaluated in an academic pediatric primary care clinic and screened at least once for FI between January 2017 and August 2021. American Community Survey Data provided additional insight into neighborhood-level information such as home ownership and poverty level. Household FI was screened using two validated questions. Various combinations of predictor variables and modeling approaches, including logistic regression, random forest, and gradient-boosted machine, were used to build and validate prediction models.
Results:
A total of 25,214 encounters from 8521 unique patients were included, with FI present in 3820 (15%) encounters. Logistic regression with a 12-month look-back using census block group neighborhood variables showed the best performance in the test set (C-statistic 0.70, positive predictive value 0.92), had superior C-statistics to both random forest (0.65, p < 0.01) and gradient boosted machine (0.68, p = 0.01), and showed the best calibration. Results were nearly unchanged when coding missing data as a category.
Conclusions:
Although our models could predict FI, further work is needed to develop a more robust prediction model for pediatric FI.
More Related Videos
Related Concept Videos
01:30Methods of Documentation VII: EMR
01:19Purpose of Health Records II