Pediatric Respiratory Infection Risk Factors from a 10-Year Cohort

A cohort of 6,026 children was followed from 2016 to 2025 to study recurrent respiratory tract infections. The dataset, created by Hehua Cai and shared under CC-BY-4.0, includes 1,227 cases and was used to develop a predictive nomogram based on clinical history and nutritional indicators. The model was trained on 4,219 children and tested on 1,807.

Use Cases

• Developing predictive models for recurrent respiratory infections based on clinical history and nutritional indicators.
• Validating risk stratification tools using a real-world cohort with a defined training and testing split.
• Analyzing the association between factors like vitamin levels, birth weight, and passive smoking with pediatric infection risk.
• Conducting decision-curve analysis to evaluate the clinical utility of a predictive nomogram.

Strengths

• Dataset includes 6,026 children with 1,227 events, providing a substantial sample size.
• Model development used a 70/30 split with a training set of 4,219 and a testing set of 1,807 for validation.
• Specific odds ratios for independent predictors like history of allergy (OR 5.187) and vitamin D levels (OR 0.556 per 10 ng/mL) are reported.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Data is stored in a DOCX file (12.8 KB), which suggests a summary table rather than a full raw dataset, limiting direct analysis.
• The geographic origin of the cohort is unknown, which may limit generalizability.

Provenance

Source

Hehua Cai via figshare

Collection Method

Real-world cohort study following predefined screening and follow-up criteria.

Time Range

2016–2025

Freshness

Last updated 2026-05-12 05:43:16; freshness should be verified.