Data mining for identifying novel associations and temporal relationships with Charcot foot

Area of Science:

• Diabetology
• Neurology
• Rheumatology

Background:

• Charcot foot is a severe diabetes complication with debated etiology.
• Understanding associated disorders is key to understanding pathogenesis.
• Novel associations could reveal new insights into disease development.

Purpose of the Study:

• To identify novel associations and temporal relationships with Charcot foot using a large-scale data mining approach.
• To explore potential new pathways in Charcot foot pathogenesis.
• To uncover potential targets for earlier diagnosis and prevention.

Main Methods:

• Utilized a large health system database (1.6 million patients) with time-stamped ICD-9 codes.
• Conducted association analysis focusing on 388 patients diagnosed with Charcot foot (ICD-9 713.5).
• Analyzed 41.2 million time-stamped diagnosis codes to identify significant associations and temporal links.

Main Results:

• Identified 710 associations with Charcot foot, with 95.2% having P < 1.0 × 10⁻⁵ and 84.9% showing an odds ratio > 5.0.
• Found 111 (15.6%) associations with a significant temporal relationship (P < 1.0 × 10⁻³).
• Highlighted cardiac dysrhythmia, pulmonary eosinophilia, and volume depletion disorder as novel, temporally associated conditions.

Conclusions:

• Novel associations with Charcot foot suggest complex pathogenesis involving neurotrophic, neurovascular, and microtraumatic factors.
• Identified conditions like cardiac dysrhythmia offer new avenues for research into Charcot foot.
• Further investigation into these associations may lead to improved prevention and earlier diagnosis strategies for Charcot foot.