Identification of Gender Differences in Acute Myocardial Infarction Presentation and Management at Aga Khan University Hospital-Pakistan: Natural Language Processing Application in a Dataset of Patients With Cardiovascular Disease
2024

Gender Differences in Heart Attack Symptoms and Treatment in Pakistan

Sample size: 5358 publication Evidence: high

Author Information

Author(s): Mavragani Amaryllis, Pevnick Joshua, Papakonstantinou Trisevgeni, Ngaruiya Christine, Samad Zainab, Tajuddin Salma, Nasim Zarmeen, Leff Rebecca, Farhad Awais, Pires Kyle, Khan Muhammad Alamgir, Hartz Lauren, Safdar Basmah

Primary Institution: Aga Khan University Hospital

Hypothesis

This study aims to assess gender differences in the symptoms and management of patients hospitalized with acute myocardial infarction at a tertiary care hospital in Pakistan using natural language processing.

Conclusion

The study found significant gender-based differences in clinical presentation and medication management for patients with acute myocardial infarction in Pakistan.

Supporting Evidence

  • Women had higher odds of presenting with shortness of breath (OR 1.46) and lower odds of presenting with chest pain (OR 0.65).
  • Women had nearly 30% lower odds of being prescribed statins (OR 0.71) and 40% lower odds of being prescribed β-blockers (OR 0.67).
  • Statistically significant differences in symptom presentation were found, with women more likely to report atypical symptoms.

Takeaway

The study shows that women and men experience heart attacks differently, with women more likely to have symptoms like shortness of breath instead of chest pain, and they receive less medication.

Methodology

Natural language processing was used to analyze 5358 discharge summaries from patients with acute myocardial infarction between 1988 and 2018.

Potential Biases

Potential confirmation bias among the research team members could affect the model's quality.

Limitations

The study may be subject to selection bias due to missing data and lacks traditional expert medical record review for validating the NLP methodology.

Participant Demographics

The sample included 1769 women and 3589 men, with mean ages of 67.8 years for women and 63.3 years for men.

Statistical Information

Confidence Interval

95% CI 1.26-1.70 for shortness of breath; 95% CI 0.57-0.90 for statins; 95% CI 0.57-0.78 for β-blockers.

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.2196/42774

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication