Comparing two large data repositories to understand the differences in demographics, health history, and behavioral attributes in populations
2024

Comparing Two Large Data Repositories for Health Insights

Sample size: 368963 publication 10 minutes Evidence: high

Author Information

Author(s): Nasiha Maliq Nihmath, Ong Toan, Giano Zachary, Rivera William, Tiwari Tamanna

Primary Institution: University of Colorado

Hypothesis

What are the differences in demographics, health history, and behavioral attributes between the All of Us and BigMouth data repositories?

Conclusion

The study found significant differences in health outcomes and demographics between the All of Us and BigMouth datasets, highlighting the need for targeted health interventions.

Supporting Evidence

  • 80.6% of Hispanic or Latino participants in AoU reported alcohol use compared to 16.8% in BigMouth.
  • 87.9% of females in AoU reported alcohol use, contrasting with 26.0% in BigMouth.
  • Diabetes prevalence among females was 8.8% in AoU vs. 21.6% in BigMouth.
  • 49.2% of Hispanic or Latino participants in AoU reported health literacy, compared to 3.2% in BigMouth.
  • 70.1% of Hispanic or Latino respondents in AoU reported satisfactory health status, while BigMouth indicated 98.3%.

Takeaway

This study looked at two big health data sets to see how people's health and behaviors differ. It found that people reported very different things about their health depending on which data set they were in.

Methodology

The study used descriptive statistics, Chi-square tests, odds ratios, and 95% confidence intervals to analyze data from the All of Us and BigMouth datasets.

Potential Biases

Self-reported data may be subject to social desirability bias, and differences in survey methodologies could influence results.

Limitations

The datasets may not be fully representative of the wider population, and there are risks of omitted variable bias and inherent biases in data collection.

Participant Demographics

The study included diverse participants from both datasets, with notable differences in race, ethnicity, and gender distributions.

Statistical Information

Confidence Interval

(20.36, 21.10)

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.3389/froh.2024.1427109

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication