Assessing community variation and randomness in public health indicators
2011

Assessing Community Variation in Public Health Indicators

Sample size: 8301 publication 10 minutes Evidence: moderate

Author Information

Author(s): Arndt Stephan, Laura Acion, Kristin Caspers, Ousmane Diallo

Primary Institution: University of Iowa

Hypothesis

How much of the diversity among communities in public health indicators is due to random variation versus real differences?

Conclusion

The Pearson's chi-square measure is more effective for assessing community variation in public health indicators than the alternative method.

Supporting Evidence

  • The study found that the Pearson's chi-square method provided better estimates of true community variance.
  • More than 60% of community differences in the BRFSS data were attributed to random variability.
  • The IYS data indicated that over 85% of community variation was due to actual differences.

Takeaway

This study looks at how different communities show varying health needs and whether those differences are real or just random chance.

Methodology

The authors compared two statistical measures of heterogeneity to assess community differences in health indicators using simulations and real data.

Potential Biases

Potential bias due to random sampling errors affecting community health indicators.

Limitations

The study may not generalize to indices measured on fewer observations, and it assumes independent observations among communities.

Participant Demographics

Participants included individuals from 99 counties in Iowa, with varying sample sizes and demographics.

Statistical Information

P-Value

0.0002

Confidence Interval

0.075 to 0.234

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1478-7954-9-3

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication