Efficiency of two-phase methods with focus on a planned population-based case-control study on air pollution and stroke
2007

Study on Air Pollution and Stroke Risk

Sample size: 1600 publication Evidence: high

Author Information

Author(s): Anna Oudin, Jonas Björk, Ulf Strömberg

Primary Institution: Lund University Hospital

Hypothesis

Does exposure to nitrogen dioxide (NO2) increase the risk of stroke?

Conclusion

Method 4 showed the best performance for improving efficiency while adjusting for varying participation rates across areas.

Supporting Evidence

  • The study simulates a case-control study based on planned research on air pollution and stroke.
  • Method 4 generally outperformed other methods in terms of efficiency.
  • The study uses individual disease status and residential address linked to register data on NO2 exposure.

Takeaway

The study looks at how air pollution might cause strokes and tests different ways to collect data to get the best results.

Methodology

A two-phase case-control study was simulated with varying numbers of first- and second-phase subjects, using different estimation methods.

Potential Biases

Participation bias may occur if control participation decreases in areas with high pollution.

Limitations

The study assumes no selection bias arises from the recruitment of first-phase cases and controls.

Participant Demographics

The study involves a large population sample from Scania, southern Sweden, with individual-level data on stroke occurrence and exposure.

Digital Object Identifier (DOI)

10.1186/1476-069X-6-34

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication