Multivariate Markovian Modeling of Tuberculosis: Forecast for the United States
2000

Modeling Tuberculosis Incidence in the U.S.

publication Evidence: moderate

Author Information

Author(s): Sara M. Debanne, Roger A. Bielefeld, George M. Cauthen, Thomas M. Daniel, Douglas Y. Rowland

Primary Institution: Case Western Reserve University

Hypothesis

Can a Markov chain model accurately project tuberculosis incidence in the United States across different demographic groups?

Conclusion

The model predicts a decline in tuberculosis cases in the U.S., with varying rates of decline among different racial and ethnic groups.

Supporting Evidence

  • The model projects an intermediate increase in national TB incidence followed by a decline.
  • Projections indicate that the rate of decline among Hispanics will be slower than for white non-Hispanics and black non-Hispanics.
  • Model performance is best for large subgroups with high case rates.

Takeaway

The researchers created a computer model to predict how many people will get tuberculosis in the U.S. over the next few years, showing that some groups will have more cases than others.

Methodology

A multivariate Markov chain model was developed using population data disaggregated by race, ethnicity, age, and geography to project TB incidence from 1980 to 2010.

Potential Biases

Potential inaccuracies in initial conditions and model parameters could lead to projection errors.

Limitations

The model does not account for changes in TB control measures and may not accurately reflect geographic disease patterns.

Participant Demographics

The model considers various demographic groups defined by age, race, ethnicity, and geographic location.

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication