Early prediction of median survival among a large AIDS surveillance cohort
2007

Predicting Median Survival in AIDS Patients

Sample size: 96373 publication Evidence: moderate

Author Information

Author(s): Enanoria Wayne T, Hubbard Alan E, van der Laan Mark J, Chen Mi, Ruiz Juan, Colford John M Jr

Primary Institution: University of California at Berkeley

Hypothesis

Can early survival quantiles accurately predict median survival for AIDS patients?

Conclusion

The study suggests that early prediction of median survival time after AIDS diagnosis may be possible using early quantiles of the survival distribution.

Supporting Evidence

  • Using early quantiles, median survival for four cohorts was predicted to be 34, 34, 31, and 29 months.
  • The actual median survival for these cohorts was found to be 32, 40, 46, and 80 months.
  • The methodology allows predictions to be made almost four years earlier than traditional methods.

Takeaway

This study found a way to guess how long people with AIDS might live after they are diagnosed, using early data to make predictions.

Methodology

The study used AIDS surveillance data to estimate early survival quantiles and predict median survival using a linear model.

Potential Biases

Potential biases due to delays in reporting deaths and diagnosis.

Limitations

The methodology may not work well during periods of rapid changes in survival rates due to new treatments.

Participant Demographics

Individuals diagnosed with AIDS in California from 1983 to 1996.

Digital Object Identifier (DOI)

10.1186/1471-2458-7-127

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication