Inference on healthcare indicators for the treatment of HIV/AIDS (2001)

Author(s): Walshaw D; Heiner K

    Notes: The authors develop Bayesian models for binary healthcare indicators, which represent certain key procedures carried out in the care of patients with HIV/AIDS. These allow for the assessment of a newly introduced procedure called HIVQUAL, which incorporates self-reporting of the indicators by the institutions involved. The models treat the success probabilities as a Markov Chain, with each year's success probability being used to inform the subsequent year's prior. Covariates, such as institution type, are easily incorporated as covariates. The models developed here have been adopted by the Aids Institute of New York to inform its public data releases.

      • Conference Name: Bayesian Methods with Applications to Science, Policy and Official Statistics
      • Pages: 213-222
      • Publication type: Conference Proceedings (inc. abstract)
      • Bibliographic status: Published

        Dr David Walshaw