Control charts for health care monitoring: the heterogeneous case

Willem/Wim Albers

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    18 Downloads (Pure)

    Abstract

    Attribute data from high quality processes can be monitored adequately by using negative binomial charts. The optimal choice for the number r of failures involved depends on the expected rate of change in failure rate during Out-of-Control. To begin with, such results have been obtained for the case of homogeneous data. But especially in health care monitoring, (groups of) patients will often show large heterogeneity. In the present paper we will present an overview of how this problem can be dealt with. Two situations occur: the underlying structure is either unknown (the overdispersion case) or known (risk adjustment feasible). An additional complication to be dealt with is the fact that in practice typically all parameters involved are unknown. Hence estimated versions of the new proposals need to be discussed as well.
    Original languageUndefined
    Title of host publicationProceedings of the Joint Research Conference on Statistics in Quality, Industry & Technology, QPRC 2010
    Place of PublicationAlexandria, VA, USA
    PublisherAmerican Statistical Association
    Pages5837-5846
    Number of pages10
    ISBN (Print)not assigned
    Publication statusPublished - 2010
    EventJoint Research Conference on Statistics in Quality, Industry & Technology, QPRC 2010 - Gaithersburg, MD
    Duration: 24 May 201028 May 2010

    Publication series

    NameJSM Proceedings
    PublisherAmerican Statistical Association
    Volume2010

    Conference

    ConferenceJoint Research Conference on Statistics in Quality, Industry & Technology, QPRC 2010
    Period24/05/1028/05/10
    Other24-28 May, 2010

    Keywords

    • IR-75593
    • METIS-275843
    • Statistical Process Control
    • Estimated parameters
    • EWI-19298
    • High quality processes
    • Geometric charts
    • Average run length
    • Heterogeneity

    Cite this