Browsing by Subject "misclassification"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Estimation of a selectivity model with misclassified selection
    (Taylor & Francis, 2022) Tennekoon, Vidhura S. B. W.; Caudill, Steven B.; Economics, School of Liberal Arts
    Despite the great interest in models of self-selection and models with misclassification, there have been few studies combining the two. Notable exceptions are given by McCarthy, Millimet, and Roy and Shiu. None of these models have been developed in a contingent valuation setting that we are interested in. The goal of this note is to add to this literature by presenting a model for estimating willingness to pay using data collected through a contingent valuation survey. We examine the case of a selectivity model in which the outcome equation is interval censored but the decision indicator is not observed.
  • Thumbnail Image
    Item
    High Unreported Mortality in Children and Youth (<25 Years) Living With HIV Who Were Lost to Care From Antiretroviral Therapy Programs in Southern Africa: Results From a Multicountry Tracing Study
    (Wolters Kluwer, 2022-12-15) Nyakato, Patience; Christ, Benedikt; Anderegg, Nanina; Muhairwe, Josephine; Jefferys, Laura; van Dijk, Janneke; Vinikoor, Michael J.; van Lettow, Monique; Chimbetete, Cleophas; Phiri, Sam J.; Egger, Matthias; Ballif, Marie; Yiannoutsos, Constantin T.; Schomaker, Michael; Kassanjee, Reshma; Davies, Mary-Ann; Cornell, Morna; International epidemiology Databases to Evaluate AIDS Southern Africa (IeDEA-SA); Biostatistics and Health Data Science, Richard M. Fairbanks School of Public Health
    Background: Antiretroviral therapy program mortality maybe underestimated if deceased patients are misclassified as lost. Methods: We used two-stage inverse probability weighting to account for probability of being: sampled for tracing and found by the tracer. Results: Among 680 children and youth aged <25 years on antiretroviral therapy who were lost and traced in Southern Africa between October 2017 and November 2019, estimated mortality was high at 9.1% (62/680). After adjusting for measured covariates and within-site clustering, mortality remained lower for young adults aged 20–24 years compared with infants aged <2 years [adjusted hazard ratio: 0.40 (95% confidence interval: 0.31 to 0.51)]. Conclusions: Our study confirms high unreported mortality in children and youth who are lost and the need for tracing to assess vital status among those who are lost to accurately report on program mortality.