Analysis of Zero Altered & Inflated Data

Zero altered models were fitted and applied to two datasets (one from Malawi and another from Zambia). Malawi data were collected in a cluster randomized study, in Chikhwawa district in 2004, with 18 villages randomized to intervention and control arms with a total of 1642 participants. Zambia data were collected from school children in a cross-sectional study in Lusaka province in 2004 with a total of 2040 participants. Results from the study showed that Negative Binomial Logit Hurdle (NBLH) model offered best-fit to data inflated with zeros; with capability to handle over-dispersion, excess zeros and capture true zeros in the data. Its implementation and interpretation, ease of components, and its direct link with observed data make it a valuable alternative for analyzing zero inflated count data. Conclusions drawn from the study indicated that Helminths were highly localized, with small section of people harboring parasites; showing heterogeneous infection risk for both Malawi and Zambia settings. Joint modeling approach allowed identification of risk factors for infection presence and severity hence provide a platform to design combative control efforts.

Autorentext

Michael Give Chipeta, BSc., MSc Biostatistics: Studied Mathematical sciences (Statistics and Computing) at the University of Malawi, Malawi Polytechnic. He went on to do Masters in Bio statistics at University of Malawi, Chancellor College. Currently working as a Data Manager/Statistician at Malawi-Liverpool-Wellcome Trust in Blantyre - Malawi.