You are invited to Yuridia Leyva's MS Statistics defense this afternoon.

Yuridia Leyva
MS Statistics defense
Mon 4/27/2015, 2-3pm
SMLC 352

Adivsors:
Erik Erhardt, Mathematics and Statistics, UNM
Kimberly Page, Division Chief of Epidemiology, Internal Medicine, UNM
Title:
Per-contact infectivity of HCV associated with injection exposures in a prospective cohort of
young injection drug users in San Francisco, CA (UFO Study)

Abstract:
Sharing needles and ancillary injection drug equipment places injection drug users (IDU) at risk
for Hepatitis C Virus (HCV), a highly infectious blood-borne virus. A limited number of studies
have analyzed the per-contact infectivity of HCV associated with the use of previously-used
needles, but per-contact infectivity of ancillary injecting equipment has not been previously
investigated. Our goal is to estimate the per-contact infectivity of HCV associated with (1)
injecting with another person's previously-used needle, classi ed as receptive needle sharing
(RNS), and (2) using another person's previously-used ancillary injecting equipment, such as
cookers to melt drugs and cottons to strain impurities from the melted drugs, termed
receptive equipment sharing (RES). Estimates of per-contact probabilities were calculated based on
self-reported exposures to RNS and RES. A probabilistic exposure model was used on the UFO (yoU
Find Out) dataset composed of 784 IDU under the age of 30 who were surveyed quarterly between
2003-2008 and 2010-2014. For each participant, we selected the rst survey with an HCV seronegative
status up through their next seropositive survey, leaving us with 505 subjects on whom to conduct
the analysis. A marginal maximum likelihood estimate (MLE) considering only RNS gives
a per-contact infectivity of HCV as 0.39% (95% CI: 0.188% - 0.679%). A joint MLE gives RNS as
0.44% (95% CI: 0.0001% - 0.600%) and RES as 0% (95% CI: 0.0001% - 0.690%), thus needles are a much
bigger cause of concern than equipment. Though both probabilities are small, 13% (65/505) of the
subjects studied seroconverted to an HCV-postitive status. Strategies for reducing RNS, and RES to
a lesser extent, are important for reducing the spread of HCV and it's related maladies.