Files in this item



application/pdfImelda_Moise.pdf (5MB)
(no description provided)PDF


Title:Health risk differentials: implications of neighborhood conditions on various health outcomes in New Orleans, 2004-2009
Author(s):Moise, Imelda
Director of Research:Ruiz, Marilyn O.
Doctoral Committee Chair(s):Kalipeni, Ezekiel
Doctoral Committee Member(s):Ruiz, Marilyn O.; McLafferty, Sara L.; Wang, Shaowen; Grigsby-Toussaint, Diana S.; Allan, Brian F.
Department / Program:Geography & Geographic InfoSci
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):New Orleans
Abandoned Swimming Pools
Mental Health
Geographic Information System (GIS)
Abstract:This dissertation examines the effects of neighborhood conditions on health outcomes in New Orleans pre- and post-hurricane Katrina, an important topic in the context of community recovery following a disaster event. This dissertation follows the “three chapter (paper) journal format”. The first paper (Chapter 2) examines hot spots for unattended swimming pools and associated neighborhood predictors from 2006 to 2009, a period during which the city experienced widespread devastation from hurricanes Katrina and Rita. A surfeit of unattended pools on properties abandoned by displaced residents has a number of potential direct and indirect negative health impacts for returning local residents including elevated transmission rates due to a punctuated increase in the relative abundance of mosquito breeding sites. Based upon the Broken Windows theory, spatial variables describing physical and socioeconomic neighborhood characteristics hypothesized to influence the density of unattended pools in the study area were identified to answer the following core research questions: 1) what is the spatial structure of unattended pools over time during post-Katrina New Orleans? 2) What neighborhood predictors are associated with this variability? 3) How do demographic characteristics and neighborhood structural deterioration affect the odds of a pool being unattended in 2006 immediately after Katrina compared to the longer period from 2006 to 2008 when recovery dynamics may reflect more than the physical effects of the hurricane. The results indicate that traditional variables associated with neighborhood deterioration (e.g. vacant houses, properties with major structural damages); do not provide as robust an explanation for presence of unattended pools in study area. This suggests that it may be more of the effect of neighborhood demographics and underlying preexisting characteristics of neighborhood conditions influencing unattended pool presence rather than the physical deterioration of neighborhoods of which unmaintained pools contribute. The second paper (Chapter 3) is an observational study whose aim is to identify neighborhood and socio-demographic factors associated with changes in mental health disorders in New Orleans, Louisiana for three time points (2004, 2008 and 2009). In addition, this study examines the extent to which the rate of hospitalization varies across individual characteristics such as gender, race, marital status and income. Cases include all hospital admissions reported to the Louisiana Department of Health and Hospitals’ database but exclude military and veteran administration hospitals. Predictor and explanatory (patient data) variables were first geocoded and aggregated to the block group using geographical information systems (GIS) ArcGIS 9.3 software. Logistic regression was used to explore the relationship between predictor and outcome variables. Results indicate that (1) the hospitalization rates of mood and anxiety and psychosomatic disorders have decreased over time. The rate in 2008 was higher than that in 2004(for mood and anxiety and for psychosomatic disorders (P=0.000), (2) No statistically significant change was observed in substance abuse hospitalizations over time. (3) Protective factors for mental health disorders included recovery factors and population white. (4) Risk factors include neighborhood decay variables and demographic characteristics. More men than women were admitted overall. Gender, race, marital status and income affected how patients received treatment for stress-related mental disorders. In the third paper (Chapter 4) I correlate temporal and spatial changes in mosquito abundance with neighborhood, socio-demographic, and land cover factors (2006 to 2010) with Culex(Cx.) quinquefasciatus species abundance. Neighborhood and socio-demographic covariates recorded for each trap location included abandoned swimming pools, imminent health threat listed properties, public nuisance property listings, completed property demolitions, population, household income and number of households and a disaster impact variable water depth. Both spatial (GIS and remote sensing) and non-spatial techniques (logistic regression) are used. Predictors of Cx. quinquefasciatus species abundance in New Orleans include land cover variables, temperature, large water bodies and highly developed areas, and a neighborhood variable or completed property demolitions. The findings of this study suggest that land cover and neighborhood characteristics predict mosquito abundance in the study area more than socio-demographic characteristics. The current findings add to a growing body of literature on vector-borne disease risk associated with disasters. The findings of this chapter have a number of important implications for mosquito control programs and supports earlier research findings that vector-borne disease risks increase rapidly post-disaster representing a greater public health threat.
Issue Date:2013-02-03
Rights Information:Copyright 2012 Imelda Moise
Date Available in IDEALS:2013-02-03
Date Deposited:2012-12

This item appears in the following Collection(s)

Item Statistics