Document Type

Thesis

Date of Degree

Spring 2016

Access Restrictions

Access restricted until 07/13/2018

Degree Name

MS (Master of Science)

Degree In

Epidemiology

First Advisor

Marizen Ramirez

Abstract

Areas of high crime often have an increased need for mental health services for the victims and perpetrators of violent crime. Access to these types of services can vary significantly throughout the US, and geographic proximity is one barrier that an individual in need of treatment must overcome, which is especially difficult in rural areas.

The long term goal of this study was to inform communities on gaps in the need for mental health services as they relate to crime rates in the state of Iowa. The objective is to measure the association between crime and mental health service geographic proximity at the census block level using ESRI CrimeRisk index scores and US Census data. Geographic proximity to mental health care facilities in Iowa was defined as the ratio of the number facilities within a 5, 10, and 30 miles radius of the center of all block groups in Iowa to the population size of that block group.

There are disparities in crime and geographic proximity to mental health facilities. While more than 70% of Iowa’s block groups have a total crime index score below the national average, about 18% have total crime levels above the national average and some are very high at six or seven times the national average. There are also disparities in geographic proximity to mental health facilities, as much of the state has low geographic proximity to provider within 5 or 10 miles of their home and moderate geographic proximity within 30 miles of their home.

Relative ratios produced by negative binomial models using a variety of different crime type index scores (total crime, murder, rape, assault) and predicting the facility-to-population ratio within a specified distance of block groups (5, 10, and 30 miles), showed that overall, the ratio of facilities-to-population increases as crime increases. For example, when predicting the facility-to-population ratio at 5 miles using the murder index scores, there was a relative ratio of 1.93 (CL= 1.76, 2.12), or a near doubling (93% increase) of the number of facilities when increasing each crime category

Rurality is an effect measure modifier for the association between crime index scores and the facility-to-population ratio. The relative ratios tended to be highest in urban block groups and lower in isolated rural block groups. Additionally, the facilities-to-population relative ratios tend to be lower overall in areas of high violent crime like rape and murder, compared to areas of lower violent crime.

From the geographically weight regression model, there was a large range in crime index scores and geographic proximity to mental health facilities in Iowa. The facility-to population relative ratios show that, despite that increases in crime in Iowa are associated with an increase in available mental health providers, increases are not equitable throughout the state and often depend on how rural an area is. The relative ratios also show that some types of violent crime, notably rape and assault, have smaller increases in geographic proximity to mental health facilities than other types of crime, which is concerning given the high rates of victimization and trauma associated with those crimes. By using model selection and geographically weighted regression, we can see that the model fit varies over space as does crime’s association with provider access, giving more resolution to these conclusions.

Keywords

Crime, Geography, Mental health, Public health

Pages

x, 64 pages

Bibliography

Includes bibliographical references (pages 55-60).

Copyright

Copyright © 2016 Elizabeth Mello

Available for download on Friday, July 13, 2018

Share

COinS