Defense Date

2008

Document Type

Thesis

Degree Name

Master of Public Health

Department

Epidemiology & Community Health

First Advisor

Gonzalo Bearman

Abstract

Background: Schizophrenia is a chronic, debilitating mental disorder characterized by positive (e.g., hallucinations, delusions) and negative (e.g., catatonia, flat affect) signs and symptoms. Many studies suggest that individuals born in winter or spring months are at increased risk for schizophrenia. Study Objectives: 1) To determine whether season of birth affects risk for schizophrenia in the Irish Study of High Density Schizophrenia Families (ISHDSF). 2) To examine, by computer simulation, power to detect genetic associations with schizophrenia under a variety of conditions and using different analytic strategies. 3) To test whether specific genes are associated with schizophrenia in the Irish Case Control Schizophrenia Study (ICCSS), using different analytic strategies to account for season of birth. Methods: A case-control design was used to examine the relationship between schizophrenia and season of birth. Cases were individuals from the ISHDSF diagnosed with schizophrenia. Controls were the general population of Ireland, with data provided by Ireland’s Central Statistics Office (CSO). The birth frequencies for each month or quarter were compared in the two groups by a chi square test. Computer simulations were conducted to examine power to detect schizophrenia susceptibility loci using either all cases or only cases born in high-risk months, under different conditions and models for how genetic risk and season of birth interact to influence risk for schizophrenia. Data for six genetic markers from the ICCSS were analyzed for evidence of association, using all cases, and then using only cases born in high-risk months. Setting and Study Participants: ISHDSF families were ascertained through inpatient psychiatric care facilities serving >95% of the Irish population. Diagnoses were established using DSM-III-R criteria, and birthdates were recorded for all individuals. The Irish CSO provided aggregate, population-level data for number of births in Ireland by month for the years 1976-2000 and by quarter for the years 1900-2000. The ICCSS is a sample of schizophrenic and control individuals who have been genotyped at many loci across the genome. Schizophrenics were ascertained through in- and outpatient psychiatric facilities, had diagnoses verified by an expert, and their birthdates recorded. Controls were selected from several sources, e.g. blood donation centers, and denied any lifetime history of schizophrenia. For each subject in the ICCSS, all four grandparents were born in Ireland or the United Kingdom. Results: Number of births in each month was compared for schizophrenics in the ISHDSF and general population controls, resulting in a chi square of 19.44 (p value ~ 0.054, 11 df). Simulations revealed that, in some circumstances, power to detect genetic associations was increased by restricting cases to those born in high-risk months. Analysis of data from the ICCSS revealed that restricting cases to high-risk birth months increased the evidence for association for three of six markers tested, two of which were associated with the gene FBXL21. Conclusions: Birth in the months of March, April, or May appears to be associated with elevated risk for schizophrenia in the ISHDSF. In attempting to find susceptibility loci for schizophrenia, restricting genetic association analyses to schizophrenics born in high-risk months may result in increased power to detect genetic association in some circumstances.

Rights

© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

December 2008

Included in

Epidemiology Commons

Share

COinS