HMC Supplement Progress Report -
DEPRESSION
February 2006
Supplement
Title:
The Role of Depression and Context in Health Behavior
An Administrative Supplement to:
Preventing
Suicide
Risk Behaviors: Long Term Change
Supplement Chair/
Affiliation:
Brooke
Randell,
DNSc,
CS -
University of Washington School of Nursing
Purpose:
To apply an ecological framework of neighborhood effects on individual
health by exploring the
relationship between depression and health change for a broad set of
health outcomes – suicide risk,
sexual practices, diet adherence, and smoking cessation – and
specifically exploring how the context
in which an individual is embedded contributes to and moderates these
effects.
Depression
Sharepoint
Group:
Sharepoint Contact – Karen
Snedker
(In addition to HMCRC Members)
Progress to Date:
The lead
site has received approval from the University of Washington’s Human
Subjects Division for the supplemental grant. Other sites are at various
stages of the process of receiving human subjects
approval.
The lead site has
created a geocoding protocol that details
the procedure for address matching and adding neighborhood-level
variables. Each site is currently getting their data in the appropriate
format as detailed in the protocol.
The necessary
geographic information systems and mapping software,
ArcGIS version 9.1 (ESRI), has been
installed on computers at the lead site. In preparation for the
geocoding process, the lead site is
currently collecting US Census TIGER (Topologically Integrated
Geographic Encoding and Referencing system) street files for each county
in all of the studies.
Challenges to Date:
Minor
challenges associated with data formatting procedures (data encryption,
multiple addresses and multiple assessments) have been resolved.
Next Steps:
Addresses
will be geocoded, home addresses will be
paired with a unique geocode and the census
tract, using ArcMap by the lead site. In the
process of geocoding, troubleshooting for
unmatched addresses will be shared between the lead site and each site.
Neighborhood-level variables (at the census tract level or zip code
level depending upon availability) from a variety of public datasets
including the U.S. Census Bureau will be collected for each census tract
in the counties in the study in preparation for being matched to the
individual-level anonymous data from each site. The lead site will
explore the operationalization of additional
neighborhood-level variables beyond census measures for inclusion in the
model. Discussion about these neighborhood variables will be made as we
proceed throughout the process.