10-intro.qmd

## Introduction

### Background

Sudden Unexpected Infant Death (SUID) exacts a severe, lasting toll.
The National Institute of Child Health and Human Development defines SUID as "the death of an infant younger than 1 year of age that occurs suddenly and unexpectedly" [@CommonSIDSSUID].
At the familial level, loss of an infant may put bereaved caregivers--especially birth mothers--at heightened risk for Prolonged Grief Disorder where the process of healing and acceptance evades realization [@goldsteinGriefMothersSudden2018].
At the societal level, Sudden Infant Death Syndrome (SIDS), a subset of SUID, contributes to 20% of all post-neonatal infant mortality in the United States [@GBDCompare].

Previous research has demonstrated that SUID varies unequally in incidence across communities within major metropolitan regions [@brikerModifiableSleeprelatedRisk2019; @drakeDescriptiveGeospatialAnalysis2019; @caitline.feeGeographicalAnalysisSudden2017].
Within Cook County, Guest et al. (1998) demonstrated that this geographic variation was significantly associated with the community-level factors of racial segregation and unemployment [@guestEcologyRaceSocioeconomic1998].

To counteract these negative socioeconomic influences and reduce community incidence of SUID in a just manner, public health practitioners must precisely target their interventions in space and time.
Without spatiotemporal precision, practitioners risk exacerbating inequity by bolstering communities already well-resourced, while neglecting the communities in highest need [@hortonOfflineDefencePrecision2018]. 

### Objectives

In this study of Cook County, IL, we sought to precisely describe where SUID occurred in the recent past (2015-2019) and to project where these deaths would occur in the near future (2021-2025).
Geospatially, previous analyses have not provided sufficient detail to name specific community areas of highest priority [@brikerModifiableSleeprelatedRisk2019; @drakeDescriptiveGeospatialAnalysis2019; @caitline.feeGeographicalAnalysisSudden2017; @grimsonSearchingHierarchicalClusters1981].
We afforded greater detail by aggregating information at the census tract level into "communities" as our primary unit of analysis.
We wanted public health practitioners to be able to take full advantage of this greater detail by generating interactive maps that could pan and zoom to areas of high risk.

<!-- Census tracts were also useful for predictive modeling because publicly-available datasets like the American Community Survey from the U.S. Census provide a rich catalog of prospective socioeconomic covariates at a yearly cadence of time intervals [@AmericanCommunitySurvey]. -->

While these analytic efforts were specific to Cook County, we designed an open source coding workflow that could be adapted to generate maps and predictive models for other jurisdictions.
We hope these tools will enable communities most afflicted by SUID to get assistance in a timely, targeted manner.