Kruger 2007.
| Methods |
Study design: controlled before‐after cross‐sectional study (multiple time points) Sampling frame: Michigan state vital statistics Sampling method: All recorded infant mortality outcomes Collection method: state records Description of the community coalition: The Genesee County REACH 2010 partnership was established in the early 1990s with private foundation funding and continued without outside financial assistance for 3 years until Prevention Research Center and REACH 2010 grants were awarded in 1998. Partnership included the local medical center, health department, and university, and multiple community‐based organizations |
|
| Participants |
Communities: 5 Michigan counties Country: USA Ages included in assessment: infants Reasons provided for selection of intervention community: Partnership was already established in the community Intervention community (population size): Genesee County (436,000) Comparison community (population size): 4 comparison groups: Saginaw, Berrien, Oakland, Wayne Counties (population sizes not reported) |
|
| Interventions |
Name of intervention: Genesee County REACH Initiative Theory: not reported Aim: to effect community and systems change sufficient to improve local infant mortality problem, particularly the disparity in rates between African Americans and European Americans Description of costs and resources: funded at $988,968 for first year of implementation and $912,062 for subsequent years Components of the intervention: included community dialogue sessions, “Undoing Racism” workshops, university course in cultural competency, maternal/infant health advocates, and a community media campaign Start date: 2000 Duration: ongoing |
|
| Outcomes |
Outcomes and measures: infant mortality rate as measured by vital statistics data Time points: pre‐intervention (1999) and follow‐up (2005) |
|
| Notes | Funding source: government | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | High risk | Intervention not randomly assigned |
| Allocation concealment (selection bias) | High risk | No allocation concealment |
| Baseline outcome measurement similar | High risk | Intervention county had poorer outcomes relative to comparison counties at baseline |
| Baseline characteristics similar | High risk | Intervention county selected for disadvantaged status relative to comparison county at baseline |
| Blinding of outcome assessment (detection bias) All outcomes | Low risk | State vital statistics records |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | State vital statistics data |
| Blinding of participants and personnel (performance bias) All outcomes | High risk | No blinding |
| Protection against contamination | Unclear risk | Did not discuss proximity of counties |
| Selective reporting (reporting bias) | Low risk | Main outcomes reported |