Rothman 1999.

Methods	Study design: controlled before‐after study (independent samples) Sampling frame: children age 6 years of age and younger living in predominantly African American census tracts in Philadelphia Sampling method: not clearly stated, but appears to be associated with where children were brought for blood lead level testing (self selection). Control census tracts matched by demographics and housing older than 1950 Collection method: elevated blood lead levels reported to the City of Phildelphia Description of the community coalition: The coalition was formed to address lead awareness in North Central Philadelphia; it originated from a pre‐existing community‐academic partnership. The group was convened by the City Lead Poisoning Prevention Program and the Temple Health Connection, a community‐based academic nursing practice, and included the Department of Health, a university, the local housing authority, community residents, the Salvation Army, and other grassroots community‐based organizations
Participants	Communities: Philadelphia neighborhoods with a high proportion of residents who were African American and below the poverty level, with a high proportion of housing built before 1950 Country: USA Ages included in assessment: ≤ 6 years old (n = 890) Reasons provided for selection of intervention community: Targeted neighborhood was economically disadvantaged, underserved, and at risk because of lead in the environment Intervention community (population size): 4 census tracts of North Central Philadelphia (population size not reported) Comparison community (population size): 4 additional census tracts within Philadelphia that matched as closely as possible the experimental population on % housing built before 1950, % African American, and % below poverty level (population size not reported)
Interventions	Name of intervention: Lead Awareness: North Philly Style Theory: not reported Aim: to implement and evaluate community‐developed, community‐based strategies that address childhood lead poisoning Description of costs and resources: after‐school and camp programs with a focus on lead education provided for children through participating agencies. ‘Block captains’ who were residents of the targeted community invited neighbors to block parties and provided educational materials, lead remediation materials, and gifts. Cost data not reported Components of the intervention: educational after‐school programs and camps for children; informational “block parties” in local churches and recreation centers for adults, with educational materials and free lead‐remediation materials Start date: 1997 Duration: 3 years
Outcomes	Outcomes and measures: change in elevated blood lead levels (> 10 micrograms/Dl) in children (proportion of blood lead levels over certain thresholds) (f/u intervention: n = 360; comparison: n = 530) Time points: baseline (1997) and follow‐up (3 years later)
Notes	Funder: US National Institute for Nursing Research 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	Unclear risk	Outcome (proportion above lead level cutoff) shifted during study and baseline levels not presented with final results
Baseline characteristics similar	High risk	Control census tracts matched on housing built before 1950, % African American, and poverty, with some differences evident
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Mandatory reporting of elevated blood lead levels to City of Phildelphia. Objectively measured outcome should not be affected by lack of blinding
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Completeness of reporting elevated blood lead levels to City of Phildelphia not described
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Objectively measured outcome should not be affected by lack of blinding
Protection against contamination	Unclear risk	No statement regarding protection against contamination. Proximity of experimental and control census tracts unclear
Selective reporting (reporting bias)	High risk	Exact results for blood lead level ≥ 20 described in initial paper as main outcome variable reported only in brief; no statistical testing provided