Appendix 4.  Detail on Residential Lead Hazard Control on Blood Lead Levels

Although newer residential hazard control methods can effectively reduce exposure to lead paint and lead-contaminated dust,¹ compared to older strategies that often increased lead exposure during the intervention, these newer techniques can still result in an elevation of blood lead in a subset of children immediately following lead control interventions (Tables 3 and 4). In an evaluation of HUD-sponsored lead control interventions among fourteen state and local governments, 81 of 869 children (9.3%) had an elevation of > 5 µg/dL. Risk factors associated with post-intervention increases were the number of exterior paint deteriorations, the educational level of the female parent or caregiver, and younger age of the child.⁶⁹

Before 1996, retrospective cohort studies, case series, and uncontrolled experiments suggested a modest decline (4-10 µg/dL) in mean blood lead levels in children with initial blood lead levels > 25 µg/dL. More recent studies of newer lead-based paint hazard control techniques that included an untreated comparison group, however, found more modest beneficial effects^70,71 or no effects.^72,73

A meta-analysis of four randomized controlled trials conducted in 1996-2000 found that interventions had no effect on mean blood levels (-0.62 µg/dL, 95% CI -1.55 to 0.32), but there were significant reductions in the proportion of children who had blood lead concentrations exceeding 15 µg/dL (6% vs. 14%, p=0.008) and 20 µg/dL (2% vs. 6%, p=0.024) in the intervention group compared with controls.⁷⁴

Two of these four trials evaluated dust control and two evaluated providing education and equipment to families. The earlier of the two trials of dust control (1998) evaluated one-time professional dust control and window-sill-paint sealing in homes of children aged four or younger, with mean blood lead of 16.9 µg/dL.⁷²  There were similar reductions in blood levels in the intervention and control groups (-6.2 vs. -5.9 µg/dL) six months after abatement.  In the second randomized trial (1999), conducted in Jersey City, N.J., investigators recruited children aged 6 to 36 months who had lead paint in the home.  Families (n=113) were randomized to a lead exposure reduction group or to an accident prevention control group.  In the lead exposure reduction group, staff members visited the home every two weeks and spent about two hours cleaning up dust.  After one year, there was a small but statistically significant difference in blood lead change between intervention and control groups, adjusted for baseline lead levels (-2.1 vs. +0.1 µg/dL, p<0.05).⁷⁰

A followup study in urban children participating in the TLC trial examined the effects of a second professional lead dust cleaning of homes 18 months after an initial cleaning and therapy commencement.⁷⁵  All homes in the Philadelphia site (n=165) of the TLC trial were offered a second professional cleaning. Participation in the followup intervention was voluntary rather than randomized.  The mean BLL at study initiation was 26 µg/dL.  The mean BLL was 15.7 µg/dL at the second cleaning visit, but six months later there was no difference in blood lead levels between children whose homes were cleaned (n=73) and those whose homes were not cleaned (n=86).  The report did not stratify results by the original treatment assignment of the subjects (chelation vs. placebo), so the effects of the combined interventions cannot be compared with an untreated group.

A 2003 retrospective cohort study identified children listed in the New York City child blood lead registry and compared blood levels before and 10-14 months after remediation with those of a control group that did not have remediation.⁷³  Mean blood levels declined significantly from 24.3 µg/dL to 12.3 µg/dL at followup, regardless of remediation.  After adjusting for confounders, the remediation effect was 11% (p=ns). Race was identified as the only confounding factor, and white and Asian children had an adjusted mean followup blood lead level 30% lower than African American children (p<0.01).  The effect of remediation appeared to be stronger in younger children (10 - <36 months) than in older children (36-72 months.)  Another retrospective cohort study that evaluated in-home counseling, combined with professional lead paint remediation, compared lead levels in children aged six months to six years with mean blood lead of 28.8 µg/dL with similar children who did not receive the intervention.⁷¹  Followup blood lead was measured on average 69 days after abatement, 172 days after the initial sample.  After adjusting for season and age of the child, the treatment group blood lead decreased 6.0 µg/dL from 28.8 to 22.8, and the effect of treatment was significant (p<0.05). The comparison group mean blood lead decreased 1.6 µg/dL from 31.1 to 29.5 (p=ns). 

In a retrospective study that measured blood lead levels in children whose homes were abated from 1987 to 1990, before and after abatement policies in Massachusetts became more stringent in 1988, the mean blood lead decreased from 26.0 µg/dL at baseline to 21.2 µg/dL (p<0.001) measured between two weeks to six months post abatement.  Reductions were only seen, however, among children whose baseline blood lead levels were greater than 20 µg/dL.  This study found no meaningful change in pre- to post abatement levels by calendar year of intervention.⁷⁶  The effect of different housing policies on the risk of subsequent lead exposure in homes where a child with elevated blood lead resided in the past was demonstrated in adjacent geographic regions of two northeastern states. Approximately eight years later, the risk of identifying at least one child with an elevated blood lead level (>10 µg/dL) was four times greater in the state with less stringent housing-based lead poisoning prevention policies.⁷⁷

A study of 1212 HUD dwellings that received interior treatment for lead hazard control in thirteen states from 1994 to 1998 reported a mean 2.8 µg/dL reduction in children's (n=240) blood lead levels at 12 months post-intervention, from a median level of 10 µg/dL at baseline.⁷⁸  The effect of treatment in these studies was not compared with an untreated population.  Another study of HUD dwellings in four Massachusetts communities found a significantly larger decline in blood lead levels between 1993 and 2002 among children in treated homes than in untreated homes, matching on pre-intervention BLL.  Children's BLLs decreased from 7.07 and 6.62 µg/dL to 3.59 and 4.28 in the treated and untreated homes respectively (p=0.015).  The study adjusted for time and seasonality to account for the downward trend in BLLs observed among children in the general Massachusetts population, from 5.9 µg/dL in 1994 to 3.2 µg/dL in 2002.⁷⁹

These trials highlight the difficulties of lead-paint hazard control as a method to reduce lead exposure.  Poor, inner-city families tend to move frequently, so that treating the current residence may have limited long-term benefit to the individual child, although benefit accrues to subsequent children moving into that residence. In the Jersey City, NJ, study, for example, approximately 30% of the randomized families moved during the 12-month followup period.⁷⁰  Residential lead-paint hazard control can be costly and labor-intensive, limiting the availability of intervention, especially in poor communities.¹  Recontamination by nearby lead sources, including soil lead, may occur after lead-paint hazard control efforts in a dwelling^1,80  These limitations demonstrate the need for effective comprehensive individual interventions, as well as community-based interventions, to reduce household lead exposure. Unfortunately, available data about programs that employ multiple interventions are sparse.^69,81

Return to Document