Whyatt, RM, V Rauh, DB Barr, DE Camann, HF Andrews, R Garfinkel, LA Hoepner, D Diaz, J Dietrich, A Reyes, D Tang, PL Kinney and FP Perera. 2004. Prenatal insecticide exposures, birth weight and length among an urban minority cohort. Environmental Health Perspectives, on line 22 March 2004.

Whyatt et al. report that pregnant women exposed to higher levels of two insecticides had smaller babies than other mothers, but that following an EPA phase-out of the insecticides not only were exposures substantially lower but the association between newborn size and exposure disappeared. Newborn size is an important predictor of health later in life.

Their results are important both for confirming the impact of chlorpyrifos and diazinon exposure on fetal growth and development, and for demonstrating that public health steps taken by the EPA to reduce exposures can quickly yield positive results for health.

What did they do? Whyatt et al. assessed exposure of pregnant minority women living in upper Manhattan to two insecticides, diazinon and chlorpyrifos in two different ways: by taking a sample of umbilical cord blood and by having subjects wear a small "personal ambient air monitor" to collect air samples of what they were breathing.

Mothers also responded to a questionnaire about home pesticide use, demographics, personal histories, maternal education and income level, etc.

The team of researchers then carried out multivariate statistical analyses of the relationship between these exposure measurements and various indicators of child health, including birth weight, length, head circumference.

Women who reported smoking during pregnancy or whose blood cotinine levels indicating active smoking were excluded from the study. A variety of covariates were used in the statistical analysis, including race/ethnicity, gestational age, maternal pre-pregnancy weight and pregnancy weight gain and season. Other potential covariates were considered but discarded because they were neither associated with birth outcome statistically nor improved the fit of the multivariate analysis.

The study began in 1997 and continued during and after EPA's phaseouts of chlorpyrifos (December 2001) and diazinon (December 2002). Whyatt et al. compared exposure levels and health effects before and after the bans to see if exposure levels were reduced and whether that had any effect on infant health.

What did they find? The study cohort included 314 babies.

Pesticide levels in air and blood samples were substantially lower in infants born after 1 January 2001, except for diazinon in blood.

Examined separately, higher umbilical cord blood levels of the pesticides were associated with small birth weight and length.

• "Birth weight decreased by 42.6 grams (95% CI -81.8 to -3.8, p=0.03) and birth length decreased by 0.24 cm (95% CI -0.47 to -0.01, p = 0.04) for each log unit increase in cord plasma chlorpyrifos levels."
• The pattern for diazinon was similar in size but there was more variability in the relationship and it was not statistically significant.

When Whyatt et al. used an EPA technique for examining the combined effect of chlorpyrifos and diazinon together, the significance of the relationship between exposure and impact grew stronger.

The largest impacts on birth weight and length were in the most heavily exposed infants. On average they weighed 186 grams less than the least exposed infants (p = 0.01) and their birth length averaged 0.8 cm less (p =0.056).

Separate analysis of the babies born before vs. after the pesticide bans showed the effect of phasing out the chemicals.

In babies born before the phase-outs, birth weight and birth height were strongly associated with chlorpyrifos exposure and the combined exposure assessment (chlorpyrifos plus diazinon). After 1 January 2001, which marked the end of the chlorpyrifos phase-out and was part way through the diazinon phase-out, no relationships were discernable. This was the result of the fact that higher exposures were longer taking place, and as noted above, the largest impacts were seen in the most heavily exposed infants.

Analysis of data from the questionnaire found no significant associations between pesticide use reported by the mother and any birth outcome. This negative finding provides a cautionary note about the limited value of questionnaires and the importance of effective biomarkers such as cord blood measurements.

None of the associates between birth outcome and pesticide measurements from the air monitors were significant. Whyatt et al. observe that dietary exposures can be important sources of diazinon and chlorpyrifos that would not be detected by air sampling but would by the cord blood measurements. Crop use of the two pesticides continue and residues are found regularly in produce.

What does it mean? EPA made the right call in deciding to phase diazinon and chlorpyrifos out of residential use. The results reported by Whyatt et al. reveal a striking and immediate decrease in exposure and improvement in the health of newborns in this minority community in Upper Manhattan, following the EPA decision.

.