Document wr8g3q4ZKg8NN3R0pMMJwGKOJ

Marshall reports on criticisms directed against die pediatric lead studies of Needleman ef al. by consultants to EPA. Needleman et al. (J) evaluated the neuropsychologic performance of 58 children with high dentine lead concen trations and of 100 children with low concentrations. The children with the high lead concentrations achieved a mean IQ score on the Wechsler Intelli gence Scale for Children which was ap proximately four points below that of the children with low concentrations. Analy sis of variance indicated that this differ ence could not be explained by any of 39 covariates, such as age, sex, race, socio economic status, past medical history, or child-rearing practices. Needleman at al. (2) demonstrated, subsequently that the downward shift in IQ was associated with a significant reduction in the num ber of children in their high lead group with IQ's above 125 and with a trebling in die number having IQ's below 80, These findings have been generally cor roborated in studies by Emhart et cil. (5), Lansdown et al. (4), Winneke et al. (5), and Smith.et al, (6). A recent review by Rutter (7) concluded that those studies of the association between lead and IQ in which close heed has been paid to methodologic issues have consistently found a dose-response association between lead exposure and neuropsychologic def icit. This effect appears to have been indisputably established at blood lead concentrations above 40 micrograms per deciliter, and there is growing evidence to suggest that, even at lower concentra tions, lead can cause a subtle, but irre versible, reduction in children's intelli gence. Atmospheric lead emissions have in creased 2000-fold since the pre-Roman era, and even then the atmosphere may have been contaminated With lead by a factor of 3 over natural background (8). In 1978, the second National Health and Nutrition Examination Survey (9) esti mated that 660,000 children age 6 months through 5 years in the United States had blood lead concentrations above 30 micrograms per deciliter, the concentration considered by the Centers for: Disease Control (.10) to represent excessive lead absorption. Taken together with the data of Needleman et al. (1, 2), these findings have profoundly disturbing implications, not only for the future development of the affected children, hut for the whole fabric of modern American society. For tunately, the number of children in the United States with acute high-dose lead poisoning has declined precipitously dur ing the past decade as the result of implementation of the Lead-Paint Poi- soiling Prevention Act. Also the back ground rate of pediatric lead absorption has been reduced substantially, most likely as a result of the EPA-mandated phasedown in the lead content of gaso line (11). This latter accomplishment is, in no small measure, a consequence of Needleman's work. Ph il ip J. La n d r ig a n Division ofSurveillance, Hazard Evaluations and FieldStudies, National Institute for Occupational Safety and Health, 4676 Columbia Parkway, Cincinnati, Ohio 45226 Ve r n o n N. Ho u r Centers for Disease Control, Atlanta, Georgia 30333 References 1. H. L. Needleman ef al.. New Engl. J. Med. 300, 689(1979). 2. H. L. Needleman, A. Leviton, D. Bellinger, ibid. 306, 367 (1982). 3. C. B. Ernhart, B. Landa, N. B. Schell, Pediat rics 67, 911 (1981). 4. R. Lansdown et al., in Lead Versus Health, M. Rutterand R. R. Jones, Eds. (Wiley, New York, 1983), pp. 267-296. 6. G. Winneke et al., m ibid., m. 249-266. 6. M. Smith et al., Dev. Med. Child Neurol. 47 (Spec. Suppl.) (1983). 7. M. Rutter, in Lead Versus Health, M. Ratter and R. R. Jones, Eds. (Wiley, New York, .1983), pp. 333-370. 8. M. Murozomi, T. J. Chow, C. Patterson, Geochirn. Cosmochtm. Acta 33, 1247 (1969). 9. K. R. Mahaffey et at.. New Engl. J. Med. 307, 573 (1982). 10. Centers for Disease Control, J. Pedlatr. 93, 709 (1978). 11. J. L. Annest et at.. New Engl. J. Med.WE, 1373 (1983). With Marshall's report, we finally have an independent published report, in an important journal, of criticism of the "classic" Needleman study of lead ef fects. Some of the problems identified by the Expert Committee on Pediatric Neurobehavioral Evaluations appointed by EPA have been mentioned in my public statements. Other problems, such as biases associated with missing data, were barely discernible in the published reports and are now substantiated through the committee's limited access to the data. Since my role as adversary is men tioned, I would like to clarify a few points. Needlemen attributes criticisms of his work to industry affiliation; how ever, at the time of the first publication (/) in which I refer to problems in his work, I had no Connection with the lead industry. (I now have a modest research grant from the International Lead Zinc Research Organization.) Marshall states that I have maintained "that there is no proof of a causal rela tionship between high levels of lead in children's blood and low scores on IQ tests, , . .''This is misleading. There is a definite and tragic risk of intellectual deficit associated with high levels of exposure, and I have never stated other wise. My assertion is that there is no convincing evidence of an association of low level lead exposure and measurable deficits in psychological development when other relevant factors (such as parent intelligence, home environment, and so forth) are controlled. Unfortu nately, efforts to blame the problems of many of our disadvantaged children on lead exposure tend to detract our atten tion from the amelioration of the difficult conditions in which these children are reared. My "docility" about the committee's findings results from a respect for the competence and courtesy of its members and a recognition of the validity of the major features of its review. My asso ciates and I cooperated in supplying EPA with complete sets of our data. We also conducted reanalyses as requested. (An additional reanalysis has since been requested.) The corrected results did not support the conclusion that low level lead exposure is related to deficits in intelligence. This applies to the earlier work mentioned by Marshall, as well as the data from the 1981 report. The sam ple sizes, however, do not provide the statistical power necessary for a reason ably definitive statement that there is little or no effect. Recent, well-controlled studies (2, 3) in the United Kingdom also faffed to support the Needleman position. One study (3), directed by Maijorie Smith, is particularly important because it repli cated Needleman's procedures with a large sample and sound methodology. After appropriate control of confounding variables, no significant association of dentine lead and intelligence was found. Cl a ir e B. Er n h a r t Departments of Psychiatry and Reproductive Biology, Case Western University, Cleveland Metropolitan General Hospital-Highland View Hospital, 3395 Scranton Road, Cleveland, Ohio 44109 References 1. C. B. Enihart, B. Landa, N. B. Schell, Pediat rics SJ, 911 (1981), 2. P. Harvey, M. Hamlin, R. Kumar, paper pre sented at the Annual Conference of the British Psychological Society, Vork, United Kingdom, April 1983; W. Yule and R. Lansdown, ibid. 3. M. Smith, T. Delves, R. Lansdown, B. Clayton, P. Graham, The Effects of Lead Exposure on Urban Children: The Institute of Child Health/ Southhampton Study (Department of the Envi ronment, London, 1983). Erratum: In the Research News article "New test of variable gravitational constant" (23 Dec., p. 1316), in the third paragraph of column two on rage 1317, a minus sign was left out of an exponent. The correct value for the drift ofatomic clocks relativeto gravitations! clocks is (0.1 0.8) x Hr" per year. l is SCIENCE, VOL. 223 TEH 0532675