Public Health Classics: Collecting children’s baby teeth to identify lead poisoning

The pediatrician suspected that something wasn’t quite right with the youngster.  He’d met the teen as part of his North Philadelphia community health center’s psychiatry outreach program.  “He was a very nice kid…[but] he had trouble with words, with propositions and ideas,” the pediatrician remembered.  It made him wonder, “how many of these kids who are coming to the clinic are in fact missed cases of lead poisoning?”

That’s the story recalled by Herbert Needleman, MD and shared in 2005 with historians David Rosner and Gerald Markowitz about the pediatrician’s initial inquiries into the effects of childhood lead poisoning.  For years, furniture, walls, and toys were coated in lead-based paint, and the soil and air contained the heavy metal from leaded-gasoline emissions.  Did these sources of lead provide exposures so grave that the cognitive functions of children were impaired?

Because Needleman had training in psychiatry and psychology, he was familiar with the standard tests for measuring neuropsychologic performance and IQ.  The trick would be identifying a measure for the children’s past exposure to lead.  A blood-lead test, for example, would reveal recent exposure to the toxin, but not shed light on the child’s past exposure, such as during critical developmental periods.

Lead is absorbed by bone and resides there.  Needleman couldn’t subject the children to bone biopsies.  They are painful, invasive and expensive.  But a tooth—a simple deciduous tooth that children shed beginning about age six—would be a painless way to get the exposure data.

In a preliminary paper published in a January 1972 issue of Nature, Needleman reported the lead levels identified in 109 deciduous teeth collected from dental clinics in Philadelphia, Pennsylvania.  Sixty-nine of the teeth were collected from inner city clinics where the environment was known to contain numerous sources of lead.  Forty teeth came from suburban clinics.  The figure below (from the original paper) shows the distribution of lead levels (ppm) in the teeth, stratified by the suburban (control) and inner city (experimental) samples.  The mean tooth lead level for the control teeth was 11.1 ppm (+/- 14.8).  The mean lead level for the inner city teeth was nearly five times higher, at 51.1 ppm (+/- 109).

Figure 1, Needleman et al. "Lead levels in deciduous teeth of urban and suburban American children," Nature, Janaury 14, 1972.

The last sentence of the paper foretold what would come next:

“Studies are now underway to examine neurological and psychological function in subjects with elevated tooth lead levels.”

Between 1975 and 1978 Needleman and his colleagues recruited first and second graders (and their parents) who lived in Chelsea and Somerville, Massachusetts.   The children’s teachers assisted in several ways: they verified and collected the students’ baby teeth, and rated the children’s classroom behavior using an 11-item scale.  The questions included “is this child easily distracted during her work?” “Can he follow a sequence of directions?”

The parents completed a medical and social history, a questionnaire assessing parental attitude, and a vocabulary IQ test.  These variables were used to help to assess other factors that might influence the children’s development. Each child participated in a battery of 12 neuropsychological tests, from reading comprehension and reaction time, to the Wechsler Intelligence Scale and the Wepman Auditory Discrimination test.

More than 2,100 children submitted at least one tooth.  Depending on the dentine lead level, the children were divided into six groups.  There were 58 children in the highest dentine lead level group (27 ppm or greater) and 100 children in the lowest level (less than 5.1 ppm).   These two groups were used in the analysis which appeared in the March 29, 1979 issue of the New England Journal of Medicine.

The children in the two groups were similar on most non-lead variables, such as race, weight at birth, and parent’s IQ.  Table 1 presents some of the data from the original paper, including the key findings on neuropsychological performance between the two groups of children.

Needleman and his colleagues reported that the children with the high lead levels performed less well on the Weschler Intelligence Scale, particularly on measures of verbal ability and attention.  The teachers’ ratings of the children’s classroom behavior was also markedly different between with two groups.  On nine of the 11 items evaluated by the teachers, the children with the high dentine lead levels were rated significantly poorer than the children with low dentine lead levels.  The authors concluded:

“The impaired function of children with high lead levels, demonstrated in the neuropsychologic laboratory, mirrored by disordered classroom behavior, appears to be an early adverse effect of exposure to lead.  Permissible exposure levels of lead for children deserve re-examination in the light of these data.”

Dr. Needleman would go on to write dozens of papers on the adverse effects of lead exposure on children’s health.  His more recent research examined the association between bone lead levels in adolescents and delinquency.  His research, however, greatly irritated (and threatened) the lead industry.  They launched a full-scale legal and public relations assaults against him, including trumped-up allegation of scientific misconduct.   The leadership at his academic institution, the University of Pittsburgh, did not come to his defense.  Needleman wrote about those trying years in “Salem comes to the National Institutes of Health: notes from inside the crucible of scientific integrity,” (Pediatrics. 1992;90:977-981.)

Despite the personal, professional and financial trials the attacks caused, Dr. Needleman still urges young researchers to explore controversial topics even those that might provoke powerful interests.   That makes him and his papers on childhood lead poisoning, public health classics.

[Update 11/26/12:  TPH reader David Kriebel, ScD, Chair of the Department of Work Environment at University of Massachusetts Lowell reminded me that recently deceased scientist Barry Commoner also collected baby teeth to assess public health consequences.  Commoner and colleagues in St. Louis used deciduous teeth to measure Strontium-90 and assess the impact of radioactive fallout from above-ground nuclear weapons testing.  The Pauling Blog’s “Baby Tooth Survey” describes the project.  Kriebel reminded me of this terrific example of “science for the people.”



  1. #1 Nancy Evans
    San Francisco
    October 26, 2012

    Baby teeth were a major factor in the Nuclear Test Ban Treaty–strontium 90 levels were measured. Here’s a link to the story

  2. [...] some of the classic studies and reports that have shaped the field of public health.” From lead poisoning to the Surgeon General’s 1964 report on the implications of cigarette smoking, the series makes [...]