POISONED DEVELOPMENT: ASSESSING CHILDHOOD LEAD EXPOSURE AS A CAUSE OF CRIME IN A BIRTH COHORT FOLLOWED THROUGH ADOLESCENCE

Date01 May 2018
DOIhttp://doi.org/10.1111/1745-9125.12171
Published date01 May 2018
POISONED DEVELOPMENT: ASSESSING CHILDHOOD
LEAD EXPOSURE AS A CAUSE OF CRIME IN A BIRTH
COHORT FOLLOWED THROUGH ADOLESCENCE
ROBERT J. SAMPSON and ALIX S. WINTER
Department of Sociology, Harvard University
KEYWORDS: lead poisoning, crime, antisocial behavior, impulsivity, toxic inequality
The consequences of lead exposure for later crime are theoretically compelling,
but direct evidence from representative, longitudinal samples is sparse. By capitaliz-
ing on an original follow-up of more than 200 infants from the birth cohort of the
Project on Human Development in Chicago Neighborhoods matched to their blood
lead levels from around age 3 years, we provide several tests. Through the use of four
waves of longitudinal data that include measures of individual development, family
background, and structural inequalities in how lead becomes embodied, we assess the
hypothesized link between early childhood lead poisoning and both parent-reported
delinquent behavior and official arrest in late adolescence. We also test for mediating
developmental processes of impulsivity and anxiety or depression. The results from
multiple analytic strategies that make different assumptions reveal a plausibly causal
effect of childhood lead exposure on adolescent delinquent behavior but no direct link
to arrests. The results underscore lead exposure as a trigger for poisoned development
in the early life course and call for greater integration of the environment into theories
of individual differences in criminal behavior.
Exposure to dangerous levels of lead was extensive for long stretches of the twenti-
eth century (Feigenbaum and Muller, 2016; Troesken, 2006). Although environmental
reforms such as the bans on lead in gasoline and paint in the 1970s were deemed victories
for public health at the time (Markowitz and Rosner, 2013; Needleman, 2004), lead toxi-
city is far from a hazard of the past. High levels of lead have recently been found in thou-
sands of cities in the United States (Pell and Schneyer, 2016) and in both developed and
developing countries around the world (Tong, von Schirnding, and Prapamontol, 2000).
Additional supporting information can be found in the listing for this article in the Wiley Online
Library at http://onlinelibrary.wiley.com/doi/10.1111/crim.2018.56.issue-2/issuetoc.
Support for this article was provided in part by the Project on Race, Class and Cumulative Adver-
sity at Harvard University funded by the Ford Foundation and the Hutchins Family Foundation, as
well as by the Hymen Milgrom Supporting Organization. We thank Chris Muller and the reviewers
and Editor of Criminology for an unusually detailed and helpful set of comments.
Direct correspondence to Robert J. Sampson, Harvard University, 33 Kirkland Street, Cambridge,
MA 02138 (e-mail: rsampson@wjh.harvard.edu).
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-
NoDerivs License, which permits use and distribution in any medium, provided the original work is prop-
erly cited, the use is non-commercial and no modifications or adaptations are made.
C2018 The Authors. Criminology published by Wiley Periodicals, Inc. on behalf of American Society of
Criminology. doi: 10.1111/1745-9125.12171
CRIMINOLOGY Volume 56 Number 2 269–301 2018 269
270 SAMPSON & WINTER
The poisoning of the water in Flint, Michigan, in late 2015 and the evacuation in 2016 of
an entire neighborhood contaminated by a smelting plant in East Chicago, Indiana, shone
bright public lights on the contemporary perils of lead (Goodnough, 2016a, 2016b).
The legacy of lead, both historic and contemporary, carries important theoretical impli-
cations for the study of crime over the life course. An association between lead exposure
and cognitive impairment has been documented in medical research (Canfield et al., 2003;
Lanphear et al., 2000; Reuben et al., 2017), and many criminologists have long argued
that there is a connection between cognitive ability and delinquency (Farrington, 1998).
Similarly, there is evidence that lead aggravates hyperactivity, impulsive behavior, and
mental health problems (Braun et al., 2006; Winter and Sampson, 2017), which have been
shown to predict delinquency (Elliott, Huizinga, and Menard, 1989; Moffitt, 1990). More
directly, in a small body of research, an association has been found between lead exposure
and later measures of delinquency and crime (e.g., Reyes, 2015; Wright et al., 2008).
Also important theoretically is that exposure to lead is unequally distributed, likely
more so today than in the past when lead was “in the air” and thus almost everywhere.
Contemporary lead exposure is linked, for example, to minority status and poverty at
the individual level, as well as to racial segregation and concentrated poverty at the
neighborhood level primarily because of the unequal distribution of dilapidated hous-
ing that contains remnants of lead paint (Lanphear et al., 1998; Oyana and Margai, 2010;
Sampson and Winter, 2016). Similarly, the highest concentrations of lead in Flint were
located in its most disadvantaged neighborhoods (Hanna-Attisha et al., 2016), and the
evacuated neighborhood in East Chicago was both low income and predominantly Black
(Goodnough, 2016b). Lead may thus be a mechanism that connects inequality to crime
(Muller, Sampson, and Winter, 2018).
Despite the scientific consensus that lead exposure inflicts serious damage (National
Research Council, 1993) and the fact that lead is still a contemporary threat to society,
especially among the poor and in racially segregated areas, our knowledge of childhood
lead exposure and the developmental course of crime is surprisingly sparse. As simply
stated in a recent review of the literature, “there is a dearth of criminological research on
this topic” (Narag, Pizarro, and Gibbs, 2009: 954). In particular, the results of our review
reveal that longitudinal data from representative samples with a long-term follow-up of
individuals are rare, as are studies in which lead exposure is integrated with mediating de-
velopmental processes or in which measures of alternative explanations, such as poverty,
are included at both the individual and ecological levels. Conceptual integration of the
age-graded mechanisms of lead’s damage with criminological theory is also lacking.
We address these limitations in this article and argue that exposure to lead in infancy
can poison development and exert negative long-term consequences. We also integrate
research on the contextual pathways of childhood lead exposure with life-course crimino-
logical theory, motivating a core set of hypotheses on the consequences of lead exposure.
The result of our theoretical integration supports an analytic strategy that allows us to
assess the long-term link between childhood lead exposure and later delinquent behavior
among adolescents and to estimate whether the link is plausibly causal in nature. Further-
more, we assess several explanatory mechanisms that are hypothesized to connect early
exposure to later behavior.
The data supporting our analytic strategy come from an original follow-up of the birth
cohort from the Project on Human Development in Chicago Neighborhoods (PHDCN).
We match four waves of longitudinal data from the PHDCN that include measures of
LEAD POISONING AND CRIME 271
individual development, family background, and neighborhood context with childhood
blood lead level (BLL) tests and the locations of smelting plants in Chicago, Illinois. By
capitalizing on this comprehensive linked data set, we estimate the effect of childhood
lead poisoning on adolescent delinquency—both parent-reported and official criminal his-
tories. Prior research is assessed in the next section, and then we build on it to formulate
a theoretical framework for testing the lead–crime hypothesis.
STATE OF KNOWLEDGE
The criminological significance of lead is at once provocative and understudied. In a few
longitudinal cohort studies, scholars have assessed the relationship between lead exposure
and delinquent behavior at the individual level, with uncertain results. For example, in
a lead smelting community in Australia, lifetime BLL above 15 µg/dL was associated
with increased aggressive behavior at ages 11 to 13 years relative to children with lower
levels of exposure (Burns et al., 1999). By contrast, Beckley et al.’s (2017) study based
on a 1972 birth cohort in Dunedin, New Zealand, revealed that lead exposure at age
11 (average BLL of 11 µg/dL) had an imprecise or weak relationship with self-reported
offending in adolescence (but see Farrington, 2017). It is uncertain how either of these
findings translates to the contemporary United States, where the population average BLL
is around 2 µg/dL (Jones et al., 2009).
Within the United States, Dietrich et al. (2001) found a positive relationship between
pre- and postnatal lead exposure and both self- and parent-reported delinquent behavior
at ages 15 to 17 in a socioeconomically disadvantaged sample in Cincinnati, Ohio, with
high levels of lead exposure. In a nationally representative sample, childhood lead expo-
sure was associated with antisocial behavior from ages 4 to 12; childhood lead exposure
also predicted oppositional behavior and having hurt someone badly by age 17 (Reyes,
2015). Although carefully executed, Reyes (2015) estimated individuals’ lead exposure,
and measured ecological confounding, such as poverty, at the state level rather than at
the more proximate levels, such as neighborhood, associated with exposure.
The results of several cross-sectional studies are nonetheless broadly consistent with
these longitudinal findings. Associations have been reported between lead exposure and
aggressive or antisocial behavior at ages 6 to 9 in Edinburgh, Scotland (Thomson et al.,
1989), as well as at ages 7 to 11 in Pittsburgh, Pennsylvania (Needleman et al., 1996);
conduct disorder among American children ages 8 to 15 (Braun et al., 2008); and rule-
breaking behavior at ages 14 to 18 in an impoverished area of Brazil (Olympio et al.,
2010). In a meta-analysis of 19 studies, both longitudinal and cross-sectional, Marcus,
Fulton, and Clarke (2010) reported an association between lead exposure and conduct
disorder (including aggressive and externalizing behavior).
Whereas the research described thus far has been focused on delinquent behavior re-
ported by subjects, parents, and/or teachers, a few scholars have considered the link be-
tween lead exposure and involvement with the criminal justice system. Wright and col-
leagues (2008) followed the aforementioned disadvantaged cohort in Cincinnati to ages
19 to 24 and found that pre- and postnatal lead exposure were associated with total ar-
rests and arrests for violent crime, whereas Denno (1990) found a positive relationship
between lead exposure and arrests in a disadvantaged minority sample in Philadelphia,
Pennsylvania. Fergusson, Boden, and Horwood (2008) studied a New Zealand cohort,
showing associations between lead exposure and both violent and property offenses,

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT