CATHERINE COOK-COTTONE

Pediatric exposure to polychlorinated biphynels (PCBs) is a national health concern with significant

implications for school psychologists. According to the healthcare collaboration model, the

school psychologist plays a key role in the provision of services to children affected by environmental

teratogens. To effectively function as healthcare collaborators, school psychologists must

have an understanding of the nature of PCBs, the current state of PCB research, and implications

for practice. This article provides a brief and critical update of empirical findings and posited

developmental implications as well as an empirically guided overview of PCB-related school

psychology practice. © 2004 Wiley Periodicals, Inc.

Pediatric exposure to potential neurotoxins such as polychlorinated biphynels (PCBs) is a

continuing national health concern with significant implications for school psychologists. As

researchers work to understand, document, and explicate the effects of PCBs on the developing

child, educators strive to provide educational services responsive to children’s needs. According to

the healthcare collaboration model (e.g., Phelps, 1999), the school psychologist plays a key role in

the provision of services to children affected by environmental teratogens. This role includes: (a)

collaborating with medical personnel, (b) assisting in the development of appropriate support

services, (c) providing in-service training with educational personnel, (d) partnering with ongoing

toxicology research teams, and (e) advocating for prevention and early identification. To effectively

function as healthcare collaborators, school psychologists must have an understanding of

nature of PCBs, the current state of PCB research, and implications for practice. Currently, knowledge

of the effects of PCBs on children is limited and, at times, confusing. A brief and critical

update of empirical findings and posited developmental implications as well as an empirically

guided overview of PCB-related school psychology practice is provided in this article.

Overview

Polychlorinated biphynels (PCBs), or synthetic chlorinated hydrocarbon compounds, were

first synthesized in 1881 resulting in wide commercial production in the United States for several

decades (1929–1977; Cicchetti, Kaufman, & Sparrow, this issue; Friedrich, 2000). According to

the U.S. Environmental Protection Agency (U.S. Environmental Protection Agency, 2003), the

unique physical properties of PCBs (e.g., nonflammability, chemical stability, high boiling point,

and electrical insulating properties) made them valuable for many industrial and commercial

applications (e.g., as insulators in electrical, heat transfer, and hydraulic equipment; as plasticizers

in paints, plastics and other rubber products; and as additives in pigments, dyes, and carbonless

copy paper). Before they were banned in most industrial countries, more than 1.5 billion pounds of

PCBs were manufactured in the United States alone (U.S. Environmental Protection Agency,

2003). Because of their resistance to natural chemical and biological processes, these persistent

and fat-soluble chemicals have remained in the environment long after their initial introduction

and continue to bioaccumalate in the food chain (Friedrich, 2000; Wolff & Landrigan, 2002).

Literature reviews report PCB exposure to be associated with significant cognitive deficits in

mice, rats, and monkeys (e.g., Faroon, Jones, & De Rosa, 2000). However, studies of human

exposure are the source of ongoing debate (e.g., Cicchetti et al., this issue; Daniels et al., 2003;

Correspondence to: Catherine Cook-Cottone, Department of Counseling, School, and Educational Psychology, 409

Baldy Hall, SUNY at Buffalo, Buffalo, NY 14260–1000. E-mail: cpcook@buffalo.edu

Psychology in the Schools, Vol. 41(6), 2004 © 2004 Wiley Periodicals, Inc.

Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/pits.20012

709

Wolff & Landrigan). Scientific reviews of human studies criticize PCB research completed to date

as seriously flawed and failing to provide convincing evidence of significant and/or meaningful

cognitive effects (e.g., Cicchetti et al., this issue).

Research Difficulties

Assessing the potential effects of PCBs is very complex for many reasons. First, PCBs are a

class of compounds that consists of 209 congeners that differ in toxicity level (Daniels et al., 2003;

Kim, 1999). Many studies measure the presence of one, or a few of the congeners believed to be

the most toxic. This toxic diversity complicates exposure assessment and reduces the validity of

statements made about PCBs in general. Also, there are several sources of human exposure, and

researchers are at the early stages of assessing exposure type and toxicity relationships (e.g.,

Jacobson & Jacobson, 2002). Exposure sources include: prenatal placental exposure; postnatal

breast milk exposure; ingestion of fatty fish, beef, and dairy products subject to bioaccumulation;

accidental ingestion due to industrial or waste management accidents (as in the Japan and Taiwan

contaminated cooking oil exposures); as well as simple background exposure (Cicchetti et al., this

issue; Ribas-Fito, Sala, Kogevinas, & Sunyer, 2001; Stein, Schettler,Wallinga, & Valeenti, 2002).

To date, the relationships between exposure type and subsequent toxicity remain unclear.

Second, toxicity and neurological effects may vary with age at exposure (e.g., Jacobson &

Jacobson, 2002). While no steadfast conclusions regarding age at exposure can be made as a result

of the many methodological shortcomings in PCB research, an inverse relationship between age

and exposure effects has been postulated. That is, the younger the child the more significant the

effects. To illustrate, selected findings implicate prenatal exposure as more damaging than breast

milk and other later exposures (e.g., Ribas-Fito et al., 2001).

Third, researchers struggle to accurately quantify, or measure, actual PCB exposure for reasons

beyond toxic diversity and age-related effects (Ribas-Fito et al., 2001). Each exposure type

presents with its own measurement problems. For example, while umbilical cord blood is considered

the best direct measure of prenatal exposure, PCBs are lipophilic and cord blood has low fat

content making cord blood a potential under indicator of exposure (Kim, 1999). Further, while

PCBs are readily detectible in human breast milk, due to its high fat content, this measure may also

inaccurately indicate actual maternal or child exposure (Kim, 1999). Another problem lies in the

use of self-report measures (e.g., maternal report of contaminated fish consumption), which depends

on subjective recall of experience and is potentially unreliable (Cicchetti et al., this issue). The

reliability of self-reported consumption is further attenuated as researchers use approximate PCB

averages for different species of fish to calculate reported exposure (Cicchetti et al., this issue).

Fourth, there are no set guidelines for the classification of safe and toxic exposure levels,

which further complicates the process. To date, research and literature review descriptions suggest

an emerging consensus of exposure-level classification. Specifically, the Japan and Taiwan contaminated

cooking oil incidents are believed to have resulted in what are often considered high

levels of prenatal exposure as a result direct maternal ingestion of PCBs; maternal ingestion of

fish, in some studies, has resulted in comparatively moderate prenatal exposure due to bioaccumulation

(Ribas-Fito et al., 2001); and background exposure in the general community is thought

to be responsible for relatively lower levels of exposure (Ribas-Fito et al., 2001). Recently, the

Center for Disease Control (CDC; Center for Disease Control, 2003) released a set of lipidadjusted

serum levels of PCBs found in the general population which include levels found in

children and adolescents 12–19 years of age. The CDC recommends comparing individual levels

to national averages to estimate potential elevated exposure. As these reports are new and list data

only for the adolescent age group, conclusions based on normative comparisons are lacking. While

some have attempted to make distinctions among exposure levels and related effects (e.g., Daniels

710 Cook-Cottone

et al., 2003; a study of low-level exposure), it will be important for future studies to compare

sample level to reported norms and carefully analyze data and make conclusions for respective

levels of exposure.

Finally, it is difficult, and in some cases impossible, to control for the many confounding and

potentially interactive variables such as parent IQ, home environment, other prenatal factors (e.g.,

alcohol consumption and smoking), and the potential additive or synergistic effects of other neurotoxic

agents (e.g., methylmercury; Cicchetti et al., this issue). Consequently, until researchers can

more clearly assess and report PCB effects, school psychologists should continue to read PCB

research cautiously. As with other potential risk factors, PCB exposure should be considered

within the context of a complete assessment of other background variables, developmental patterns,

and current psychoeducational variables.

Current Model for PCB Exposure and Cognitive, Neuropsychological,

and Behavioral Effects

Some researchers believe that exposure to PCBs results in mild behavior disorder and cognitive

deficits (e.g., Lai et al., 2002). While the exact biological mechanism of neurotoxicity

remains unclear, studies of animals and primates illustrate potential neurological outcomes (Lai

et al., 2002). Results of such studies have suggested the following sequelae: (a) changes in hippocampal

long-term potentiation, (b) alterations in brain corticosterioid levels during development

which may change brain organization and affect dopenergic systems, and (c) interference

with thyroid hormone and estrogen signaling (Abelsohn, Gibson, Sanborn, &Weir, 2002; Lai et al,

2002; Ribas-Fito et al., 2001). As compared to other species, developmental disabilities in humans

are the result of an appreciably more complex and sophisticated set of interactions among genetic,

social, and toxologic factors (Stein et al., 2002). Consequently, human phenotypic manifestation

of posited neurological outcomes of PCB exposure has, and will be difficult to document (e.g.,

Daniels et al., 2003). Further, research outcomes demonstrating small but significant group average

effect (e.g., 3 IQ points) may not be clinically significant at the individual level (Cicchetti

et al., this issue; Ribas-Fito et al., 2001). Still, some researchers pose that developmental evidence

of PCB-related neurological dysfunction exists (e.g., Ribas-Fito et al., 2001; Stein et al., 2002).

Assessing the developmental effects of prenatal exposure to PCBs in infants and newborns is

particularly troublesome due to the high number of confounding factors inherent to infant development

(Ribas-Fito et al., 2001). Beyond difficulties assessing potential exposure, newborn assessments

mainly assess characteristics such as muscle tone, reflexes, alertness, responsiveness, and

state regulation (Ribas-Fito et al., 2001) and are not considered measures of cognitive development.

Further, as indicated earlier, researchers’ reliance on maternal self-report as well as other

serious empirical limitations suggest that outcomes be read with a cautious and critical eye (Cicchetti

et al., this issue). In newborns, reviews of effects of prenatal exposure have reported the

following potential adverse outcomes: decreased birth weight, head circumference, and gestational

age (Stein et al., 2002); motor immaturity and poor ability (Abelsohn et al., 2002; Ribas-

Fito et al., 2001; Stein et al., 2002); poorer lability of states (Ribas-Fito et al.); increased startle

response (Ribas-Fito et al.; Stein et al.); and decreased or hypoactive reflexes (Abelsohn et al.;

Ribas-Fito et al.; Stein et al.).

While infant and early childhood assessments provide a more direct measure of cognitive

development (Ribas-Fitro et al., 2001), outcomes must also be interpreted guardedly as studies

PCB Implications 711

reviewed are manifest with empirical problems. In early childhood, PCB exposure has been inconsistently

associated with cognitive impairments with significant findings reported mainly in studies

of highly exposed infants and children. For example, a recent study assessing the development

of 1,207 8-month-old children did not observe a relation between low-level PCB exposure and

mental or psychomotor scores (Daniels et al., 2003).

The cognitive impairments reported include: mental impairment (Abelsohn et al., 2002; Ribas-

Fito, 2001); reduced memory (Ribas-Fito et al., 2001; Stein et al., 2002); decreased verbal ability

(Stein et al.); and impaired information processing (Stein et al.). Inconsistently reported developmental

delays associated with PCB exposure include reduced psychomotor development (Ribas-

Fito et al.; Stein et al.). Further, some studies suggest adverse behavioral and emotional effects

such as: decreased sustained activity; decreased high-level play; increased withdrawn and depressed

behavior; and increased activity level (e.g., Stein et al.). Finally, it is believed that posited behavioral

consequences may only be associated with instances of extremely high exposure (e.g., the

Taiwan incident; Ribas-Fito et al.).

While reviews and lay reports suggest school-age outcomes, there are little data on the longterm

effects of prenatal and postnatal PCB exposure. For example, a review by Stein and colleagues

(2002) associates PCB exposure with preteen cognitive difficulties including decreased

word and reading comprehension, decreased full-scale and verbal IQ, reduced memory, and reduced

attention. However, empirical support for such statements is lacking. For example, Vreugdenhil,

Lanting, Mulder, Boesma, and Weisglas-Kuperus (2002) reported subtle negative effects of prenatal

PCB and dioxin exposure in children whose parental and home characteristics were less

optimal. Of note, findings are based on the critically reviewed Dutch cohort data set. Also, Jacobson

and Jacobson (1996) conducted a follow-up study of children born to mothers reporting

ingestion of Lake Michigan fish and indicated deficit general intellectual functioning, memory

weaknesses, and attention difficulties at age 11 years. However, this data set is marked with

serious methodological shortcomings (Cicchetti et al., this issue). Further, using Raven’s Coloured

Progressive Matrices (Raven, Court, & Javen, 1985), Lai, Guo, Guo, and Hsu (2001) found

that highly prenatally exposed children (i.e., the Taiwan cohort) showed significantly lower general

cognitive functioning than controls at ages 7, 8, 11 and 12. Yet, at ages 13, 14, and 15 exposed

children seemed to gradually “catch up” with controls showing no significant cognitive differences.

In sum, sound, converging empirical evidence of significant long-term effects is not available.

Implications for School Psychologists

The health-care collaboration model (e.g., Phelps, 1999) places the school psychologist in a

key role in the provision of services to children affected by potential environmental teratogens.

The psychoeducational implications for exposure to PCBs remain unclear with much of the research

on children failing to show consistent, negative outcomes. While practitioners await more conclusive

evidence of outcomes, tentative practice guidelines can be recommended.

First, in great contrast to documented teratogens (e.g., alcohol and associated Fetal Alcohol

Syndrome and Fetal Alcohol Effects), PCB research does not support the use of a syndrome or

symptom cluster label. As with other similarly debated potential neurotoxins (e.g., lead), exposure

should be viewed as a possible risk factor considered along with each child’s ontogenetic and

ecological context (Phelps, 1999). Likewise, in the school setting children exposed to PCBs should

be evaluated for services as a result of their cumulative academic, social. and emotional presentation,

rather than as the singular result of an elevated PCB blood level. When developmental or

behavioral difficulties are suspected, assessment can be easily conducted in the school setting

712 Cook-Cottone

using a traditional psychoeducational assessment battery (i.e., cognitive, academic, and behavioral

measures). Assessment should be based on multiple sources of input and using multiple

formats (i.e., direct observation, oral-report, test, interview, and questionnaire). Subsequently,

recommendations should be driven by and responsive to the child’s needs as determined by a

synthesis of assessment data. In short, the referral and evaluation process for a child exposed to

PCBs should mirror that of any at-risk child.

Second, as with other health-related risk factors the school psychologist may be asked to

collaborate with medical personnel (Phelps, 1999). Such collaboration might include educational

consultation for medical personnel, release of assessment findings, and ongoing monitoring of

suspected behavioral or academic concerns potentially related to PCB exposure.

Third, the school psychologist may also assist in the development of appropriate support

services at school and home (Phelps, 1999). For example, a cross-environmental academic support

plan or behavioral intervention might be designed to respond to academic or behavioral

needs.

Fourth, as a healthcare collaborator, the school psychologists may also meet school or districtwide

needs by providing in-service training regarding PCB exposure, research limitations, and

suspected outcomes.

Fifth, by partnering with ongoing toxicology research teams the school psychologist can play

a critical role in the ongoing effort to understand PCB exposure and its effects on children.

Finally, if in fact PCB exposure proves to play a significant role in the development of

academic, social or emotional difficulties, the school psychologist will have a responsibility to

advocate for exposure prevention and early identification of exposed children (Phelps, 1999).

Overall, the current state of PCB research warrants a cautious awareness of the potential risks

involved with exposure. While methodologically sound and convincing research is awaited, adherence

to the healthcare collaboration model assures that school psychologists remain responsive to

children’s needs while helping to support efforts to identify and ameliorate potential environmental

risks.

PCB Implications 713