Familial Autoimmune Thyroid Disease as a Risk Factor
for Regression in Children with Autism Spectrum Disorder:
A CPEA Study
Cynthia A. Molloy,1,14Ardythe L. Morrow,1Jareen Meinzen-Derr,1Geraldine Dawson,2
Raphael Bernier,2Michelle Dunn,3Susan L. Hyman,4William M. McMahon,5
Julie Goudie-Nice,5Susan Hepburn,6Nancy Minshew,7Sally Rogers,8Marian Sigman,9
M. Anne Spence,10Helen Tager-Flusberg,11Fred R. Volkmar,12and Catherine Lord13
A multicenter study of 308 children with Autism Spectrum Disorder (ASD) was conducted
through the Collaborative Programs of Excellence in Autism (CPEA), sponsored by the
National Institute of Child Health and Human Development, to compare the family history of
autoimmune disorders in children with ASD with and without a history of regression. A history
R). Family history of autoimmune disorders was obtained by telephone interview. Regression
wassignificantly associated with afamily history ofautoimmune disorders (adjusted OR=1.89;
95% CI: 1.17, 3.10). The only specific autoimmune disorder found to be associated with
regression was autoimmune thyroid disease (adjusted OR=2.09; 95% CI: 1.28, 3.41).
KEY WORDS: Autism; ASD; family history; autoimmune; regression.
Autism Spectrum Disorder (ASD) is a neuro-
developmental disorder occurring at a rate of
approximately 4/1000, and affecting four times more
boys than girls (Fombonne, 1999; Yeargin-Allsopp
et al., 2003). The diagnosis of autism (autistic
disorder) requires that specific clinical criteria be
met, as outlined in the Diagnostic and Statistical
Manual of the American Psychiatric Association,
Fourth Edition within the overarching category of
13University of Michigan, Ann Arbor, MI, USA.
14Correspondence should be addressed to: Cynthia A. Molloy,
Center for Epidemiology and Biostatistics, Cincinnati Children’s
Hospital Medical Center, University of Cincinnati College of
Medicine, 3333 Burnet Avenue, MLC 4051, Cincinnati, OH,
45229–3039, USA; Tel: (513) 636-7301; Fax:
1Center for Epidemiology and Biostatistics, Cincinnati Children’s
Hospital Medical Center, University of Cincinnati College of
Medicine, Cincinnati, Ohio, USA.
2University of Washington, Seattle, Washington, USA.
3Albert Einstein College of Medicine of Yeshiva University,
Bronx, NY, USA.
4School of Medicine and Dentistry, University of Rochester,
Rochester, NY, USA.
5School of Medicine, University of Utah, Salt Lake City, Utah,
6Health Sciences Center, University of Colorado, Denver,
7School of Medicine, University of Pittsburgh, Pittsburgh, PA,
8University of California – Davis, Davis, CA, USA.
9University of California – Los Angeles, Los Angeles, CA, USA.
10University of California – Irvine, Irvine, CA, USA.
11Boston University, Boston, MA, USA.
12Yale University, New Haven, CT, USA.
Abbreviation: CPEA, Collaborative Programs of Excellence in
Autism; NICHD, National Institute of Child Health and Human
Development; NIDCD, National Institute on Deafness and Other
Communication Disorders; ADI-R, Autism Diagnostic Interview-
Revised; ADOS, Autism Diagnostic Observation Schedule; ASD,
Autism Spectrum Disorder; PDD-NOS, Pervasive Developmental
Disorder-Not Otherwise Specified; AITD, Autoimmune Thyroid
? 2006 Springer ScienceþBusiness Media, Inc.
Journal of Autism and Developmental Disorders (? 2006)
(American Psychiatric Association, 2000). These
reciprocal social interaction, and the presence of
restricted interests or stereotypic behaviors. The
term Autism Spectrum Disorder (ASD) is a broader
category, inclusive of individuals with autism and
those having the same core deficits to a lesser degree
of severity (Tanguay, 2000). This latter group
includes children diagnosed with Pervasive Devel-
opmental Disorder– Not
(PDD-NOS) and Asperger’s Syndrome (American
Psychiatric Association, 2000).
While the etiology of ASD is unknown, there is
some evidence for an association with immune dys-
function (van Gent, Heijnen, & Treffers, 1997). There
are reports that children with ASD have abnormalities
in the peripheralblood cells and cytokines that mediate
both innate and adaptive responses (Connolly et al.,
1999; Croonenberghs, Bosmans, Deboutte, Kenis, &
Maes, 2002a; Gupta, Aggarwal, Rashanravan, & Lee,
1998; Jyonouchi, Sun, & Itokazu, 2002; Singh, 1996;
Warren, Foster, & Margaretten, 1987). Within the
adaptive immune response, elevated levels have been
reported for both TH1 cytokines, commonly seen in
autoimmune disorders (Singh, 1996), and TH2 cyto-
kines, as seen in atopic diseases (Gupta, Aggarwal, &
Heads, 1998). Elevated serum immunoglobulins
(Croonenberghs et al., 2002b) and autoantibodies to
neuronal elements (Singh, Warren, Averett, & Gha-
ziuddin, 1997) have also been reported in children
with autism. In addition, Comi et al. reported that
66% of children with autism had at least one relative
with an autoimmune disorder, compared to 50% of
normally developing controls (Comi, Zimmerman,
Frye, Law, & Peeden, 1999). Sweeten and colleagues
also reported a significantly higher mean number of
autoimmune disorders in families of children with
PDDs (1.87±1.6) compared to families of healthy
autoimmune disorders (1.44±1.5) (Sweeten, Bowyer,
Posey, Halberstadt, & McDougle, 2003). Croen and
colleagues reported an increased occurrence of
asthma and allergy in the mothers of children with
ASD, as well as two autoimmune disorders, psoriasis
and Type I diabetes (Croen, Yoshida, Odouli, &
The evidence for an association between the
immune system and ASD is increasing, but no
consistent pattern has yet been identified. While there
is evidence to support ASD as an autoimmune
phenomenon which is commonly associated with a
cell-mediated, TH1 response (Comi, Zimmerman,
Frye, Law, & Peeden, 1999; Singh, Warren, Averett,
Ghaziuddin, 1997; Sweeten, Bowyer, Posey, Halbers-
tadt, & McDougle, 2003) there is also evidence that
the disorder is associated with an allergic (TH2)
immune response (Lucarelli, et al., 1995; Gupta,
Aggarwal, Rashanravan, & Lee, 1998), or that ASD
is associated with general dysregulation of the
immune system (Gupta, Aggarwal & Heads, 1996;
Jyonouchi, Sun, & Le, 2001). These conflicting results
may be due in part to the underlying heterogeneity
within ASD, a diagnosis based on observable clinical
and historical findings. Within this clinically defined
disorder there are likely to be distinctive subgroups
that can provide more homogeneous samples for
etiologic studies. Categorizations for some of these
subgroups have been based on language skills
(Bradford et al., 2001), neurocognitive abilities
(Tager-Flusberg & Joseph, 2003) or the presence of
Another clinical feature with potential immunologic
associations is a history of regression (Richler, et al.,
Regression is a phenomenon in which a child
loses already established skills in communication or
social interaction (Goldberg, et al., 2003). It is
reported to occur in 20–30% of ASD cases, most
often between 18 and 24 months of age (Davidov-
itch, Glick, Holtzman, Tirosh, & Safir, 2000;
Fombonne, 1999; Rapin & Katzman, 1998; Shinnar
et al., 2001), and is rarely seen in other developmental
disorders. However, regression itself is, as yet, not
clearly defined. Regression may involve loss of lan-
guage, loss of other socio-communicative behaviors,
or both (Goldberg et al., 2003). The majority of
children with ASD who regress do not have com-
pletely normal development prior to the regression
(Lord, Shulman, & DiLavore, 2004; Siperstein &
Volkmar, 2004) and classification into regression–
nonregression groups is dependent upon the defini-
tion of regression.
Findings have been mixed on whether or not
measurable differences exist between children who
experience regression and those who do not. One
retrospective review of home videotapes indicated
that, at 12–18 months, children who experienced
regression had better social behavior, gaze, expressive
language and level of play when compared to children
with ASD who did not experience regression (Gold-
berg et al., 2003). In the Collaborative Programs of
Excellence in Autism (CPEA) sample used for the
current study, Luyster et al. found that children with
& Hillman, 2000).
Molloy et al.
regression were more likely than children with no
regression to have early prosocial skills in the typical
range (Luyster et al., 2005).
Several investigators have reported a difference
in cognitive ability between regression and nonre-
gression groups at school age, finding lower mean IQ
scores in children with a history of regression (Burack
& Volkmar, 1992; Kobayashi & Murata, 1998;
Kurita, Kita, & Miyake, 1992). In their report citing
no evidence for a new variant of MMR-induced
autism, Fombonne and Chakrabarti found no differ-
ences between regression and nonregression groups
on symptom severity as measured by ADI scores, but
they did report a difference in cognitive functioning
that approached significance, with 46.1% of children
in the regression group having an IQ<70 compared
to 21.5% of the nonregression group (p=0.08,
Fisher’s exact test) (Fombonne & Chakrabarti,
2001). Brown and Prelock (1995) also reported that
individuals with autism and regression had poorer
social skills later in life compared to those with
autism and no regression.
Taken together, these findings suggest that a
significant minority of children with ASD who
experience regression may represent a reasonably
research. The purpose of this study was to determine
if children with ASD and regression differed from
those with ASD and no regression in the frequency of
familial autoimmune disorders. Such group differ-
ences would support the continued investigation of
regression as a marker for a subgroup within ASD
with potential immunologic associations.
for further etiologic
This study was conducted as part of a larger
project within the Collaborative Programs of Excel-
lence in Autism (CPEA). The CPEA is a nationwide
collaboration established by The National Institute
of Child Health and
(NICHD) and the National Institute on Deafness
and Other Communication Disorders (NIDCD).
Ten sites participate in the CPEA and were collab-
orators in the present study: the Albert Einstein
College of Medicine, Boston University, University
of California – Irvine Medical Center, University of
California – Los Angeles, University of California –
Colorado, University of Rochester Medical Center,
and University of
Utah Autism Project,
University of Washington, University of Pittsburgh
– Western Psychiatric Institute and Clinic, and the
Yale Child Study Center, which included families
recruited through the University of Chicago, the
University of North Carolina and the University of
Investigators at each of the CPEA sites enroll
children with ASD into research studies on an
ongoing basis. Data that are part of the standard
CPEA evaluation include the Autism Diagnostic
Interview – Revised (ADI-R) (Lord, Rutter, &
Le Couteur, 1994), the Autism Diagnostic Observa-
tion Schedule (ADOS) (Lord et al., 2000), standard
measures of verbal and nonverbal IQ, including the
Differential Abilities Scale and the Mullen Scales of
Early Learning, and demographic information.
Subjects for this study were identified throughthe
CPEA site databases, which included a total of 1592
children with an ASD diagnosis. Children were
eligible for the study if they had a diagnosis of ASD,
available records of assessments performed at or near
the time of diagnosis, complete ADI-R, ADOS, verbal
and nonverbal IQ scores within the last five years.
Caretakers had to be available for a follow-up
telephone interview, and the children had to be
between the ages of 4 and 15 at the time of this
interview. Specific selection criteria and methods for
standardization across sites have been described
previously (Luyster et al., 2005). Since regression is
reported to occur in approximately 30% of children
with ASD, the design of the study required over
sampling children who experienced regression to
result in roughly equal groups of children with a
history of regression and children with no history of
regression. Attempts were made at each CPEA site to
recruit all eligible children identified as having a
history of word loss on their initial ADI-R. When
enrollment of children with word loss exceeded
enrollment of children with no word loss, the coor-
dinating site (Michigan) maintained balance between
group sizes by intermittent compensatory over sam-
pling of children with no word loss at that site.
The caretakers of potential subjects were invited
to participate in a standardized telephone interview
designed for this study and administered by trained
interviewers blind to the child’s regression history.
The interview collected additional information about
the child’s social and communication skill develop-
ment and the medical history of the child and family.
If there was a discrepancy between the initial ADI-R
coding and the telephone interview in reported loss
history, the original ADI-R protocol notes were
Familial Autoimmune Thyroid Disease
reviewed and compared with the interview results. If
the loss had been miscoded in the ADI-R (protocol
notes confirmed loss but item coded as no loss) the
child’s classification for regression status was chan-
ged. If no supporting evidence in the original ADI-R
was found to confirm the telephone interview, the
classification was maintained.
A positive history of regression was defined as a
loss of any previously acquired social or communi-
cation skills, based on all items from the initial ADI-
R pertaining to loss of language or other social skills.
A subset of the regression group was identified as
experiencing word loss. This was defined as having at
least three meaningful words used spontaneously on a
daily basis for at least a month, followed by a period
of at least a month in which no spontaneous language
at all was produced.
In addition to other medical history items
reported elsewhere (Richler, Luyster et al., 2005),
respondents were asked who, if anyone, in the child’s
biological family had a diagnosis of specific autoim-
mune disorders, including Type I Diabetes, Autoim-
mune Thyroid Disease (AITD), Connective Tissue
Autoimmune Disease, Juvenile Rheumatoid Arthri-
tis, Rheumatoid Arthritis, Multiple Sclerosis, Vitiligo
or Inflammatory Bowel Disease (Crohn’s Disease or
Ulcerative Colitis) or any other autoimmune disor-
der. Only first and second degree relatives were
included in the final analysis since caregivers were
more likely to have accurate knowledge regarding the
medical histories of immediate family members.
Two groups were defined by a history of
regression established in accordance with the CPEA
protocol. The Case group was defined as children
with ASD and a history of regression. The Control
group was defined as children with ASD and no
history of regression. Within the case group we
examined separately those children with and without
word loss. Proportions in the two groups were
compared using the v2statistic. Mean scores on
standardized tests were compared using t-tests and
Analysis of Variance (ANOVA). Unadjusted odds
ratios were calculated for familial history of autoim-
mune disorder in general and for specific disorders in
cases compared to controls. Potential covariates and
confounders were identified, including age at ADI-R,
race, gender, diagnosis (autism vs. other PDD),
verbal IQand maternal education.Maternal
education was reported as (1) college graduate, (2)
some college or associate degree, (3) high school
graduate or (4) did not finish high school, GED. After
demonstrating that groups 1 and 2 and groups 3 and 4
were similar, a dichotomous variable was created for
maternal education: some college or no college.
Logistic Regression models were built to evalu-
ate the contribution of each of the variables to the
outcome measure of regression status. Variables that
were associated with the regression outcome at the
p £ 0.1 level, maternal education and diagnosis, were
kept in the final models to calculate the adjusted odds
ratios for familial autoimmunity. The potential effect
modification (interaction) of familial autoimmunity
by maternal education and familial autoimmunity by
diagnosis were tested in the model and found not to
be significant in relation to the regression outcome
(p>0.2). Because 70% of the sample was recruited
from the Michigan/Chicago/NC site, with a higher
proportion (76%) of no word loss subjects (Table I),
CPEA site of enrollment was also included in the final
model. To evaluate the impact of the classification
status of the 35 children who experienced regression
without word loss, analyses were repeated with those
children removed from the dataset. All statistical
analyses were performed using Statistical Analysis
System (SAS) statistical package, version 8e (SAS
Institute, Inc. 1999).
Complete data on regression status and familial
autoimmune disorders were available on 308 children
with ASD from 9 CPEA sites. Demographic infor-
mation for this sample is shown in Table I. The
majority of the subjects (70%) were recruited through
studies conducted at the Chicago/NC/Michigan sites
as part of the Yale CPEA. The response rate was
evaluated at the Michigan site, where 55% of the
families who were contacted chose to participate in
the interview portion of the study (Luyster et al.,
2005). There were no significant differences between
responders and nonresponders with regard to age,
race, gender, diagnosis, maternal education or history
of regression from original ADI-R. As reported by
Luyster et al. 19% of children had discrepant infor-
mation between the original ADI-R and the tele-
phone interview. Forty-three (12.3%) children were
reported to have word loss in the interview only. On
review of original ADI-R data, 6 of these were
reclassified. Twenty-threechildren(6.6%) were
Molloy et al.
reported to have no word loss at the time of
interview. After review, 5 of these were reclassified.
Thus, following review of all available data, only 3%
of the children had a change in regression classifica-
tion from their initial ADI-R.
By design, one half of the children in this study
(n=155) experienced regression. For 120 (39%)
children, this regression included word loss. There
were no significant differences between the 155
children with regression and the 153 children with
no regression in regard to gender, race, IQ scores,
diagnosis or age at first ADI. A history of regression
was reported significantly more often by mothers
with at least some college education (p=0.02), and
the larger proportion of children experiencing regres-
sion who were diagnosed with autistic disorder
approached significance (p=0.06) (Table I).
A total of 175 of the 308 families (57%) in the
sample reported at least one first or second degree
relative with an autoimmune disorder. Children with
ASD and a history of regression were significantly
more likely than children with no history of regres-
sion to have a first or second degree relative with an
autoimmune disorder (Table II). Among the pro-
bands themselves, one child had a history of JRA
with no history of regression and one child had AITD
with a history of regression. Six children had a
diagnosis of inflammatory bowel disease. Of these, 5
Table I. Characteristics of Sample by History of Regression or NRegression
No regression n=153
Maternal education includes some college
Chicago/NC/Michigan CPEA site
Verbal IQ Mean±SD
Nonverbal IQ Mean ± SD
Age at first ADI (mos) Mean ± SD
aOther PDD includes ASD, PDD-NOS and Asperger’s Syndrome.
Table II. Family History of Autoimmune Disorders in a First or Second Degree Relative as a Risk Factor for Regression in Children with
Regression (any loss)
(95% Confidence Interval)p value
Family History of 1 or more relatives with
an Autoimmune Disorderb
Type I Diabetesc
Autoimmune Thyroid Diseasec
Connective Tissue Diseasec
Juvenile Rheumatoid Arthritisc
Inflammatory Bowel Diseasec
Family History of 2 or more relatives with
an Autoimmune Disorderb
108 (70)81 (55)1.89 (1.17, 3.10) 0.009
1.23 (0.73, 2.10)
2.09 (1.28, 3.41)
1.17 (0.55, 2.48)
0.60 (0.21, 1.72)
1.14 (0.68, 1.90)
1.76 (0.70, 4.42)
2.61 (0.66, 10.28)
1.25 (0.60, 2.63)
1.26 (0.77, 2.06)
aAdjusted for diagnosis, maternal education and CPEA site.
bMissing data on family history of any autoimmune disease n=9.
cMissing data on specific autoimmune diseases range n=10–15.
Familial Autoimmune Thyroid Disease
had a history of regression. The association between
regression and bowel symptoms in this study is
reported elsewhere (Richler et al., 2004).
Of the eight specific autoimmune disorders
included in the telephone survey, Type I Diabetes,
AITD, Connective Tissue Disease, JRA, Rheumatoid
Arthritis, Multiple Sclerosis, Vitiligo and Inflamma-
tory Bowel Disease, only familial AITD was found to
be a significant risk factor for regression in children
with ASD with an odds ratio of 2.09 (95% CI=1.28,
3.41) adjusted for maternal education, diagnosis and
site (Table II). The odds ratio increased to 2.40 (95%
CI=1.38, 4.20) if the AITD was reported to occur in
a maternal first or second degree relative, ie. mother,
maternal aunt or uncle, maternal half sibling or
maternal grandparent. (Table III). The relationship
between regression and AITD in a maternal relative
did not vary by the gender of the proband.
With increased specificity, comparing children
with word loss (n=120) to children with no regres-
sion (n=153), the strength of association for any
autoimmune disorder increased to an adjusted odds
ratio of 2.16 (95% CI: 1.24, 3.56). For AITD the
adjusted odds ratio increased to 2.36 (95% CI: 1.40,
3.99) for any relative and 2.96 (95% CI: 1.65, 5.30;
p=0.0003) for a maternal relative.
In this large sample of well characterized cases of
ASD where the definition of regression was stan-
dardized across established autism research centers,
familial autoimmunity, specifically AITD was a
significant risk factor for regression as defined here.
Previous case–control studies have found familial
autoimmunity to be a risk factor for autism (Comi
et al., 1999; Sweeten et al., 2003). In our study we
found that, in children diagnosed with ASD, a history
of familial autoimmunity, specifically AITD is a
significant risk factor for regression, suggesting that
this group of children with ASD and regression may
contribute substantially to the associations that have
In this study, an association was found between a
specific autoimmune disorder, AITD, and regression
in children with ASD. Among all autoimmune disor-
ders, AITD is the most common, with a prevalence
estimate of 2–4% in women and 1% in men (Vaidya,
Kendall-Taylor, & Pearce, 2002). We did not find
associations with any other specific autoimmune
disorders, but even with an estimated 3100 total
relatives in this study, we may not have had sufficient
power to detect differences in the less common
autoimmune disorders. The reports of an increased
prevalence of AITD and other immune disorders
among relatives of patients with rheumatoid arthritis
(Walker, Griffiths, & Griffiths, 1986), Type I Diabetes
(Payami, Joe, & Thomson, 1989), idiopathic inflam-
(Broadley, Deans Sawcer, Clayton, & Compston,
2000)and JRA (Prahalad,
Giannini, & Glass, 2002) have been interpreted as
evidence for a common polygenic risk factor that
confers susceptibility to autoimmunity (Ginn et al.,
The increased risk for regression in children with
ASD having a relative with AITD may also result
from these genetic risk factors common to autoim-
mune disorders. Alternatively, there may be disease
Table III. Family History of Autoimmune Thyroid Disease as a Risk Factor for Regression in Children with ASD
Regression (any loss)
(95% Confidence Interval)
Autoimmune Thyroid Disease in any
first or second degree relativeb
Autoimmune Thyroid Disease in a
maternal first or second degree relative
Autoimmune Thyroid Disease in a
paternal first or second degree relative
Autoimmune Thyroid Disease in mother
Autoimmune Thyroid Disease in father
64 (41)38 (27) 2.09 (1.28, 3.41)0.003
52 (34) 25 (18)2.40 (1.38, 4.20)0.002
18 (12) 17 (12)1.06 (0.51, 2.18)0.88
2.01 (0.78, 5.17)
0.99 (0.24, 4.17)
aAdjusted for diagnosis, maternal education and CPEA site.
bMissing data n=12.
Molloy et al.
specific susceptibility allele(s) common to AITD and
regression in ASD, or there may be alleles contrib-
uting separately to the disorders that are in linkage
disequilibrium and track together in families. As
these and other hypotheses of immune function in
ASD are explored, the results of this study support
the segregation of ASD cases by a history of
regression as a way to improve sample homogeneity.
In this study, the history of regression and
through caregiver report, which is a limitation.
However, the history of regression was based on the
ADI-R that was conducted at the CPEA site at the
time of diagnosis, minimizing reliance on recall of
past events for classification of cases and controls.
The history of familial autoimmunity was obtained
through direct interview of the caregiver by tele-
phone. The issue of selective recall of family history
that can introduce a potential bias when comparing
children with a condition such as ASD to healthy
controls is also minimized in this study since both
cases and controls had ASD, and were subject to the
increased awareness of family history that may result
from the diagnosis of a developmental disorder.
Because the nature of regression is not well
understood, potential misclassification of the children
with regression without word loss is another limita-
tion of the study. When the 35 children with
regression and no word loss were eliminated from
the analysis, the proportion of children with familial
AITD increased from 41 to 46% in the case group
characterized strictly by word loss (n=120). This
increased the adjusted odds ratio to 2.36 (95% CI:
Even as regression itself becomes better charac-
terized through ongoing research, attention to the
subgroup of children with ASD and a history of
immune function. With the diagnostic instruments
now available, it is possible to confidently characterize
and without regression that can be compared for
evidence of immune system dysregulation.
We are grateful to the patients and families at all
the CPEA sites who participated in this research. We
thank Dr. Mekibib Altaye for his critical review of
the statistical analyses. We also thank Dr. Thayne
Sweeten for his thoughtful reading and comments on
the manuscript and Deb Anderson for her assistance
in data management.
American Psychiatric Association. (2000). Diagnostic and Statisti-
cal Manual of Mental Disorders (4th ed.). Washington, DC:
American Psychiatric Association.
Bradford, Y., Haines, J., Hutcheson, H., Gardiner, M., Braun, T.,
Sheffield, V., Cassavant, T., Huang, W., Wang, K., Vieland,
V., Folstein, S., Santangelo, S., & Piven, J. (2001). Incorpo-
rating language phenotypes strengthens evidence of linkage to
autism. American Journal of Medical Genetics, 105(6), 539–
Broadley, S. A., Deans, J., Sawcer, S. J., Clayton, D., & Compston,
D. A. (2000). Autoimmune disease in first-degree relatives of
patients with multiple sclerosis A UK survey. Brain, 123(Pt 6),
Brown, J., & Prelock, P. A. (1995). Brief report: The impact of
regression on language development in autism. Journal of
Autism and Developmental Disorders, 25(3), 305–309.
Burack, J. A., & Volkmar, F. R. (1992). Development of low- and
high-functioning autistic children. Journal of Child Psychology
and Psychiatry, 33(3), 607–616.
Comi, A. M., Zimmerman, A. W., Frye, V. H., Law, P. A., &
Peeden, J. N. (1999). Familial clustering of autoimmune dis-
orders and evaluation of medical risk factors in autism.
Journal of Child Neurololgy, 14(6), 388–394.
Connolly, A. M., Chez, M. G., Pestronk, A., Arnold, S. T., Mehta,
S., & Deuel, R. K. (1999). Serum autoantibodies to brain in
Landau-Kleffner variant, autism, and other neurologic disor-
ders. Journal of Pediatrics, 134(5), 607–613.
Croen, L. A., Yoshida, C. K., Odouli, R., & Grether, J. K. (2004)
Maternal autoimmune and allergic diseases and childhood
autism. Presented at the International Meeting for Autism
Research, Sacramento, CA. May 7–8.
Croonenberghs, J., Bosmans, E., Deboutte, D., Kenis, G., & Maes,
M. (2002a). Activation of the inflammatory response system in
autism. Neuropsychobiology, 45(1), 1–6.
Croonenberghs, J., Wauters, A., Devreese, K., Verkerk, R.,
Scharpe, S., Bosmans, E., Egyed, B., Deboutte, D., & Maes,
M. (2002b). Increased serum albumin, ( globulin, immuno-
globulin IgG, and IgG2 and IgG4 in autism. Psychological
Medicine, 32, 1457–1463.
Davidovitch, M., Glick, L., Holtzman, G., Tirosh, E., & Safir, M.
P. (2000). Developmental regression in autism: Maternal per-
ception. Journal of Autism and Developmental Disorders, 30(2),
Fombonne, E. (1999). The epidemiology of autism: A review.
Psychological Medicine, 29, 769–786.
Fombonne, E., & Chakrabarti, S. (2001). No evidence for a new
variant of measles-mumps-rubella-induced autism. Pediatrics,
108(4), E58 .
Ginn, L.R., Lin, J.P., Plotz, P.H., Bale, S.J., Wilder, R.L.,
Mbauya, A., & Miller, F.W. (1998). Familial autoimmunity in
pedigrees of idiopathic inflammatory myopathy patients sug-
gests common genetic risk factors for many autoimmune dis-
eases. Arthritis and Rheumatism, 41(3), 400–405.
Goldberg, W., Osann, K., Filipek, P., Laulhere, T., Jarvis, K.,
Modahl, C., Flodman, P., & Spense, M. (2003). Language and
other regression: Assessment and timing. Journal of Autism
and Developmental Disorders, 33, 607–616.
Gupta, S., Aggarwal, S., & Heads, C. (1996). Dysregulated immune
system in children with autism: Beneficial effects of intrave-
nous immune globulin on autistic characteristics. Journal of
Autism and Developmental Disorders, 26, 439–452.
Familial Autoimmune Thyroid Disease
Gupta, S., Aggarwal, S., Rashanravan, B., & Lee, T. (1998). Th1- Download full-text
and Th2-like cytokines in CD4+ and CD8+ T cells in autism.
Journal of Neuroimmunology, 85(1), 106–109.
Jyonouchi, H., Sun, S., & Itokazu, N. (2002). Innate immunity
production against common dietary proteins in patients with
autism spectrum disorder. Neuropsychobiology, 46(2), 76–84.
Jyonouchi, H., Sun, S., & Le, H. (2001). Proinflammatory and
regulatory cytokine production associated with innate and
adaptive immune responses in children with autism spectrum
disorders and developmental regression. Journal of Neuroim-
munology, 120(1–2), 170–179.
Kobayashi, R., & Murata, T. (1998). Setback phenomenon in
autism and long-term prognosis. Acta Psychiatrica Scandi-
navica, 98(4), 296–303.
Kurita, H., Kita, M., & Miyake, Y. (1992). A comparative study of
development and symptoms among disintegrative psychosis
and infantile autism with and without speech loss. Journal of
Autism and Developmental Disorders, 22(2), 175–188.
Lord, C., Risi, S., Lambrecht, L., Jr. Cook, E. H., Leventhal, B. L.,
DiLavore, P. C., Pickles, A., & Rutter, M. (2000). The autism
diagnostic observation schedule-generic: A standard measure
of social and communication deficits associated with the
spectrum of autism. Journal of Autism and Developmental
Disorders, 30(3), 205–223.
Lord, C., Rutter, M., & Le Couteur, A. (1994). Autism diagnostic
interview-revised: A revised version of a diagnostic interview
for caregivers of individuals with possible pervasive develop-
mental disorders. Journal of Autism and Developmental Dis-
orders, 24(5), 659–685.
Lord, C., Shulman, C., & DiLavore, P. (2004). Regression and
word loss in autism spectrum disorder. Journal of Child Psy-
chology and Psychiatry, 45(5), 936–955.
Lucarelli, S., Frediani, T., Zingoni, A. M., Ferruzzi, F., Giardini,
O., Quintieri, F., Barbato, M., D’Eufemia, P., & Cardi, E.
(1995). Food allery and infantile autism. Panminerva Medicine,
Luyster, R., Richler, J., Risi, S., Hsu, W. L., Dawson, G., Bernier,
R., Dunn, M., Hepburn, S., Hyman, S. L., McMahon, W. M.,
Goudie-Nice, J., Minshew, N., Rogers, S., Sigman, M.,
Spence, M. A., Goldberg, W. A., Tager-Flusberg, H., Volk-
mar, F. R., & Lord, C. (2005). Early regression in social
communication in autism spectrum disorders: A CPEA study.
Developmental Neuropsychology, 27(3), 311–336.
Miles, J. H., & Hillman, R. E. (2000). Value of a clinical mor-
phology examination in autism. American Journal of Medical
Genetics, 91(4), 245–253.
Payami, H., Joe, S., & Thomson, G. (1989). Autoimmune thyroid
disease in type I diabetic families. Genetic Epidemiology, 6(1),
Prahalad, S., Shear, E. S., Thompson, S. D., Giannini, E. H., &
Glass, D. N. (2002). Increased prevalence of familial autoim-
munity in simplex and multiplex families with juvenile rheu-
matoid arthritis. Arthritis & Rheumatism, 46(7), 1851–1856.
Rapin, I., & Katzman, R. (1998). Neurobiology of autism. Annals
of Neurology, 43(1), 7–14.
Richler, J., Luyster, R., Risi, S., Hsu, W., Dawson, G., Bernier, R.,
Dunn, M., Hyman, S., McMahon, W., Goudie, J., Minshew,
N., Rogers, S., Sigman, M., Spence, M., Tager-Flusberg, H.,
Volkmar, F., & Lord, C. (2004). Is there a regressive ‘‘phe-
notype’’ of Autism Spectrum Disorder associated with the
Measles-Mumps-Rubella Vaccine: A CPEA Study. Presented
at the International Meeting for Autism Research, Sacra-
mento, CA. May 7–8, 2004.
Shinnar, S., Rapin, I., Arnold, S., Tuchman, R. F., Shulman, L.,
Ballaban-Gil, K., Maw, M., Deuel, R. K., & Volkmar, F. R.
(2001). Language regression in childhood. Pediatric Neurol-
ogy, 24(3), 183–189.
Singh, V. K. (1996). Plasma increase of interleukin-12 and inter-
feron-gamma Pathological significance in autism. Journal of
Neuroimmunology, 66(1–2), 143–145.
Singh, V. K., Warren, R., Averett, R., & Ghaziuddin, M. (1997).
Circulating autoantibodies to neuronal and glial filamentpro-
teins in autism. Pediatric Neurology 17, 88–90.
Siperstein, R., & Volkmar, F. (2004). Parental reporting of
regression in autism. Journal of Autism and Developmental
Disorders, 34(6), 731–734.
Sweeten, T. L., Bowyer, S. L., Posey, D. J., Halberstadt, G. M., &
McDougle, C. J. (2003). Increased prevalence of familial
autoimmunity in probands with pervasive developmental dis-
orders. Pediatrics, 112(5), e420.
Tager-Flusberg, H., & Joseph, R. M. (2003). Identifying neuro-
cognitive phenotypes in autism. Philosophical Transactions of
the Royal Society of London. Series B. Biological Sciences,
Tanguay, P. E. (2000). Pervasive developmental disorders: a 10-
year review. Journal of the American Academy of Child and
Adolescent Psychiatry, 39(9), 1079–1095.
Vaidya, B., Kendall-Taylor, P., & Pearce, S. H. (2002). The
genetics of autoimmune thyroid disease. Journal of Clinical
Endocrinology and Metabolism, 87(12), 5385–5397.
van Gent, T., Heijnen, C. J., & Treffers, P. D. (1997). Autism and
the immune system. Journal of Child Psychology and Psychi-
atry, 38(3), 337–349.
Walker, D. J., Griffiths, M., & Griffiths, I. D. (1986). Occurrence of
autoimmune diseases and autoantibodies in multicase rheu-
matoid arthritis families. Annals of Rheumatic Diseases, 45(4),
Warren, R. P., Foster, A., & Margaretten, N. C. (1987). Reduced
natural killer cell activity in autism. Journal of the American
Academy of Child and Adolescent Psychiatry, 26(3), 333–335.
Yeargin-Allsopp, M., Rice, C., Karapurkar, T., Doernberg, N.,
Boyle, C., & Murphy, C. (2003). Prevalence of autism in a US
metropolitan area. Journal of the American Medical Associa-
tion, 289(1), 49–55.
Molloy et al.