Introduction

ORIGINAL PAPER

Psychosocial Adjustment and Sibling Relationships in Siblings of Children with Autism Spectrum Disorder: Risk and Protective Factors

Katherine M. Walton1,2 • Brooke R. Ingersoll2

� Springer Science+Business Media New York 2015

Abstract This study compared sibling adjustment and

relationships in siblings of children with Autism Spectrum

Disorder (ASD-Sibs; n = 69) and siblings of children with

typical development (TD-Sibs; n = 93). ASD-Sibs and

TD-Sibs demonstrated similar emotional/behavioral ad-

justment. Older male ASD-Sibs were at increased risk for

difficulties. Sibling relationships of ASD-Sibs involved less

aggression, less involvement, and more avoidance than

those of TD-Sibs. Partial support for a diathesis–stress

conceptualization of sibling difficulties was found for

ASD-Sibs. For TD-Sibs, broader autism phenotype (BAP)

was related to psychosocial difficulties regardless of family

stressors. For ASD-Sibs, BAP was related to difficulties

only when family stressors were present. This suggests that

having a sibling with ASD may be a protective factor that

attenuates the negative impact of sibling BAP.

Keywords Siblings � Relationship � Adjustment � Diathesis–stress � Risk factors

Introduction

In recent years, a number of research studies have examined

the psychosocial adjustment and sibling relationships of

siblings of children with Autism Spectrum Disorder (ASD).

Findings from this literature have been varied. Several

studies have found that siblings of children with ASD ex-

perience more behavioral and emotional problems (Fisman

et al. 1996; Gold 1993; Griffith et al. 2014; Hastings 2003a;

Rodrigue et al. 1993; Ross and Cuskelly 2006; Verté et al.

2003) and poorer quality sibling relationships (Kaminsky

and Dewey 2001) compared to siblings of children with

typical development or siblings of children with other dis-

abilities. In contrast, other studies have found that siblings

of children with ASD do not show elevated levels of be-

havioral or emotional difficulties (Dempsey et al. 2012;

Hastings 2003b; Hastings and Petalas 2014; Kaminsky and

Dewey 2002; Pilowsky et al. 2004).

Recently, research on siblings has recognized the vari-

ability in sibling outcomes and has begun to focus on

identifying factors that may place some siblings at higher

risk for negative outcomes than others. For example, sev-

eral studies have found that male siblings are at higher risk

for difficulties (e.g., depression, low levels of prosocial

behavior) compared to female siblings (Hastings 2003a;

Macks and Reeve 2007). In contrast, Orsmond and Seltzer

(2009) found that sisters were more likely to report de-

pression and anxiety symptoms than brothers. Other studies

have found that older siblings of children with ASD may be

at increased risk for difficulties (e.g., externalizing prob-

lems, depression) in comparison to younger siblings

(Macks and Reeve 2007; Rodrigue et al. 1993). However,

the majority of these studies have been limited by small

sample sizes of children with ASD (with a maximum of 51

children with ASD in the largest study; Macks and Reeve

2007). In addition, the only study to specifically examine

how sibling gender and birth order may interact included

only 19 total children in the ASD group (Rodrigue et al.

1993), making it extremely difficult to detect an interaction

effect given the small sizes in each gender and age group.

& Katherine M. Walton

Katherine.walton@osumc.edu

1 Present Address: The Ohio State University, 371F

McCampbell Hall, 1581 Dodd Drive, Columbus, OH 43210,

USA

2 Michigan State University, Psychology Building, 316 Physics

Road Room 105B, East Lansing, MI 48824, USA

123

J Autism Dev Disord

DOI 10.1007/s10803-015-2440-7

Several studies have also examined how the character-

istics and behaviors of other family members (i.e., mother,

child with ASD) may influence sibling adjustment. Several

studies have found that greater ASD severity or more be-

havior problems in the child with ASD is associated with

more challenges (e.g., behavioral/emotional difficulties) for

typically developing siblings (Benson and Karlof 2008;

Hastings 2003b, 2007; Meyer et al. 2011). Meyer et al.

(2011) found that this relationship between child ASD

severity and sibling behavioral/emotional difficulties was

mediated by maternal depression. Furthermore, maternal

depression has been linked to child behavioral and emo-

tional difficulties in a number of populations (see Downey

and Coyne 1990 for review). This literature suggests that

characteristics of several family members may interact to

predict sibling adjustment problems.

Other research has proposed that variability in sibling

outcomes may be related to presence of sub-threshold

symptoms related to ASD (e.g., subtle social-communica-

tion deficits, behavioral rigidity) present in typically-

developing siblings. This pattern of behaviors is known as

the broader autism phenotype (BAP) and has been found to

more frequently occur in family members of individuals

with ASD compared to the general population (e.g., Bailey

et al. 1998); given this family linkage, researchers have

conceptualized BAP as a phenotypic representation of ge-

netic risk for ASD (Piven 2001). Several recent studies

have found that siblings who exhibit more characteristics

of the BAP have more emotional or behavioral difficulties

(Meyer et al. 2011; Mohammadi and Zarafshan 2014; Pe-

talas et al. 2012). In addition, several researchers have

found support for a diathesis–stress conceptualization of

sibling difficulties, in which sibling BAP characteristics

(genetic diathesis) interact with various family stressors to

predict sibling difficulties. For example, several research

groups have found that sibling BAP interacts with behav-

ioral problems in the child with ASD to predict sibling

difficulties (Meyer et al. 2011; Mohammadi and Zarafshan

2014; Petalas et al. 2012). Similarly, Orsmond and Seltzer

(2009) found that sibling BAP interacted with number of

stressful life events to predict depression and anxiety

symptoms in siblings. Interestingly, possible interactions

between maternal depression and sibling BAP have not

been explored in previous studies, despite the well-

documented relationship between maternal depression and

child behavioral and emotional difficulties.

Taken together, this body of research suggests that while

some siblings of children with ASD may be at increased

risk for behavioral and emotional difficulties, other siblings

are functioning well. In addition, sibling age and gender,

sibling BAP characteristics, child with ASD characteristics,

maternal depression, and stressful life events may all

contribute to predicting which siblings of children with

ASD are more likely to experience behavioral or emotional

difficulties. Despite the recent progress in identifying risk

patterns for siblings, this literature continues to suffer from

several significant weaknesses. First, while a literature

describing siblings’ psychological adjustment and func-

tioning has begun to develop, there are still very few

studies that have described sibling relationship quality in

siblings of children with ASD or examined factors that may

predict relationship quality. The few studies that have

looked at sibling relationships have not taken into account

how sibling BAP may influence sibling relationship quality

(e.g., Hastings and Petalas 2014; Kaminsky and Dewey

2001). Second, most previous studies have been limited by

small sample sizes, making it difficult to break down de-

mographic comparisons in a way that looks at meaningful

risk groups by both age and gender.

The current study sought to add to the literature on

adjustment and sibling relationships in siblings of children

with ASD by examining several questions in a sample of

children with ASD and children with typical development.

First, are there differences in mother-rated BAP charac-

teristics, behavioral/emotional difficulties, and sibling re-

lationship characteristics in a sample of siblings of children

with ASD versus siblings of children with typical devel-

opment? Second, do sibling gender and birth order influ-

ence risk for behavioral/emotional difficulties or sibling

relationship difficulties for children with ASD? Third, is a

diathesis–stress model (interaction between sibling BAP

and family stressors) predicting sibling behavioral/emo-

tional problems or sibling relationship problems supported

in this sample? If so, is this diathesis–stress conceptual-

ization unique to families with a child with ASD, or also

present in families with typically developing children?

Methods

Participants and Procedure

Participants included 163 mothers of two or more children

under the age of 18. Parents of children with ASD were

recruited through the Interactive Autism Network (IAN)

Research database. IAN Research is an on-line service that

links families of children with ASD in the United States to

on-going research studies. Over 25,000 families were

registered with IAN at the time of the current study. To

register, children must have received a diagnosis of

Autistic Disorder/Autism, Asperger Syndrome, Childhood

Disintegrative Disorder, Pervasive Developmental Disor-

der-Not Otherwise Specified, Pervasive Developmental

Disorder, or Autism Spectrum Disorder from a profes-

sional. An email explaining the purpose of the study and a

link to the on-line survey was sent to all registered parents.

J Autism Dev Disord

123

A description of the study with a link to the survey was also

provided on the IAN website under its research studies

page. Mothers of typically developing children were re-

cruited through StudyResponse (Stanton and Weiss 2002),

an online service that links researchers with participants

willing to complete surveys. The sample was recruited

from a panel of potential participants (n = 95,574) who

could earn five dollars for participation. Mothers of chil-

dren with ASD (ASD group) were asked to complete a

number of parent report measures on their child with ASD

(target child) and his or her typically developing sibling

(target sibling). If there was more than one typically de-

veloping sibling in the targeted age range (4–17 years),

mothers were asked to choose one sibling to report on.

Mothers of typically developing children (TD sample)

were asked to select both a target child and target sibling of

their choice to report on.

Informed consent was obtained and all data were col-

lected anonymously. Ninety mothers of children with ASD

and 144 mothers of children with typical development

began the online survey. Forty-seven participants (21 from

the ASD group and 24 from the TD group) were excluded

because they did not complete all measures in the survey or

left more than three items blank on any one measure. One

parent in the TD group reported ‘‘not true’’ to all behavioral

items (regardless of positive or negative valence of the

item), and was eliminated from the sample. An additional

26 participants from the TD group were excluded because

they reported that an immediate or extended family

member had been diagnosed with ASD (22 participants) or

that the target child or sibling had been diagnosed with

another developmental or language disorder (4 par-

ticipants). Children in the TD group who had been diag-

nosed with Attention/Deficit-Hyperactivity Disorder or an

emotional or behavioral disorder (16 target children and 12

target siblings) were not excluded from participation. In-

formation about sibling emotional/behavioral diagnoses

was not available for the ASD group. This yielded a final

sample of 69 mothers in the ASD group and 93 mothers in

the TD group.

Measures

Sociodemographic Characteristics

Mothers were asked to report several family demographic

characteristics. These included maternal age, ethnicity,

level of education, annual income, and marital status. They

were also asked to report on several sibling characteristics

that, based on previous research, were expected to affect

sibling adjustment and the sibling relationship, including

sibling gender, age, and birth order (older or younger than

target child).

Sibling Psychological Adjustment

Mothers completed the Strengths and Difficulties Ques-

tionnaire (SDQ; Goodman 1997) on the target sibling. The

SDQ is a 25-item behavioral screening questionnaire for

children between the ages of 4 and 17, and has been used in

adolescents up to age 19 (Van Roy et al. 2008). It is divided

into five scales comprised of 5-items each: Emotional

Symptoms, Conduct Problems, Hyperactivity/Inattention,

Peer Relationship Problems, and Prosocial Behavior. It also

includes a Total Difficulties composite score made up of

the four problem scales. The SDQ has strong psychometric

properties (Goodman 1997) and has been used in a range of

studies of psychological adjustment in children and ado-

lescents (e.g., Goodman and Goodman 2011; Mathai et al.

2002). Coefficient alpha for the SDQ Total Problems score

in this sample was .88.

Sibling Broader Autism Phenotype

Mothers completed the Social Responsiveness Scale (SRS;

Constantino 2002) on the target sibling. The SRS is a

65-item dimensional measure of symptoms associated with

a diagnosis of autism for children between the ages of 4

and 18. Raw scores on the SRS are converted to T-scores

based upon child gender. T-scores of 59 or less are in the

normal range. T-scores between 60 and 75 are in the mild

to moderate range of autism-related characteristics, and

T-scores of 76 or higher are in the severe range of autism-

related characteristics. This measure has been shown to

discriminate between children with ASD and those with

typical development and other psychopathology (e.g.,

Constantino et al. 2000). It has also been used to identify

autistic traits in unaffected siblings of children with ASD

(e.g., Constantino et al. 2006). Internal consistency for the

SRS in this sample was excellent (a = .97).

Sibling Relationship Quality

Mothers completed the Sibling Inventory of Behavior (SIB;

Schaefer and Edgerton 1981) for the target sibling and target

child. The SIB was developed to measure sibling relationship

quality in families with and without a child with special

needs. The SIB is made up of 32 items which comprise six

scales: Empathy, Rivalry, Aggression, Avoidance, Teach-

ing/Directiveness, and Companionship/Involvement. The

SIB also includes a Positive Behaviors Composite (sum

of Teaching/Directiveness, Empathy, and Companionship/

Involvement) and a Negative Behaviors Composite (sum of

Rivalry, Aggression, and Avoidance). The SIB has been

shown to have strong psychometric properties in

preadolescent and adolescent children (Hetherington and

Clingempeel 1992) and has been used in a number of studies

J Autism Dev Disord

123

investigating sibling relationships in this age range (e.g.,

Hetherington et al. 1999). Internal consistency for the SIB

Positive (a = .93) and SIB Negative (a = .93) scales in this

sample were both excellent.

Maternal Depression

Mothers completed the Center for Epidemiologic Studies

Depression Scale (CES-D; Radloff 1977) to report about

their own depressive symptoms. The CES-D is a 20-item

self-report scale that asks an individual to rate how often

she has experienced different symptoms of depression

during the past week. Higher scores on the CES-D indicate

greater depressive symptomatology. Scores of 16 or above

are suggestive of concern about possible clinical levels of

depressive symptoms. Internal consistency of the CES-D in

this sample was excellent (a = .95).

Impact of Child on Family

Mothers completed the Family Impact Questionnaire (FIQ;

Donenberg and Baker 1993) to provide information about

the impact of the child with ASD on the family, as com-

pared to other children of the same age. The FIQ includes

50 items rating the child’s impact on different areas of

family life. Caregivers are asked to rate the impact of their

child compared to other children the same age on a 4-point

scale from ‘‘Not at all’’ to ‘‘Very Much.’’ For example,

‘‘Compared to children and parents with children the same

age as my child, my child’s behavior embarrasses me in

public more.’’ Items are divided into six scales: impact on

social relationships (11 items), negative feelings about

parenting (9 items), positive feelings about parenting (7

items), financial impact (7 items), impact on marriage (7

items), and impact on siblings (9 items). For this study, the

impact on social relationships and negative feelings about

parenting subscales were summed to create the FIQ

Negative Composite, which was used as a measure of a

child’s negative impact on the family. The FIQ Negative

Composite has shown good reliability and validity in sev-

eral previous studies of children with developmental dis-

abilities (e.g., Baker et al. 2002; Neece and Baker 2008).

Internal consistency for the FIQ Negative Composite in

this sample was excellent (a = .95).

Statistical Methods

Missing Data

Since the final analysis sample consisted of participants

who left no more than three items blank on any one mea-

sure, missing data were not pervasive. However, since

scoring of the measures required complete responses, a

very small amount of missing items had to be singly im-

puted. For the SRS, a total of 15 items were left blank by

11 subjects (0.14 % of all SRS data). The median score

from the norming sample was entered for each the missing

items as instructed in the scoring manual for this instru-

ment. One item from one respondent on the SDQ (0.02 %

of all SDQ data), 13 items from 12 respondents on the SIB

(0.25 % of all SIB data), 2 items from 2 respondents on the

CES-D (0.06 % of all CES-D data), 7 items from 7 re-

spondents on the FIQ Negative Composite (0.22 % of all

FIQ data), and 19 items from 17 respondents on the SIB

(0.28 % of all SIB data) were left blank. For these ques-

tionnaires, the mean score of the other items completed on

the subscale containing the missing item (rounded to the

nearest whole number) was entered for the missing item.

Outliers

Data for sibling outcome measures were inspected for

outliers, defined as values greater than 3.5 standard de-

viations (SD) units from the sample mean for a given

variable. Two outliers on the SRS (one from the typical

group and one from the ASD group) and one outlier on the

SIB were found to not have high residuals or leverage

(outside the top 10 observations in the dataset) when pre-

dicted by group or by sibling age and birth order (the main

predictors in subsequent analyses), so were retained in the

dataset. There were no outliers on the SDQ.

Results

Sample Demographics

Sample characteristics were summarized using means/

standard deviations for continuous measures or counts/

percentages for categorical measures. Tests for differences

in demographic makeup between groups were examined

using independent samples t tests or Fisher’s exact tests as

appropriate. Descriptive statistics for the sociodemographic

and study variables are presented in Tables 1 and 2. The

groups did not differ on any of the measured demographic

variables, with the exception that the ASD-Sibs group

contained more male target children than the TD-Sibs

group (p \ .001), which is consistent with the finding that

ASD is approximately three to four times more common in

males (e.g., Fombonne 2003). Since the groups differed on

this one demographic variable, sibling data on the SDQ,

SRS, and SIB with a male target child were compared to

sibling data with a female target child to assess the po-

tential for confounding. None of these comparisons were

significant (all ps [ .10). Thus, target child gender was not

considered in any further analyses.

J Autism Dev Disord

123

Group Comparisons

Maternal Depression and Child Impact on Family

An independent samples t test indicated that mothers of

ASD-Sibs demonstrated significantly higher depression

scores on the CES-D compared to mothers of TD-Sibs (ASD-

Sibs mothers M = 17.74, SD = 12.75; TD-Sibs mothers

M = 13.49, SD = 14.13), p \ .05. In addition, mothers in

the ASD-Sibs group reported significantly higher negative

impact scores on the FIQ Negative Composite (M = 31.49,

SD = 14.13) compared to mothers in the TD-Sibs group

Table 1 Participant

demographics by group Variables Mean (SD)/percentage p value

ASD siblings

N = 69

Non-ASD siblings

N = 93

Parent age 40.53 (6.61) 40.18 (6.99) n.s.a

Parent race/ethnicity (%) n.s.b

White 94.2 81.7

Black 2.9 6.5

Hispanic 2.9 4.3

Asian 0 6.5

Native American 0 1.1

Parent education level (%) n.s.b

Some high school 2.9 0

High school diploma 4.3 0

Some college/spec. training 33.3 36.6

4-year college degree 39.1 48.4

Graduate degree 20.3 15.1

Marital status (%) n.s.b

Never married 1.4 5.4

Live-in partner 2.9 3.2

Married 81.2 78.5

Divorced/separated 13.0 12.9

Widowed 1.4 0

Annual family income (%) n.s.b

$24,999 or less 5.8 6.5

$25–$49,999 21.7 17.2

$50–$74,999 27.5 33.3

$75–$99,999 15.9 15.1

$100,000 or more 29.0 28.0

Target child age 9.35 (3.75) 8.98 (5.04) n.s.a

Target child gender (%) \.001b

Male 85.5 50.5

Female 14.5 49.5

Target sibling age 10.43 (3.57) 10.99 (3.86) n.s.a

Target sibling gender (%) n.s.b

Male 47.8 47.3

Female 52.2 52.7

Target sibling birth order (%) n.s.b

Younger 43.5 32.3

Older 50.7 65.6

Same age 5.8 2.2

a Independent samples t test, n.s. indicates not significant at the p \ .05 level b Fisher’s exact test, n.s. indicates not significant at the p \ .05 level

J Autism Dev Disord

123

(M = 13.97, SD = 11.00). This large difference in FIQ

scores in unsurprising, given that the FIQ instructions ask

parents to compare their child’s characteristics to an ‘‘aver-

age’’ child of the same age and gender. See Table 2.

Broader Autism Phenotype

On the SRS, 9.3 % of the siblings (8.7 % ASD-Sibs, 9.7 %

TD-Sibs) scored in the ‘‘severe’’ range of ASD symptoms

and 18.5 % (14.5 % ASD-Sibs, 21.5 % TD-Sibs) scored in

the ‘‘mild to moderate’’ range. A Chi square indicated that

there were no significant differences between the number of

ASD-Sibs versus TD-Sibs who scored in each risk/severity

range on the SRS, v2(2, N = 162) = 1.44, p = .49. An

independent samples t test detected no group differences in

total standard score on the SRS, t(160) = -1.20, p = .23.

A MANOVA detected no significant group differences on

the subscales of the SRS, F(5, 156) = 1.309, p [ .10.

Sibling Adjustment

Overall, 69.1 % of siblings (68.1 % ASD-Sibs, 69.9 %

TD-Sibs) scored in the ‘‘typical’’ range on the SDQ.

Among the remaining siblings, 8.0 % (5.8 % ASD-Sibs,

9.7 % TD-Sibs) had ‘‘slightly raised’’ scores, 6.8 % (7.2 %

ASD-Sibs, 6.5 % TD-Sibs) had ‘‘high’’ scores, and 16.0 %

(18.8 % ASD-Sibs, 14.0 % TD-Sibs) had ‘‘very high’’

scores. A Chi square test indicated that there was no sig-

nificant difference between the number of ASD-Sibs versus

TD-Sibs who scored in each severity group on the SDQ,

v2(3, N = 162) = 1.382, p = .71. An independent sam-

ples t test indicated no group differences in total difficulties

score on the SDQ, t(160) = 0.059, p = .95. However, a

MANOVA did indicate significant group differences on the

subscales of the SDQ, F(5, 156) = 2.288, p \ .05. Follow-

up tests indicated that ASD-Sibs demonstrated higher

scores (indicating better functioning) on the Prosocial

Table 2 Participant

characteristics by group ASD siblings

Mean (SD)

n = 69

Typical siblings

Mean (SD)

n = 93

Sample statistic p value

SDQ total difficulties 10.35 (7.31) 10.28 (7.35) 0.06a .95

Emotional problems 2.83 (2.54) 2.47 (2.62) 0.74b .39

Conduct problems 1.71 (2.08) 2.18 (2.13) 1.99b .16

Hyperactivity 3.81 (3.14) 3.58 (2.68) 0.25b .62

Peer problems 2.00 (1.92) 2.04 (1.79) 0.02b .88

SDQ pro-social behavior 8.36 (2.00) 7.62 (2.22) 4.77b .03*

SIB negative composite 37.72 (10.43) 38.20 (13.39) -0.25a .81

Rivalry 16.19 (5.77) 16.74 (6.20) 0.34b .56

Aggression 12.29 (3.80) 13.39 (4.32) 2.83b .10

Avoidance 9.25 (3.64) 8.08 (4.13) 3.52b .06

SIB positive composite 51.91 (11.49) 53.11 (12.33) -0.63a .53

Involvement 19.70 (5.52) 21.59 (5.46) 4.73b .03*

Empathy 19.72 (4.23) 19.00 (4.71) 1.03b .31

Teaching 12.49 (3.97) 12.52 (3.86) 0.00b .97

SRS total (T-score) 51.57 (15.74) 54.48 (15.04) -1.20a .23

Social awareness (T) 49.99 (13.84) 53.22 (11.23) 2.46b .12

Social cognition (T) 51.06 (14.72) 54.24 (14.09) 1.76b .19

Social communication (T) 49.88 (14.59) 52.83 (14.12) 1.56b .21

Social motivation (T) 53.09 (13.95) 54.97 (13.33) 0.72b .40

Autistic mannerisms (T) 53.58 (15.56) 54.34 (17.03) 0.04b .84

CES-D total score 17.74 (12.75) 13.49 (14.13) 2.20a .03*

FIQ negative composite 31.49 (14.13) 13.97 (11.00) 8.87a \.001*

SDQ Strengths and Difficulties Questionnaire, SIB Sibling Inventory of Behaviors, SRS Social Respon-

siveness Scale, CES-D Center for Epidemiologic Studies Depression Scale, FIQ Family Impact

Questionnaire

* p \ .05 a Independent samples t test b F-statistic, MANOVA

J Autism Dev Disord

123

Behaviors scale compared to TD-Sibs, F(1, 160) = 4.765,

p = .031.

Sibling Relationship

No significant differences between groups were detected by

independent samples t-tests for the SIB Positive Compos-

ite, t(160) = -.628, p = .53, or the SIB Negative Com-

posite, t(160) = -.247, p = .81. However, a MANOVA

did indicate significant differences between ASD-Sibs and

TD-Sibs on SIB subscale scores, F(6, 155) = 4.055,

p = .001. Follow-up tests indicated that ASD-Sibs dis-

played lower levels of Involvement, F(1, 160) = 4.732,

p = .031, as compared to TD-Sibs. In addition, there was a

trend toward ASD-Sibs displaying less Aggression toward

their siblings, F(1, 160) = 2.828, p = .095, and higher

levels of Avoidance of their siblings, F(1, 160) = 3.520,

p = .062, as compared to TD-Sibs.

Demographic Risk Factors

Strengths and Difficulties Questionnaire

A 2 9 2 9 2 (Group 9 Sibling Gender 9 Sibling Birth

Order) ANOVA was conducted with SDQ Total Score as

the dependent variable. Siblings reported to be the same

age (2 TD-Sibs and 4 ASD-Sibs) were excluded from this

and all other analyses including Birth Order due to inability

to determine birth order/twin status from the data collected.

This analysis revealed a significant main effect of sibling

gender, F(1, 148) = 3.933, p \ .05, such that brothers had

higher total problems scores on the SDQ than sisters

(Males M = 11.50, SD = 6.95; Females M = 9.43,

SD = 7.42). This main effect was modified by a significant

Group by Gender by Birth Order interaction, F(1, 148) =

6.585, p \ .05. Follow-up pairwise tests with Bonferroni

correction indicated significant differences between ASD-

Sibs and TD-Sibs for older brothers. Older brothers in the

ASD group (n = 15) had significantly higher total problem

scores on the SDQ (indicating more difficulties) than both

older sisters (n = 20) and younger brothers in the ASD

group (n = 14); older brothers in the TD group did not

show this pattern (See Fig. 1).

Given this interesting finding of elevated problem levels

in older male ASD-Sibs, this subgroup was examined in

more detail. The mean age of older male ASD-Sibs was

11.2 years (SD = 3.59, range 6–17). These older brothers

were a mean of 3.2 years older than their siblings with

ASD (SD = 1.32, range 1–5), who averaged 8.0 years old

(SD = 3.91, range 4–16). Examination of older brothers’

SDQ data in more detail revealed that 60 % of older male

ASD-Sibs scored above the clinical cutoff for SDQ total

score (compared to 24 % of the remaining ASD-Sibs).

Independent-samples t-tests comparing older male ASD-

Sibs to all other ASD-Sibs revealed that older male ASD-

Sibs had higher SRS scores [t(67) = -2.81, p \ .01] and

higher scores on the SIB Negative Composite [t(67) =

-2.07, p \ .05] than other ASD-Sibs. However, there were

no significant differences between these groups for SIB

Positive Composite scores [t(67) = 0.47, p = .63] moth-

ers’ CES-D scores [t(67) = -0.71, p = .48] or FIQ

Negative Impact score [t(67) = -1.03, p = .31]. This

suggests that older brothers’ difficulties were driven pri-

marily by their own characteristics, rather than by con-

founds with maternal depression or child with ASD impact

on the family.

A 2 9 2 9 2 (Group 9 Sibling Gender 9 Sibling Birth

Order) MANOVA was conducted with the SDQ subscales

as dependent variables. This MANOVA indicated sig-

nificant main effects of Group, F(5, 144) = 2.883, p \ .05,

Gender, F(5, 144) = 4.262, p \ .01, and Birth Order, F(5,

144) = 2.593, p \ .05. These main effects were modified

by a significant Group by Gender by Birth Order interac-

tion, F(5, 144) = 4.379, p \ .01. Follow-up tests indicated

that the overall main effect of Gender was driven by a

significant main effect of Gender for the Hyperactivity

scale, F(1, 148) = 14.228, p \ .001, such that boys scored

higher than girls overall. ASD-Sibs demonstrated higher

levels of Prosocial Behavior compared to TD-Sibs, F(1,

148) = 4.233, p = .041. There was a significant Group by

Gender by Birth Order interaction for both the Peer Prob-

lems subscale, F(1, 148) = 8.037, p \ .01, and the

Hyperactivity subscale, F(1, 148) = 11.059, p \ .01. Fol-

low-up paired comparisons indicated that older brothers in

the ASD group had more difficulties than other groups of

Fig. 1 Sibling behavioral/emotional problems by group, birth order,

and gender. *Significantly different from all other groups, p \ .05

following Bonferroni correction. Error bars represent SE. SDQ

Strengths and Difficulties Questionnaire

J Autism Dev Disord

123

siblings. They were at a significant disadvantage compared

to older sisters in the ASD group (Hyperactivity p \ .001,

Peer Problems p \ .01), younger brothers in the ASD

group (Hyperactivity p \ .05, Peer Problems p \ .001),

and older brothers in the TD group (Hyperactivity p \ .01;

Peer Problems p = .01).

Sibling Inventory of Behaviors

A 2 9 2 9 2 (Group 9 Sibling Gender 9 Sibling Birth

Order) MANOVA predicting Positive and Negative Com-

posite scores on the SIB revealed a significant Gender

by Birth Order interaction, F(2, 147) = 3.735, p \ .05.

Follow-up comparisons indicated that this effect was dri-

ven by a significant Gender by Birth Order interaction for

the SIB Negative Composite, F(1, 148) = 6.476, p = .01,

such that younger sisters in both groups displayed more

negative relationship behaviors compared to younger

brothers.

A second MANOVA examining the SIB subscales indi-

cated significant main effects of Group, F(6, 143) = 3.747,

p \ .01, and Birth Order, F(6, 143) = 5.291, p \ .001. The

main effect of Birth Order was modified by a significant

Gender by Birth Order interaction, F(6, 143) = 2.648,

p \ .05. Follow-up tests examining individual subscales did

not detect a significant main effect of Group for any sub-

scale, but there was a trend toward ASD-Sibs demonstrating

lower levels of Aggression, F(1, 148) = 3.475, p = .064,

higher levels of Avoidance, F(1, 148) = 3.457, p = .065,

and lower levels of Involvement, F(1, 148) = 4.183,

p = .043, compared to TD-Sibs. In addition, these tests re-

vealed a significant main effect of sibling Birth Order for the

Teaching subscale, F(1, 148) = 15.615, p \ .001, such that

older siblings (M = 13.43, SD = 3.44) were likely to dis-

play more teaching behavior compared to younger siblings

(M = 10.88, SD = 4.19). Finally, these tests revealed that

the Gender by Birth Order interaction was driven by a sig-

nificant interaction for the Rivalry subscale, F(1, 148) =

9.997, p \ .01. Follow-up comparisons indicated that

younger sisters across both the ASD and TD groups

(M = 19.30, SD = 6.15) displayed more Rivalry toward

their siblings compared to both younger brothers (M =

14.53, SD = 4.67; p \ .01) and older sisters (M = 15.94,

SD = 6.73; p \ .01).

Diathesis–Stress Model of Sibling Adjustment

Difficulties and Relationship Quality

Maternal Depression

A series of regression analyses were conducted to test a

diathesis–stress model of sibling adjustment and

relationship difficulties, in which sibling broader autism

phenotype (the proxy for genetic diathesis; SRS Total

T-Score) might interact with maternal depression (CES-D

score) and Group to predict sibling adjustment difficulties

and sibling relationship quality. Main effects (SRS Total

T-Score, Group, and CES-D score), all two-way interac-

tions, and the three-way interaction were all entered si-

multaneously. In this model, an interaction between SRS

and CES-D would indicate support for a diathesis–stress

conceptualization. All predictor variables were centered to

facilitate interpretation of main effects.

The regression predicting sibling SDQ Total Problems

detected a significant three-way (SRS by CES-D by

Group) interaction (b = 0.17, p \ .05). See Table 3 for

the full regression model. To examine this three-way

interaction, the final regression model was repeated for

the low CES-D (below the clinical depression cutoff of

16; ASD-Sibs n = 34, TD-Sibs n = 62) and the high

CES-D (scores of 16 or greater; ASD-Sibs n = 35, TD-

Sibs n = 31) groups separately. There was a significant

SRS 9 Group interaction for the Low CES-D group only

(b = -0.29, p \ .01). In the High CES-D group, there

was a main effect of SRS (b = 0.73, p \ .001). Gra-

phical examination suggested that, when mothers were

clinically depressed, there was a strong positive rela-

tionship between sibling SRS and sibling SDQ Total

Problems in both groups. However, when mothers were

not depressed, the relationship between SRS score and

SDQ Total Problems score was stronger for the TD-Sibs

than the ASD-Sibs. See Fig. 2.

The regression analysis predicting SIB Positive Com-

posite also detected a three-way SRS by CES-D by Group

interaction (b = -0.19, p \ .05). To examine this three-

way interaction, the final regression model was repeated

for the low CES-D and high CES-D groups separately. A

significant SRS 9 Group interaction was detected for the

Low CES-D group only (b = 0.40, p \ .01). For the High

CES-D group, main effects of SRS (b = -0.36, p \ .05)

and Group (b = -0.26, p \ .05) were detected. Graphical

examination of the data at high and low CES-D suggested

that, when mothers were depressed, there was a negative

relationship between sibling SRS score and SIB Positive

for both groups (i.e., sibling relationships were less

positive when siblings had high levels of BAP charac-

teristics). However, when mothers were not depressed,

this relationship held only for TD-Sibs. See Table 3 and

Fig. 2.

The regression predicting SIB Negative Relationship

behaviors detected a significant SRS by Group interaction

(b = -0.26, p \ .01). Graphical examination of SRS

scores by group suggested that SRS scores were more

strongly related to sibling negative relationship behaviors

for TD-Sibs than for ASD-Sibs. See Table 3.

J Autism Dev Disord

123

Sibling Impact

A second series of regression models were used to examine

whether sibling BAP (SRS score), target child impact on

family (FIQ Negative Composite score), and Group would

interact to predict sibling adjustment and relationship dif-

ficulties. Main effects (SRS Total T-Score, Group, and FIQ

Negative Composite score), all two-way interactions, and

the three-way interaction were all entered simultaneously.

In these analyses, interactions between SRS and FIQ would

indicate support for a diathesis–stress conceptualization.

All predictor variables were centered to facilitate inter-

pretation of main effects.

The regression predicting sibling SDQ Total Problems

score detected a significant three-way (SRS by FIQ by

Group) interaction (b = 0.247, p \ .01). To examine this

three-way interaction, the final regression model was

repeated for the low FIQ (below the sample FIQ mean;

ASD-Sibs n = 14, TD-Sibs n = 70) and the high FIQ

(above the sample FIQ mean; ASD-Sibs n = 55, TD-Sibs

n = 23) groups separately. A significant SRS by Group

interaction (b = -0.45, p \ .001) emerged for the Low

FIQ group only. In the High FIQ group, there was a main

effect of SRS (b = 0.74, p \ .001), but no significant in-

teraction. Graphical examination of the data at high and low

FIQ suggested that, when mothers reported a high negative

impact of the child, there was a strong positive relationship

between sibling SRS score and sibling SDQ Total Problems

score for both groups. However, when mothers reported a

low negative impact of the child, there was a positive re-

lationship between sibling SRS score and sibling SDQ Total

Problems score only for TD-Sibs. See Table 4 and Fig. 3.

The regression analysis predicting SIB Positive Com-

posite also detected a significant SRS by FIQ by Group

Table 3 Multiple regressions

using group, social

responsiveness scale, and

maternal depression to predict

sibling adjustment and

relationship quality

Predictor variable R R2 Beta Sample statistica

Sibling behavioral/emotional problems (SDQ total score)

Overall model .73 .54 25.57***

Group -.004 -0.07

Sibling broader autism phenotype (SRS) .79 9.25***

Maternal depression (CES-D) .15 1.72

Group 9 SRS -.18 -2.31*

Group 9 CES-D .08 .95

SRS 9 CES-D -.14 -1.89

Group 9 SRS 9 CES-D .17 2.51*

Positive sibling relationship (SIB positive composite)

Overall model .43 .19 5.02***

Group -.03 -.44

Sibling broader autism phenotype (SRS) -.58 -5.15***

Maternal depression (CES-D) .16 1.43

Group 9 SRS .31 2.90**

Group 9 CES-D -.24 -2.21*

SRS 9 CES-D .19 1.93

Group 9 SRS 9 CES-D -.19 -2.15*

Negative sibling relationship (SIB negative composite)

Overall model .60 .35 12.07***

Group .01 .09

Sibling broader autism phenotype (SRS) .66 6.60***

Maternal depression (CES-D) .03 .28

Group 9 SRS -.26 -2.76**

Group 9 CES-D .15 1.56

SRS 9 CES-D .09 .99

Group 9 SRS 9 CES-D -.04 -.46

SDQ Strengths and Difficulties Questionnaire, SRS Social Responsiveness Scale Total T-Score, CES-D

Center for Epidemiological Studies Depression Scale Total Score, SIB Sibling Inventory of Behaviors

* p \ .05; ** p \ .01; *** p \ .001 a Sample statistic = F for overall models, t for individual predictors

J Autism Dev Disord

123

interaction (b = -0.46, p \ .001). To examine this three-

way interaction, the final regression model was repeated for

the low versus high FIQ scorers. In the Low FIQ group

there was a significant SRS by Group interaction

(b = 0.32, p \ .05). No significant predictors emerged in

the High FIQ group. Graphical examination of the data at

high and low FIQ suggested that, when FIQ Negative was

high, sibling relationships were less positive overall, and

the relationship between SRS score and SIB Positive

composite for both groups was relatively weak. The rela-

tionship between SRS score and SIB Positive remained

weak for ASD-Sibs when FIQ negative was low. However,

there was a strong negative relationship between SRS score

and SIB Positive score for TD-Sibs when FIQ Negative

was low, suggesting that sibling BAP characteristics had a

detrimental impact on sibling relationships for TD-Sibs

when negative impact of the child on the family was low.

See Table 4 and Fig. 3.

For the regression predicting SIB Negative Relationship

behaviors, a significant SRS by Group interaction was

detected (b = -0.21, p \ .05). There was no significant

SRS by FIQ by Group interaction. Graphical examination

of SRS scores by group suggested that SRS scores were

more strongly related to sibling negative relationship be-

haviors for TD-Sibs than for ASD-Sibs.

Cumulative Risk

To examine whether the combination of a number of risk

factors may accumulate to predict risk for sibling adjust-

ment and relationship difficulties, a ‘‘risk scale’’ was cre-

ated. Siblings were assigned one point on this risk scale for

each of the following: (1) male gender, (2) older than the

sibling with ASD, (3) maternal CES-D score of C16 (above

clinical cutoff), (4) child FIQ score above the sample mean,

(5) SRS score above the sample mean. Risk scores could

range from 0 to 5. Bivariate correlations were calculated

separately for each group to examine whether this ‘‘risk

scale’’ was significantly correlated with siblings’ scores on

the SDQ and SIB. Medium to large correlations were found

between risk scale scores and the SDQ Total Difficulties

score, the SDQ Pro-social Behavior score, and the SIB

Negative Composite score for both groups. Significant

correlations between risk scale score and several subscales

were also found for both groups. All correlations were in

the expected direction. See Table 5.

Discussion

Group Comparisons

The first goal of this study was to examine whether there

were differences in sibling adjustment, sibling relationship

quality, and sibling BAP characteristics in a sample of

siblings of children with ASD and siblings of children with

typical development. In the current sample, there were no

differences in BAP characteristics (as measured by ma-

ternal report on the Social Responsiveness Scale) between

ASD-Sibs and TD-Sibs. This finding is consistent with

recent research indicating relatively low rates of BAP

characteristics in family members when only one indi-

vidual in the family is affected by ASD (e.g., Virkud et al.

2009).

This study also failed to find elevated levels of behav-

ioral/emotional difficulties in siblings of children with

ASD compared to a comparison group. While a substantial

minority of ASD-Sibs (about 32 %) did have at least mild

elevations in total problems on the Strengths and Diffi-

culties Questionnaire, this percentage was similar to that

found in the group of TD-Sibs. This finding of relatively

elevated scores in comparison to a normative sample, but

comparability to a comparison group, suggests that there

Fig. 2 Sibling adjustment and positive relationship by maternal

depression, group, and sibling broader autism phenotype. High and

Low SRS scores calculated at ±1 SD based on separate regression

equations for Depressed versus Non-Depressed Mothers (using

recommended clinical cutoff of C16 on CES-D to indicate depres-

sion); SDQ Strengths and Difficulties Questionnaire, SRS Social

Responsiveness Scale, SIB Sibling Inventory of Behaviors

J Autism Dev Disord

123

may be a selection bias toward families with concerns

about their children participating in research studies about

sibling adjustment/functioning. Given this finding of

elevations in behavioral issues in both groups, previous

findings noting elevations in sibling behavioral/emotional

difficulties in comparison to normative samples (e.g.,

Hastings 2003a; Meyer et al. 2011) should be considered

with caution. The one group difference that was found in

this study favored siblings in the ASD group; mothers re-

ported higher levels of prosocial behavior in ASD-Sibs

compared to TD-Sibs. This finding is consistent with other

research indicating that siblings of individuals with ASD

show positive outcomes, such as increased competence and

self-concept, increased empathy, and prosocial behaviors

(Verté et al. 2003).

In regard to sibling relationship quality, overall differ-

ences in positive and negative sibling relationship behav-

iors were not found across groups. However, several

differences in individual subscales of the Sibling Inventory

of Behavior were found. ASD-Sibs demonstrated slightly

less aggression toward their sibling. However, ASD-Sibs

were also reported to be slightly more avoidant and less

involved with their siblings compared to TD-Sibs. This

pattern may reveal both a level of understanding of the

sibling with ASD’s limitations (as demonstrated by lower

levels of aggression), but also a level of discomfort with

the sibling or difficulty having successful interactions (as

demonstrated by lower involvement and greater avoidance

in siblings of children with ASD). Another possible ex-

planation for this pattern could be that siblings who spend

less time together (greater avoidance) may have fewer

opportunities to display aggressive behaviors. However, as

avoidance and aggression were highly positively correlat-

ed, (r = .52, p \ .001) in the sample, this explanation is

not well-supported by the data. These findings suggest that

sibling relationships in families where one child has ASD

Table 4 Multiple regressions

using group, social

responsiveness scale, and child

negative impact on family to

predict sibling adjustment and

relationship quality

Predictor variable R R2 Beta Sample Statistica

Sibling behavioral/emotional problems (SDQ total score)

Overall model .74 .55 26.47***

Group -.11 -1.49

Sibling broader autism phenotype (SRS) .73 8.63***

Child negative impact (FIQ) .17 1.47

Group 9 SRS -.21 -2.50*

Group 9 FIQ .11 1.08

SRS 9 FIQ -.09 -1.03

Group 9 SRS 9 FIQ .24 2.63**

Positive sibling relationship (SIB positive composite)

Overall model .49 .24 7.10***

Group .19 2.05*

Sibling broader autism phenotype (SRS) -.28 -2.61*

Child negative impact (FIQ) -.34 -2.34*

Group 9 SRS .17 1.60

Group 9 FIQ .03 .20

SRS 9 FIQ .40 3.45**

Group 9 SRS 9 FIQ -.46 -3.93***

Negative sibling relationship (SIB negative composite)

Overall model .65 .42 15.77***

Group -.17 -2.05*

Sibling broader autism phenotype (SRS) .57 5.98***

Child negative impact (FIQ) .41 3.22**

Group 9 SRS -.20 -2.12*

Group 9 FIQ -.05 -.48

SRS 9 FIQ .18 1.74

Group 9 SRS 9 FIQ -.11 -1.04

SDQ Strengths and Difficulties Questionnaire, SRS Social Responsiveness Scale Total T-score, FIQ Family

Impact Questionnaire Negative Composite Total Score, SIB Sibling Inventory of Behaviors

* p \ .05; ** p \ .01; *** p \ .001 a Sample statistic = F for overall models, t for individual predictors

J Autism Dev Disord

123

may not be better or worse overall, but may be charac-

terized by different relational patterns than those of

typically developing siblings. These findings also highlight

the need to support siblings by helping them understand

how to engage successfully with their sibling with ASD.

Sibling Risk Factors

Examination of demographic risk factors for sibling be-

havioral/emotional difficulties revealed that older brothers

of children with ASD were at higher risk for overall dif-

ficulties, especially in the areas of Hyperactivity and Peer

Problems. In fact, 60 % of older brothers in the ASD-Sibs

group scored in the elevated range for overall SDQ score

(while only 24 % of other children in the ASD-Sibs group

scored in the elevated range). For comparison, only 33 %

of older brothers in the TD-Sibs group scored in the

elevated range for overall SDQ score. This finding is

consistent with several previous studies suggesting that

male siblings (Hastings 2003a; Macks and Reeve 2007)

and older siblings (Macks and Reeve 2007; Rodrigue et al.

1993) may be at increased risk for psychosocial difficulties.

In combination with these previous findings, the current

results indicate that older brothers of children with ASD

may be at a particular risk for psychosocial difficulties.

Therefore, monitoring psychosocial adjustment may be

especially important in this subgroup in order to identify

and support siblings who are having difficulties.

Diathesis–Stress Model

Partial support was found for a diathesis–stress conceptu-

alization of sibling behavioral/emotional and relationship

difficulties. Interestingly, this finding only held true for

ASD-Sibs. For TD-Sibs, elevated BAP in siblings was

consistently related to poorer behavioral/emotional func-

tioning as well as poorer sibling relationship quality (fewer

positive behaviors and more negative behaviors). For TD-

Sibs, this relationship between BAP and sibling psy-

chosocial and relationship difficulties remained strong re-

gardless of family stress variables (parent depression, high

impact of child on family). However, this was not the case

for ASD-Sibs. In ASD-Sibs, elevations in sibling BAP did

confer risk for poorer behavioral/emotional functioning

and fewer positive sibling relationship qualities. However,

this was only true when family stressors (maternal de-

pression, high child impact on family) were also present.

When these family stressors were absent, ASD-Sibs

Fig. 3 Sibling adjustment and positive relationship by family impact,

group, and sibling broader autism phenotype. High and Low SRS

scores calculated at ±1 SD based on separate regression equations for

High FIQ (above mean) and Low FIQ (below mean) families. SDQ

Strengths and Difficulties Questionnaire, FIQ Family Impact Ques-

tionnaire, SRS Social Responsiveness Scale, SIB Sibling Inventory of

Behaviors

Table 5 Correlations between risk score and sibling SDQ and SIB

scores

ASD siblings

n = 69

Typical siblings

n = 93

SDQ total difficulties .578*** .484***

Emotional problems .359** .333**

Conduct problems .397** .388***

Hyperactivity .510*** .375**

Peer problems .463*** .476***

SDQ pro-social behavior -.303* -.208*

SIB negative composite .333** .296**

Rivalry .115 .198

Aggression .253* .272**

Avoidance .442*** .379***

SIB positive composite -.191 -.167

Involvement -.280* -.244*

Empathy -.120 -.216*

Teaching -.036 .075

SDQ Strengths and Difficulties Questionnaire, SIB Sibling Inventory

of Behaviors

* p \ .05; ** p \ .01; *** p \ .001

J Autism Dev Disord

123

appeared to be protected from the deleterious impacts of

high BAP. That is, under conditions of low stress and high

sibling BAP characteristics, siblings of children with ASD

were actually functioning better than siblings of typically

developing children. We hypothesize that parents of chil-

dren with ASD may be better prepared to manage sub-

threshold autism characteristics in their typically-develop-

ing children and are able to employ appropriate support

strategies so that these characteristics do not interfere with

their children’s psychosocial functioning. In contrast, par-

ents of children with typical development may not have

developed strategies for managing these difficulties, and

therefore sibling BAP results in increased psychosocial

difficulties for these children. However, under conditions

of high stress (i.e., maternal depression, high negative

impact of the child with ASD on the family), parents of

children with ASD may be less effective in utilizing these

protective strategies or overall risk may accumulate,

negating the protective effect seen for siblings of children

with ASD under low stress conditions.

It is notable that previous studies that have found sup-

port for a diathesis–stress conceptualization of sibling BAP

and family stressors (e.g., Meyer et al. 2011; Mohammadi

and Zarafshan 2014; Orsmond and Seltzer 2009; Petalas

et al. 2012) have examined these patterns in samples that

only include siblings of children with ASD (without a

typically developing comparison group). Therefore, it is

unknown whether the interactions between BAP and family

stressors seen in these previous studies would differ in

families of children with typical development, as was seen

in the current sample.

Cumulative Risk

The moderate to high correlations between number of risk

factors (male gender, older than sibling with ASD, high

broader autism phenotype, high child FIQ score, maternal

depression) and siblings’ scores on the SDQ and SIB

indicates that these risks may operate in a cumulative

manner, both for siblings of children with ASD and sib-

lings of children with typical development. These findings

suggest that the number of risk factors may be important in

predicting sibling outcomes, over and above the presence

of any single risk factor.

Limitations

This study has several limitations. First, the researchers did

not independently confirm diagnoses for children with

ASD in this study. Therefore, it is possible that some

children who were reported by parents to have ASD in this

study did not meet full diagnostic criteria for the disorder.

It is also possible that some children or siblings in the TD

sample may have met criteria for ASD or another devel-

opmental disorder that was not yet diagnosed or was not

reported by the parent. However, previous research on

validity of parent-reported ASD status suggests that inter-

net parent report of ASD status is valid (Rosenberg et al.

2009). Furthermore, characterization of a representative

sample of children registered in the Interactive Autism

Network (from which the ASD sample was recruited) has

indicated that 98 % of the children in the sample were

given best-estimate ASD diagnoses by expert clinicians

(Lee et al. 2010). Nonetheless, future research using a more

fully characterized sample of individuals with and without

ASD would be helpful to confirm these findings.

An additional limitation of this study is that all infor-

mation about children was provided by mothers. Therefore,

it is possible that correlations between measures tapping

different maternal and child behaviors may have been ar-

tificially inflated by positive or negative reporting bias for

individual mothers. For example, mothers with high levels

of depression may have been more likely to rate their

children negatively on a variety of behavioral and emo-

tional outcomes due to their own negative reporting bias,

rather than due to true child deficits across several areas. It

will be important for future research to obtain ratings from

multiple informants (e.g., parents, teachers, typically de-

veloping siblings) to ensure that the relationships between

different child characteristics found in this study were not

primarily related to bias of a single reporter.

Finally, the sampling strategy for this study may have

resulted in a sample that is not representative of all

families/siblings of children with ASD. Response rates in

both groups were very low; this sample was highly self-

selected. It is notable that among participants in the TD-

Sibs group, a relatively large proportion of siblings were

excluded due to presence of a child with ASD in the

family. In addition, TD-Sibs demonstrated elevated levels

of difficulties on the SDQ, confirming that this comparison

group has a different makeup than representative samples

used for questionnaire norming. The current sample con-

sisted of primarily Caucasian, well-educated, and middle to

upper class families. Given this sampling bias, these results

may not be generalizable to all families. Future studies

should examine these questions in a more representative

sample of families.

In addition, when the child with ASD had multiple

siblings, mothers were given the choice of which typically

developing sibling to report on. Mothers of typically de-

veloping children were asked to choose a ‘‘target child’’

and ‘‘target sibling’’ without additional instruction. This

strategy was intended to produce a range of responses

about children of different ages and birth orders (and re-

sulted in similar mean ages for both the target child and

J Autism Dev Disord

123

target sibling across groups). However, mothers may have

chosen to report on the sibling who they were most con-

cerned about, or the sibling closest in age to the child with

ASD. The experiences of these siblings may not be char-

acteristic of all siblings of children with ASD (Hodapp

et al. 2005). Given these sampling limitations, findings

from this study may not generalize to all families with

children with ASD, or to all siblings of children with ASD

within a single family.

Conclusions and Future Directions

Findings from this study point to several important issues

for clinicians and researchers working with families of

individuals with ASD to consider. As a group, siblings of

children with ASD did not show elevated rates of adjust-

ment difficulties compared to siblings of children with

typical development. However, some subgroups of siblings

did show high rates of behavioral/emotional difficulties.

This was particularly true for older brothers of children

with ASD; in this subgroup, 60 % of siblings were reported

to have elevations in behavioral/emotional difficulties.

Clinicians working with families of children with ASD

should be alert for possible difficulties in siblings (par-

ticularly older brothers) and should be prepared to offer

additional family-based services or make appropriate re-

ferrals when indicated. Findings from this study also sug-

gested that siblings of children with ASD may be less

involved and more avoidant of their siblings compared to

siblings of typically developing children. It is not surpris-

ing that siblings of children with ASD may have difficulties

relating to their sibling, as their sibling’s social difficulties

may lead to unsuccessful play or interaction attempts. This

suggests that siblings are in need of resources and in-

struction about how to appropriately engage their sibling

with ASD in play in order to foster more frequent and

successful sibling interactions.

Despite the challenges experienced by some siblings of

children with ASD, this study also points to several po-

tential strengths for siblings of children with ASD. Siblings

of children with ASD demonstrated slightly higher levels

of prosocial behavior and lower levels of aggression in

their sibling relationships in comparison to siblings of

typically developing children. These findings suggest that

the experience of having a sibling with ASD may sensitize

typically developing siblings to the needs of others, and

that they are aware of and adjust to their siblings’ unique

needs within the sibling relationship. In addition, under

conditions of low stress (low maternal depression, low

child impact on the family), having a sibling with ASD

appeared to confer some protection from detrimental psy-

chosocial impacts associated with broader autism

phenotype characteristics. Future research should focus on

exploring how parents may effectively support these sib-

lings in order to encourage and foster this resilience.

Acknowledgments This research was supported by Autism Speaks

Grant #4726 to the first and second authors. We would like to thank

the families who participated in this research. We would also like to

thank Rebecca Andridge for her comments on an earlier draft of this

manuscript.

Conflict of interest The authors declare that they have no conflict

of interest.

Ethical standard This research was approved by the Michigan

State University Institutional Review Board. All participants provided

informed consent prior to participating in this research.

References

Bailey, A., Palferman, S., Heavey, L., & LeCouteur, A. (1998).

Autism: The phenotype in relatives. Journal of Autism and

Developmental Disorders, 28, 369–392.

Baker, B., Blacher, J., Crnic, K., & Edelbrock, C. (2002). Behavior

problems and parenting stress in families of three-year-old

children with and without developmental delays. American

Journal on Mental Retardation, 107, 433–444.

Benson, P., & Karlof, K. (2008). Child, parent, and family predictors

of latter adjustment in siblings of children with autism. Research

in Autism Spectrum Disorders, 2, 583–600.

Constantino, J. (2002). The social responsiveness scale. Los Angeles:

Western Psychological Services.

Constantino, J., Lajonchere, C., Lutz, M., Gray, T., Abbacchi, C.,

McKenna, K., et al. (2006). Autistic social impairment in the

siblings of children with pervasive developmental disorders.

American Journal of Psychiatry, 163, 294–296.

Constantino, J., Przybeck, T., Friesen, D., & Todd, R. (2000).

Reciprocal social behavior in children with and without perva-

sive developmental disorders. Journal of Developmental Behav-

ioral Pediatrics, 21, 2–11.

Dempsey, A., Llorens, A., Brewton, C., Mulchandani, S., & Goan-

Kochel, R. (2012). Emotional and behavioral adjustment in

typically developing siblings of children with autism spectrum

disorders. Journal of Autism and Developmental Disorders, 42,

1393–1402.

Donenberg, G., & Baker, B. L. (1993). The impact of young children

with externalizing behaviors of their families. Journal of

Abnormal Child Psychology, 21, 179–198.

Downey, G., & Coyne, J. C. (1990). Children of depressed parents:

An integrative review. Psychological Bulletin, 108, 50–76.

Fisman, S., Wolf, L., Ellison, D., Gillis, B., Freeman, T., & Szatmari,

P. (1996). Risk and protective factors affecting the adjustment of

siblings of children with chronic disabilities. Journal of the

American Academy of Child and Adolescent Psychiatry, 35,

1532–1541.

Fombonne, E. (2003). Epidemiological surveys of autism and other

pervasive developmental disorders: An update. Journal of

Autism and Developmental Disorders, 33, 365–382.

Gold, N. (1993). Depression and social adjustment in siblings of boys

with autism. Journal of Autism and Developmental Disorders,

23, 147–163.

Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A

research note. Journal of Child Psychology and Psychiatry, 38,

581–586.

J Autism Dev Disord

123

Goodman, A., & Goodman, R. (2011). Population mean scores

predict child mental disorder rates: Validating SDQ prevalence

estimators in Britain. Journal of Child Psychology and Psy-

chiatry, 52, 100–108.

Griffith, G., Hastings, R., & Petalas, M. (2014). Brief report: Fathers’

and mothers’ ratings of behavioral and emotional problems in

siblings of children with autism spectrum disorder. Journal of

Autism and Developmental Disorders, 44, 1230–1235.

Hastings, R. P. (2003a). Behavioral adjustment of siblings of children

with autism. Journal of Autism and Developmental Disorders,

33, 99–104.

Hastings, R. (2003b). Behavioral adjustment of siblings of children

with autism engaged in applied behavior analysis early inter-

vention programs: The moderating role of social support.

Journal of Autism and Developmental Disorders, 33, 141–150.

Hastings, R. (2007). Longitudinal relationships between sibling

behavioral adjustment and behavior problems of children with

developmental disabilities. Journal of Autism and Developmen-

tal Disorders, 37, 1485–1492.

Hastings, R., & Petalas, M. (2014). Self-reported behaviour problems

and sibling relationship quality by siblings of children with

autism spectrum disorder. Child: Care, Health, and Develop-

ment, 40, 833–839.

Hetherington, E. M., & Clingempeel, W. G. (1992). Coping with

marital transitions: A family systems approach. Monographs of

the Society for Research in Child Development, 57 (2–3, Serial

No. 227).

Hetherington, E. M., Henderson, S. H., Reiss, D., et al. (1999).

Adolescent siblings in stepfamilies: Family functioning and

adolescent adjustment. Monographs of the Society for Research

in Child Development, 64(4), 222.

Hodapp, R., Glidden, L., & Kaiser, A. (2005). Siblings of persons

with disabilities: Toward a research agenda. Intellectual and

Developmental Disabilities, 43, 334–338.

Kaminsky, L., & Dewey, D. (2001). Sibling relationships of children

with autism. Journal of Autism and Developmental Disorders,

31, 399–410.

Kaminsky, L., & Dewey, D. (2002). Psychosocial adjustment in

siblings of children with autism. Journal of Child Psychology

and Psychiatry, 43, 225–232.

Lee, H., Marvin, A., Watson, T., Piggot, G., Law, J. K., et al. (2010).

Accuracy of phenotyping of autistic children based on internet

implemented parent report. American Journal of Medical

Genetics, Part B, 153B, 1119–1126.

Macks, J. R., & Reeve, R. E. (2007). The adjustment of non-disabled

siblings of children with autism. Journal of Autism and

Developmental Disorders, 37, 1060–1067.

Mathai, J., Anderson, P., & Bourne, A. (2002). The Strengths and

Difficulties Questionnaire (SDQ) as a screening measure prior

to admission to a Child and Adolescent Mental Health Service

(CAMHS). Mental Health: Australian e-Journal for the Advance-

ment of. 1.

Meyer, K., Ingersoll, B., & Hambrick, D. (2011). Factors influencing

adjustment in siblings of children with autism spectrum disor-

ders. Research in Autism Spectrum Disorders, 5, 1413–1420.

Mohammadi, M., & Zarafshan, H. (2014). Family function, parenting

style and broader autism phenotype as predicting factors of

psychological adjustment in typically developing siblings of

children with autism spectrum disorders. Iranian Journal of

Psychiatry, 9, 55–63.

Neece, C., & Baker, B. (2008). Predicting maternal parenting stress in

middle childhood: The roles of child intellectual status,

behaviour problems, and social skills. Journal of Intellectual

Disability Research, 52, 1114–1128.

Orsmond, G., & Seltzer, M. (2009). Adolescent siblings of individuals

with an autism spectrum disorder: Testing a diathesis–stress

model of sibling well-being. Journal of Autism and Develop-

mental Disorders, 39, 1053–1065.

Petalas, M., Hastings, R., Nash, S., Hall, L., Joannidi, H., & Dowey,

A. (2012). Psychological adjustment and sibling relationships in

siblings of children with autism spectrum disorders: Environ-

mental stressors and the broad autism phenotype. Research in

Autism Spectrum Disorders, 6, 546–555.

Pilowsky, T., Yirmiya, N., Doppelt, O., Gross-Tsur, V., & Shalev, R.

S. (2004). Social and emotional adjustment in siblings of

children with autism. Journal of Child Psychology and Psy-

chiatry, 45, 855–865.

Piven, J. (2001). The broad autism phenotype: A complementary

strategy for molecular genetic studies of autism. American

Journal of Medical Genetics, 105, 34–35.

Radloff, L. S. (1977). The CES-D scale: A self-report depression

scale for research in the general population. Applied Psycho-

logical Measurement, 1, 385–401.

Rodrigue, J. R., Geffken, G. R., & Morgan, S. B. (1993). Perceived

competence and behavioral adjustment of siblings of children

with autism. Journal of Autism and Developmental Disorders,

23, 665–674.

Rosenberg, R. E., Law, J. K., Yenokyan, G., McGready, J.,

Kaufmann, W. E., & Law, P. A. (2009). Characteristics and

concordance of autism spectrum disorders among 277 twin pairs.

Archives of Pediatrics and Adolescent Medicine, 163(10),

907–914.

Ross, P., & Cuskelly, M. (2006). Adjustment, sibling problems, and

coping strategies of brothers and sisters of children with autistic

spectrum disorder. Journal of Intellectual and Development

Disability, 31, 77–86.

Schaefer, E. S., & Edgerton, M. (1981). The sibling inventory of

behavior. Chapel Hill, NC: University of North Carolina.

Stanton, J. M., & Weiss, E. M. (2002). Online panels for social

science research: An introduction to the study response project.

Technical report No. 13001. Syracuse, NY: Syracuse University,

School of Information Studies.

Van Roy, B., Veenstra, M., & Clench-Aas, J. (2008). Construct

validity of the five-factor Strengths and Difficulties Question-

naire (SDQ) in pre-, early, and late adolescence. Journal of Child

Psychology and Psychiatry, 49, 1304–1312.

Verté, S., Roeyers, H., & Buysse, A. (2003). Behavioural problems,

social competence, and self-concept in siblings of children with

autism. Child: Care, Health, and Development, 29, 193–205.

Virkud, Y., Todd, R., Abbacchi, A., Ahzng, Y., & Constantino, J.

(2009). Familial aggregation of quantitative autistic traits in

multiplex versus simplex autism. American Journal of Medical

Genetic Part B: Neuropsychiatric Genetics, 3, 328–334.

J Autism Dev Disord

123