Research in Developmental Disabilities 51-52 (2016) 116125

Contents lists available at ScienceDirect

Research in Developmental Disabilities

Anxiety and depression symptomatology in adult siblings of

individuals with different developmental disability diagnoses
Linda P. ONeill *, Lindsay E. Murray 1
Department of Psychology, University of Chester, Parkgate Road, Chester CH1 4BJ, UK

A R T I C L E I N F O

A B S T R A C T

Article history:
Received 23 February 2015
Received in revised form 12 November 2015
Accepted 30 December 2015
Available online 25 January 2016

Factors predicting the emotional well-being of adult siblings of those with developmental
disability (DD) remain under-researched. In this study adult siblings of individuals with
Downs syndrome (DS), autism (ASD), Prader-Willi syndrome (PWS) and those with DD
but with unknown aetiology (DUA) were compared with each other and a closely-matched
control group to ascertain if sibling disability type made a difference to anxiety and/or
depression levels. Also considered was the interactive effect of gender, age, parental and
sibling educational attainment levels, socio-economic status and birth order on anxiety
and depression outcomes. With the exception of siblings of those with DS, adult siblings of
those with ASD, PWS and DUA reported signicantly higher levels of anxiety and
depression than the control group. There were some predictive effects of the demographic
variables upon anxiety and depression but none common to all disability types and no
moderating effects of demographic factors were found. Consequently other solutions must
be found as to why this important group of people have elevated rates of anxiety and
depression in comparison to the general population.
2016 Elsevier Ltd. All rights reserved.

Keywords:
Adult siblings
Developmental disabilities
Anxiety
Depression
Demographics

1. Introduction
Sibling relationships can offer one of the most mutually supportive and enduring bonds over the course of a lifetime (Bank
& Kahn, 1997; Voorpostel & Blieszner, 2008), improving emotional, cognitive and social growth (Brody, 2004). It is a
relationship that exists through circumstance and because of shared inuences it is often a relationship of relative equals
(Cicirelli, 1995). However, when one of the siblings has a developmental disability (DD) the relationship can be viewed as a
caring function as opposed to being reciprocally supportive (McHale & Gamble, 1989; Turnbull & Turnbull, 1990); this may
result in affective (anxiety and depression) consequences for the sibling without disabilities, particularly as many with DD
retain a lifelong reliance upon their families (Wolfe, Song, Greenberg, & Mailick, 2014). The characteristics of the type of DD
can also affect sibling outcome, with, for example, siblings of individuals with Downs syndrome (DS) usually having better
overall health, including fewer reported depression symptoms than those who have siblings with autistic spectrum disorder
(ASD) (e.g., Hodapp & Urbano, 2007). Research to date has usually focused on sibling affective outcomes in childhood;
consequently, there is relatively little information available on adult sibling affective outcome, and in particular, if it is the
characteristics of the disability or if it is the interactive effect of demographic factors that account for the outcomes.

* Corresponding author. Tel.: +44 1244 513193.

E-mail addresses: [email protected] (L.P. ONeill), [email protected] (L.E. Murray).
1
Tel.: +44 1244 511945.
http://dx.doi.org/10.1016/j.ridd.2015.12.017
0891-4222/ 2016 Elsevier Ltd. All rights reserved.

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

117

1.1. Inuences of disability types on sibling affective outcome

Adult siblings of individuals with DS, in comparison to ASD, have better sibling relationships, health, affective health and
more contact with their sibling with intellectual disabilities (Hodapp & Urbano, 2007). This is a trajectory that appears to be
continued from childhood because, as children, siblings of individuals with DS can emerge as better adjusted than normative
control samples (Hastings, 2007). Conversely, 40% of children whose siblings have ASD report scores on affective disorders
that place them on the at-risk or clinical range, compared to between 6 and 13% in typically developing samples (Ross &
Cuskelly, 2006). Smith and Perry (2005) found that 36% of children and adolescents whose sibling had ASD reported
borderline or clinically signicant internalising problems. Siblings of children with ASD and an intellectual disability have
more emotional problems than children whose siblings have only an intellectual disability or a control sample (Petalas,
Hastings, Nash, Lloyd, & Dewey, 2009). Possible reasons for this increased vulnerability include the increasing age of the child
with autism, being younger than the child with autism and the possibility of phenotypic similarities. However, when
Orsmond, Kuo, and Seltzer (2009) specically compared adolescents and adults whose sibling had ASD there was no
difference in depressive symptoms found between adolescence and adulthood; noticeably, neither age group reported
symptoms near to the cut-off point for clinical depression. The differences in these ndings may be due to comparing
children with adolescents and adults and possibly some ameliorative affective mechanism occurs as the sibling ages.
Similar to individuals with ASD, those diagnosed with Prader[9_TD$IF]-Willi syndrome (PWS) are characterised by maladaptive
behavioural tendencies (Dimitropoulos & Schultz, 2007; Veltman et al., 2004), often associated with mild mental retardation
(Cassidy & Driscoll, 2009). PWS is caused by a sporadic chromosomal disorder and tends to be diagnosed in infancy (Cassidy
& Driscoll, 2009). However, despite many disabilities such as DS, ASD and PWS having a clear diagnosis, in approximately
40% of individuals who have moderate to severe intellectual disability there is no known diagnosis (Knight et al., 1999). This
can be particularly stressful because, without a diagnosis, there can be no realistic prognosis. It also limits participation in
specic support groups, which can be isolating for the family.
1.2. Demographic inuences on sibling affective outcome
There is a greater general propensity for females to report higher affective symptoms than males (e.g., McLean, Asnaani,
Litz, & Hofmann, 2011); however, research has been inconsistent in reporting gender outcomes for siblings. Hannah and
Midlarsky (1999) reported that sisters of children with disabilities scored higher on affective disorders than brothers,
possibly because they play a larger part in their siblings caretaking; however, a meta-analysis found no gender differences in
affective disorders (Rossiter & Sharpe, 2001). Conversely, Hastings (2003) reported that brothers of children with ASD had
more adjustment difculties than sisters.
Higher parental educational attainment and levels of household income can help family adjustment and ameliorate
stress, due to a greater ability to access and pay for support (Bramlett & Mosher, 2002; Hastings, 2003). However, decrements
in mathematical and language achievements were witnessed as early educational experiences for siblings, alongside an
increased likelihood to have to repeat a grade (Fletcher, Hair, & Wolfe, 2012). Conversely, Wolfe et al. (2014) found normative
levels of education and employment in adult siblings. Lower levels of socio-economic status (SES) tend to be associated with
increased adjustment problems in children whose siblings have disabilities but this risk can be ameliorated by lower levels of
parental stress (Giallo & Gavidia-Payne, 2006; Macks & Reeve, 2007). Hartley, Barker, Baker, Seltzer, and Greenberg (2012)
found in families who have a child with ASD that household income signicantly affected maternal marital satisfaction,
which general studies show can impact upon the whole family (e.g., Gudmunson, Beutler, Israelsen, McCoy, & Hill, 2007).
1.3. Familial inuence on sibling affective outcome
In addition, being older than the sibling with disabilities is associated with better adjustment scores (Hastings, 2003).
Younger siblings often worry about overtaking their older sibling with disabilities either academically or physically
(Turnbull & Turnbull, 1990), they may suffer from more emotional problems (Petalas et al., 2009), and, due to extra care
giving responsibilities, may be more likely to have elevated rates of affective disorders (Gold, 1993). Conversely, Macks and
Reeve (2007) report siblings emotional adjustment patterns were more at risk when older than their sibling with ASD.
1.4. The present study
Research into the affective condition of adult siblings of individuals with disabilities is relatively limited and conicting. It
is known that adult homo- or heterotypic patterns of affective issues are best predicted by previous disorders, often from
childhood or adolescence (Copeland et al., 2013). The majority of studies that have taken place to date have mainly focused
on childhood adjustment (e.g., Cuskelly & Gunn, 2006; Petalas et al., 2009); however, with the progression of medical
technology, it is now common for individuals with disabilities to have much longer lifespans (Seltzer, Greemberg, Orsmond,
& Lounds, 2005). Consequently, there is a requirement to examine the long-term outcomes for these siblings because in
many instances adult siblings will assume responsibility for their sibling with disabilities once the parents are no longer able
to do so (Davys, Mitchell, & Haigh, 2015b). It is of paramount importance that the siblings are psychologically healthy in
order for them to be able to cope with the undertaking of responsibility for their sibling with disabilities, while potentially

118

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

looking after children of their own, maintaining careers and sometimes caring for elderly parents (Davys, Mitchell, & Haigh,
2015a). Our study therefore aimed to question whether elevated rates of anxiety and /or depression actually did exist in
adult siblings in comparison to the general population, or whether affective levels stabilised once the individual was no
longer as inuenced by the day-to-day challenges of living with their sibling with disabilities. Having a sibling with
disabilities is not in itself sufcient cause to experience higher affective levels; consequently, we also examined potential risk
factors as informed by previous research. Therefore, based upon reviewing the above research, the hypotheses for our study
were that:
1. There would be a difference in the levels of anxiety and depression symptomatology reported by the sibling group versus
the control group; however, the levels reported may vary depending upon sibling disability type.
2. Gender would be associated with affective symptomatology.
3. Participant educational attainment, parental education and income levels would be predictive of affective
symptomatology.
4. Being born before or after the sibling with disabilities would have an inuence on adult sibling symptomatology for each
of the DD sibling groups.
5. There would be individual moderating effects of gender, age, SES, participant educational level, maternal and paternal
educational levels inuencing the relationship between having a sibling with disabilities and anxiety and depression.

2. Methods
2.1. Participants
2.1.1. Participants who have a sibling with a disability
The participants were selected on the basis of being non-disabled siblings of individuals with disabilities and were
over 18 years old. The participants were identied via UK disability charities, including Sibs, the Downs Syndrome
Association, the National Autistic Society and the Prader[9_TD$IF]-Willi Society. For reasons of condentiality we were not
permitted direct access to the databases but the charities agreed to post questionnaires directly to potential participants
or advertise for participants via their newsletters or websites. The voluntary sample under study included 132 adults who
had a sibling with DS (n = 59), ASD (n = 31), PWS (n = 26) or DUA (n = 16). The age range varied from 19 to 71. The sex split
was 34 (25.8%) male and 98 (74.2%) female. A majority of 82 participants were married or cohabiting (62.1%), with 45
(34.1%) single and 5 (3.8%) divorced or separated. Mean employment level was working in intermediate occupations, with
an average salary of under 30,000. An undergraduate university degree was the mean level of education acquired. The
participants were split into groups depending upon their siblings disability type. The characteristics for each group are
shown in Table 1.
2.1.2. Control group participants
A further 193 participants responded to a questionnaire distributed through the University, friends, families, colleagues
and local businesses; 132 were included on the basis of their matching the sibling group and having at least one sibling but
no sibling with disabilities. Previous research has highlighted the importance in disability research of matching on a case-bycase basis in order to ensure that demographic differences do not skew group data (Stoneman, 1989). Consequently, the
control group was matched to the sibling group on a one-to-one basis; gender was always applied. Age was applied to within
ten years, with the mean age for the sibling group being 33.83 ( 10.69) and the control group 37.03 ( 12.34). Marital status
was applied in the majority of cases. Furthermore, SES was usually applied to within one point on employment status and annual
income. The full range of data for the control group can be seen in Table 1.
Table 1
Characteristics of the groups by sibling disability type shown as means (SDs) or frequencies (%).

Siblings
Age (mean years)
Sex (% female)
Employment
Annual income
EducLevel
MatEdLevel
PatEdLevel
Sib with disabilities
Age
Severity disability
Older than Sib %
Primary carer %

DS (N = 59)

ASD (N = 31)

PWS (N = 26)

DUA (N = 16)

Control (N = 132)

Test statistic

33.22 (9.01)
66.1%
4.16 (2.19)
3.68 (2.05)
3.95 (1.12)
2.98 (1.36)
3.25 (1.52)

32.19 (11.52)
90.3%
3.44 (2.06)
2.87 (1.85)
3.55 (1.12)
2.39 (1.45)
2.30 (1.61)

30.65 (9.70)
69.2%
3.35 (2.17)
2.77 (2.14)
3.42 (90.90)
2.46 (.88)
2.13 (1.32)

42.50 (14.59)
81.3%
3.75 (2.86)
3.13 (1.77)
3.13 (1.77)
1.00 (1.12)
1.42 (1.56)

37.03 (12.34)
74.2%
3.36 (2.26)
3.11 (1.96)
3.04 (1.30)
1.96 (1.47)
1.92 (1.56)

F
F
F
F
F
F
F

(4,
(4,
(4,
(4,
(4,
(4,
(4,

254) = 4.42,
259) = 1.82,
237) = 1.35,
258) = 1.45,
254) = 5.96,
200) = 7.42,
195) = 7.37,

32.57 (6.36)
2.47 (.71)
55.9%
1.7%

29.66 (12.48)
2.57 (1.55)
58.1%
6.5%

28.92 (11.02)
2.65 (.63)
53.8%
.0%

40.98 (13.20)
2.69 (.79)
31.3%
6.3%

F
F
F
F

(3,
(3,
(3,
(3,

127) = 5.76, p = .001

125) = .48, p = .70
127) = 1.22, p = .31
127) = 1.01, p = .39

p = .002
p = .13
p = .25
p = .22
p < .001
p < .001
p < .001

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

119

2.2. Measures
2.2.1. Demographic information
Employment status was rated from 0 (not working) to 7 (working in a highly qualied professional area), according to the
UK National Statistics Socio-economic Classication (Hall, 2006). Annual income was scored from 1 to 7, respectively: under
10,000, 10,00115,000, 15,00120,000, 20,00130,000, 30,00145,000, 45,00170,000 or 70,001+. The highest
level of educational attainment was requested for the participant and their mother and father, with the choice of none,
technical, GCSE or equivalent, A levels, undergraduate degree or postgraduate qualication, scoring from 0 to 5,
respectively.
2.2.2. The Hospital Anxiety and Depression Scale (HADS, Zigmond & Snaith, 1983)
The HADS, used to assess affective symptoms, is a 14-point, self-report scale, which assesses both depression and anxious
symptomatology, with scores above 7 indicating borderline problems and 11+ scores indicating clinical levels. HADS is
suitable for use in a non-psychiatric setting (Snaith, 2003) and has proven reliability and validity (Bjelland, Dahl, Haug, &
Neckelmann, 2002). This study had high reliability for both anxiety and depression (Cronbachs alpha reliabilities were .85
and .74, respectively). Also pertinent is that anxiety and depression are clearly delineated, making it easier to assess which
affective symptoms were evident within the different participant groups.
2.3. Procedure
After acquiring approval from the departmental ethics committee, paper or electronic questionnaires were sent to
participants, depending upon preference. Participants then privately completed the questionnaire and returned it in preaddressed envelopes or electronically. Postal responses had no identifying markers and thus were guaranteed anonymity.
Anonymity was ensured in the electronic versions because, upon submission, the questionnaire was allocated a random
number, then stored in a separate folder and the original email was deleted, thus no identifying markers were present.
2.4. Statistical analysis
This was a cross-sectional, between-subjects study design, completed by self-report. For all variables, mean values
and standard deviations for each of the groups were calculated, and these were then tested for differences between the
groups by one-way analyses of variance (ANOVAs), followed by Tukeys post hoc tests. Thereafter, moderation analyses
were carried out, using correlational analyses and multiple regression, to explain whether the demographic variables of
gender, age, SES and participant or parental educational attainment moderate the causal effect of having a sibling with
disabilities on affective symptomatology (Frazier, Tix, & Barron, 2004). Certain criteria for design, measurement and
analysis were considered due to the characteristic problem of low statistical power in moderated multiple regression
analyses (Aguinis & Gottfredson, 2010). The potential design issue of sample size was considered; the problem mainly
arises when the sample subgroup proportions are unequal (a 50/50 population base is optimum); this was addressed by
having exactly the same amount of participants in the sibling and control groups so, therefore, there should be no
decrease in power. Continuous variables were standardised (Frazier et al., 2004) and dichotomous variables were
transformed into dummy variables. Unstandardised Beta coefcients, F change values and signicance levels were
included for each variable, with R2[9_TD$IF] change values and signicance values presented for each step. The alpha level was set
at p < .05.
3. Results
3.1. Demographic variables for each of the groups
Table 1 presents demographic information on the groups. There was a predominance of female respondents within each
group. Those whose siblings had DUA were older than the other groups; specically, Tukeys post hoc analyses showed they
were signicantly older than the groups whose siblings had DS (mean difference = 9.28, SE = 3.23, p = .04), ASD (mean
difference = 10.31, SE = 3.53, p = .03) and PWS (mean difference = 11.85, SE = 3.65, p = .01). There were no signicant
differences in marital status, employment status or annual income between the groups. However, those whose siblings had
DS had signicantly higher educational qualications than the control group (mean difference = .91, SE = .19, p < .001).
Further, those whose siblings had DS had mothers with signicantly higher educational qualications than the DUA sibling
group (mean difference = 1.98, SE = .44, p < .001) and also the control group (mean difference = 1.03, SE = .30, p < .001);
similarly, those whose siblings had DS had fathers with signicantly higher educational qualications than the groups whose
siblings had PWS (mean difference = 1.12, SE = .38, p = .03) or DUA (mean difference = 1.83, SE = .49, p = .002) and the control
group (mean difference = 1.34, SE = .27, p < .001).
Table 1 also shows characteristics that relate solely to the sibling groups. The only signicant difference between the
groups was that siblings with DUA were signicantly older than the siblings with DS (mean difference = 8.36, SE = 2.80,
p = .02), ASD (mean difference = 11.27, SE = 3.07, p = .02) and PWS (mean difference = 12.01, SE = 3.15, p = .01).

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

120

Table 2
Differences in anxiety and depression symptomatology between the groups.
Siblings

DS (N = 59)

ASD (N = 31)

PWS (N = 26)

DUA (N = 16)

Control (N = 132)

Test statistic

Anxiety
Depression

8.12 (4.54)
3.85 (2.80)

10.23 (4.45)
4.81 (2.94)

10.23 (4.14)
4.73 (3.34)

9.50 (2.97)
5.75 (3.36)

6.94 (4.03)
3.19 (3.06)

F (4, 258) = 6.24, p < .001

F (4, 259) = 4.44, p = .01

Table 3a
Correlations among study variables (siblings of individuals with DS below the diagonal/siblings of individuals with ASD above).

1. Anxiety
2. Depression
3. Gender
4. Age
5. Employment
6. AnnIncome
7. EducLevel
8. MatEdLevel
9. PatEdLevel
10. Birth order

.56**
.27*
.09
.04
.11
.17
.28*
.17
.28*[8_TD$IF]

.69**

.02
.04
.05
.01
.12
.19
.18
.09

3
.35
.21

.06
.25
.31*
.01
.04
.01
.06

10

.01
.07
.09

.01
.01
.15
.13
.01
.43**

.14
.05
.22
.34

.79**
.34**
.30*
.16
.02

.04
.04
.14
.44*
.58*

.25
.28*
.09
.17

.00
.12
.04
.11
.22
.32

.29*
.41**
.07

.02
.12
.07
.03
.01
.18
.48**

.45**
.07

.06
.10
.04
.05
.20
.08
.50**
.41*

.02

.32
.14
.05
.27
.01
.02
.04
.31
.01

Key: ** Correlation is signicant at the .01 level, * Correlation is signicant at the .05 level; EducLevel = educational qualications; MatEdLevel = maternal
educational qualications; PatEdLevel = paternal educational qualications.

Table 3b
Correlations among study variables (siblings of individuals with PWS below the diagonal/siblings of individuals with DUA above).

1. Anxiety
2. Depression
3. Gender
4. Age
5. Employment
6. AnnIncome
7. EducLevel
8. MatEdLevel
9. PatEdLevel
10. Birth order

10

.18
.08
.25
.18
.09
.03
.41*
.05
.22

.42

.10
.06
.09
.16
.27
.36
.27
.50**[8_TD$IF]

.14
.09

.21
.13
.19
.06
.07
.07
.05

.10
.68**[1_TD$IF]
.20

.37
.43*
.19
.54**
.04
.04

.28
.27
.02
.32

.70**
.09
.34
.35
.07

.15
.04
.15
.02
.92**

.01
.37
.31
.07

.19
.54*
.16
.39
.41
.56*

.03
.20
.26

.11
.12
.19
.60*
.24
.01
.26

.07
.35

.17
.03
.32
.55
.57
.16
.07
.87**

.05

.16
.20
.37
.41
.29
.49
.22
.61**
.48

Key: ** Correlation is signicant at the .01 level, * Correlation is signicant at the .05 level; EducLevel = educational qualications; MatEdLevel = maternal
educational qualications; PatEdLevel = paternal educational qualications.

3.2. Differences in anxiety and depression symptomatology between the groups

To assess whether there were any affective differences between the sibling groups and the control group a series of oneway ANOVAs and Tukeys post hoc analyses were again carried out (see Table 2). Participants whose siblings had ASD or PWS
had signicantly higher anxiety levels than the control group (mean difference = 3.18, SE = .84, p = .002; mean
difference = 3.18, SE = .89, p = .04, respectively). It should be noted that the mean level for all of the sibling groups was
in the anxiety borderline category, in contrast to the control group, which was below the borderline range. Tukeys post hoc
analysis also showed the group whose siblings had DUA had signicantly higher depression levels than the control group
(mean difference = 2.56, SE = .80, p = .01) and those whose siblings had ASD showed a non-signicant trend to reporting
higher depression levels than the control group (mean difference = 1.62, SE = .61, p = .06). However, every group was below
the borderline range for depression.
3.3. Correlations within each sibling group
We computed bivariate correlations to ascertain if there were any signicant associations between the main study
variables. From Table 3a, it can be seen that siblings of individuals with DS displayed signicant positive correlations
between anxiety and depression, anxiety and being a sister, and high levels of maternal education and being younger than
the sibling with DS. Siblings of those with ASD also showed a strong association between anxiety and depression while, for
siblings of those with PWS, the main ndings were high associations between anxiety and high levels of maternal education,
and between depression and being younger than the sibling with PWS. Table 3b also shows for siblings of those with DUA
there were no associations between the variables chosen and anxiety; however, there were associations between depression
symptomatology and increasing age, having an older sibling with disabilities and low levels of education.

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

121

Table 4
Moderating effects of gender, age, SES, participant and parental educational levels on the relationship between having a sibling with disabilities and
standardised scores for anxiety.
Step and variable
Gender
Step1
Step 2
Age
Step 1
Step 2
SES
Step 1
Step 2
Educ level
Step 1
Step 2
MatEducLevel
Step 1
Step 2
PatEducLevel
Step 1
Step 2

SE B

95% CI

(sig)

R2[1_TD$IF]D

(sig)

SibWithDisbl
Gender
Gender  Sib

.45
.45
.19

.11
.14
.27

.23, .67
.19, .72
.34, .72

.23
.19
.09

<.0001
.001
.49

.09
.00

<.0001
.49

SibWithDisbl
Age (z score)
Age  Sib

.47
.03
.20

.12
.06
.12

.24, .70
.08, .15
.42, .03

.23
.03
.13

<.0001
.57
.10

.05
.01

<.0001
.10

SibWithDisbl
SES (z score)
SES  Sib

.45
.08
.02

.11
.06
.11

.23, .67
.20, .03
.20, .25

.23
.08
.02

<.0001
.14
.85

.06
.00

<.0001
.85

SibWithDisbl
EducLevel (z score)
EducLevel  Sib

.49
.04
.09

.12
.06
.12

.26, .72
.16, .07
.15, .32

.42
.04
.06

<.0001
.47
.47

.06
.00

<.0001
.47

SibWithDisbl
MatEdLevel (z)
MatEdLevel  Sib

.43
.01
.20

.12
.06
.13

.17, .69
.13, .12
.06, .45

.22
.01
.15

.001
.93
.12

.05
.01

.004
.12

SibWithDisbl
PatEdLevel (z)
PatEdLevel  Sib

.53
.06
.10

.13
.07
.13

.28, .79
.19, .07
.16, .35

.27
.06
.07

<.0001
.37
.46

.07
.01

<.0001
.46

3.4. Predictors of anxiety and depression within each sibling group

Ordinary least squares regression models were carried out separately for each of the sibling disability types; for
parsimonious reasons the predictors were chosen if they signicantly correlated with anxiety or depression for any of the
sibling groups. Therefore, due to no signicant correlations between the variables tested, no regression analyses were carried
out for depression for the siblings of those with DS or ASD or for anxiety for siblings of those with ASD or DUA. Please note
that due to sample size constraints, the predictor variables were entered individually in the groups whose siblings had PWS
or DUA. The regression model for anxiety symptomatology for siblings of individuals with DS was signicant (F (3, 48) = 6.94,
p = .001); the individual signicant predictors for anxiety were being a sister, higher levels of maternal education and being
younger than the sibling with DS. The regression model for depression in siblings of individuals with PWS was not signicant
(F (4, 21) = 2.34, p = .09); however, being younger than the sibling with disabilities did uniquely show signicance in
predicting depression. The anxiety and depression models for the DUA group showed no signicance; however, the
individual predictors of age (F (1, 14) = 12.17, p = .004), age of the sibling with disabilities (F (1, 14) = 7.12, p = .02) and level of
education achieved (F (1, 13) = 5.47, p = .04) were all signicant predictors of depression.
3.5. Moderation analyses
For the moderation analyses, we pooled the sibling groups together in order to increase statistical power. We wanted to
ascertain if there were any interactions between the variables under study. We found that the unstandardised regression coefcient for the whole cohort who had siblings with disabilities was signicant throughout, albeit slightly lower for maternal
education levels for anxiety and maternal and paternal education levels for depression. None of the unstandardised
regression coefcients for the interaction term were signicant, indicating there was no interaction or moderating effect for
any of the variables under study upon the relationship between having a sibling with disabilities and anxiety or depression.
Further the R2D maximum scores associated with the interaction terms was .01 indicating that the interaction terms did
nothing to increase the explained variance of anxiety or depression in addition to the variance explained by the independent
effects of having a sibling with disabilities and the demographic variables (see Tables 4 and 5).
4. Discussion
All of the sibling groups reported higher levels of anxiety and depression symptoms than the control group; however,
while all the sibling groups were in the borderline category for anxiety, all of the groups were below the borderline range for

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

122

Table 5
Moderating effects of gender, age, SES, participant and parental educational levels on the relationship between having a sibling with disabilities and
standardised scores for depression.
B
Gender
Step1
Step 2
Age
Step 1
Step 2
SES
Step 1
Step 2
Educ level
Step 1
Step 2
MatEducLevel
Step 1
Step 2
PatEducLevel
Step 1
Step 2

SE B

95% CI

(sig)

R2D

(sig)

SibWithDisbl
Gender
Gender  Sib

.43
.38
.19

.11
.14
.27

.21, .65
.12, .65
.72, .34

.21
.16
.09

<.0001
.005
.48

.07
.00

<.0001
.48

SibWithDisbl
Age (z score)
Age  Sib

.49
.17
.06

.11
.06
.12

.27, .71
.06, .28
.29, .17

.25
.17
.04

<.0001
.003
.60

.08
.00

<.0001
.60

SibWithDisbl
SES (z score)
SES  Sib

.43
.08
.13

.11
.06
.11

.20, .65
.19, .03
.10, .35

.21
.08
.09

<.0001
.16
.26

.05
.00

<.0001
.26

SibWithDisbl
EducLevel (z score)
EducLevel  Sib

.49
.16
.17

.12
.06
.12

.26, .71
.27, .05
.06, .40

.25
.16
.11

<.0001
.006
.14

.07
.00

<.0001
.14

SibWithDisbl
MatEdLevel (z)
MatEdLevel  Sib

.39
.11
.10

.13
.06
.13

.14, .64
.24, .01
.15, .34

.20
.12
.07

.003
.08
.45

.05
.01

.005
.45

SibWithDisbl
PatEdLevel (z)
PatEdLevel  Sib

.45
.10
.15

.12
.06
.12

.21, .79
.22, .07
.09, .40

.24
1.65
.12

<.0001
.10
.12

.06
.01

.001
.22

SibWithDisbl = sibling with disabilities; Sib = sibling with disabilities; EducLevel = educational attainment level, MatEdLevel = maternal educational levels;
PatEdLevel = paternal educational levels; CI = condence interval; Sex Dummy Coding (males coded 0, females coded 1); Sibling with disabilities Dummy
Coding (sibling without disabilities coded 0, sibling with disabilities coded 1).

depression. Those who had siblings with either ASD or PWS scored signicantly higher than the control group for anxiety,
and those whose siblings had ASD or DUA scored signicantly higher than the control group for depression. Individuals
whose siblings had DS were not signicantly different from the control group in their reporting of anxiety or depression.
Interestingly, being a sister of an individual with DS was a signicant predictor of anxiety but not for the other sibling
groups. Similarly, lower educational levels predicted higher levels of depression but only for those participants whose sibling
had DUA. Likewise, higher levels of maternal education predicted anxiety only for those individuals whose siblings had DS.
However, being younger than the sibling with DS was predictive of anxiety and being younger than a sibling with PWS
signicantly predicted depressive symptoms. Further, in most sibling groups, being the younger sibling showed a trend
towards predicting anxiety and /or depression. The interaction between having a sibling with disabilities with gender, age,
SES, and participant, mother and father educational attainment levels were all non-signicant and did not explain the
increased levels of affective disorders. These results indicate that there is no over-riding demographic variable that can
explain the increased affective symptomatology in siblings.
A growing body of research indicates that higher levels of educational attainment can act as a protective factor against
affective disorders (Bjelland et al., 2008; Gale, Hatch, Batty, & Deary, 2009); possibly with a cumulative effect as the
individual ages (Bjelland et al., 2008). Gale et al. (2009) suggested that having a higher IQ meant it was less likely that the
individual would experience socio-economic disadvantage, which can be indicative of affective disorders. However, despite
lower educational attainment being predictive of depressive symptoms in those whose siblings had DUA, they had the
second highest levels of SES and annual income when compared to the other groups. These ndings might be due to their
increased age and therefore longer period of working life or they might be anomalous due to the small sample size; however,
they are worthy of further study.
Another interesting atypical nding regards high maternal education levels which are associated with anxiety for the DS
and PWS sibling groups; previous research usually suggests that higher maternal educational levels have an ameliorative
inuence upon affective problems (Hastings, 2003). However, it has been suggested that siblings of individuals with
disabilities often implicitly perceive a need from their parents to excel in order to compensate for the lack of achievement in
the sibling with disabilities (Moore, Howard, & McLaughlin, 2002); this was not under study in our current research but we
could speculate that implicit pressure might have caused undue stress within the sibling group, resulting in heightened
anxiety levels that have been maintained into adulthood. Interestingly the groups whose siblings had DS and PWS were
amongst the highest academic achievers in this cohort and so our nding might be worthy of further exploration.

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

123

In line with previous research (Hastings, 2003; Petalas et al., 2009) our ndings show that being younger than the sibling
with disabilities, especially if the sibling has DS or PWS, is predictive of affective symptomatology. Research has raised
concerns from younger siblings regarding overtaking their older disabled siblings either academically or physically (Turnbull
& Turnbull, 1990). Younger siblings might consequently suppress negative feelings regarding their disabled sibling due to
guilt (Stoneman, Brody, Crapps, & Malone, 1991). This is an area that we feel deserves further research and gives potential
scope for intervention strategies, particularly in light of our previous comments that those who have siblings with DS or PWS
are among the highest achievers.
Interestingly, there were some predictive effects of symptomatology where siblings of individuals with DUA were more
likely to report higher levels of depressive symptomatology and, whilst not in the clinical range, it raises a theoretical point
about the effects of public awareness and acceptance of disability. The mean age of these participants was approximately ten
years older than the other sibling groups and, even in that short generation span, disabilities have become much more widely
accepted. It could be that the raising of public awareness and subsequent reduction in discrimination through disability
campaigns (e.g., http://www.worlddownsyndromeday.org/2013) have had a positive inuence, not just on the individual
with disabilities but also their siblings and should therefore be encouraged. However, perhaps the higher reporting of
depressive symptoms in siblings of DUA individuals was due to lack of access to specic disability support groups or
intervention schemes because of the absence of diagnosis and this too is worthy of further exploration.
The exact reasons underlying the increased sibling propensity towards affective issues as shown in this study remain
unclear and do not appear to be as a result of the demographic factors we considered. Falk, Norris, and Quinn (2014) reported
how socio-economic support is an important contributor to parental mental health; however, in direct contrast, no
independent effects of SES were found in the present study, potentially indicating that higher levels of affective problems in
the sibling groups were due to cognitive or emotional rather than demographic reasons. This argument follows through as no
independent effects were found for age or education attainment levels and the relationship with anxiety; however, being
female was independently reported as an indicator of affective disorders. Due to the independence of the effect, however,
this result may be due to the general increased propensity of women to report affective symptoms (e.g., McLean et al., 2011)
and little to do with the experience of having a sibling with disabilities. Further research is necessary to establish if sisters are
more vulnerable to affective disorders as a direct result of sibling disability.
The current results support previous research indicating that having a sibling with DS does not inevitably cause poor
levels of adjustment, with the present group outcome being very similar to the control group (Cuskelly & Gunn, 2006;
Hodapp & Urbano, 2007). Cuskelly and Gunn (2006) suggested that individuals whose siblings had DS should be addressed
on a within-family basis in order to identify specic families at risk. We would advocate that this approach be taken with all
families that include a disabled member, particularly addressing the specic risk and resilience factors we have identied,
namely, gender, age, birth order position, educational level and maternal educational level. This could have important
implications for the future roles of the sibling relationship; if higher quality relationships can be maintained, it should help to
promote increased positive contact as the siblings become older, resulting in increased emotional well-being for both
siblings.
The reasons behind the ASD and PWS sibling groups raised tendency towards adult anxiety require further investigation.
New research might focus on the lifelong behavioural difculties of individuals with ASD or PWS; whereby the lower
functioning individuals often have intensive, long-term specic care-needs to ensure their well-being (Cassidy & Driscoll,
2009; Strunk, Pickler, McCain, Ameringer, & Myers, 2014), which can often result in feelings of social isolation for the carer
(Strunk et al., 2014). This would be expected to have either a subliminal or overt affective inuence upon the sibling,
depending upon how involved they are in their siblings day-to-day care needs. If the involvement is on a regular basis, it
might take an exhaustive, emotional toll, leading to affective disorders symptomatology as it can do in mothers of autistic
individuals (Baker, Seltzer, & Greenberg, 2011). If, however, the sibling plays a more distant role, it might be feelings of guilt
that lead to affective symptomatology (Atkins, 1991).
Limitations to this study include the cross-sectional study design, which did not allow us to examine pre-existing
developmental factors that might have an enduring inuence on the individuals. Also we had no baseline marker and so we
cannot know whether affective symptoms had increased or decreased since the siblings reached adulthood. Further, we did
not examine whether there was a subtle presence of characteristics of the broader autism phenotype which might impact on
sibling affective outcomes. Another limitation was the sample size of the sub-groups; limited numbers did not permit
detailed analyses on individual groups. Finally, this study was carried out on a Caucasian, mainly middle-class, volunteer
sample and consequently may not be generalisable to the wider population.
However, a strength of our research was that we used a very closely matched control group, which ensured that we were
measuring like-for-like and allowed for very tight control between the variables under investigation. We also had a relatively
high number of participants, allowing individual variation in results and increased statistical power. The exception was for
those with siblings with DUA but this is typical of this type of study (Hastings, 2003).
In conclusion, our study has provided empirical evidence that adult siblings of individuals with disabilities do appear
more vulnerable to affective disorders than a closely matched group. We have also shown that the level of vulnerability is
dependent upon the type of disability that the sibling has. The exception to this appears to be siblings of those with DS, who
reported similar levels of affective symptomatology to the control group. The emotional wellbeing of siblings is a very
important area of further study, due to more people with disabilities surviving into late adulthood. This means that the
siblings may well have to assume responsibility for their welfare (Davys et al., 2015a,b), perhaps alongside looking after

124

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

children of their own and potentially caring for elderly parents. The affective state of the sibling therefore needs to be
safeguarded if they are going to carry out these roles effectively.
Previous suggestions have indicated that low levels of economic security are predictive of low levels of mental wellbeing
in parents of children with disabilities (e.g., Falk et al., 2014); however, we could nd no overall underlying demographic
factor that could predict symptomatology in all of the adult sibling groups. Whilst this research has demonstrated the
importance of eliminating factors that can cause symptomatology, we suggest that a focus on emotional cognitions and
behaviours may prove a more productive area for future research into the causes of higher sibling levels of anxiety. While it is
not expected that a single variable will explain the increased sibling tendency towards affective symptomatology, our study
does highlight the importance of further research to try to ascertain which factors play a part in maintaining vulnerability;
without an understanding of specic factors inuencing vulnerability it is difcult to put effective intervention procedures in
place.
Ethical standards
The research associated with this article was subjected to the stringent standards set down by the Department of
Psychologys Ethical Committee, which fully adheres to the ethical guidelines set out by the British Psychological Society.
Informed consent
All participants gave their fully informed consent prior to participating in this research.
Conict of interest statement
There are no conicts of interest[2_TD$IF].
Acknowledgements
We would like to sincerely thank Professor Mike Boulton and Dr David Scott who offered advice on aspects of the study.
References
Aguinis, H., & Gottfredson, R. K. (2010). Best-practice recommendations for estimating interaction effects using moderated multiple regression. Journal of
Organisational Behavior, 31, 776786. http://dx.doi.org/10.1002/job.686
Atkins, S. P. (1991). Siblings of learning disabled children: Are they special too? Child and Adolescent Social Work, 8, 525533. http://dx.doi.org/10.1007/
BF00755239
Baker, J. K., Seltzer, M. M., & Greenberg, J. S. (2011). Longitudinal effects of adaptability on behavior problems and maternal depression in families of adolescents
with autism. Journal of Family Psychology, 25, 601609. http://dx.doi.org/10.1037/a0024409
Bank, S. P., & Kahn, M. D. (1997). The sibling bond. New York, NY: Harper Collins.
Bjelland, I., Dahl, A. A., Haug, T. T., & Neckelmann, D. (2002). The validity of the Hospital Anxiety and depression scale: An updated literature review. Journal of
Psychosomatic Research, 52, 6977. http://dx.doi.org/10.1016/S0022-3999(01)00296-3
Bjelland, I., Krokstad, S., Mykletun, A., Dahl, A. A., Tell, G. S., & Tambs, K. (2008). Does a higher educational level protect against anxiety and depression? The HUNT
study. Social Science and Medicine, 66, 13341345. http://dx.doi.org/10.1016/j.socscimed.2007.12.019
Bramlett, M. D., & Mosher, W. D. (2002). Cohabitation, marriage, divorce and remarriage in the United States National Centre for Health Statistics. Vital and Health
Statistics, 23, 1103.
Brody, G. H. (2004). Siblings direct and indirect contributions to child development. Current Directions in Psychological Science, 13, 124126. http://dx.doi.org/
10.1111/j.0963-7214.2004.00289.x
Cassidy, S. B., & Driscoll, D. J. (2009). Prader[9_TD$IF]-Willi syndrome. European Journal of Human Genetics, 17(1), 313. http://dx.doi.org/10.1038/ejhg.2008.165
Cicirelli, V. G. (1995). Sibling relationships across the lifespan. New York, NY: Plenum Press.
Copeland, W. E., Adair, C. E., Smetanin, P., Stiff, D., Briante, C., Colman, I., et al. (2013). Diagnostic transitions from childhood to adolescence to early adulthood.
Journal of Child Psychology and Psychiatry, 54(7), 791799. http://dx.doi.org/10.1111/jcpp.12062
Cuskelly, M., & Gunn, P. (2006). Adjustment of children who have a sibling with Down syndrome: Perspectives of mothers, fathers and children. Journal of
Intellectual Disability Research, 50, 917925. http://dx.doi.org/10.1111/j.1365-2788.2006.00922.x
Davys, D., Mitchell, D., & Haigh, C. (2015a). Futures planning: Adult life perspectives. British Journal of Learning Disabilities, 43, 219226. http://dx.doi.org/10.1111/
bld.12099
Davys, D., Mitchell, D., & Haigh, C. (2015b). Adult siblings consider the future: Emergent themes. Journal of Applied Research in Intellectual Disabilities. http://
dx.doi.org/10.1111/jar.12172
Dimitropoulos, A., & Schultz, R. T. (2007). Autistic-like symptomatology in PraderWilli syndrome: A review of recent ndings. Current Psychiatry Reports, 9, 159
164. http://dx.doi.org/10.1007/s11920-007-0086-7
Falk, N. H., Norris, K., & Quinn, M. G. (2014). The factors predicting stress, anxiety and depression in the parents of children with autism. Journal of Autism and
Developmental Disorders, 44, 31853203. http://dx.doi.org/10.1007/s10803-014-2189-4
Fletcher, J., Hair, N., & Wolfe, B. (2012). Am I my brothers keeper? Sibling spillover effects: The case of developmental disabilities and externalising behaviour In
NBER working paper no.18279. Cambridge, MA: National Bureau of Economic Research.
Frazier, P. A., Tix, A. P., & Barron, K. E. (2004). Testing moderator and mediator effects in counselling psychology research. Journal of Counseling Psychology, 51, 115
134. http://dx.doi.org/10.1037/0022-0167.5101.115
Gale, C. R., Hatch, S. L., Batty, G. D., & Deary, I. J. (2009). Intelligence in childhood and risk of psychological distress in adulthood: The 1958 National Child
Development Survey 1970 British Cohort Study. Intelligence, 37, 592599. http://dx.doi.org/10.1016/j.intell.2008.09
Giallo, R., & Gavidia-Payne, S. (2006). Child, parent and family factors as predictors of adjustment for siblings with a disability. Journal of Intellectual Disability
Research, 50, 937948. http://dx.doi.org/10.1111/j.1365-2788.2006.00928.x
Gold, N. (1993). Depression and social adjustment in siblings of boys with autism. Journal of Autism and Developmental Disorders, 23, 147163. http://dx.doi.org/
10.1007/BF01066424

L.P. ONeill, L.E. Murray / Research in Developmental Disabilities 51-52 (2016) 116125

125

Gudmunson, C. G., Beutler, I. F., Israelsen, C. I., McCoy, J. K., & Hill, E. J. (2007). Linking nancial strain to marital instability: Examining the roles of emotional
distress and marital interaction. Journal of Family and Economic Issues, 28, 357376. http://dx.doi.org/10.1007/s10834-007-9074-7
Hall, C. (2006). A picture of the United Kingdom using the national statistics socio-economic classication. Retrieved from the Ofce for National Statistics website:
hhttp://www.statistics.gov.uk/article.asp?id=1663i.
Hannah, M. E., & Midlarsky, E. (1999). Competence and adjustment of siblings of children with mental retardation. American Journal on Mental Retardation, 104,
2237. http://dx.doi.org/10.1352/0895-8017(1999)
Hartley, S. L., Barker, E. T., Baker, J. K., Seltzer, M. M., & Greenberg, J. S. (2012). Marital satisfaction and life circumstances of grown children with autism across 7
years. Journal of Family Psychology, 26(5), 688697. http://dx.doi.org/10.1037/a0029354
Hastings, R. P. (2003). Behavioural adjustment of siblings of children with autism engaged in applied behaviour analysis early intervention programs: The
moderating role of social support. Journal of Autism and Developmental Disorders, 33, 141150 doi:0162-3257/03/0200-0099/0.
Hastings, R. P. (2007). Longitudinal relationships between sibling behavioural adjustment and behaviour problems of children with disabilities. Journal of Autism
and Developmental Disorders, 37, 14851492. http://dx.doi.org/10.1007/s10803-006-0230-y
Hodapp, R. M., & Urbano, R. C. (2007). Adult siblings of individuals with Down syndrome versus autism: Findings from a large-scale US survey. Journal of
Intellectual Disability Research, 51, 10101029. http://dx.doi.org/10.1111/j.1365-2788.2007.00994.x
Knight, S. J., Regan, R., Nicod, A., Horsley, S. W., Kearney, L., Homfrey, T., et al. (1999). Subtle chromosomal rearrangements in children with unexplained mental
retardation. Lancet, 354, 16761681.
Macks, J. R., & Reeve, R. E. (2007). The adjustment of non-disabled siblings of children with autism. Journal of Autism and Developmental Disorders, 37, 10601067.
http://dx.doi.org/10.1007/s10803-006-0249-0
McHale, S., & Gamble, W. C. (1989). Sibling relationships of children with disabled and nondisabled brothers and sisters. Developmental Psychology, 25, 421429.
McLean, C. P., Asnaani, A., Litz, B. T., & Hofmann, S. G. (2011). Gender differences in anxiety disorders: Prevalence, course of illness, comorbidity and burden of
illness. Journal of Psychiatric Research, 45, 10271035. http://dx.doi.org/10.1016/j.psychires.2011.03.006
Moore, M. L., Howard, V., & McLaughlin, T. F. (2002). Siblings of children with disabilities: A review and analysis. International Journal of Special Education, 17, 4963.
Orsmond, G. I., Kuo, H.-Y., & Seltzer, M. M. (2009). Siblings of individuals with an autistic spectrum disorder: Sibling relationships and wellbeing in adolescence
and adulthood. Autism, 13(1), 5980. http://dx.doi.org/10.1177/1362361308097119
Petalas, M. A., Hastings, R. P., Nash, S., Lloyd, T., & Dowey, A. (2009). Emotional and behavioural adjustment in siblings of children with intellectual disability with
and without autism. Autism, 13, 471483. http://dx.doi.org/10.1177/1362361309335721
Ross, P., & Cuskelly, M. (2006). Adjustment, sibling problems and coping strategies of brothers and sisters of children with autistic spectrum disorder. Journal of
Intellectual & Developmental Disability, 31, 7786. http://dx.doi.org/10.1080/13668250600710864
Rossiter, L., & Sharpe, D. (2001). The siblings of individuals with mental retardation: A quantitative review of the literature. Journal of Child and Family Studies, 10,
6584. http://dx.doi.org/10.1023/A: 1016629500708
Seltzer, M. M., Greenberg, J. S., Orsmond, G. I., & Lounds, J. (2005). Life course studies of individuals with developmental disabilities. Mental Retardation, 43, 354359.
Smith, T., & Perry, A. (2005). A sibling support group for brothers and sisters of children with autism. Journal of Developmental Disabilities, 11, 7788
doi:10.1.1.473.566.
Snaith, R. P. (2003). The hospital anxiety and depression scale. Health and Quality of Life Outcomes, 1, 29. http://dx.doi.org/10.1186/1477-7525-1-29
Stoneman, Z. (1989). Comparison groups in research on families with mentally retarded members: A methodological and conceptual review. American Journal of
Mental Retardation, 95, 537550.
Stoneman, Z., Brody, G. H., Davis, C. H., Crapps, J. M., & Malone, D. M. (1991). Ascribed role relations between children with mental retardation and their younger
siblings. American Journal on Mental Retardation, 95, 537550.
Strunk, J. A., Pickler, R., McCain, N. L., Ameringer, S., & Myers, B. J. (2014). Managing the health care needs of adolescents with autistic spectrum disorder: The
parents experience. Families, Systems and Health, 32(3), 328337. http://dx.doi.org/10.1037/a0037180
Turnbull, A. P., & Turnbull, H. R., III (1990). Families, professionals, and exceptionality: A special partnership (2nd ed.). New York, NY: Macmillan.
Veltman, M. W. M., Thompson, R. J., Roberts, S. E., Thomas, N. S., Whittington, J., & Boulton, P. F. (2004). PraderWilli syndrome: A study comparing deletion and
uniparental disomy cases with reference to autistic spectrum disorders. European Child & Adolescent Psychiatry, 13, 4250.
Voorpostel, M., & Blieszner, R. (2008). Intergenerational solidarity and support between adult siblings. Journal of Marriage and Family, 70, 157167. http://
dx.doi.org/10.1111/j.1741-3737.2007
Wolfe, B., Song, J., Greenberg, J. S., & Mailick, M. R. (2014). Ripple effects of developmental disabilities and mental illness on nondisabled adult siblings. Social
Science and Medicine, 108, 19. http://dx.doi.org/10.1016/j.socscimed.2014.01.021
Zigmond, A. S., & Snaith, R. P. (1983). The hospital anxiety and depression scale. Acta Psychiatrica Scandinavica, 67, 361437.