Parental Perspectives on the Causes of an Autism Spectrum Disorder in their Children
- First Online:
- Cite this article as:
- Mercer, L., Creighton, S., Holden, J.J. et al. J Genet Counsel (2006) 15: 41. doi:10.1007/s10897-005-9002-7
- 1.1k Views
Autism Spectrum Disorders (ASDs) are complex neurodevelopmental disorders with many biological causes, including genetic, syndromic and environmental. Such etiologic heterogeneity impacts considerably upon parents’ needs for understanding their child's diagnosis. A descriptive survey was designed to investigate parental views on the cause(s) of ASD in their child. Among the 41 parents who replied to the questionnaire, genetic influences (90.2%), perinatal factors (68.3%), diet (51.2%), prenatal factors (43.9%) and vaccines (40.0%) were considered to be the most significant contributory factors. Parents reported inaccurately high recurrence risks, misperceptions of the contribution of various putative factors, feelings of guilt and blame regarding their child's diagnosis, as well as a lack of advocacy for genetic counseling by non-geneticist professionals. This study offers clinicians and researchers further insight into what parents believe contributed to their child's diagnosis of ASD and will help facilitate genetic counseling for these families.
KEY WORDS:autism spectrum disorderASDautismpervasive developmental disorderPDDgenetic counselingparental perspectivesetiologygeneticsrecurrence riskfamily history.
The autism spectrum disorders (ASDs) represent a group of pervasive developmental disorders (PDDs), characterized by impairments in communication and social interactions, as well as restricted, repetitive and stereotyped patterns of behavior. Autistic disorder (commonly referred to as autism), pervasive developmental disorder-not otherwise specified (PDD-NOS), Asperger disorder, Rett disorder and childhood disintegrative disorder are found under the DSM-IV classification for PDDs (American Psychiatric Association, 1994). The first three of these diagnoses are collectively referred to as ASDs, which as a group affect at least 1/250 (Gillberg and Wing, 1999) to 1/500 (Fombonne, 1999) individuals with a male:female ratio of 4:1 (Bryson et al., 1988; Smalley, 1991).
The precise etiological factors which predispose a child to ASDs are not fully understood although extensive investigations using the traditional methods of behavioral genetic analysis (family, twin and adoption studies; quantitative trait loci (QTL); biometric model fitting) have clearly substantiated a predominant role for genetic influences in autism over the years (Lewis, 2002; Muhle et al., 2004). Individuals with an ASD often have a positive family history of the same (Piven et al., 1990). In addition, positive family histories are often noted for autoimmune (Comi et al., 1999; Sweeten et al., 2003) and/or psychiatric disorders (Piven et al., 1990; Smalley et al., 1995; Bolton et al., 1998; Piven and Palmer, 1999), implicating a potential common biologic link with the etiology of ASDs. Theories purporting environmental etiologies have been raised including prenatal and perinatal factors (Comi et al., 1999; Juul-Dam et al., 2001; Zwaigenbaum et al., 2002), diet (Reichelt et al., 1990; Whiteley et al., 1999; Knivsberg et al., 2002), and childhood vaccinations (Wakefield et al., 1998). However, scientific evidence has either been inconclusive (Wing and Potter, 2002) or in the case of the MMR vaccine, has excluded them as likely etiologies (Peltola et al., 1998; Kaye et al., 2001; DeWild et al., 2001; Hviid et al., 2003; Honda et al., 2005).
To date there have been two papers (Elder, 1994; Gray, 1995) and two abstract publications (Rosen et al., 2000; Rosen-Sheidley et al., 2002) that have examined the beliefs of parents regarding the possible causes of ASD in their children. These beliefs may adversely impact parent–child relationships and acceptance of their child's condition (Elder, 1994; Gray, 1995). In the first descriptive paper on this subject, Elder (1994) observed that genetic transmission, prenatal factors or exposures affecting the fetus, and postnatal factors in infancy (such as the occurrence of high fevers) were commonly believed by parents to underlie the etiology of autism. Gray's (1995) qualitative analysis of the views of parents of autistic children found that they attributed their child's ASD to a variety of potential factors including birth trauma, heredity, illnesses, perinatal damage, and vaccinations, with most parents offering more than one explanation for the presence of autism in their child/children. In a web-based survey, which evolved as a more extensive version of an earlier study by Rosen et al. (2000), Rosen-Sheidley et al. (2002) reported that genetics (85%), vaccinations (31%) and other environmental factors (33%) were implicated as possible causes of autism by relatives (parents as well as other relatives) of an affected individual.
The aims of the current study were to assess current parental beliefs regarding the etiology of ASDs in their offspring, as well as general perceptions regarding the role of genetics and estimated recurrence risks for the ASDs. Results of this pilot study will inform the development of a subsequent survey to be administered to a larger group of parents. It is anticipated that information from the study may ultimately help health care providers and researchers identify gaps in parental knowledge as well as misconceptions surrounding the causes, genetic influences and recurrence risk for the ASDs. In addition, we aim to identify areas of research felt to be priorities for parents of children with ASDs.
The study surveyed a group of parents with at least one child diagnosed with an ASD. Parents were recruited between October 2003 and February 2004 through various organizations in Canada and the United States who advertised the study on their websites and included the URL link to the online questionnaire. These organizations included the Autism Society of New Brunswick, the Autism Calgary Association, Autism Society Ontario (Halton Chapter), Autism Society of BC, and Autism NWT (Northwest Territories). Families were also invited to participate online through an announcement in a newsletter to families participating in research being carried out by ASD-CARC (Autism Spectrum Disorders—Canadian-American Research Consortium; www.autismresearch.ca). The online results were held within a database accessible to the primary author. The study was approved by the Behavioral Research Ethics Board of the University of British Columbia.
Parental Perspectives on ASD Questionnaire
The pilot questionnaire was developed on the basis of literature review of current etiological theories, published information on beliefs of parents regarding their child's ASD, and input from researchers from the ASD-CARC (including a Parent Advisory Group). The questionnaire was reviewed with the Parent Advisory Group to determine if the questions were understandable, non-ambiguous, and that the content adequately reflected the issues felt to be important (in their experience) for parents of children with an ASD. Questions in the survey consisted of various format types such as fill in the blank, dichotomous, Likert response, cumulative/Guttman, nominal, and open-ended. The questionnaire was divided into the following sections: (i) Demographics, (ii) Parental beliefs on the etiology of ASDs: genetics and family history, prenatal and perinatal factors, vaccinations, dietary factors (iii) Perceived recurrence risks and (iv) Research.. The etiologies represented in the questionnaire included current theories, despite some being inconclusive or unlikely. The authors included a statement noting that their inclusion did “not imply that they play a role in ASD.” Space for comments was provided at the end of some sections.
Quantitative information was entered into a database. For nominal data (i.e. presence or absence of a particular event), a chi-square analysis was performed. SPSS was used to analyze results and generate p values. A chi-square test could not be used to establish relationships amongst questions, as the sample size was too small.
Many respondents provided brief responses to open-ended questions and comments. The responses were analyzed by content analysis, a method of classifying the numerous words of the text into a few content categories (Kelly and Sime, 1990) which are then quantified. Content analysis, because it uses the data in its original form, is context sensitive, so that the data are not separated from the symbolic meaning they have for the subject. This process was undertaken separately by two members of the team to ensure non-bias and to examine inter-rater reliability.
Forty-one surveys were completed. Of the 41 children about whom the responses were received, 33 (80.0%) were male and 8 (20.0%) were female, in keeping with an expected 4:1 male to female sex ratio. The age of the children ranged from 1 to 25 years (only three were adults: two aged 18 years and one aged 25 years), with a median age of 8.9 years. The breakdown of diagnoses provided by the families was as follows: Autistic disorder: 53.7%; Asperger disorder: 14.6%; “High functioning autism”: 9.8%; PDD-NOS: 9.8% and an otherwise undefined ASD in 12.2%. None of the children had been diagnosed with a genetic condition; however none had been assessed by a geneticist. Four were born with congenital birth defects (two with congenital heart disease, one with arthrogryposis and one with pyloric stenosis). In all but two families, no specific cause for autism had been identified by a health care professional. In the two families where a cause was reported to be known, one reported that their child's physician felt he was “vaccine damaged” and the other reported that complications of a twin pregnancy and extreme premature delivery were most likely causal factors.
Thirty-eight of the questionnaires were completed by mothers, and an unspecified biological parent completed the remainder. The primary language spoken in all homes was English, with one family also speaking French. Thirty-seven (90.2%) of the respondents were Caucasian, and the remainder were Caucasian/African American, Caucasian/Arab and Filipino. Participants included those from Ontario (19), Nova Scotia (11), Alberta (3), USA (3), British Columbia (2), New Brunswick (2) and the Northwest Territories (1).
Not all participants answered all the questions; therefore the total number of respondents for any given question varies. For questions pertaining to perinatal factors, postnatal factors and vaccinations, parents were permitted to provide multiple responses.
Parental Beliefs Regarding the Etiology of ASD
Multiple etiologies were believed to be involved in ASD, as reported by the majority of parents (37/41, 90.2%) in the study.
Genetics and Genetic Counseling Issues
Genetic Counseling Issues
Reproductive risks (five parent reports): Recurrence risks appeared to have altered reproductive decision making in these families, not just for the parents but for other family members as well. For example: “…everyone on father's side of the family is terrified to have children”; “My youngest sister (who has some ASD traits) is concerned about her risk and may not have children”; “The child's father had a vasectomy.”
Transference (five parent reports): Following the diagnosis of an ASD, there was transference of ASD signs and symptoms to other family members. For example: “After my son's diagnosis, I realized my father's condition”; “My family has started to question the REAL reasons for my father's drug/alcohol addition and whether it can shed light on my son”; “I believe [my husband] also sees some of the traits in himself.”
Guilt (four parent reports): Another common feeling among family members was guilt that they may have contributed genes causing their child's ASD. For example: “… everyone is upset that they could have contributed to my son's problem,” “It makes my husband very uncomfortable because he thinks he might be blamed.” (This comment was not felt to represent “blame,” which we consider an externally directed factor).
Blame (one parent report): For one family, it was felt that each side of the family was blaming the other.
Prenatal and Perinatal Factors
Eighteen parents (43.9%) felt that prenatal maternal risk factors contributed to their child's ASD, including advanced maternal age (6), smoking (4), influenza (4), prescription medications (3), maternal vaccination (3), uterine bleeding (2), toxemia (2), street drugs (1), upper respiratory infection (1) and genital herpes (1).
Twenty-eight parents (68.3%) indicated that specific perinatal events contributed to their child's ASD. These included severe trauma or injury during birth (8), fetal distress or anoxia (8), premature labor or premature rupture of membrane (6), low birth weight/premature birth (6), induced labor (5), hyperbilirubinemia (5), requirement for oxygen after birth (3), emergency cesarean section (3), blood group incompatibility (3), resuscitation/ventilation (2), respiratory distress syndrome (2), anemia (2) and infection after birth (1).
All but 1 of the 41 children with ASD had received vaccinations. Sixteen (40.0%) of the immunized children's parents believed that vaccines were responsible or contributed to their child's ASD versus 21 (52.5%) who did not (two did not answer the question and two were unsure). Parents were then asked to indicate which vaccinations they thought could have played a role in their child's ASD (they could choose more than one). The vaccines included measles/mumps/rubella (MMR), Polio (IPV), hemophilus influenza B (HiB), Hepatitis B, diphtheria/pertussis/tetanus (DTaP/DTP), diphtheria/tetanus (DT) and meningitis. One third of parents (13/40, 32.5%) felt the MMR vaccine played a role although, interestingly, 10 of these parents felt that other vaccines were also responsible. Each of the other vaccines listed in the questionnaire, except the chicken pox vaccine, was implicated at least once by parents. With the exception of four families (three citing MMR, one citing DTaP/DTP), another vaccine was also implicated. One parent commented that “vaccinations in general could stress an immune system that has abnormalities.”
Twenty-one parents (51.2%) believed that diet was a contributory factor to their child's ASD. Given three options regarding food containing casein and gluten, 21/41 (51.2%) answered “I believe intolerance to these foods is a contributory factor in ASD”; 7/41 (17.1%) answered “I believe my child is intolerant to these foods because of ASD”; and the remaining 13/41 (31.7%) chose, “I do not believe there is a connection between these foods and ASD.” Twenty-seven parents (65.9%) said they had altered their child's diet and of those, 10 (37.0%) subjectively found a difference in behavior. One parent noted that her child experienced panic attacks on the diet and another indicated she believed the diet was responsible for an eating disorder.
Perceived Recurrence Risks
Perception of Recurrence Risk for ASDs Compared with Degree of Relationship of Affected Relatives
Perceived recurrence risk (%)
Families with first-degree relativea (N)
Families with other affected relativeb (N)
Families with no affected relatives (N)
When asked if a health care professional caring for their child had ever raised the prospect of a genetic basis for ASD, 17/38 (44.7%) said it had been addressed. Developmental pediatricians and psychologists were the professionals found most likely to raise the possibility of a genetic basis for an ASD with the parents. However, without exception the families indicated that they had not been referred to either a medical geneticist or a genetic counselor for consultation after their child had been diagnosed.
Perception of Recurrence Risk for ASDs Compared with the Number of Affected Relatives in a Family
Perceived recurrence risk (%)
No affected relatives (N)
One affected relative (N)
Two affected relatives (N)
Three affected relatives
More than three affected relatives
The need to understand and make sense of a disability or serious illness is critical as a coping mechanism for the individual, their family, and others in the illness experience. The reasons for wanting a diagnosis have been previously reported in published surveys of parents whose children were diagnosed with various medical conditions (Lippman-Hand and Fraser, 1979; Sorenson et al., 1981; Strauss and Munton, 1985; Elder, 1994; Gray, 1995; Rosenthal et al., 2001). The findings have indicated that parents want to know about 1) etiology, in order to establish responsibility and to prevent recurrence; 2) natural history (prognosis); 3) treatment options; 4) access to special services; 5) contact with other parents of affected children; 6) how to seek advice and information and means for discussing the condition with others; and 7) appropriate parenting. Obtaining the diagnosis of ASD in a child satisfies some of these needs. However, what remains notably unanswered is the issue of etiology.
Parents in our study, as well as those of Gray (1995) and Elder (1994), felt that perinatal and prenatal factors (including maternal illness, substance abuse and maternal vaccination) were implicated in ASD causation. Several studies have suggested that there is a higher rate of adverse prenatal and postnatal events in children later diagnosed with autism than in the general population (Comi et al., 1999; Juul-Dam et al., 2001; Zwaigenbaum et al., 2002). Although minor obstetric complications may be more common in individuals with ASD, there is little evidence to suggest that such perinatal complications play an etiological role in the development of ASD. Recent studies suggest that such complications are more likely to be secondary consequences related to the determinants underlying ASD causation (Bolton et al., 1997; Zwaigenbaum et al., 2002). In the present study, the most common prenatal factor of concern expressed by parents was advanced maternal age. Whether this reflects the parent's knowledge of findings suggesting that the risk of autism in a child may be increased with increasing maternal age (Tsai and Stewart, 1983; Bolton et al., 1997; Croen et al., 2002), or that obstetrical complications are more common in women who have an advanced maternal age (Abu-Heija et al., 2000; Weerasekera and Udugama, 2003), is unclear.
The excess opioid theory suggests that ASD is caused by the incomplete breakdown and excessive absorption of casein and gluten from foods (Whiteley et al., 1999). While 37% of parents surveyed in the present study who had altered their child's diet felt that a casein- and gluten-free diet led to improvement in their child, other studies have indicated reports of success rates by parents in up to 67.0% of cases (Whiteley et al., 1999). At this time, the data for a dietary etiology in ASDs is inconclusive.
In stark contrast to the lack of evidence of causality in ASDs for perinatal factors, diet, prenatal factors and vaccinations, the evidence for genetics playing a significant role in the etiology of ASDs is substantial, although the precise genetic mechanisms have yet to be delineated (Lewis, 2002; Muhle et al., 2004). In our study, just under half of parents (41.5%) indicated a genetic basis had been raised by a health care professional, possibly reflecting a lack of awareness of the genetic basis of ASDs. Clearly, and despite it not having been explained to them, most of the families in our study (90.2%) felt genetic factors were important, and nearly two-thirds felt this contribution was high. Several associated a positive family history of ASD, and in some cases a family history of psychiatric disorders or autoimmune disorders, with their child's ASD. Interestingly, researchers have previously identified a positive family history of the following disorders as important in the etiology of ASDs: (i) ASD: Piven et al., 1990; Gillberg et al., 1992; Piven et al., 1997; Bolton et al., 1998; (ii) autoimmune disorders: Comi et al., 1999; Sweeten et al., 2003; and (iii) psychiatric disorders: Piven et al., 1990; Smalley et al., 1995; Bolton et al., 1998; Piven and Palmer, 1999). In addition, the recurrence risk for ASDs was likely overestimated by many, and grossly overestimated by some of the participants in this study, irrespective of the number or degree of relationship ot other affected family members. While a recurrence risk of 2–8% for autistic disorder is generally accepted (Muhle et al., 2004), Folstein et al. (1999) estimate the risk for having one or more of the features of autism (i.e. the broader autism phenotype) in a subsequent child to be as high as 30%. Syndromic causes of a Mendelian nature and/or the birth of a second child would alter these risks. Furthermore, in at least 10% of cases, specific etiologies such as chromosomal abnormalities (dup15q11–13), metabolic (PKU), syndromic (Fragile X syndrome), genetic (tuberous sclerosis) or other non-genetic disorders (valproate embryopathy) may be identified (Muhle et al., 2004), and are critical to distinguish from idiopathic autism through careful dysmorphologic evaluation and investigation, as the recurrence risks are quite different. It is thus surprising that none of these families were referred to a medical geneticist or genetic counselor to fully assess the family, rule-out syndromic causes for ASDs, and discuss recurrence risks that are specifically evaluated for each family. While most respondents supported research on the genetics of ASDs, the primary motivating factors for the majority was improved diagnosis. It is possible that families may feel that medical genetics services offer little advantage for them until there are well-defined genetic markers and/or tools that can be applied toward improved diagnosis and management.
Some parents indicated feelings of guilt, a finding also noted by both Gray (1995) and Elder (1994). This was especially evident in mothers in the latter two studies. Elder (1994) noted that there was often an element of blame between sides of the family, similar to what was expressed by one parent in the present study. Blame and guilt among family members are common findings in families in which a genetic disorder is present (Weil, 2000) and can negatively impact individuals and/or the relationship among family members, including that between parent and child. If such issues are raised by a family, appropriate intervention and education would assist in mitigating any negative consequences.
Limitations due to ascertainment bias in the current study should be considered, as the surveys were primarily made available through support and parent advisory groups. Parents who are members of a support group often have access to academic research and may have links to researchers in the community and, therefore, are more likely to have access to accurate and reliable data.
In summary, this pilot study offers health-care professionals and researchers further insight into specific issues that parents believe contributed to their child's ASD. Parental perspectives are invaluable for informing and developing new research initiatives, and providing greater awareness of parent sensitivities, perceptions and means for improving practice essential for ongoing care and genetic counseling. Studies such as the present one also provide a critical perspective on the extensive breadth and depth of factors that parents believe led to their child's ASD. A medical genetics assessment by a genetic counselor and/or a clinical geneticist provides an accurate means to help families understand genetics, etiology, and inheritance patterns in a supportive and informative environment. On the basis of the results of this study, the involvement of medical genetics as a component in the assessment of an individual with an ASD is clearly needed to assist and guide parents and family members toward resolving misinformation and misunderstandings. However, the sheer magnitude of individuals believed to be affected with an ASD (at least 1/250 (Gillberg and Wing, 1999) to 1/500 (Fombonne, 1999) individuals) would make genetic assessment and counseling for each and every family a daunting task. The means to achieve this would be an important focus for future research and development of medical genetics services.
The authors would like to thank all the participants of the study. We wish to express special thanks to Anita Acheson and other members of the ASD-CARC Parent Advisory Group, and to participating ASD-CARC regional teams who assisted in the creation of the questionnaire and aided in its distribution. This work was supported by funds from the Canadian Institutes for Health Research (CIHR) through an Interdisciplinary Health Research Team grant (RT-43820) to the Autism Spectrum Disorders Canadian-American Research Consortium (ASD-CARC: www.autismresearch.ca) (JJAH, principal investigator). MESL is supported by a CIHR Institute of Genetics Clinician Investigator Award (2003–2005) and Michael Smith Foundation for Health Research Career Investigator (Scholar) Award.