VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION

Psychological Well-Being and Coping in Mothers of Youths With Autism, Down Syndrome, or Fragile X Syndrome

Leonard Abbeduto, Marsha Mailick Seltzer, and Paul Shattuck University of Wisconsin-Madison Marty Wyngaarden Krauss Brandeis University Gael Orsmond Boston University Melissa M. Murphy University of Wisconsin-Madison

Abstract The psychological well-being of mothers raising a child with a developmental disability varies with the nature of the disability. Most research, however, has been focused on Down syndrome and autism. We added mothers whose adolescent or young adult son or daughter has fragile X syndrome. The sample was comprised of mothers of a child with fragile X syndrome (n ϭ 22), Down syndrome (n ϭ 39), or autism (n ϭ 174). Mothers of individuals with fragile X syndrome displayed lower levels of well-being than those of individuals with Down syndrome, but higher levels than mothers of individuals with autism, although group differences varied somewhat across different dimensions of well-being. The most consistent predictor of maternal outcomes was the adolescent or young adult’s behavioral symptoms.

Parents raising a child with a developmental group has seldom been studied from the perspec- disability face challenges that are not shared by tive of parental psychological well-being, and its parents of typically developing children (Stone- inclusion yields additional insights about the im- man, 1997). There is, however, substantial evi- pact of developmental disabilities on parents. We dence that the challenges parents face and the examined group differences in parental psycho- ways in which they deal with these challenges vary logical well-being and coping, their association with the nature of the child’s disability (Dunst, with characteristics of the family and characteris- Trivette, & Cross, 1986; Gallagher, Beckman, & tics and behavior of the son or daughter, and the Cross, 1983; Walker, Van Slyke, & Newbrough, potential buffering effects of coping. In doing so, 1992). This evidence comes primarily from studies we focused on mothers because, despite changing of two groups of parents: those having a child gender roles, mothers still tend to have primary with Down syndrome and those having a child responsibility for child care and are, thus, most with autism (Stoneman, 1997). In the present subject to the challenges associated with their study, we added a third group: parents whose son child’s disability (Gray & Holden, 1992). In ad- or daughter has fragile X syndrome. This latter dition, we focused on mothers whose son or

᭧ American Association on Mental Retardation 237 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. daughter was an adolescent or young adult (ages asek, 1986), have more extensive and more satis- 10 to 23) because this is a time of transition and fying networks of support (Hauser-Cram, War- new challenges for families as well as being a pe- field, Shonkoff, & Krauss, 2001; Shonkoff, Hau- riod in which behavioral differences between per- ser-Cram, Krauss, & Upshur, 1992), and perceive sons with Down syndrome, autism, and fragile X their children to have less difficult temperaments syndrome may become more pronounced. (Kasari & Sigman, 1997). Parents of children with It is important to make explicit why parental Down syndrome also report less pessimism about psychological well-being would be expected to dif- their children’s future than do parents of children fer as a function of the son or daughter’s diag- with either Williams or Smith-Magenis syndrome nosis. Certainly, there is nothing inherent in any (Fidler, Hodapp, & Dykens, 2000), they report particular diagnosis that would in and of itself fewer family problems than do parents of children elicit a particular set of parental reactions. Instead, with Smith-Magenis syndrome (Fidler et al., it is likely that the physical and behavioral chal- 2000), and they are less likely to interpret child lenges posed by the child will be critical, and to noncompliance negatively compared to mothers the extent that those challenges vary across diag- of children who, as a group, are heterogeneous noses, differences in parental reactions are possi- with respect to etiology (Ly & Hodapp, 2002). Al- ble. In the case of syndromes that have a genetic though the vast majority of studies have focused origin, we might also expect different parental re- on young children, there is evidence that the ‘‘ad- actions if the parental genes that are the origin of vantage’’ of having a son or daughter with Down the child’s condition also affect the parent’s own syndrome continues throughout life (Seltzer, characteristics, behaviors, and vulnerability to the Krauss, & Tsunematsu, 1993). However, not all challenges of parenting in different ways across researchers have found Down syndrome to be as- syndromes. Finally, we could expect different pa- sociated with higher levels of psychological well- rental reactions across diagnoses to the extent that being in mothers compared to other forms of those diagnoses are correlated with variations in mental retardation (Gath & Gumley, 1986; Han- social acceptance of, and support for, children son & Hanline, 1990; Ryde-Brandt, 1991), sug- with special needs and their families. These points gesting that the extent to which Down syndrome have been demonstrated empirically by Cahill is perceived as less stressful will depend on the and Glidden (1996), who found that differences nature of the comparison group. in parental functioning according to child diag- In contrast to mothers of children with Down noses were minimized when the groups were syndrome, mothers of individuals with autism re- matched on various child characteristics. The im- port greater stress, lower levels of social support, plication is that demonstrating that parents of and less positive views of their child than do children with different diagnostic conditions are mothers of individuals with mental retardation themselves different on measures of parental psy- who, as a group, are heterogeneous in terms of chological well-being is merely a first step, with etiology (Donovan, 1988). This result emerges the source of those differences still to be deter- across variations in geographical region and cul- mined. In the present study, we were interested in ture, child age, and child IQ (Koegel et al., 1992). first determining whether there are differences in Moreover, direct comparisons between mothers of maternal psychological well-being across the three children with Down syndrome and mothers of groups of mothers and in beginning to under- children with autism are quite consistent in sug- stand the sources of those differences among gesting that mothers of children with Down syn- mothers. drome experience lower levels of stress (Holroyd Previous research on the impact of a child’s & McArthur, 1976; Kasari & Sigman, 1997) and disability on parental psychological well-being has have more positive views of their children’s char- demonstrated that mothers of children with acteristics (Kasari & Sigman, 1997). Down syndrome often fare better than do moth- These differences in psychological well-being ers of children with other forms of mental retar- between mothers of children with autism and dation. Compared to mothers of children who, as those with Down syndrome can be traced, in part, a group, are heterogeneous with respect to etiol- to variations in the uncertainty attending the two ogy, mothers of children with Down syndrome diagnoses. The long history of research on Down report lower levels of stress (Kasari & Sigman, syndrome and the certainty surrounding its diag- 1997; Marcovitch, Goldberg, MacGregor, & Lojk- nosis and cause provide parents with a clear and

238 ᭧ American Association on Mental Retardation VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. coherent explanation of their child’s condition higher rates of depressive symptomatology in par- and with a reasonable expectation of what the fu- ents (Seltzer, Greenberg, & Krauss, 1995). Even ture holds, hence, fewer worries about the future the physical appearance that is characteristic of (Beavers, Hampson, Hulgus, & Beavers, 1986; Fi- individuals with Down syndrome, which is per- dler, Hodapp, & Dykens, 2002). In addition, the ceived as highly child-like, elicits positive emo- fact that the diagnosis of Down syndrome is made tional responses from mothers and other adults so soon after birth allows parents to begin the (Fidler & Hodapp, 1999). process of adaptation to their child’s condition The challenges faced by mothers of individ- when their children are only a few hours old. In uals with autism also extend beyond those arising contrast, the diagnosis of autism is often substan- from the degree of certainty attending the child’s tially delayed relative to the onset of parental rec- condition or his or her particular characteristics ognition of developmental problems (Lord & Ris- and behaviors. In particular, there is evidence of si, 2000). Moreover, this diagnosis is made on an increased risk of a host of problems in other strictly behavioral grounds rather than through a family members, including siblings and parents, definitive genetic test as in the case of Down syn- with this risk being reflective of the broader au- drome, which may leave many parents wondering tism phenotype (Piven, 2001). Thus, there is a whether the diagnosis of autism is ‘‘correct’’ (Lord higher recurrence risk for siblings of children with & Rissi, 2000). In addition, the life course mani- autism compared to the general population (Jorde festation of autism has only recently been studied, et al., 1991; Piven et al., 1990), and the rates of resulting in parents greater uncertainty about the cognitive, linguistic, and social impairments and future compared to parents of individuals with psychiatric disturbances are greater in siblings of Down syndrome (Seltzer, Krauss, Orsmond, & affected individuals than in siblings of individuals Vestal, 2000). with Down syndrome (August, Stewart, & Tsai, The differences in psychological well-being 1981; Bolton et al., 1994; Piven et al., 1990). Fur- between mothers of children with autism and ther, parents of children with autism display ele- mothers of children with Down syndrome may vated rates (compared to the general population) also be due to variations in the children’s char- of social impairments and personality character- acteristics and behavior, with autism being asso- istics known to lead to social impairments (Piven, ciated with fewer positive and more negative be- Palmer, Jacobi, Childress, & Arndt, 1997; Piven, havioral propensities. Thus, the sociability that Palmer, Landa et al., 1997), psychiatric distur- characterizes Down syndrome (Kasari & Baumin- bances (DeLong, 1994; Delong & Nohria, 1994; ger, 1998) is, by definition, not characteristic of Lainhart & Folstein, 1994), learning disabilities individuals with autism. Individuals with autism (Bolton et al., 1994; Folstein & Rutter, 1977; Piv- also display more maladaptive behaviors than do en et al., 1990), and deficits in communication persons with Down syndrome (Pueschel, 1996; (Landa, Folstein, & Isaacs, 1991). In short, moth- Seltzer et al., 2000), and these behaviors may in- ers of persons with autism may not only face terfere with the parent–child relationship. In fact, greater challenges from their child’s condition, the rate of child maladaptive behavior is a predic- they may also be less equipped to deal with those tor of parental psychological well-being (Fidler et challenges than are mothers of individuals with al., 2000; Friedrich, Wilturner, & Cohen, 1985; Down syndrome (Seltzer et al., 2000), who, as a Hodapp, Dykens, & Masino, 1997; Hodapp, Fi- group, are not at increased risk of psychiatric, lan- dler, & Smith, 1998). Moreover, in contrast to guage, or cognitive symptoms. Mothers of indi- Down syndrome, individuals with autism display viduals with autism may also face the challenge features such as impulsivity, hyperactivity, irrita- of raising multiple children with special needs. bility, and aggressiveness (Diagnostic and Statis- Fragile X syndrome is second only to Down tical Manual–IV DSM-IV, American Psychiatric syndrome as a genetic cause of mental retardation, Association, 1994), which are characterized by un- affecting 1 in 4,000 males and 1 in 8,000 females predictability and volatility. The predictability of (Dykens, Hodapp, & Finucane, 2000). The impact child maladaptive behavior has been found to be of raising a child with fragile X syndrome on par- an important determinant of stress and coping in ents has been investigated in only a single pub- parents, with less predictability associated with lished study (Franke et al., 1996). Nevertheless, higher levels of stress, greater reliance on emotion- there are reasons to believe that mothers of chil- focused rather than problem-focused coping, and dren with fragile X syndrome, like mothers of per-

᭧ American Association on Mental Retardation 239 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. sons with autism, may face greater parenting chal- 1999), which may have phenotypic effects and, lenges than do those of individuals with Down thus, affect their ability to deal with parenting syndrome. Similar to parents of children with au- challenges. Many women who carry the full mu- tism, parents often do not receive the diagnosis of tation have mental retardation, and those who fragile X syndrome until long after developmental have normal-range IQs often have impairments in delays have been of concern (Carmichael, Pem- executive functions, discourse comprehension, brey, Turner, & Barnicoat, 1999), which is stress- and math ability (Mazzocco, 2000; Simon, Keen- ful for parents (Roy, Johnsen, Breese, & Hager- an, Pennington, Taylor, & Hagerman, 2001). man, 1995). Bailey, Skinner, Hatton, and Roberts Women with the full mutation also display in- (2000) found that parents did not receive a diag- creased rates of social anxiety, schizotypal symp- nosis of fragile X syndrome until their child was toms (e.g., illogical reasoning, odd perceptual ex- near the age of 3 years, more than a year after the periences), and autistic-like behaviors compared to child had been diagnosed with a developmental IQ matches (Feinstein & Reiss, 2001; Mazzocco, delay. Receipt of the fragile X syndrome diagnosis Kates, Baumgardner, Freund, & Reiss, 1997; So- is even later in the cohort of families whose chil- besky, Porter, Pennington, & Hagerman, 1995). dren are adolescents or young adults today (Poehl- The finding of elevated rates of depression in mann, Clements, Farsad, Abbeduto, & Murphy, women with the full mutation, however, are in- 2003). Parents of individuals with fragile X syn- consistent (Sobesky, Pennington, Porter, Hull, & drome may have little help in resolving their un- Hagerman, 1994). Although there has been lim- certainty because knowledgeable and effective ited research on women who carry the premuta- professional supports are currently less extensive tion (Keyser & Mazzocco, 2002), there is some and effective for families affected by the syn- evidence of elevated rates of affective disorders drome than for other developmental disorders, in- (Franke et al., 1996; Hagerman & Hagerman, cluding Down syndrome and autism (York, von 2002), but no consistent evidence of cognitive ef- Fraunhofer, & Sedgwick, 1999). fects has been obtained to date (Hagerman & Hag- Psychological well-being in mothers of indi- erman, 2000; Simon et al., 2001; Sobesky et al., viduals with fragile X syndrome might also be 1995). In addition, because the syndrome is in- threatened because sons and daughters with the herited, many mothers are raising more than one syndrome, like those with autism, display relative- affected child (Hagerman, 1999), which may lead ly high rates of maladaptive behaviors (Hagerman, to greater challenges compared to mothers caring 1999; Keysor & Mazzocco, 2002), such as hyper- for only one child with special needs. activity and social anxiety, which often isolates Taken together, the foregoing observations them from others. In fact, a substantial number suggest that the challenges faced by mothers of of individuals with fragile X syndrome meet di- children with fragile X syndrome should be more agnostic criteria for autism, and those who do not similar to those of mothers raising children with often display autistic-like behaviors (Dykens & autism than those of mothers raising children Volkmar, 1997; Feinstein & Reiss, 2001). Such be- with Down syndrome. In fact, Franke et al. (1996) haviors are seen in both males and females, al- found no difference in the rates of clinically di- though on average they are more frequent and agnosed affective disorders between mothers of severe in males (Mazzocco, 2000). Moreover, par- children with fragile X syndrome and mothers of ents anecdotally report that their sons and daugh- children with autism, although both groups were ters with fragile X syndrome go through alternat- characterized by higher rates than seen in the gen- ing periods of ‘‘calm’’ and ‘‘storm,’’ the latter be- eral population. It is possible, however, that dif- ing characterized by anxiety-related behaviors, ferences in psychological well-being between these tantrums, and aggressive outbursts. As noted pre- two groups of mothers might emerge on measures viously for autism, such unpredictability may be of problems that are at subclinical levels but are, a source of substantial stress for parents. nonetheless, serious enough to interfere with in- As is the case in autism, the challenges faced dividual and family functioning. Moreover, Fran- by mothers of individuals with fragile X syndrome ke et al. did not examine maternal psychological extend beyond the affected child. In particular, well-being in relation to child behaviors and char- the mothers of children with fragile X syndrome acteristics. This makes it impossible to distinguish are themselves carriers of either the full mutation between those aspects of maternal functioning re- or premutation of the fragile X gene (Hagerman, sulting from the challenges of parenting an af-

240 ᭧ American Association on Mental Retardation VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. fected child and those resulting from maternal ge- ferences in the nature of the son or daughter’s netic status. More generally, although there have disability. In particular, we examined maternal been numerous studies of socioemotional func- feelings of pessimism about their son or daugh- tioning in women with either the full mutation or ter’s future, views on the quality of their relation- the premutation, little is known about the impact ship with their son or daughter, and self-reported on psychological well-being of parenting a son or depressive symptomatology. We hypothesized daughter with fragile X syndrome. that of the three groups of mothers, those par- In examining differences in psychological enting an adolescent or young adult with Down well-being between mothers of children with dif- syndrome would report the least pessimism about ferent types of developmental disabilities, it is im- their son or daughter’s future, the greatest feelings portant to recognize that the challenges facing of closeness in their relationship with the son or mothers may change at different stages of the fam- daughter, and the lowest rate of depressive symp- ily life cycle. The period of adolescence and toms. We also predicted that mothers of adoles- young adulthood may be especially challenging cents and young adults with Down syndrome for parents because of the impending transition would report the greatest use of problem-focused from school to work and a heightened concern coping and the lowest use of emotion-focused about their son or daughter’s future (Lueckling & coping, the former generally seen as more effec- Fabian, 1997). In fact, there is evidence of a de- tive in buffering mothers from parenting chal- crease in parental psychological well-being with lenges than the latter (Seltzer et al., 1995). We also increased age of the child, from childhood expected that the mothers of adolescents and through adolescence (Seltzer et al., 2000; Stone- young adults with fragile X syndrome would be man, 1997) among families of children with men- more similar to mothers of individuals with au- tal retardation (Wikler, 1986) or autism (Fong, tism than to mothers of individuals with Down Wilgosh, & Sobsey, 1993). syndrome, although more precise predictions were In adolescence, there are differences in the de- not possible due to the lack of research on par- velopmental trajectories for Down syndrome, au- enting and fragile X syndrome. We examined the tism, and fragile X syndrome that may create very factors that predicted maternal psychological well- different levels and types of challenges for moth- being, including the behavioral symptoms of the ers. In particular, persons with fragile X syndrome adolescent or young adult, with the aim of ex- display age-related declines in their rate of devel- plaining diagnosis-related differences among opment in both cognitive and adaptive skills dur- mothers. We also examined the possibility that ing adolescence (Dykens et al., 2000). In contrast, coping strategies would moderate the effects of persons with Down syndrome display gains in so- the son or daughter’s behavioral problems on ma- cial skills during this period relative to persons ternal psychological well-being. with mental retardation of unknown or hetero- geneous etiology or persons with motor impair- Method ments (Hauser-Cram et al., 2001). The problem behaviors of persons with autism often intensify The present analysis was based on two on- during adolescence and further challenge their going studies of families of individuals with de- parents (Seltzer et al., 2000). Such differences in velopmental disabilities, one focused on families the developmental trajectories of affected off- of individuals with either fragile X or Down syn- spring during adolescence and young adulthood drome and the other on families of individuals may require different adaptations and, thus, also with autism. As there was considerable overlap contribute to differentiation of mothers of indi- across studies in the age range of the individuals viduals with fragile X syndrome, autism, and with disabilities and a common core of measures Down syndrome. completed by their mothers, it was possible to In summary, the present study was designed conduct the present comparative analysis. The dif- to examine the psychological well-being and cop- fering goals of these projects required different ing of mothers caring for an adolescent or young sample sizes and, therefore, the number of moth- adult with Down syndrome, fragile X syndrome, ers available for inclusion in the present analyses or autism. We focused on those dimensions of was substantially greater for the autism group (n maternal psychological well-being that in previous ϭ 174) than for the fragile X syndrome (n ϭ 22) research have been shown to be sensitive to dif- or Down syndrome group (n ϭ 39).

᭧ American Association on Mental Retardation 241 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al.

Sample ucational professional and (b) their Autism Diag- Recruitment. Mothers of participants with nostic Interview-Revised (Lord, Rutter, & Le- Down syndrome or fragile X syndrome were re- Couter, 1994) lifetime algorithm profile had to be cruited through advertisements in newspapers consistent with their reported autism spectrum throughout Wisconsin, Iowa, and Illinois; news- disorder diagnosis. letters distributed by regional and national profes- A standard short form of the Autism Diag- sional organizations; university-based research reg- nostic Interview-Revised (C. Lord, personal com- istries of families having a child with a disability; munication, February 1999), consisting of the 37 and brochures mailed to special educators and ge- items that comprise the diagnostic algorithm of netics clinics in the region. Because of its lower this instrument, was administered by appropriate- prevalence, participants with fragile X syndrome ly trained interviewers to confirm that all partici- were recruited nationally, whereas those with pants met the current diagnostic criteria for au- Down syndrome lived in Wisconsin and sur- tism spectrum disorders. The Autism Diagnostic rounding states. The families of the participants Interview-Revised is a standardized face-to-face in- with autism were recruited in Wisconsin and Mas- terview conducted with a primary caregiver that is sachusetts through agencies, schools, diagnostic based on the DSM-IV and International Classifi- clinics, postings on the Internet, mailing lists and cation of Diseases ICD-10 (World Health Orga- registries of families having a son or daughter with nization, 1990) criteria for autism. In the present autism sample, 164 of the 174 of the individuals a disability maintained by the investigators’ uni- met all of the Autism Diagnostic Interview-Re- versity research center, and via the media. vised criteria for Autistic Disorder. The profiles of Confirmation of the diagnosis. The parents of all the remaining 10 cases were reviewed and found participants with Down syndrome reported etiol- to be consistent with their autism spectrum dis- ogy as being trisomy 21. We were able to obtain order diagnosis (i.e., PDD-NOS or Asperger’s). reports confirming the karyotype for most partic- Exclusionary criteria. None of the adolescents ipants with Down syndrome. Reports of DNA or young adults with Down syndrome or fragile confirmation of the fragile X full mutation were X syndrome in the present sample had autism. available for all but 2 of the participants with frag- The autism status of participants in these groups ile X syndrome, with only cytogenetic confirma- was confirmed by first asking the mother, father, tion available for them. All participants with frag- and teacher of each target adolescent and young ile X syndrome had the full mutation, although adult to complete the Autism Behavior Checklist cases of mosaicism (i.e., premutation plus full mu- (Krug, Arick, & Almond, 1980). This checklist tation) were included. (which is described in more detail below) is a com- Mothers of participants with fragile X syn- monly used screening measure in which the in- drome were asked to provide the results of any formant is asked to indicate which of the com- genetic testing that had been conducted to deter- mon behavioral manifestations of autism the tar- mine their status as carriers of the fragile X pre- get individual displays. Following Volkmar et al. mutation or full mutation. However, less than (1988), we used a score of 44 or higher as reflect- half of the women responding reported having ing the possible appropriateness of an autism di- had genetic testing, and many of those did not agnosis. Any participant who met or exceeded this have copies of the results. Of those who had the score according to the responses of at least two of results, only one mother had undergone DNA the three informants was then referred for clinical testing. We were unable, therefore, to consider evaluation by a licensed psychologist at a devel- maternal pre- full-mutation carrier status. opmental disabilities clinic. The psychologist then Unlike Down syndrome or fragile X syn- evaluated the participant against DSM-IV criteria drome, in which laboratory genetic testing is used for autistic spectrum disorder. Note that infor- diagnostically, the diagnosis of autism is based on mants seldom disagreed in their Autism Behavior behavioral criteria. All participants in the autism Checklist scoring (e.g., only 10% of all mother sample met two criteria: (a) they had to have re- and teacher pairs differed in terms of whether the ceived a diagnosis on the autism spectrum (Autis- target individuals exceeded the threshold score of tic Disorder; Asperger’s Disorder; or Pervasive De- 44, with no consistent pattern with regard to velopmental Disorder, Not Otherwise Specified which of the two informants reported higher PDD-NOS) from a medical, psychological, or ed- scores).

242 ᭧ American Association on Mental Retardation VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al.

None of the mothers of the target adolescents The age of the target adolescents and young and young adults in the present autism sample adults averaged near 16 years, with a range of 10 reported that their child had Down syndrome, to 23 years in each of the diagnostic groups. All fragile X syndrome, Rett’s Disorder, Childhood targeted individuals lived at home with their par- Disintegrative Disorder, or tuberous sclerosis. ents. There were more males among the adoles- These co-morbid conditions were excluded be- cents and young adults with autism than among cause of the distinct developmental trajectories those with Down syndrome. As expected, the characteristic of each of them. three groups also differed with respect to their to- Thus, the participants in the present study tal scores on the Autism Behavior Checklist (de- consisted of three nonoverlapping diagnostic scribed subsequently), which provided a measure groups. Other criteria for inclusion in this analysis of problem behaviors, particularly those reflective were that the mother had to be the biological par- of the symptoms of autism. Those who had au- ent of the child with the disability and, to de- tism scored substantially higher than those with crease the number of factors to consider in the fragile X syndrome, with those with Down syn- analyses, the mother had to be currently married. drome having the lowest scores. Characteristics of the participants. Table 1 pre- sents the characteristics of the participants. The Measures mothers averaged in age from the mid- to late 40s, Mothers in the three groups all completed with the mothers of individuals with Down syn- measures of their psychological well-being, which drome significantly older than the other two formed the dependent variables; coping; and the groups of mothers. This difference would be ex- Autism Behavior Checklist, which provided an as- pected in light of the fact that the risk of having sessment of their son or daughter’s problem be- a child with Down syndrome is known to increase haviors and competence in several domains that with maternal age. Family income averaged be- have been shown previously (e.g., Kasari & Sig- tween $50,000 and $60,000 and more than two man, 1997) to be sources of stress for parents. thirds of the mothers in each group were em- Maternal depressive symptoms were assessed by ployed outside the home, with no diagnostic the Center for Epidemiologic Studies Depression group differences in family income or maternal Scale hereafter called the Depression Scale (Rad- employment status. Approximately half of the loff, 1977). The scale consists of 20 items in which mothers had graduated from college or had an the frequency of depressive symptoms during the advanced degree, with no difference across diag- preceding week is rated on a 4-point scale ranging nostic groups. The families of individuals with from 0 (rarely)to3(most of the time). A score of Down syndrome were larger in size than were the 16 or higher indicates the risk of clinical depres- families in the other two groups (3.4 children per sion. Alpha reliability for this instrument was .90 family in the group with Down syndrome versus for the present sample. just under 3 children for the other two groups). Pessimism felt about the individual with dis- The families who had a son or daughter with frag- abilities was measured using the 11-item Pessi- ile X syndrome, however, had significantly more mism subscale of the Questionnaire on Resources children with a disability per family than did fam- and Stress (QRS-F; Friedrich, Greenberg, & Crnic, ilies in the other two groups. 1983). According to Friedrich et al., ‘‘the central In the fragile X syndrome sample, there were characteristic of this subscale is an immediate and several families with more than one child with the future pessimism about the child’s prospects of full mutation and a disability. In such cases, we achieving self-sufficiency’’ (p. 44). Mothers indi- selected as the target adolescent or young adult cate their agreement (coded 1) or disagreement the one who was between the ages of 10 and 23 (coded 0) with the items, which are summed, with years. If more than one child qualified, the target higher scores indicating a greater degree of pessi- was selected at random. In the Down syndrome mism. Alpha reliability for the sample was .72. and autism cases, no family had more than one Closeness of the mother–child relationship was child with Down syndrome or autism who fell measured by the Positive Affect Index (Bengtson into the age range of interest, although some fam- & Black, 1973). In this 10-item scale, the first 5 ilies in these two groups had additional children items assess the level of understanding, trust, fair- with those disabilities outside of the age range of ness, respect, and affection that the mother feels interest. toward her son or daughter with the disability,

᭧ American Association on Mental Retardation 243 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. 2 *** *** *** ␹ ** *** *** *** * * * a,b,c, b,c, a,b,c, a,c, a,b,c, a,c c, a,b, c, a,c, or 3.88 0.58 0.74 1.46 3.29 9.11 6.00 3.80 44.13 25.53 55.61 16.98 48.09 .74 3.00 0.03 5.97 1.30 5.93 8.60 7.09 6.87 5.54 25.64 SD F 174) 18,938 ϭ n 2.75 1.39 7.86 8.85 7.73 73.6 70.7 48.3 45.75 53.44 16.98 12.02 Autism ( 56,968 Mean or % 39) .43 ϭ 3.43 16.32 7.02 1.80 2.31 4.35 3.68 3.47 2.97 13.45 SD n 18,283 Down syndrome versus Autism. Chi-squares in boldface. c 3.37 1.16 1.69 2.31 2.26 2.95 3.54 51.3 76.9 59.0 47.91 12.74 53,378 Mean or % Down syndrome ( .84 .79 Fragile X versus Autism. 3.14 16.37 3.70 8.60 4.84 5.89 5.55 5.80 b 24.25 22) SD ϭ 22,239 n 4.23 6.32 6.41 8.18 2.95 1.81 72.7 68.2 50.0 16.18 35.14 10.00 43.52 Fragile X ( 57,273 Mean or % Fragile X versus Down syndrome. a .001. Ͻ p .01. *** Ͻ p Sample Characteristics .05. ** Significant post hoc comparisons: Ͻ Age Gender (% male) Total score Sensory subscale Relating subscale Body Use subscale Language subscale Social & Self-Help subscale Age Family income ($ midpoints) Employment status (% employed) Education level (% college grad) No. of children in family No. of disabled children p * Child Autism Behavior Checklist Table 1. Characteristic Maternal Note.

244 ᭧ American Association on Mental Retardation VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. whereas the next 5 items assess the mother’s per- stones in language and the use of language in so- ception of reciprocated closeness (i.e., the extent cially inappropriate ways); and Social and Self- to which the son or daughter is understanding, Help (failures in acquiring adaptive behaviors as trusting, etc., of the mother). Each item is rated well as destructive and aggressive behavior). We on a 6-point scale, with higher scores indicting a focused on autism spectrum behaviors rather than better relationship quality. Alpha reliability for other types of behavioral problems (e.g., anxiety, the mother’s rating of closeness toward her child depression) because the former were expected to was .78 for our sample, and for reciprocated close- maximally distinguish the three groups of adoles- ness perceived by the mother, it was .83. cents and young adults and to be highly salient Coping refers to the ways in which people re- for mothers. Alpha reliability across the 57 items spond to stressful events. Coping was measured of the Autism Behavior Checklist was .91 for the with Carver, Scheier, and Weintraub’s (1989) present sample of mothers. Multidimensional Coping Inventory, which con- In analyzing the contribution of behavioral sists of 13 four-item subscales. Each item is rated symptoms to maternal psychological well-being, in terms of how often (1 ϭ not at all to 4 ϭ a lot) we report those involving only the Autism Behav- the respondent uses a particular coping strategy ior Checklist total score. In general, the three when experiencing a difficult or stressful event. groups of adolescents and young adults were dis- We chose the eight subscales representing theo- tinguished by their scores on each of the sub- retical aspects of problem- or emotion-focused scales, although the participants with fragile X coping (Pearlin & Schooler, 1978). Four subscales syndrome did not differ significantly from the measured problem-focused coping: active coping group with Down syndrome on the Sensory sub- refers to direct attempts to remove, circumvent, scale or from the group with autism on the Body or reduce the effects of a stressor; planning in- Use and Language subscales (see Table 1). More- volves planning an action to deal with a stressor; over, the pattern of results for predicting maternal suppression of competing activities refers to putting outcomes is similar across the subscales of the Au- other things aside to focus on the stressful situa- tism Behavior Checklist as well as for individual tion; and positive reinterpretation and growth in- items focused on particular high-salience mal- volves reframing a problematic event in a positive adaptive behaviors (e.g., aggression), important light (Carver et al., 1989). Four other subscales developmental milestones, or indicators of com- measured emotion-focused coping: denial involves petence (e.g., use of multiword utterances). (The denying that a stressor exists or is real; focusing on data are available from the authors.) Thus, use of and venting of emotions is the focusing on the stress- the Autism Behavior Checklist total score provid- or and ventilating those feelings; behavioral disen- ed increased reliability without obscuring the re- gagement involves the reduction of efforts to deal lationships between child behavior and maternal with a stressor; and mental disengagement involves outcome compared to the use of subscales or in- turning to thoughts and activities to take one’s dividual items. mind off of the problem (Carver et al., 1989). Al- A missing response for an item in any ques- pha reliability for the present sample was .89 for tionnaire was replaced by the participant’s mean problem-focused coping and .74 for emotion-fo- score across all relevant items for the scale or sub- cused coping. scale, with the requirement that no more than The behavioral symptoms manifested by the 75% of the items for a scale (or subscale) could target adolescent or young adult were indexed by have missing values. In the case of missing de- maternal ratings on the 57-item Autism Behavior mographic data (e.g., family income), the diag- Checklist. The items on this checklist are differ- nostic group mean was substituted for the missing entially weighted to reflect the centrality of the value. Missing values were infrequent, with no problem behavior to the diagnosis of autism. The more than 4% of the values being coded as miss- items are also grouped into five subscales: Sensory ing for any measure. (displays of nonresponsiveness or heightened re- sponsiveness to sensory stimuli); Relating (failures Results to engage with or react to other people); Body and Object Use (stereotyped body movements, pref- Our first research question concerned differ- erence for routines, and self-injury); Language ences among the mothers with respect to psycho- (failures to achieve typical developmental mile- logical well-being and coping. This question was

᭧ American Association on Mental Retardation 245 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. addressed by conducting a series of one-way AN- It is important to note that the average Depression OVAs for each dependent measure, with Fisher’s Scale score for all three groups was below the cut- least significant difference technique (Levin, Ser- off for clinical depression (16); however, a signif- lin, & Seaman, 1994) used for post hoc compari- icantly higher proportion of mothers of adoles- sons to maintain a family-wise alpha level of p Յ cents and young adults with autism had Depres- .05 (see Table 2). The three groups of mothers sion Scale scores in the clinical range than did the differed with respect to their feelings of pessimism other two groups (33.3% of mothers of adoles- about the adolescent or young adult’s future. As cents and young adults with autism versus 18.2% hypothesized, mothers of those with Down syn- of mothers of those with fragile X syndrome and drome were significantly less pessimistic than were 10.3% of mothers of those with Down syndrome, mothers in the other two groups. ␹2(2, N ϭ 154) ϭ 6.46, p ϭ .01. Counter to our The groups also differed in their perceptions hypothesis, the three groups of mothers did not of the quality of the mother–child relationship. differ in their use of either problem-focused or Consistent with our hypothesis, mothers of ado- emotion-focused coping. lescents and young adults with Down syndrome Next, we asked whether the group differences reported more closeness in the relationship than in maternal pessimism, closeness in the relation- did mothers of those with autism, but the mothers ship with the son or daughter with the disability, of those with fragile X syndrome did not differ and depressive symptoms could be explained, in from the other two groups. Regarding the moth- part, by differences in demographic characteristics er’s perceptions of her child’s feelings of recipro- of the family, the behavioral symptoms of the ad- cated closeness, the three groups were significantly olescent or young adult, and maternal coping. We divergent, with mothers of those with (a) Down focused on family demographic factors because syndrome reporting the most perceived recipro- they constitute the social context in which non- cated closeness, (b) autism reporting the least, and normative parenting is experienced. We focused (c) fragile X syndrome in the middle. on the adolescent or young adult’s level of behav- The three groups of mothers differed in their ioral symptoms because such symptoms consti- level of depressive symptoms, with mothers of ad- tute the immediate challenge of nonnormative olescents and young adults with autism reporting parenting and on coping because past research has significantly higher levels of depressive symptoms shown that coping can buffer the effects of be- than did mothers of those with Down syndrome. havior problems in mothers of adults with devel- In contrast, mothers of those with fragile X syn- opmental disabilities. We hypothesized that once drome did not differ from the other two groups. contextual factors, behavioral symptoms, and ma-

Table 2. Means and SDs of Maternal Psychological Scale Scores Fragile X Down Autism (n ϭ 22) (n ϭ 39) (n ϭ 174) Scale Mean SD Mean SD Mean SD F Pessimism (QRS)d 6.05 2.38 4.82 1.92 6.65 2.34 10.38a,c,*** Closeness in the mother/adolescent relationship Maternal rating of 25.27 2.35 26.13 3.19 24.30 3.49 5.10c,** closeness Perceived reciprocated 24.05 2.98 26.26 2.69 21.62 4.44 21.95a,b,c,*** closeness Depressive symptoms 11.00 7.65 6.74 6.31 12.60 9.10 7.47c,*** Emotion-focused coping 11.81 5.42 12.50 5.46 13.29 5.28 0.97 Problem-focused coping 30.09 8.82 31.95 7.82 32.33 7.34 0.86 Note. Significant post hoc comparisons: aFragile X vs. Down syndrome. bFragile X vs. Autism. cDown syndrome vs. Autism. dQuestionnaire on Resources and Stress. **p Ͻ .01. ***p Ͻ .001.

246 ᭧ American Association on Mental Retardation VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. ternal coping patterns were controlled, the group tered in Step 2, diagnostic group differences can differences would become nonsignificant. We also be explained by contextual, behavioral, or coping hypothesized that level of behavioral symptoms factors. would be the strongest predictor of maternal out- In these regressions, the two variables mea- comes across diagnostic groups. Finally, we hy- suring diagnostic groups were dummy variables pothesized that coping would buffer the effects of indexing whether the child had autism or Down behavioral symptoms on maternal outcomes. syndrome. By including these two variables in the Table 3 presents the findings of the four re- regression models, each of these diagnostic groups gression models, one for each of the maternal out- was contrasted with fragile X syndrome, which come variables: pessimism, feelings of closeness to was the omitted category (Aiken & West, 1991). son or daughter, perceptions of son or daughter’s A significant coefficient would signify that the reciprocated feelings of closeness, and level of de- particular group (autism or Down syndrome) dif- pressive symptoms. The findings for each out- fers from those with fragile X syndrome with re- come variable are reported in two steps. In Step spect to the dependent variable. 1, two variables representing diagnostic group are Regarding maternal pessimism about the fu- the only predictors, whereas in Step 2, variables ture of the son or daughter with the disability, measuring contextual background characteristics, mothers of those with Down syndrome were sig- behavioral symptoms, and maternal coping are nificantly less pessimistic than mothers of those added. To the extent that diagnostic group coef- with fragile X syndrome (Step 1), but this differ- ficients that were significant in Step 1 became ence disappeared after all other factors in the nonsignificant after the other variables were en- model were controlled (Step 2). There was no dif-

Table 3. Multiple Regressions (Standardized Coefficients) Maternal rating Reciprocated of relationship relationship Depression Variable Pessimism quality quality Scale Step 1 Autism vs. Fragile X .112 Ϫ.125 Ϫ.240** .080 DSa vs. Fragile X Ϫ.193* .093 .186* Ϫ.180 Step 2 Autism vs. Fragile X .038 Ϫ.100 Ϫ.197* .040 DS vs. Fragile X Ϫ.119 Ϫ.033 .050 Ϫ.104 Contextual factors Family income Ϫ.107 Ϫ.123 Ϫ.123* Ϫ.205** Maternal education .189** Ϫ.109 Ϫ.078 .063 No. of disabled children .006 Ϫ.090 Ϫ.107 .148* Gender of target child .040 Ϫ.029 Ϫ.022 Ϫ.041 Age of child .155** .105 Ϫ.048 Ϫ.038 Behavioral symptoms Autism Behavior Checklist score .269*** Ϫ.241** Ϫ.283*** .191** Maternal coping Problem focused coping Ϫ.216*** .238*** .160** Ϫ.134* Emotion focused coping .213*** Ϫ.134* Ϫ.102 .345*** R2 (Step 1) .082 .042 .159 .061 R2 (Step 2) .262 .197 .278 .303 R2 change .180*** .155*** .118*** .243*** aDown syndrome. *p Ͻ .05. **p Ͻ .01. ***p Ͻ .001.

᭧ American Association on Mental Retardation 247 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. ference between those with autism and those with daughter’s Autism Behavior Checklist score, and fragile X syndrome at either Step 1 or Step 2. the mother’s coping style were significant predic- Among the contextual variables, more highly ed- tors. Mothers tended to be more depressed if there ucated mothers and mothers of older sons or was lower family income, other children with a daughters were more pessimistic about their disability in the family, if the adolescent had a child’s future. In addition, the number of behav- greater number of behavioral symptoms, and if ioral symptoms of the adolescent and young adult the mother reported less use of problem-focused and the mother’s coping were significant predic- and greater use of emotion-focused coping. tors. Mothers of adolescents and young adults In follow-up analyses, we examined whether with more behavioral symptoms were more pes- either problem-focused or emotion-focused cop- simistic about their son or daughter’s future. Fur- ing buffered the effects of behavioral symptoms ther, mothers who used lower levels of problem- on maternal outcomes. None of the interaction focused coping and higher levels of emotion-fo- terms (coping by Autism Behavior Checklist cused coping tended to have higher pessimism score) was significant. We also constructed three- scores than those who manifested the reverse cop- way interactions (coping by Autism Behavior ing profile. Checklist score by diagnostic group) to assess the Regarding the mother’s rating of closeness, possibility of differential coping effects by group, mothers of adolescents and young adults with but again, none was significant. Thus, our hy- fragile X syndrome did not differ at either Step 1 pothesis that coping would buffer the stressful ef- or Step 2 from the mothers in either of the other fects of behavioral symptoms on maternal out- two groups. Contextual factors were not predic- comes was not supported (results of these analyses tive of maternal perception of closeness of the re- are available from the authors). lationship. However, as in the prediction of pes- simism, mothers of adolescents and young adults Discussion with fewer behavioral symptoms tended to report greater closeness, and mothers who used problem- The results of the present analyses confirm focused coping and refrained from using emotion- past reports about the family context of individ- focused coping reported greater closeness. uals with Down syndrome and autism and pro- Regarding the mother’s perception of her ad- vide new insights regarding the family context of olescent’s reciprocated feelings of closeness to her, individuals with fragile X syndrome. Consistent mothers of adolescents and young adults with au- with past research (Kasari & Sigman, 1997; Seltzer tism perceived less reciprocated closeness than did et al., 1993), we found that mothers of adolescents mothers of those with fragile X syndrome (Step and young adults with Down syndrome reported 1). This difference remained even after controlling low levels of pessimism about their son or daugh- for contextual factors, level of behavioral symp- ter’s future, a close relationship with their son or toms, and maternal coping (Step 2). Although daughter, and low levels of depressive symptoms. there was a difference between mothers of adoles- Also consistent with past research (Holroyd & cents and young adults with Down syndrome and McArthur, 1976; Kasari & Sigman, 1997; Koegel those with fragile X syndrome in Step 1, this dif- et al., 1992), mothers of adolescents and young ference became nonsignificant when other factors adults with autism reported higher levels of pes- were controlled in Step 2. The adolescent’s be- simism, more distant relationships with the af- havioral symptoms again were predictive, with a fected son or daughter, and higher levels of de- greater number of symptoms predicting dimin- pressive symptoms than did the mothers of ado- ished perceived reciprocated closeness. Finally, lescents and young adults with Down syndrome. mothers who used higher levels of problem-fo- Given the clear difference in the extent of prob- cused coping were more likely to perceive recip- lem behaviors reported on the Autism Behavior rocated closeness from the son or daughter. Checklist for the adolescents and young adults Regarding the mother’s level of depressive with autism versus those with Down syndrome, it symptoms, mothers of adolescents and young is likely that some of the differences in maternal adults with fragile X syndrome did not differ from outcomes were the result of the divergent caregiv- either of the other two groups. Rather, the fami- ing challenges posed by these two groups of ad- ly’s income, the number of children in the family olescents and young adults. It is also possible that who had developmental disabilities, the son or genetic vulnerability in mothers whose adoles-

248 ᭧ American Association on Mental Retardation VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. cents and young adults had autism (i.e., the broad- and young adults with fragile X syndrome in the er autism phenotype; Jorde et al., 1991; Piven et present study. Nevertheless, 16% of these mothers al., 1990) contributed to their elevated levels of had levels of depressive symptoms above the cut- depressive symptoms and relationship distance. off for clinical depression and, thus, the mental How did mothers of adolescents and young health status of these women remains a concern. adults with fragile X syndrome compare with the Based on previous research (Essex, Seltzer, & other two groups? They were more pessimistic Krauss, 1999; Seltzer et. al., 1995), we had hy- about their child’s future than were mothers of pothesized that coping would buffer the stressful adolescents and young adults with Down syn- effects of behavioral symptoms; however, there drome but did not differ in this respect from their was no evidence of a buffering effect. It is possible counterparts whose adolescents and young adults that the small sample size for the fragile X syn- had autism. This pessimism may reflect not undue drome and Down syndrome groups strained the concerns and worries but rather a realistic apprais- statistical power needed to detect coping by group al of the future based on the behavioral challenges interaction effects. The possibility of Type II er- that typically accompany fragile X syndrome ror, therefore, should not be discounted. Second, (Clayton, Glidden, & Kiphart, 1994). In any we have argued that an essential ingredient in ef- event, these results suggest that these mothers fective coping is the expectation of gaining some were especially worried about their son or daugh- control over the sources of stress as a result of ter’s future. Mothers of adolescents and young coping (Seltzer et al., 1995). Because the behav- adults with fragile X syndrome also perceived less ioral symptoms of individuals with autism or frag- reciprocated closeness in the relationship with the ile X syndrome are often unpredictable, it may be adolescent or young adult than did mothers of the case that maternal coping efforts are ineffec- adolescents and young adults with Down syn- tive in changing this source of stress; hence, cop- drome, but did not perceive that this relationship ing might not have any measurable effects on reg- was as distant as did mothers of adolescents and ulating the psychological well-being of these young adults with autism. Further, mothers of ad- mothers, even though they may exert coping ef- olescents and young adults with fragile X syn- forts. drome did not differ from the other two groups The strongest and most consistent predictor of mothers in either their feelings of closeness of maternal outcomes was the extent and severity with their adolescent or young adult or their level of the behavioral symptoms of the adolescent, as of depressive symptoms. In general, mothers of measured by the Autism Behavior Checklist. adolescents and young adults with fragile X syn- Higher scores on this measure significantly pre- drome display fewer signs of compromised psy- dicted greater maternal pessimism, less closeness chological well-being than do mothers of individ- in the relationship between the mother and ado- uals with autism but more such signs than do lescent or young adult, and greater maternal de- mothers of individuals with Down syndrome. pressive symptoms. A similar effect of behavioral It is interesting to note that the mothers of symptoms across diagnostic groups has been re- adolescents and young adults with fragile X syn- ported by Floyd and Gallagher (1997) and Do- drome did not differ from the other groups of nenberg and Baker (1993). Indeed, previous re- mothers in depressive symptoms. Some previous search has shown behavior problems to be a better researchers have reported higher rates of depres- predictor of parenting stress than is IQ (Hodapp sion in females who carry the full mutation and, et al., 1997). Thus, our research confirms past albeit less consistently, even in females who carry findings about the effects of behavior problems the premutation compared to other diagnostic on parental psychological well-being, while ex- groups (Keysor & Mazzocco, 2002). In these stud- tending the findings to a new diagnostic group ies, however, researchers have typically not re- (i.e., fragile X syndrome). It also supports the cau- stricted their samples to women who have affected tionary note sounded by Cahill and Glidden children. Moreover, previous investigators have (1996), who argued that differences in child char- often focused on chronic depression in contrast acteristics must be considered when interpreting to current or recent displays of the symptoms of group differences in maternal behavior and psy- depression. These methodological differences may chological functioning. account for the relatively low levels of depressive As in previous studies, we found that contex- symptoms reported by the mothers of adolescents tual factors (e.g., maternal education) were also re-

᭧ American Association on Mental Retardation 249 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. lated to the measures of maternal well-being (Ryff outcomes, we cannot rule out a role in the present & Seltzer, 1996). Of special interest was the find- study. ing that the number of children with disabilities There are several additional limitations of the in the family was related to maternal symptoms present study. First, we lacked an adequate char- of depression. Inclusion of multiple affected chil- acterization of the genetic status (as pre- or full- dren within the family is one of the features that mutation carriers) of the mothers of the adoles- distinguishes families of children with autism or cents and young adults with fragile X syndrome. fragile X syndrome from families of children with Inclusion of genetic status as a predictor in our Down syndrome and many other forms of mental regressions would have provided a more complete retardation. This suggests that there is a need in picture of the sources of parenting stress in this future research to conduct more detailed analyses group, especially in light of recent findings of cor- of the impact on parental well-being of variations relations between molecular measures and the in family composition. clinical phenotype, even in women who carry It is important to note that despite the pow- only the premutation (Hagerman & Hagerman, erful effect of child characteristics on maternal 2002; Nolin et al., 2003). We plan to pursue this outcomes (Cahill & Glidden, 1996) and the con- issue in our future research. Nevertheless, it is im- tribution of contextual factors, such as number of portant that child characteristics, most notably affected children, differences between the three problem behaviors, predicted parenting outcome groups of mothers could not be explained solely in mothers of individuals with fragile X syndrome; by the behavioral symptoms of the target adoles- hence, it is clear that not all of the psychological cent or young adult, contextual factors, or varia- risk for these mothers resulted from their geno- tions in maternal coping strategy. In particular, type. Second, not all of the adolescents and young differences in perception of reciprocated mother– adults with autism in our sample had undergone child relationship quality between mothers of in- genetic testing to rule out fragile X syndrome. dividuals with fragile X syndrome and the other There is evidence suggesting that as many as 5% groups of mothers remained significant even after of individuals with autism may have fragile X syn- controlling for the contribution of these other fac- drome (Feinstein & Reiss, 2001). Future research- ers will need to attend more closely to the genetic tors. A task for future researchers is to identify the status of the individual with autism. Finally, we sources of this unexplained variation in maternal have focused here only on mothers. In future re- perception of relationship quality. Plausible can- search, it will be important to also study fathers didates include differences in maternal genotype because the factors that affect maternal and pater- and social supports that are correlated with child nal well-being may well be distinct (Essex et al., diagnosis. It will also be important to evaluate a 1999). wider range of the behaviors and characteristics The present findings support the notion of that distinguish between adolescents and young differential experiences for family members de- adults with Down syndrome, fragile X syndrome, pending upon the specific diagnosis of their child and autism than was measured in the present with a developmental disability. In part, these dif- study. ferences reflect the unique challenges posed by Indeed, one limitation of the present study the young people with the diagnoses of interest. was the somewhat narrow assessment of problem Thus, treating these behavioral challenges directly behaviors in the adolescents and young adults, re- may alleviate some negative outcomes for moth- flecting only autism-spectrum behaviors. Other ers. It may also be helpful, given the intractable problem behaviors that are likely to distinguish nature of some of these behavioral challenges, to the groups (e.g., hyperactivity and social anxiety) provide parents with respite and other forms of were not measured and may explain some of the social support that may buffer some of the inev- differences among the three groups of mothers or itable stress associated with these behaviors. The additional variance within each group. Moreover, present results, however, are also consistent with we did not have comparable measures of compe- the notion that some parents have a genetic vul- tence or level of functioning (e.g., IQ) across the nerability for less than optimal outcomes and that three samples. Although such measures have not this vulnerability is magnified by the challenges been found in previous research to account for of raising a son or daughter with special needs. substantial variation in maternal psychological Thus, these parents are likely to benefit from on-

250 ᭧ American Association on Mental Retardation VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al. going interventions directed at their own needs, ically based approach. Journal of Personality perhaps concentrated at life transitions known to and Social Psychology, 56, 267–283. be especially challenging (e.g., the child’s transi- Clayton, J. M., Glidden, L. M., & Kiphart, M. J. tion from high school to the workplace). In sum- (1994). The Questionnaires on Resources and mary, information gleaned from further studies in- Stress: What do they measure? American Jour- volving comparisons of families grappling with nal on Mental Retardation, 99, 313–316. different diagnoses can be of help to clinicians DeLong, G. R. (1994). Children with autistic spec- and service providers so that they may better tailor trum disorder and a family history of affective their interventions to families’ specific needs. disorder. Developmental Medicine, 36, 674–687. DeLong, G. R., & Nohria, C. (1994). Psychiatric family history and neurological disease in au- References tistic spectrum disorders. Developmental Medi- cine and Child Neurology, 36, 441–448. Aiken, L. S., & West, S. G. (1991). Multiple regres- Donenberg, G., & Baker, B. L. (1993). The impact sion: Testing and interpreting interactions. New- of young children with externalizing behav- bury Park, CA: Sage. iors on their families. Journal of Abnormal American Psychiatric Association. (1994). Diag- Child Psychology, 21, 179–198. nostic and statistical manual of mental disorders Donovan, A. M. (1988). Family stress and ways of (4th ed.). Washington, DC: Author. coping with adolescents who have handicaps: August, G. J., Stewart, M. A., & Tsai, L. (1981). Maternal perceptions. American Journal on The incidence of cognitive disabilities in sib- Mental Retardation, 92, 502–509. lings of autistic children. British Journal of Psy- Dunst, C., Trivette, C., & Cross, A. (1986). Me- chiatry, 138, 416–422. diating influences of social support: Personal, Bailey, D. B., Skinner, D., Hatton, D. D., & Rob- family, and child outcomes. American Journal erts, J. (2000). Family experiences and factors of Mental Deficiency, 90, 403–417. associated with diagnosis of fragile X syn- Dykens, E. M., Hodapp, R. M., & Finucane, B. drome. Journal of Developmental and Behavioral M. (2000). Genetics and mental retardation syn- Pediatrics, 21, 315–321. dromes: A new look at behavior and interventions. Beavers, J., Hampson, R. B., Hulgus, Y. F., & Bea- Baltimore: Brookes. vers, W. R. (1986). Coping in families with a Dykens, E. M., & Volkmar, F. R. (1997). Medical retarded child. Family Process, 25, 365–378. conditions associated with autism. In D. J. Bengtson, V. L., & Black, K. D. (1973, October Cohen & F. R. Volkmar (Eds.), Handbook of 16). Solidarity between parents and children: Four autism and pervasive developmental disorders perspectives on theory development. Paper pre- (2nd ed., pp. 388–407). New York: Wiley. sented at the Theory Development Work- Essex, E. L., Seltzer, M. M., & Krauss, M. W. shop, National Council on Family Relations, (1999). Differences in coping effectiveness Toronto. and well-being among aging mothers and fa- Bolton, P., MacDonald, H., Pickles, A., Rios, P., thers of adults with mental retardation. Amer- Goode, S., Crowson, M., Bailey, A., & Rutter, ican Journal on Mental Retardation, 104, 545– M. (1994). A case-control family study of au- 563. tism. Journal of Child Psychology and Psychiatry, Feinstein, C., & Reiss, A. L. (2001). Autism: The 35, 877–900. point of view from fragile X studies. Journal Cahill, B. M., & Glidden, L. M. (1996). Influence of Autism and Developmental Disorders, 28, of child diagnosis on family and parental 393–405. functioning: Down syndrome versus other Fidler, D. J., & Hodapp, R. M. (1999). Craniofa- disabilities. American Journal on Mental Retar- cial maturity and perceived personality in dation, 101, 149–160. children with Down syndrome. American Carmichael, B., Pembrey, M., Turner, G., & Bar- Journal on Mental Retardation, 104, 410–421. nicoat, A. (1999). Diagnosis of fragile-X syn- Fidler, D. J., Hodapp, R. M., & Dykens, E. M. drome: The experiences of parents. Journal of (2000). Stress in families of young children Intellectual Disability Research, 43, 47–53. with Down syndrome, Williams syndrome, Carver, C. S., Sheier, M. F., & Weintraub, J. K. and Smith-Magenis syndrome. Early Educa- (1989). Assessing coping strategies: A theoret- tion and Development, 11, 395–406.

᭧ American Association on Mental Retardation 251 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al.

Fidler, D. J., Hodapp, R. M., & Dykens, E. M. of parental stress and adaptation. Journal of (2002). Behavioral phenotypes and special ed- Early Intervention, 14, 234–248. ucation: Parent report of educational issues Hauser-Cram, P., Warfield, M. E., Shonkoff, J. P., for children with Down syndrome, Prader- & Krauss, M. W. (2001). Children with dis- Willi syndrome and Williams syndrome. Jour- abilities: A longitudinal study of child devel- nal of Special Education, 36, 80–88 opment and parent well-being. Monographs of Floyd, F. J., & Gallagher, E. M. (1997). Parental the Society for Research in Child Development, 66, stress, care demands, and use of support ser- Serial No. 266. vices for school-age children with disabilities Hodapp, R. M., Dykens, E. M., & Masino, L. and behavior problems. Family Relations, 46, (1997). Families of children with Prader-Willi 359–371. syndrome: Stress-support and relations to Folstein, S., & Rutter, M. (1977). Genetic influ- child characteristics. Journal of Autism and De- ences and infantile autism. Nature, 265, 726– velopmental Disorders, 27, 11–24. 728. Hodapp, R. M., Fidler, D. J., & Smith, A. C. M. Fong, L., Wilgosh, L., & Sobsey, D. (1993). The (1998). Stress and coping in families of chil- experience of parenting an adolescent with dren with Smith-Magenis syndrome. Journal of autism. International Journal of Disability, De- Intellectual Disability Research, 5, 331–340. velopment and Education, 40, 105–113. Holroyd, J., & McArthur, D. (1976). Mental re- Franke, P., Maier, W., Hautzinger, M., Weiffen- tardation and stress on the parents: A contrast bach, O., Gansicke, M., Iwers, B., Poustka, F., between Down’s syndrome and childhood au- Schwab, S. G., & Froster, U. (1996). Fragile-X tism. American Journal on Mental Deficiency, carrier females: Evidence for a distinct psy- 80, 431–436. chopathological phenotype? American Journal Jorde, L., Hasstedt, S., Ritvo, E., Mason-Brothers, of Medical Genetics, 64, 334–339. A., Freeman, B. J., Pingree, C., McMahon, W. Friedrich, W. N., Greenberg, M. T., & Crnic, K. M., Petersen, B., Jenson, W. R., & Mo, A. (1983). A short-form of the Questionnaire on (1991). Complex segregation analysis of au- Resources and Stress. American Journal of Men- tism. American Journal of Human Genetics, 29, tal Deficiency, 88, 41–48. 932–938. Friedrich, W. N., Wilturner, L. T., & Cohen, D. Kasari, C., & Bauminger, N. (1998). Social and S. (1985). Coping resources and parenting emotional development in children with mentally retarded children. American Journal mental retardation. In J. A. Burack, R. M. Ho- of Mental Deficiency, 90, 130–139. dapp, & E. Zigler (Eds.), Handbook of mental Gallagher, J. J., Beckman, P. J., & Cross, A. H. retardation and development (pp. 411–433). (1983). Families of handicapped children: New York: Cambridge University Press. Sources of stress and its amelioration. Excep- Kasari, C., & Sigman, M. (1997). Linking parental tional Children, 50, 10–19. perceptions to interactions in young children Gath, A., & Gumley, D. (1986). Family back- with autism. Journal of Autism and Develop- ground of children with Down’s syndrome mental Disorders, 27, 39–57. and of children with a similar degree of men- Keysor, C. S., & Mazzocco, M. M. M. (2002). A tal retardation. British Journal of Psychiatry, developmental approach to understanding 149, 161–171. fragile X syndrome in females. Microscopy Re- Gray, D. E., & Holden, W. J. (1992). Psycho-social search and Technique, 57, 179–186. well-being among parents of children with au- Koegel, R. L., Schreibman, L., Loos, L. M., Dir- tism. Australia and New Zealand Journal of De- lich-Wilhelm, H., Dunlap, G., Robbins, F. R., velopmental Disabilities, 18, 83–93. & Pilenis, A. J. (1992). Consistent stress pro- Hagerman, R. J. (1999). Neurodevelopmental disor- files in mothers of children with autism. Jour- ders. Oxford: Oxford University Press. nal of Autism and Developmental Disorders, 22, Hagerman, R. J., & Hagerman, P. J. (2002). The 205–216. fragile X premutation: Into the phenotypic Krug, D. A., Arick, J. R., & Almond, P. J. (1980). fold. Current Opinion in Genetics and Develop- Autism Screening Instrument for Educational ment, 12, 278–283. Planning. Examiner’s manual. Portland, OR: Hanson, M., & Hanline, M. F. (1990). Parenting ASIEP Education. a child with a disability: A longitudinal study Lainhart, J. E., & Folstein, S. E. (1994). Affective

252 ᭧ American Association on Mental Retardation VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al.

disorders in people with autism: A review of anen, M. L., von Koskull, H., & Sherman, S. published cases. Journal of Autism and Devel- L. (2003). Expansion of the fragile X CGG opmental Disorders, 24, 587–601. repeat in females with premutation or inter- Landa, R., Folstein, S. E., & Isaacs, C. (1991). mediate alleles. American Journal of Medical Spontaneous narrative-discourse performance Genetics, 72, 454–464. of parents of autistic individuals. Journal of Pearlin, L. I., & Schooler, C. (1978). The structure Speech and Hearing Research, 34, 1339–1345. of coping. Journal of Health and Social Behav- Levin, J. R., Serlin, R. C., & Seaman, M. A. ior, 19, 2–21. (1994). A controlled, powerful multiple-com- Piven, J. (2001). The broad autism phenotype: A parison strategy for several situations. Psycho- complementary strategy for molecular genetic logical Bulletin, 115, 153–159. studies of autism. American Journal of Medical Lord, C., & Rissi, S. (2000). Diagnosis of autism Genetics, 105, 34–35. spectrum disorders in young children. In A. Piven, J., Gayle, J., Chase, G. A., Fink, B., Landa, M. Wetherby & B. M. Prizant (Eds.), Autism R., Wzorek, M., & Folstein, S. (1990). A fam- spectrum disorders: A transactional developmental ily history study of neuropsychiatric disorders perspective (pp. 11–30). Baltimore: Brookes. in the adult siblings of autistic individuals. Lord, C., Rutter, M., & LeCouter, A. (1994). Au- Journal of the American Academy of Child and tism Diagnostic Interview–Revised: A revised Adolescent Psychiatry, 29, 177–183. version of a diagnostic interview for caregivers Piven, J., Palmer, P., Jacobi, D., Childress, D., & of individuals with possible pervasive devel- Arndt, S. (1997). Broader autism phenotype: opmental disorders. Journal of Autism and De- Evidence from a family history study of mul- velopmental Disorders, 24, 659–685. tiple-incidence autism families. American Jour- Lueckling, R. G., & Fabian, E. S. (1997). Transi- nal of Psychiatry, 154, 185–190. tion from school to work. In S. M. Pueschel Piven, J., Palmer, P., Landa, R., Santengelo, S., Ja- & M. Sustrova (Eds.), Adolescents with Down cobi, D., & Childress, D. (1997). Personality syndrome: Toward a more fulfilling life (pp. 235– and language characteristics in parents from 243). Baltimore: Brookes. multiple-incidence autism families. American Ly, T. M., & Hodapp, R. M. (2002). Maternal at- Journal of Medical Genetics, 74, 398–411. tribution of child noncompliance in children Poehlmann, J., Clements, M., Farsad, V., Abbe- with mental retardation: Down syndrome ver- duto, L., & Murphy, M. M. (2003). Family sus other causes. Journal of Developmental and experiences and maternal resolution of their child’s Behavioral Pediatrics, 23, 322–329. diagnosis of fragile X syndrome or Down syn- Marcovitch, S., Goldberg, S., MacGregor, D., & drome. Paper presented at the annual Gatlin- Lojkasek, M. (1986). Patterns of temperament burg Conference on Research and Theory in in three groups of developmentally delayed Mental Retardation and Developmental Dis- preschool children: Mother and father rat- abilities, Annapolis, MD. ings. Developmental and Behavioral Pediatrics, 7, Pueschel, S. M. (1996). Young people with Down 247–252. syndrome: Transition from childhood to Mazzocco, M. M. M. (2000). Advances in re- adulthood. Mental Retardation and Develop- search on the fragile X syndrome. Mental Re- mental Disabilities Research Reviews, 2, 90–95. tardation and Developmental Disabilities Research Radloff, L. S. (1977). The CES-D scale: A self- Reviews, 6, 96–106. report depression scale for research in the gen- Mazzocco, M. M. M., Kates, W. R., Baumgard- eral population. Applied Psychological Measure- ner, T. L., Freund, L. S., & Reiss, A. L. (1997). ment, 1, 385–401. Autistic behaviors among girls with fragile X Roy, J. C., Johnsen, J., Breese, K., & Hagerman, syndrome. Journal of Autism and Developmental R. (1995). Fragile X syndrome: What is the Disorders, 27, 415–435. impact of diagnosis on families? Developmental Nolin, S. L., Brown, W. T., Glicksman, A., Brain Dysfunction, 8, 327–335. Houck, G. E., Jr., Gargano, A. D., Sullivan, Ryde-Brandt, B. (1991). Defence strategies and A., Binacalana, V., Brondum-Nielsen, K., anxiety in mothers of disabled children. Eu- Hjalgrim, H., Holinski-Feder, E., Kooy, F., ropean Journal of Personality, 5, 367–377. Longshore, J., Macpherson, J., Mandel, J. L., Ryff, C. D., & Seltzer, M. M. (1996). The un- Matthijs, G., Rousseau, F., Steinbach, P., Vais- charted years of midlife parenting. In C. D.

᭧ American Association on Mental Retardation 253 VOLUME 109, NUMBER 3: 237–254 ͦ MAY 2004 AMERICAN JOURNAL ON MENTAL RETARDATION Well-being and coping of mothers L. Abbeduto et al.

Ryff & M. M. Seltzer (Eds.), The parental ex- handbook of mental deficiency, psychological theory perience at midlife (pp. 3–25). Chicago: Uni- and research (pp. 405–437). Mahwah, NJ: Erl- versity of Chicago Press. baum. Seltzer, M. M., Greenberg, J. S., & Krauss, M. W. Volkmar, F. R., Cicchetti, D. V., Dykens, E., Spar- (1995). A comparison of coping strategies of row, S. S., Leckman, J. F., & Cohen, D. J. aging mothers of adults with mental illness or (1988). An evaluation of the Autism Behavior mental retardation. Psychology and Aging, 10, Checklist. Journal of Autism and Developmental 64–75. Disorders, 18, 81–97. Seltzer, M. M., Krauss, M. W., Orsmond, G. I., Walker, L. S., Van Slyke, D. A., & Newbrough, J. & Vestal, C. (2000). Families of adolescents R. (1992). Family resources and stress: A com- and adults with autism: Uncharted territory. parison of families of children with cystic fi- In L. M. Glidden (Ed.), International Review of brosis, diabetes, and mental retardation. Jour- Research on Mental Retardation (Vol. 23, pp. nal of Pediatric Psychology, 17, 327–343. 267–294). San Diego: Academic Press. Wikler, L. M. (1986). Periodic stresses of families Seltzer, M. M., Krauss, M. W., & Tsunematsu, N. of older mentally retarded children: An ex- (1993). Adults with Down syndrome and their ploratory study. American Journal of Mental aging mothers: Diagnostic group differences. Deficiency, 90, 703–706. American Journal on Mental Retardation, 97, York, A., von Fraunhofer, N. T. J., & Sedgwick, 464–508. P. (1999). Fragile-X syndrome, Down’s syn- Shonkoff, J. P., Hauser-Cram, P., Krauss, M. W., drome and autism: Awareness and knowledge & Upshur, C. (1992). Development of infants amongst special educators. Journal of Intellec- with disabilities and their families: Implica- tual Disability Research, 43, 314–324. tions for theory and service delivery. Mono- graphs of the Society for Research in Child Devel- Received 8/30/02, accepted 8/20/03. opment, 57, Serial No. 6. Editor-in-charge: Frank Floyd Simon, J. A., Keenan, J. M., Pennington, B. F., Taylor, A. K., & Hagerman, R. J. (2001). Dis- The research reported here was supported by Na- course processing in women with fragile X tional Institutes of Health Grants P01 HD03352, syndrome: Evidence for a deficit establishing R01 HD24356, R01 AG08768, and T32 HD07489 coherence. Cognitive Neuropsychology, 18, 1– and by the University of Wisconsin-Madison 18. Graduate School Research Committee and the Sobesky, W. E., Pennington, B. F., Porter, D., Merck Scholars II program. The authors are grate- Hull, C. E., & Hagerman, R. J. (1994). Emo- ful to Loredana Bruno, Stephanie Cawthon, Sel- tional and neurocognitive deficits in fragile X. ma Karadottir, Rene Makuch, Anne O’Brien, Er- American Journal of Medical Genetics, 51, 378– ica Kesin Richmond, Dotty Robison, and Mich- 385. elle Weissman for their assistance with data col- Sobesky, W. E., Porter, D., Pennington, B. F., & lection and management. We are indebted to the Hagerman, R. J. (1995). Dimensions of shy- families who participated for their support, com- ness in fragile X females. Developmental Brain mitment, and generosity. Requests for reprints Dysfunction, 8, 280–292. should be sent to L. Abbeduto, Waisman Center, Stoneman, Z. (1997). Mental retardation and fam- University of Wisconsin-Madison, Madison, WI ily adaptation. In W. E. MacLean (Ed.), Ellis’ 53705. E-mail: [email protected].

254 ᭧ American Association on Mental Retardation