Skip to Content

Office of the Vice President for Research

puzzle pieces on chalkboard

Maternal Depressive Symptoms Predicting Development of Internalizing Behavior Problems in Boys with Fragile X Syndrome

Sara Deal & Mary Blair Dellinger



Individuals with fragile X syndrome (FXS) often exhibit increased internalizing problem behaviors that substantially impact functioning. Research suggests that maternal depressive symptoms can increase the development of internalizing problem behaviors in their children. The present study examined whether (1) maternal depressive symptoms related to the degree of internalizing behaviors exhibited in their sons with FXS, and (2) age as an influencing factor for the development of internalizing behaviors. Child internalizing behaviors were measured in 49 male children with FXS (range 1.08-5.75 yrs). Maternal depressive symptoms were measured in each of their mothers with the FMR1 premutation (range 20.55-40.74 yrs). Increased maternal depressive symptoms did not relate to increased internalizing behavior problems in their sons with FXS. Age was not a factor, and when separated by older and younger ages, the results were not significant. These findings are important for children with FXS due to the fact that maternal depressive symptoms did not impact the development of internalizing problem behaviors as seen in typically developing children.



Family systems theory suggests that individuals are understood as a part of their family and views the family as the larger social and emotional unit. The transaction that occurs between the child and their mother and the impact each has on the others wellbeing is evident in families with children who are typically developing and those with children that have disabilities. Family systems theory can be used to explain the importance of the interaction between mothers and their children and the influence this relationship can have on the child’s wellbeing.

One of the ways in which families are affected in a transactional manner is when a family member has a mental illness such as depression. Depression can be characterized by depressed mood and/or loss of interest in life activities and symptoms that cause clinically significant impairments in social, work, or other important areas of function (APA, 2000).The community prevalence rate of women who experience depression at some point during their life is high, affecting between 20% and 23%, and most often occurring during the childbearing years (Ertel, Rich-Edwards, & Koenen, 2011). Ertel and colleagues state that approximately 1 in 10 children are reported as having a mother with depression; thus maternal depression is a major public health concern.One hypothesis for the cause of depression indicates that elevated daily stress partially contributes to increased likelihood for depression (Singer, 2006). Therefore there is a correlation between the stress of women of child-bearing age and incidence of maternal depression.

Maternal depression has strong implications on the emotional and behavioral wellbeing of children with evidence documenting the adverse effects depression can have on children (Leckman-Westin, Cohen, & Stueve, 2009). Adverse effects include internalizing behavior problems such as anxiety, fear, and sadness. These problems may begin to occur in children as early as preschool. A longitudinal study with 92 mothers assessed for symptoms of depression as measured by the Centre for Epidemiological Studies – Depression Scale (CES-D) is one example of the effect maternal depression can have on children. Mother’s depressive symptoms were measured when their children were 4, 12, and 15 months, as well as 4 years old. At the 4 year old assessment, results showed that children who were exposed to chronic maternal depression were rated by their parents and teachers as having significantly more problematic internalizing behaviors as rated by the Child Behavior Checklist (CBCL; Achenbach, 1992). Based on mother reports, the rate of children scoring in the clinical range for internalizing behaviors was five times greater than population-based norms.

One notable finding comes from a study that examined psychosocial precursors and correlates of differing internalizing trajectories (Letchen, Smart, Sanson, & Toumbourou, 2008). The results indicate that infants and toddlers with increasing and high internalizing profiles have higher levels of irritability, shyness, low cooperation, early behavior problems, and mother-baby relationship difficulties, compared to children with low internalizing profiles.

Children who experience internalizing behavior symptoms at a young age may continue to experience problems as they develop, and the symptoms can become more severe as they reach adolescence. If left undetected, these children are at an increased risk for displaying greater levels of internalizing problem behaviors (Trapolini, McMahon, & Ungerer, 2007). A study by Coyne and Thompson (2011) analyzed rating scales measuring depression, locus of control, experiential avoidance, and child internalizing behaviors from 74 mothers with preschool aged children in Head Start schools. The study revealed that mothers who reported higher levels of depression were more likely to report internalizing behavior symptoms in their children.

Overall, maternal depression is one of the best-documented risk factors for childhood internalizing behavior difficulties. Newland and Crnic’s study examined longitudinal and concurrent relations between maternal negative affective behavior and child negative emotional expression in preschool age children with (n=96) or without (n=126) an early developmental risk, as well as the predictions of later behavior problems. Lab tasks were used to observe maternal negative affective behavior and child internalizing emotional expression when the children were 4 and 5 years old. Child internalizing behavior problems were then measured again at age 6 through a maternal questionnaire. Higher levels of maternal and child negativity were observed for the children with a developmental risk. These internalizing behavior problems can be detrimental for children and are associated with poorer outcomes both socially and in school (Hester, Baltodano, Gable, Tonelson, & Hendrickson, 2003). More often than not, these problems continue into adulthood resulting in lower education and social impairment.

While maternal depression is widespread in the general population, specific subgroups appear to be at an increased risk. A study comparing depression in mothers of children with and without developmental disabilities showed that mothers that have children with developmental disabilities have increased incidences of depression (Singer, 2006). Fragile X syndrome (FXS) is an x-linked heritable disorder involving an expansion of more than 200 repeats of the CGG nucleotide sequence on the FMR1 gene (Seltzer et. al., 2011). People with the FMR1 premutation have between 55 to 200 repeats of the CGG segment (Seltzer et. al., 2011) and women with the premutation are at an increased risk of having a child with FXS (Guttmacher, 2010). Research has shown that mothers with the fragile X premutation have an increased risk of displaying depressive symptoms (Seltzer et. al., 2011).  A clinical study of 33 women with the FXS premutation and 20 women without the premutation completed the Brief Carroll Depression Scale and the Multidimensional Anxiety Questionnaire. Both measures showed that women with the FMR1 premutation had a higher tendency for depression (Lachiewicz et al., 2010). Another study found that 47% of women with the FMR1 premutation had a mood disorder compared to 37% of the control group. The study also found that women with the premutation have a significantly higher frequency rate (43%) of major depressive disorder versus mothers without the premutation (32% )(n=93; aged 20-46 years; Roberts et. al., 2009). Overall, the increased maternal anxiety related to increased problem behaviors of children shows how child behavior impacts maternal psychological functioning.

Due to the fact that maternal depression is a predictor of child problem behaviors, it is important to study these relationships in samples at high risk, such as children with FXS. Male children with fragile X are often characterized by increased intellectual delays and problem behaviors (Roberts et al., 2009). A study by Roberts and colleagues (2009) examined the development of young males with FXS (n=55; aged 8-48 months) and found a significantly lower rate of overall development in males with FXS compared to typically developing males. Barajas and colleagues (2001) found that 42% of boys with FXS had social problems in the borderline or clinical range in comparison to 12% of children in the typically developing sibling group. Warren and colleagues (2010) examined the relationship between early maternal responsivity and later child communication outcomes in children with fragile X to show that maternal responsivity significantly influenced the rate of communication by 36 months and also the overall linear growth of child communication. Together, these studies express the impact that maternal behavior can have on the overall development of their children with FXS.

Previous studies have provided evidence to support the relationship between maternal depression and internalizing problem behaviors in typically developing children. However, there is little work focused on the relationship between mothers’ maternal depressive symptoms and the internalizing behaviors specifically in their sons with FXS. To our knowledge, there are no previous studies that have used a cross-sectional approach to examine this relationship. The research question guiding this work is: do maternal depressive symptoms relate to the development of internalizing behavior problems in their sons with FXS?  We hypothesized that higher maternal ratings of depressive symptoms would be related to increased child internalizing behavior problems.





The data were collected from a longitudinal study of family adaptation in FXS conducted at the University of North Carolina- Chapel Hill. The participants included 49 premutation mothers and their sons with the full mutation. The average maternal IQ was 109 (range 73-132). The study included 41 mothers that were married, 3 that were divorced or separated, and 5 mothers that were single or never married. The sample included 43 white mother-son dyads and 6 African American mother-son dyads. See table 1 for complete descriptive statistics of the participants. In total there were 98 participants comprised of 49 mother-son dyads.


Table I







Child Age in Years






  CBCL Internalizing T-score*






  MSEL Mental Age in Years*






  Age in Years












  BDI-II Total Score*







*Mental Age was measured by averaging fine motor, expressive language, receptive language, and visual reception age equivalents on the Mullen Scales of Early Learning; (MSEL; Mullen, 1995). Min = minimum; Max = maximum; CBCL = Child Behavior Checklist (Achenbach, 2000; Achenbach & Rescorla, 2001). BDI-II = Beck Inventory for Depression 2nd Edition (Beck, Steer & Brown, 1996).


Child Internalizing Behavior Problems: The Child Behavior Checklist (CBCL) is a questionnaire completed by the mother about their child and was used to measure internalizing (ex. anxious, depressive, and overcontrolled) behavior problems.The CBCL has shown a high test-re-test reliability (r=0.95), was normed on a large sample, and has been used with a wide variety of children with fragile X (Bargagna, Canepa, & Tinelli, 2002; Hatton et al., 2002; Kau et al., 2004; Wheeler et al., 2007). The CBCL also has several subareas which measure somatic complaints, anxiety and depression, and withdrawal behaviors. Two examples of questions from the measure are as follows: Afraid to try new things and Cries a lot. The CBCL contains 99 questions which each receive a 0 (not true), 1 (somewhat or sometimes true), or 2 (very true or often true), for a total possible minimum score of 0 and a possible maximum score of 198. The clinical cutoff score, the boundary between normal and clinical range, for the CBCL is a total score of greater than or equal to 70; in this study only three participants (6%) had scores higher than 70. This study focuses particularly on the internalizing behavior subscale of the CBCL for a dependent variable.

Maternal Depressive Symptoms: To test for maternal depression the Beck Inventory for Depression 2nd Version (BDI-II) was used (Beck, Steer, & Brown, 1996). This is a self-report form that measures the intensity, severity, and depth of depressive symptoms on a scale of 0 to 3 for each question. An example question is as follows: Sadness, 0- I do not feel sad, 1- I feel sad much of the time, 2- I am sad all the time, and 3- I am so sad or unhappy I can’t stand it. The BDI-II is only a screening method for depression, but can lead to further testing and a clinical diagnosis. The measure has a coefficient alpha of .80 and construct validity has been established to ensure differentiation between depressed and non-depressed participants. Clinical significance is indicated with a cumulative score of 20 or higher, with a score of 40 indicating extreme depression. Only two mothers (4%) met criteria for clinical significance and 0 mothers in the study were classified as having extreme depressive symptoms. For this study the BDI-II served as the independent variable.

 Child Mental Age: To test the mental age of each of the children in the sample the Mullen Scales of Early Learning (MSEL; Mullen, 1995) was used. There are five scales on MSEL that measure gross motor, visual reception, fine motor, expressive language, and receptive language. The measure provides the mental age of the child by averaging mental age scores from each of the sections. For this study the mental age acted as a comparison to the chronological age of each participant to give a sense of the developmental delay exhibit in these children with FXS.


From the larger extant data set, some participants had a single data point whereas others had multiple data points. Due to our interest in early development, the first datapoint was chosen for each participant where there was a concurrent child CBCL score, child MSEL score, and mother BDI-II score. Participants participated in two-day assessments in which several child and mother measures were collected. For the purposes of this study only the CBCL, MSEL, and BDI-II were used. During the assessment, the mother completed the BDI-II regarding personal feelings and experiences and also completed the CBCL corresponding to her son’s actions and behaviors. During each assessment, an examiner administered and scored the MSEL using the Mullen ASSIST computer program. Mullen ASSIST calculates and converts raw scores for easy interpretation. The program IBM SPSS Statistics 19 was used to calculate descriptive statistics and hierarchical multiple regression. Descriptive statistics include the maximum, minimum, range, mean, standard deviation, and variance of the scores. We used a hierarchical multiple regression to assess the relationship between maternal depression scores from the BDI-II and child internalizing problem behavior scores from the CBCL, while controlling for child’s age in months.



Hierarchical multiple regression was used to evaluate the relationship between maternal depression (as measured by the BDI-II) and internal behavior problems in children (as measured by the CBCL), while controlling for child’s age in months. Preliminary analyses were run to ensure that the relationship between maternal depression and child internalizing behaviors was not influenced by the child’s age. Age in months was entered in Step 1, explaining 14.4% of the variance in internalizing behavior problems, F (1,47) = 7.92, R2 = =.14, p < .001. After the BDI-II scores were entered in Step 2 the total variance of the whole model was 14.5%, F (2, 46) = 3.89, p < .001. The BDI-II explained 0.1% of the variance in internalizing problem behaviors, R2 change = .001, D R2=.01%,  F change (1, 46) = .020, p>.05.  In the final model the BDI-II (rating maternal depression) had no significant predicting effect on the internalizing problem behaviors (as rated by the CBCL).



This was a cross-sectional study based on the family systems theory that examined the relationship between maternal depressive symptoms and the development of internalizing problems behaviors in their sons with FXS. We hypothesized that higher level ratings of maternal depressive symptoms would lead to an increased development of child internalizing behavior problems as seen in typically developing samples. Contrary to our hypothesis, elevated depressive symptoms in mothers did not indicate elevated levels of internalizing behavior problems in their sons with FXS. Age was not a factor, and when split into older and younger groups of children, the results were also not significant. We expected to find the higher maternal ratings of depression to be a predicting factor of internalizing behavior problems in their sons. Although this is not what we predicted, these results indicate an important finding for the children with FXS in our study.

Factors such as family problems, child neglect/abuse, parental mental health, and substance abuse history are just a few risk factors that when present, increase a child’s likelihood of developing negative, long-term life outcomes. These negative, long-term life outcomes included internalizing behavior problems. Children with FXS are born with a developmental delay and are at a higher risk for emotional and behavior disorders (Nelson et al., 2007).

One explanation for the results in our study could be attributed to the fact that only 3 male participants (6%) had internalizing behavior scores above the clinical cutoff and only 2 (4%) of the mothers met criteria for clinically significant depressive symptoms. Another explanation could be the fact that the sample of males was very young and though we did examine age effects, there was a restricted age range in our study. Given that internalizing symptoms typically emerge later in development; a relationship between maternal depressive symptoms and child internalizing behavior problems may be present in an older aged sample. The mothers in the study completed both measures used; the CBCL and the BDI-II, which could potentially cause the collected data to be biased.

Given that internalizing symptoms typically emerge later in development; a relationship between maternal depressive symptoms and child internalizing behavior problems may be present in an older aged sample. A study that examined the relationship of toddlers’ negative affect in high- and low- threat contexts to maternal responses to their toddlers’ internalizing behaviors and emotions showed a lack of relation at age 2 (Luebbe, Kiel, & Buss, 2011) . The study states that the lack of relationship may partially be explained developmentally:  displayed negative affect has been connected to mothers’ emotion talk in 3 to 5 year olds; however this relation has not been found in infants although it has been tested.

The lack of chronicity of the depression in mothers with the FXS premutation could have influenced the results in the current study. Roberts and colleagues (2009) found that major depressive disorder (MDD) was more common in women who were not married; almost all the mothers in the present study were married which is yet another factor that could account for the results and may indicate that single mothers are more apt to encounter stress induced depression. Finally, our sample was homogenous and did not include measures of certain risk factors, such as socio-economic status, maternal education levels, or ethnicity information, all of which can also contribute in a child’s development of internalizing behavior problems.



There are currently no studies that examine the impacts of maternal depressive symptoms on the development of internalizing problem behaviors in their male children with fragile X and though we used a relatively large sample, the age range of the children was very young. The young age of the male participants, along with a restricted range made it difficult to examine the development of internalizing problem behaviors over time. Another limitation is the low levels of depressive symptoms displayed by the mothers in the study. A final limitation was the exclusion of externalizing behaviors which can also be measured by the CBCL.


Summary and Future Direction

For future studies, the examination of the relationship between maternal depression and internalizing behavior problems in children with FXS should include a larger, more diverse, and heterogeneous sample that addresses multiple potential risk factors.  The use of additional measures of child internalizing behavior problems completed by an examiner or outside third party could eliminate potential bias. A typically developing comparison group could be added to compare the differing rate of development of internalizing problem behaviors between typically developing children and children with fragile X whose mothers present symptoms of depression. Conducting a longitudinal study to examine the development of internalizing behavior problems, along with externalizing behaviors, and the impact of maternal depression across the developmental process would also produce stronger and potentially more significant results.

About the Authors

Sara DealSara Deal

I am a senior psychology major with a minor in counselor education. I received a Magellan Scholar award to study the comparison of attention disengagement between children at risk for autism and those with no known risk. This paper is an extension of our Discovery Day poster presentation, which examined the implications of maternal depression on the development of internalizing problem behaviors in their son’s with fragile X syndrome. I was responsible for running statistics on the collected data, organizing the information for the paper, and outlining the measures used. This paper has been very beneficial in helping me prepare for graduate school through having the opportunity to conduct a research project, run the appropriate analyses, and present the information in both a poster and publication format. Outside of school I enjoy spending time with family and friends, volunteering at the Autism Academy of South Carolina, and working with children in a variety of settings. My future plans include going to graduate school to work with children and families affected by autism.

Mary Blair DellingerMary Blair Dellinger

This is my senior year of college; I’m a Psychology major with my minor in Neuroscience. I am a member of Sigma Alpha Lambda and a member of PSI CHI. I was awarded the Magellan Scholarship to examine the effects of the novel use of Neurofeedback. This paper is based on the Discovery Day poster presentation presented in Spring of 2012; we transposed the poster into an article and took the work a step further. My roles included finding more scholarly resources related to our topic to enhance the introduction and write the bases of the outline, as well as help in the organization and proof-reading of the material for the paper. Working with my partner and mentor on this paper has been extremely beneficial for me academically in that it has given me a thorough understanding of how to construct a proper research paper, and working well with others to finalize the project. This experience will help me reach my goal of going to graduate school. I enjoy playing the piano, reading, and volunteering in the Applied Cognitive Neuropsychology Lab as well as in the Neurodevelopmental Disorders Lab at the University of South Carolina. My future plans include going to graduate school for Psychology and pursuing a doctorate degree, the submission of this paper relates to my future plans directly.



Achenbach, T. (1992) Manual for the Child Behavior Checklist/2-3 and 1992 profile. Burlington, VT: University of Vermont Department of Psychiatry.

American Psychiatric Association (2000). Mood disorders. In diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: American Psychiatric Association, 69-70.

Barajas, R.G., Dyer-Friedman, J., Glaser, B., Hessl, D., Reiss, A. L., Taylor, A., & Wisbeck, J. (2001). The influence of environmental and genetic factors of behavior problems and autistic symptoms in boys and girls with fragile X syndrome. American Academy of Pediatrics, 108(5), E88. doi: 10.1542/peds.108.5.e88.

Bargagna, S., Canepa, G., & Tinelli, F. (2002). Social adjustment in children with down mental retardation (MRD) and fragile X mental retardation (MRX). Panminerva Medica, 44(1), 7-10.

Beck, A.T., Steer, R.A., & Brown, G.K. (1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation.

Coyne, L. W., & Thompson, A. D. (2011). Maternal depression, locus of control, and emotion regulatory strategy as predictors of preschoolers’ internalizing problems. Journal Of Child And Family Studies20(6), 873-883. doi:10.1007/s10826-011-9455-2

Ertel, K. A., Rich-Edwards, J. W., & Koenen, K. C. (2011). Maternal depression in the United States: Nationally representative rates and risks. Journal Of Women’s Health, 20(11), 1609-1617. doi:10.1089/jwh.2010.2657

Guttmacher, A. E., (2010). National Human Genome Research Institute. Retrieved August 13, 2012, from

Hatton, D. D., Hooper, S. R., Bailey, D. B., Skinner, M. L., Sullivan, K. M., & Wheeler, A. (2002). Problem behavior in boys with fragile X syndrome. American Journal of Medical Genetics, 108, 105-116.

Hatton D. D., Sideris J., Skinner M., Mankowski J., Bailey D. B., Roberts J. et al. (2006) Autistic behavior in children with fragile X syndrome: prevalence, stability, and the impact of FMRP. American Journal of Medical Genetics Part A 140A, 1804–13.

Hester, P. P., Baltodano, H. M., Gable, R. A., Tonelson, S. W., & Hendrickson, J. M. (2003). Early intervention with children at risk of emotional/behavioral disorders: A critical examination of research methodology and practices.

Education & Treatment Of Children (ETC)26(4), 362-381. Kau, A. S., Tierney, E., Bukelis, I., Stump, M. H., Kates, W. R., Trescher, W. H., Kaufmann, W. E. (2004). Social behavior profile in young males with fragile X syndrome: characteristics and specificity. American Journal of Medical Genetics, 126A, 9-17.

Lachiewicz, A. A., Dawson, D. D., Spiridigliozzi, G. G., Cuccaro, M. M., Lachiewicz, M. M., & McConkie-Rosell, A. A. (2010). Indicators of anxiety and depression in women with the fragile X premutation: Assessment of a clinical sample. Journal Of Intellectual Disability Research54(7), 597-610. doi:10.1111/j.13652788.2010. 01290.x

Leckman-Westin, E., Cohen, P. R., & Stueve, A. (2009). Maternal depression and mother child interaction patterns: Association with toddler problems and continuity of effects to late childhood. Journal Of Child Psychology And Psychiatry, 50(9), 1176-1184. doi:10.1111/j.1469-7610.2009.02083.x

Letcher, P., Smart, D., Sanson, A., & Toumbourou, J. W. (2009). Psychosocial precursors and correlates of differing internalizing trajectories from 3 to 15 years. Social Development, 18(3), 618-646. doi:10.1111/j.1467-        9507.2008.00500.x

Luebbe, A. M., Kiel, E. J., & Buss, K. A. (2011). Toddlers’ context-varying emotions, maternal responses to emotions, and internalizing behaviors. Emotion, 11(3), 697-703. doi:10.1037/a0022994

Mullen, E. M. (1995). Mullen Scales of Early Learning (AGS ed.). Circle Pines, MN: American Guidance Service Inc.

Nelson, J., Stage, S., Duppong-Hurley, K., Synhorst, L., & Epstein, M. H. (2007). Risk factors predictive of the problem behavior of children at risk for emotional and behavioral disorders. Exceptional Children73(3), 367-379.

Newland, R. P., & Crnic, K. A. (2011). Mother-child affect and emotion socialization processes across the late preschool period: predictions of emerging behaviour problems. Infant & Child Development20(6), 371-388. doi:10.1002/icd.729

Roberts, J., Bailey, D., Mankowski, J., Ford, A., Sideris, J., Weisenfeld, L., & … Golden, R. (2009). Mood and anxiety disorders in females with the FMR1 premutation. American Journal Of Medical Genetics. Part B, Neuropsychiatric Genetics: The Official Publication Of The International Society Of Psychiatric Genetics150B(1), 130-139.

Roberts, J. E., Mankowski, J. B., Sideris, J., Goldman, B., Hatton, D. D., Mirrett, P. L., & … Bailey, D. r. (2009). Trajectories and predictors of the development of very young boys with Fragile X Syndrome. Journal Of Pediatric Psychology34(8), 827-836. doi:10.1093/jpepsy/jsn129

Seltzer, M., Barker, E. T., Greenberg, J. S., Hong, J., Coe, C., & Almeida, D. (2011). Differential sensitivity to life stress in FMR1 premutation carrier mothers of children with fragile X syndrome. Health Psychology, doi:10.1037/a0026528

Singer, G. S. (2006). Meta-Analysis of comparative studies of depression in mothers of children with and without developmental disabilities. American Journal on Mental Retardation, 111(3), 155-169. doi:10.1352/0895-8017(2006)111[155:MOCSOD]2.0.CO;2

Trapolini, T. T., McMahon, C. A., & Ungerer, J. A. (2007). The effect of maternal depression and marital adjustment on young children’s internalizing and externalizing behavior problems. Child: Care, Health And Development33(6), 794-803. doi:10.1111/j.1365-2214.2007.00739.x

Wheeler, A., Hatton, D., Reichardt, A., & Bailey, D. (2007). Correlated of maternal behaviours in mothers and children with fragile X syndrome. Journal of Intellectual Disability Research, 51(6), 447-62.

Warren, S.F., Brady, N., Sterling, A., Fleming, K., & Marquis, J. (2010). Maternal responsivity predicts language development in young children with fragile X syndrome. American Journal on Intellectual and Developmental Disabilities, 115(1), 54-75.