The Association Between Childhood Physical Abuse and Dyslexia

Introduction

Despite the purported association between dyslexia and psychosocial difficulties (Sundheim & Voeller, 2004), there have been few large-scale epidemiological studies linking dyslexia to a history of childhood physical abuse in the adult population. This is an intriguing question given the relatively high prevalence of reading problems or dyslexia during childhood (Leach, Scarborough, & Rescorla, 2003; Shaywitz, 2004), and the finding that children with various disabilities are 3 to 4 times more likely to be physically abused than nondisabled peers (Sullivan & Knutson, 2000). Conversely, individuals who have been maltreated, especially those with manifest posttraumatic stress disorder (PTSD), reportedly have lower academic skills (De Bellis, Hooper, Spratt, & Woolley, 2009; Kinard, 1999, 2001; Kurtz, Gaudin, Wodarski, & Howling, 1993; Leiter & Johnsen, 1994, 1997; Perez & Widom, 1994; Shonk & Cicchetti, 2001; Slade & Wissow, 2007) including dyslexia, even when controlling for such variables as age, sex, race, and social class (Perez & Widom, 1994). Furthermore, there appears to be significant neurological vulnerability for individuals who have been maltreated, abused, or traumatized as children, especially with respect to brain structure and associated functions (De Bellis & Thomas, 2003; Teicher et al., 2003). Specifically, maltreatment in children has been shown to produce abnormal neurodevelopment and associated dysfunction in key areas related to learning including the prefrontal cortex (Richert, Carrion, Karchemskiy, & Reiss, 2006), hippocampus (De Bellis, Hall, Boring, Frustaci, & Moritz, 2001), and corpus callosum (Villarreal et al., 2004), potentially leading to inattention, executive dysfunction, impaired memory, and slower information processing. These neurobiological factors may influence the development of reading and associated academic skills (e.g., spelling, written language).

This study was designed to examine the relationship between a reported diagnosis of dyslexia and childhood physical abuse in a large population-based epidemiological sample of adults. Based on the available literature examining the connection between maltreatment and academic performance (e.g., Leach et al., 2003; Shaywitz, 2004; Sullivan & Knutson, 2000), it is hypothesized that a significantly higher rate of dyslexia will be reported in individuals who have a reported history of physical abuse versus those who do not have such a history. If present, this relationship would provide additional support for the existence of a significant association between physical abuse and dyslexia and, more specifically, provide information on the magnitude of that association in a representative community-based sample.

Method

Results

Respondents who reported they had been physically abused during their childhood or adolescence were more likely to report they had been diagnosed by a health care professional with dyslexia. Only 7.2% of those without reported abuse indicated dyslexia in sharp contrast to 34.8% of those who reported physical abuse during their childhood (χ² = 93.7; p < .001).

As can be seen in Table 1, the initial adjustments for age, sex, and race produced an OR for dyslexia that was more than 7 times higher (OR = 7.09; 95% confidence interval [CI] = [4.42, 11.35]) for those who reported they had been physically abused in comparison with their non-abused peers. When additional adjustments were made for other adverse childhood experiences, including parental divorce, parental addictions, and long-term parental unemployment, the odds of dyslexia for those who had been physically abused decreased only slightly to sixfold that of non-abused respondents (OR = 6.09; 95% CI = [3.58, 10.35]). For each additional year of age, the odds of dyslexia decreased by 4% (OR = 0.96, 95% CI = [0.94, 0.97]). None of the other control variables (i.e., gender, race, parental divorce, parental addiction, parental long-term unemployment) was significantly associated with dyslexia.

OPEN IN VIEWER

Table 1.

Logistic Regression of Childhood Physical Abuse and Other Adverse Childhood Experiences and Demographic Factors Associated With Self-Report of a Health Professional’s Diagnosis of Dyslexia.

Variable	Adjusted for Age, Sex, and Race			Adjusted for Other Adverse Childhood Experiences, Age, Sex, and Race
	OR	Lower CI	Upper CI	OR	Lower CI	Upper CI
Physically abused as a child	7.09 ***	4.42	11.35	6.09 ***	3.58	10.35
Not physically abused	1.00	Ref	Ref	1.00	Ref	Ref
Male	1.45	0.93	2.26	1.46	0.93	2.27
Female	1.00	Ref	Ref	1.00	Ref	Ref
Age in years	0.96	0.94	0.97	0.96	0.94	0.97
White	1.10	0.62	1.95	1.10	0.62	1.96
Non-White	1.00	Ref	Ref	1.00	Ref	Ref
Parental divorce				1.47	0.86	2.49
No parental divorce				1.00	Ref	Ref
Parental addiction				1.56	0.91	2.68
No parental addiction				1.00	Ref	Ref
Parental unemployment				0.53	0.26	1.10
No parental unemployment				1.00	Ref	Ref
Constant	.02			.02

Note. Unweighted sample size = 12,750. OR = odds ratio; CI = confidence interval.

***

P < .001

Discussion

This study examined the relationship between reports of childhood physical abuse and the presence of dyslexia in a large epidemiological sample. Findings revealed sevenfold age–race–sex adjusted odds of dyslexia among adults reporting they had been physically abused during their childhood in comparison with their non-abused peers. These extremely high odds were not substantially attenuated by adjustment for several other adverse childhood experiences, age, race, or sex. These results raise significant questions with respect to the relationship between the brain changes that have been associated with physical abuse, and other types of events that contribute to brain impairment (e.g., substance abuse; Severtson, Hedden, Martins, & Latimer, 2012), and the appearance of learning problems, particularly dyslexia. For some children, the presence of dyslexia and related learning problems may place them at relatively higher risk for physical abuse, perhaps secondary to adult frustrations with chronic learning failure, child frustrations, and increased antagonistic behaviors with caregivers, associated attention-deficit hyperactivity disorder (ADHD) and related social-emotional difficulties, poor parent coping mechanisms, or increased neurological vulnerability (Gilger & Kaplan, 2008). Conversely, dyslexia is a neurodevelopmental disorder, with the core phonological processing deficits being present from early preschool (Mann & Foy, 2003). Given the known association between brain dysfunction and maltreatment (e.g., De Bellis & Thomas, 2003), it is speculated that the experience of physical abuse may also contribute to and/or exacerbate such learning problems, secondary to increased neurologic burden. At this juncture, the current findings do not point toward an etiological direction for this relationship but, from a public policy perspective, the findings do suggest that children and adults with a history of dyslexia should be screened for abuse, and those with a history of abuse should be screened for dyslexia or associated learning problems.

The large, population-based sample is a major strength of this study. The large sample size has allowed for detailed examination of the co-occurrence of two conditions (childhood physical abuse and dyslexia) and provided representative information of this association; however, there are several limitations of this study. First, for both dyslexia and physical abuse only self-report was available. The self-report of a medical diagnosis of dyslexia likely is an underrepresentation of the larger population of individuals with dyslexia, particularly as children are more likely to be diagnosed by an educational specialist rather than a health professional. The self-reported dyslexia rate of less than 1% was significantly below the 10% to 17% rates reported in either the child (Shaywitz, 2004) or adult (Katz, Goldstein, & Beers, 2001) literature, but probably represented individuals on the more severe end of the continuum. Although it is impossible to determine exactly how underreporting of dyslexia affected the association we found, it is likely it would bias the findings toward the null. Future research would benefit from verification of the dyslexia diagnosis through testing rather than through self-report of a medical diagnosis and an expansion to include other forms of reading disorder as well.

Second, childhood physical abuse was based on retrospective self-reports. Other research has indicated that retrospective reports of childhood abuse are vulnerable to underreporting (Berger, Knutson, Mehn, & Perkins, 1988; Fergusson, Horwood, & Woodward, 2000). Misclassifications in this direction would tend to bias results toward the null. An additional limitation is that the question on childhood abuse asked whether individuals, before the age of 18, had been “physically abused by someone close to” them. It is possible that some respondents categorized as “abused” may have been reporting dating violence they experienced at age 16 or 17. Consequently, the results need to be interpreted with caution and further replication is needed using other population-based studies which incorporate medical examination and access to child welfare records. Our confidence in our findings is stronger due to the similarity of our conclusions to those reported in a review of Omaha student files: That study concluded that there is a strong link between substantiated childhood maltreatment (which includes physical and sexual abuse and neglect) and learning and communication disabilities which had been verified by a multidisciplinary evaluation team (Sullivan & Knutson, 2000). Although a number of critical factors were controlled in our model, it was not possible to control for the presence of other problems, such as math disabilities, other psychiatric disorders, or pharmacological use, nor the chronicity, type, and severity of physical abuse including when the abuse occurred and who was the perpetrator, nor was information available on whether other types of childhood abuse occurred or whether the respondent had an impoverished childhood. These variables could have had an impact on our findings and will require further study.

In summary, findings from the current large-scale study document a six- to sevenfold higher odds of dyslexia among individuals who had been physically abused as a child or adolescent. Approximately one third of those who reported childhood physical abuse had dyslexia. Although the direction of this relationship will require ongoing examination using longitudinal data, the findings suggest that routine screening for physical abuse may be appropriate for children with dyslexia. This will be especially important for primary care providers and school-based practitioners, as they typically will be the first professionals to encounter dyslexia for any particular child. Future studies should focus on the developmental timing and chronicity of the physical abuse, and what specific underlying neurological mechanisms may be contributing to this relationship.