April 2020: The Association Between Poor Reading and Internalising Problems

Overview and Previous Research

Previous studies have indicated the people with general learning disabilities are at higher risk of internalizing problems (e.g., anxiety, depression) compared to the typical population (Nelson & Harwood, 2011). However, the relationship between one type of specific learning disability–poor reading–and internalizing problems is poorly understood. There have been inconsistent results reported from prior systematic reviews summarizing studies investigating if poor readers are more likely to have internalizing problems than typical readers, and if so, if that increased risk specifically pertains to anxiety, depression or both (Maughan & Carroll, 2006; Mugnaini et al., 2009). Two literature reviews have examined this relationship. One review determined that poor reading is associated with both anxiety and depression, and attention problems increased the risk for internalizing problems (Mugnaini et al., 2009). In another review, poor readers were also at greater risk for anxiety, but the relationship between poor reading and depression was unclear (Maughan & Carroll, 2006). In sum, the association between poor reading and internalizing problems is poorly understood.

Given the inconclusive findings of previous studies and literature reviews, as well as the lack of recent systematic reviews or meta-analyses of this topic, Francis and colleagues (2019) conducted a meta-analysis of the relationship between poor reading and internalizing problems, specifically, anxiety and depression. Additionally, in order to further clarify the association between poor reading and internalizing problems, the authors also sought to evaluate the relationship by examining important moderator variables, such as poor reader subtype, internalizing disorder subtype, and attention problem subtype.

Study Purpose and Research Questions

Francis, Caruana, Hudson, and McArthur (2019) conducted a meta-analysis to better understand the relationships between poor reading and two specific internalizing disorders, anxiety and depression. A meta-analysis is a statistical technique to aggregate the findings of multiple studies examining similar research questions (Petticrew & Roberts, 2008). It is commonly accepted as a way to estimate the overall effect of a specific treatment or, in this case, to estimate the relationship between two variables.

Francis and colleagues (2019) addressed the primary research question:

1. Is there a reliable association between poor reading and internalizing problems, specifically anxiety and depression?

The authors postulated that the inconsistency among results of previous studies could be due to the large degree of heterogeneity among participants’ characteristics, as well as differences between internalizing disorder subtype and measures of internalizing disorder. Therefore, the authors also addressed the following secondary research question:

2. If so, is this association moderated by the following variables:

• Anxiety disorder subtype, as poor reading may be associated with some anxiety subtypes and not others
• Poor reading subtype, as the variation in reading problems may differentiate poor readers with and without internalizing problems
  • Poor reader subtypes included those with singular or combined difficulty with phonological recoding, visual word recognition, reading fluency, and reading comprehension.
  • Poor readers met the subtype criteria if they scored at least one standard deviation below average grade level on reading tests, performed one year below grade level, scored statistically significantly lower on an intelligence test, performed statistically significantly lower than a typical reader control group, or met the diagnostic criteria for reading problems, such as those specified in the DSM.
• Attention difficulty, as poor attention is associated with both poor reading and internalizing problems
• Sex, as females tend to experience more problems with anxiety and depression than males
• Age, as internalizing disorders vary by age
• Ethnicity, as some ethnic minorities experience higher rates of internalizing disorders
• Type of informant of internalizing problems, as collected information on internalizing disorders may vary if self-reported or reported by a parent or teacher
• Type of internalizing measure, as clinical interviews and questionnaires differ in their methods of data collection on symptoms

Study Methodology 

Studies were included in Francis et al.’s (2019) meta-analysis if they included (a) participants who were 6 years or older, (b) source articles identified participants as “poor readers” based on their performance on one or more reading test, and (c) reported internalizing, anxiety or depression data with raw or standardized scores from clinical interviews or questionnaires. A total of 34 studies and 2,491 poor readers met the criteria and were included in the analysis. Results were organized according to the outcome measure of general internalizing disorder symptoms, anxiety, or depression.

Key Findings

A total of 34 studies overall met inclusion criteria: 14 studies analyzed the relation between reading difficulty and internalizing problems, 22 studies analyzed the relation between reading difficulty and anxiety, and 23 studies analyzed the relationship between reading difficulty and depression. Effect sizes in this meta-analysis were reported as Cohen’s d, which allowed for comparison across studies. More information about interpreting effect sizes can be found in this interactive resource: https://rpsychologist.com/d3/cohend. Positive effect sizes were interpreted as showing higher internalizing problems, anxiety or depression, for poor readers as compared to typical readers.

1. Is there a reliable association between poor reading and internalizing problems?

The results of the meta-analysis indicate that poor readers are, on average, at greater risk for experiencing overall internalizing problems. The effect size (d = .41) was statistically significant.
Poor readers are at a greater risk for anxiety than typical readers, a finding consistent with previous research. The effect size (d = .41) was statistically significant.

The authors reported that an effect size of .41 indicates that 66% of the poor readers in the sample will have an anxiety score above the mean of the typical reader control group. There is a 61.4% chance that a person picked at random from the treatment group will have a higher score than a person picked at random from the typical reader group.

The relationship between poor readers and depression was less pronounced (d = .23), indicating that this relationship is weaker than the relation between poor reading and anxiety. The effect size, which was statistically significant, indicates that 59% of the poor readers in the sample will have a depression score above the mean of the typical reader control group. There is only a 56% chance that a person picked at random from the poor reader group will have a higher score than a person picked at random from the typical reader group.

2. If so, is this association moderated by theoretical (e.g. anxiety disorder subtype or poor reading subtype) or methodological moderators (e.g. type of informant)?

• The planned moderator analysis could not be conducted because fewer than 10 studies qualified for each moderator subgroup. However, studies including these moderator variables were reviewed and summarized narratively by Francis and colleagues. The authors reported the following:
  • Poor reading is associated with both generalized and separation anxiety disorders.
  • Poor readers of all ages are at increased risk of anxiety.
  • Type of poor reader, attention, sex, ethnicity, type of informant, and type of internalizing problem measure were not indicated as potential moderators of the relationship between poor reading and anxiety.
• In sum, the effect sizes appear to be uniform across this set of variables, though the small number of studies make these conclusions less definitive.

Implications and Recommendations for Research and Practice

Evidence reported in this study and in previous studies suggests there is an association between internalizing problems and poor reading. The association between poor reading and depression was relatively weak (d = .23), whereas the relationship between poor reading and anxiety was more moderate (d = .41). The variation across studies suggests that although these relations are unlikely the result of chance, further research is required. The authors note that their meta-analysis does not indicate causation, suggesting the relationship between poor reading and internalizing problems could be bidirectional in nature. Research by the Texas Center for Learning Disabilities suggests there is a bidirectional relation between anxiety and poor reading (Grills-Taquechel et al., 2012), wherein anxiety perpetuate difficulties with learning, and such negative experiences with learning maintain or increase anxious thoughts and behaviors. However, as Francis and colleagues note, future research is required to better understand the causal relations between reading performance and internalizing behaviors.

What does this mean in practical terms? There is accumulating evidence that poor readers are more likely than typically performing readers to report experiencing internalized disorders, such as anxiety than depression. With regard to depression, the difference in depression between poor readers and typical readers is quite small, suggesting that poor readers would be less likely to experience depression than anxiety. Further research is needed to better understand the magnitude of these effects and their practical implications for children.

However, given the real impact of internalized disorders on mental wellness and daily activities, even a small increased likelihood of anxiety and depression should be considered when supporting poor readers. It is the opinion of the authors of this summary that, based on the results of the meta-analysis suggesting internalized disorders do co-occur with poor reading, internalized disorders should be considered within the context of intervention for poor readers.  There is valuable opportunity for collaboration between general and special education teachers, school psychologists, clinicians, and parents to more broadly support students with both reading difficulties and internalized disorders, including anxiety and depression. School psychologists could assess poor readers for anxiety and thereby provide helpful information and support in the treatment of anxiety. Furthermore, general and special education teachers who support poor readers might consider providing treatment that targets both poor reading and anxiety. For example, Francis et al. (2019) suggest that teachers could provide strategies for how to lower anxiety before engaging in reading. It should be noted that the authors uncovered an overall lack of intervention research in the treatment of internalizing disorders such as anxiety. With new studies examining the effects of reading interventions that incorporate anxiety management strategies for students with reading difficulties (Vaughn et al., 2020), perhaps future meta-analyses may be able to draw more definitive conclusions about the efficacy of these treatments.

Interested in ways you can support the well-being of your child or students during the COVID-19 public health crisis? Consider taking advantage of the resources freely available at the National Association of School Psychologists COVID-19 Resource Center. 

References

Grills, A. E., Fletcher, J. M., Vaughn, S., Barth, A., Denton, C. A., & Stuebing, K. K. (2014). Anxiety and response to reading intervention among first grade students. Child & Youth Care Forum, 43, 417–431. https://doi.org/10.1007/s10566- 014-9244-3.

Grills-Taquechel, A. E., Fletcher, J. M., Vaughn, S. R., & Stuebing, K. K. (2012). Anxiety and reading difficulties in early elementary school: Evidence for unidirectional-or bi-directional relations? Child Psychiatry & Human Development, 43(1), 35-47.

Magnusson, K. (2020). Interpreting Cohen’s d effect size: An interactive visualization (Version 2.1.1) [Web App]. R Psychologist. https://rpsychologist.com/d3/cohend

Maughan, B., & Carroll, J. (2006). Literacy and mental disorders. Current Opinion in Psychiatry, 19, 350–354. https://doi.org/10.1097/01.yco.0000228752.79990.41.

Mugnaini, D., Lassi, S., La Malfa, G., & Albertini, G. (2009). Internalising correlates of dyslexia. World Journal of Pediatrics, 5, 255–264. https://doi.org/10.1007/s12519-009-    0049-7.

Nelson, J. M., & Harwood, H. (2011). Learning disabilities and anxiety: A meta-analysis. Journal of Learning Disabilities, 44(1), 3-17.

Petticrew, M., & Roberts, H. (2008). Systematic reviews in the social sciences: A practical guide. Blackwell.

Vaughn, S., Grills, A. E., Capin, P, Roberts, G., Mari-Fall, A. M., Daniel, J. (2020). Examining the Effects of Integrating Anxiety Management Instruction Within a Reading Intervention for Upper Elementary Students with Reading Difficulties. [Manuscript submitted for publication]. Department of Special Education, The University of Texas at Austin.