Psychopathology among Emerging Adults with Learning Disabilities in Canada
Round 1
Reviewer 1 Report
Individuals with learning disabilities (LDs) are more likely to have a mental illness, but there is a paucity of studies exploring this association in emerging adulthood. This is an important epidemiological study that take advantage of a large population study in Canada. The results of the study are quite relevant in terms of prevention of psychological distress of emerging adults with LDs. The study report new valuable data on this regard.
Author Response
Specific Comment 1: Individuals with learning disabilities (LDs) are more likely to have a mental illness, but there is a paucity of studies exploring this association in emerging adulthood. This is an important epidemiological study that take advantage of a large population study in Canada. The results of the study are quite relevant in terms of prevention of psychological distress of emerging adults with LDs. The study report new valuable data on this regard.
Response: We thank the reviewer for their positive feedback on the importance of our investigation into LDs and psychological distress among emerging adults.
Reviewer 2 Report
Thank you for the opportunity to review this manuscript. It is interesting, although I do have some questions/suggestions:
- In the abstract, when the results for "male emerging adults" and "those aged 25-29" are presented, the comparison is not clear. These groups had higher distress relative to what otherwise equivalent groups?
- Line 47 - make "have shown" --> "have been shown"
- Lines 77-83 - Typically the results are not presented at the end of the introduction of an article. I would suggest removing this content.
- Lines 127-130 - The authors should discuss the advantages and disadvantages of covarying for lifetime major depressive disorder and generalized anxiety disorder where the outcome of the study is a measure of symptoms of these disorders. The authors' choice to covary for this history may be defensible but it requires discussion and argument. More generally, I'm not sure why anything other than demographics was held constant. The presence of severe psychological distress could cause some of the covariates, such as using medication, receiving help, etc.
- At several points in the manuscript (e.g., Line 121), the authors mention items that asked if family members had "emotions...disorders" Do the authors mean "emotional"?
- Were corrections for familywise Type I error used? If not, why not?
- Lines 252-256 - the authors find it surprising that LDs were associated with distress in males but not females. But this is not the same as finding that males had a generally higher level of distress than females, and so the authors' explanations seem irrelevant, right?
- The authors briefly acknowledge that the self-reported LD diagnosis variable may be vulnerable to "reporting bias" but quickly say that the bias was likely minimal. I find this implausible. The term "learning disability" is not even used consistently across different professionals, let alone laypeople. If the authors wish to make this claim, they would need far more evidence.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 3 Report
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
I appreciate the authors' attention to my initial review report. My remaining suggestions:
1. In their reply letter, the authors defend failing to correct for multiple comparisons by citing a paper by someone who suggests not performing such corrections. Since this is a minority position, the authors should include this explication, defense and citation IN THE PAPER ITSELF so that readers can be aware of it.
2. The authors continue to express surprise that LD was more associated with psychological distress in females, "Given that females are more likely to display internalizing disorders and experience gendered stressors in the workplace and family." This makes no sense. Just because females have a higher level of psychological distress generally, that does not mean that the connection between LD and psychological distress would be stronger in females. The authors should remove this pseudo-explanation.
3. The authors have softened their confidence a bit regarding the accuracy of self-reported LD diagnoses, but not nearly enough. This is a major, major limitation, and I have serious concerns that the authors don't appreciate that. Where is the evidence that people can recall accurately whether they were ever given an LD diagnosis? Or that they understand what the term means? They might refer to ADHD as a learning disability, for instance. Some think that the term refers to intellectual disability/mental retardation and would not report dyslexia as a learning disability. The authors' allusion to the "use of validated measures" is not reassuring - it strikes me as evasive and dismissive of what are very genuine concerns. The authors should not try to defend this; just admit the limitation and move on!
Author Response
Specific Comment 1: In their reply letter, the authors defend failing to correct for multiple comparisons by citing a paper by someone who suggests not performing such corrections. Since this is a minority position, the authors should include this explication, defense and citation IN THE PAPER ITSELF so that readers can be aware of it.
Response: The fact that the reviewer refers to Dr. Kenneth Rothman as “someone” demonstrates a lack of appreciation for the impact Dr. Rothman has had on modern methods in epidemiology and epidemiological research. This is a seminal paper and we have now included a brief statement and citation in the limitations section of the manuscript (lines 319-320). Though we have addressed the reviewer’s concern, we are compelled to task the reviewer to provide their sources of robust evidence that not adjusting for multiple comparisons is a minority position. A proper review would have the reviewer justify their position with the same integrity as the authors.
Specific Comment 2: The authors continue to express surprise that LD was more associated with psychological distress in females, "Given that females are more likely to display internalizing disorders and experience gendered stressors in the workplace and family." This makes no sense. Just because females have a higher level of psychological distress generally, that does not mean that the connection between LD and psychological distress would be stronger in females. The authors should remove this pseudo-explanation.
Response: This statement was not, as the reviewer stated, a “pseudo-explanation”, but rather a speculative statement. In the absence of research in the field of LD and psychological distress among emerging adults, our statement is reasonable. Nonetheless, we have removed this statement from the manuscript.
Specific Comment 3: The authors have softened their confidence a bit regarding the accuracy of self-reported LD diagnoses, but not nearly enough. This is a major, major limitation, and I have serious concerns that the authors don't appreciate that. Where is the evidence that people can recall accurately whether they were ever given an LD diagnosis? Or that they understand what the term means? They might refer to ADHD as a learning disability, for instance. Some think that the term refers to intellectual disability/mental retardation and would not report dyslexia as a learning disability. The authors' allusion to the "use of validated measures" is not reassuring - it strikes me as evasive and dismissive of what are very genuine concerns. The authors should not try to defend this; just admit the limitation and move on!
Response: We do not appreciate the hostile tone of the reviewer as it pertains to this comment. We responded to the reviewer’s comment about ADHD in the first round of feedback. Additionally, line 31 of the manuscript text addresses that the definition of LD uses the North American characterization, and the data are from the CCHS-MH, which is a Canadian study; thus, the response that people may use this term to mean intellectual disability would be more relevant had the study taken place outside of the North American context (i.e., the UK, where the term “learning disabilities” is used to identify intellectual disabilities). Just as the reviewer asks, “Where is the evidence that people can recall accurately whether they were ever given an LD diagnosis?” We ask the reviewer, “Where is the evidence that people cannot accurately recall such a diagnosis.” Such a rebuttal can be applied to each of the reviewer’s comments/questions. As we noted previously, reviewers must justify their position/comments with the same integrity as authors. The peer-review process is meant to be a dialogue to strengthen the knowledge base and not merely the communication of demands from a reviewer to authors. We continue to defend the methods used in the study, as this is the appropriate action, as opposed to the reviewer’s dismissive remark to, “just admit the limitation and move on!” We have further added to the limitations that despite the use of self-reported diagnoses, concerns about this approach are negligible. The manuscript now reads (lines 324-328):
“Validating LD diagnoses using administrative or physician records was not possible in this study. We note that the use of self-reported diagnoses may be problematic for chronic conditions that have very low prevalence in the population [54], which can lead to potential misclassification and the attenuation of measures of association. However, evidence suggests that this information bias is negligible [55].”
- Muggah, E.; Graves, E.; Bennett, C.; Manuel, D.G. Ascertainment of Chronic Diseases Using Population Health Data: A Comparison of Health Administrative Data and Patient Self-Report. BMC Public Health 2013, 13, 1–8, doi:10.1186/1471-2458-13-16.
- Baumeister, H.; Kriston, L.; Bengel, J.; Clinical, M.H.-J. of; 2010, U. High Agreement of Self-Report and Physician-Diagnosed Somatic Conditions Yields Limited Bias in Examining Mental–Physical Comorbidity. J. Clin. Epidemiol. 2010, 63, 558–565.
