Personality Type and Chronic Pain: The Relationship between Personality Profile and Chronic Low Back Pain Using Eysenck’s Personality Inventory

Round 1

Reviewer 1 Report

This study aims to explore the relationship between personality, clinical factors (pain and disability) and pain-related psychological factors (fear of movement and catastrophism) among chronic low back pain patients. After reviewing the manuscript, I have the following comments that should be addressed:

1.        The introduction section is disorganized. The authors begin by discussing personality (lines 37-43), then discuss the impact of chronic pain (lines 44-50) and the contextual factors that influence pain perception, highlighting the psychological factors of anxiety and depression (lines 51-56), and then return to personality (lines 57-73).

2.        The “four temperaments theory” (lines 70-72) is an obsolete and invalid theory. Eysenck developed the so-called PEN (Psychoticism, Extraversion and Neuroticism) model. Moreover, this information is irrelevant for this study (because personality was measured, not temperaments).

3.        Not all variables of interest (i.e., disability and fear of movement) are well justified (lines 76-77).

4.        The description of the outcome measures should be improved (lines 84-88). The 11-points NPRS is a tool to rate pain intensity, the ODI to measure disability, the TSK and the PCS to measure pain-related psychological factors (in this case, fear of movement and catastrophism), and the EPI to assess personality. It is also necessary to add cite reference 18 in the list of outcome measures. 

5.        In the introductory section, the authors highlight anxiety and depression as contextual factors in pain perception, but do not mention anything about catastrophizing. Instead, in their study they assess only catastrophizing (and not anxiety and depression). Why?

6.        The authors only provide data on the reliability and validity of the EPI (lines 97-99). Why?

7.        Authors should provide more details about inclusion and exclusion criteria (lines 105-106).

8.        The authors should explain why “a Mann-Whitney U test” is run to determined differences only “in the PCS, and TSK by neurotic and stable personality types” (lines 118-120).

9.        The authors should explain why they converted continuous data into binary variables (line 121).

10.    In lines 144-145, the authors mention that “Current, best, and worst pain intensity were averaged to determine the average NPRS for each participant”. This information should be indicated when describing the tool used to measure pain intensity (lines 86-87).

11.    In lines 175-176, the authors mention that “Collectively, neuroticism has a significant positive association with catastrophizing, disability, and kinesiophobia”. This information should be before “…A Mann-Whitney U test,…” (line 172).

12.    What does it mean “…pain intensity relative to disability” (lines 184-185 and lines 186-187)?

13.    Table 1 should be improved: adding more relevant information (sex, race, years of pain), standardizing the use of decimals, including the statistics used, the significance (i.e., p value) and the effect size.

14.    Table 2 should also be improved, including information about the statistic used and placing the Spearman’s rank correlation value before the p value. The authors might consider changing the title of this table (“Results of the correlation analysis between personality, clinical factors (pain and disability) and pain-related psychological factors (fear of movement and catastrophism)”.

15.    To run the cluster analysis, the authors averaged “current, best, and worst pain intensity…”. Why is the average of pain intensity not used to perform the Spearman’s rank correlation analysis? Why in Table 2 is only “NPRS-W and NPRS-B” presented? Why is current pain intensity data not included?

16.    Table 3 should also be improved, including information about the statistic used and placing the mean and SD values before the p values and effect size. The authors might also consider changing the title of this table. The authors should specify how the effect size was calculated.

17.    Table 4 should also be improved, normalizing the way in which mean values and SD values are presented in all tables, adding more relevant information (years of pain).

18.    The authors should adapt Figure 1 and Figure 2 to Eysenck's PEN model.

19.    The section is disorganized. The authors begin by discussing EPI-N correlation results (lines 243-251), then discuss the EPI-E correlation results (lines 252-260) and the return to EPI-N.

20.    The authors include areas of future research. I strongly recommend that the authors rely on Eysenck's PEN model. What type of low back pain was included in the present study?

Author Response

1. The introduction section is disorganized. The authors begin by discussing personality (lines 37-43), then discuss the impact of chronic pain (lines 44-50) and the contextual factors that influence pain perception, highlighting the psychological factors of anxiety and depression (lines 51-56), and then return to personality (lines 57-73).

--Thank you for the input.  We have reorganized the introduction to improve the organization.    

 

2. The “four temperaments theory” (lines 70-72) is an obsolete and invalid theory. Eysenck developed the so-called PEN (Psychoticism, Extraversion and Neuroticism) model. Moreover, this information is irrelevant for this study (because personality was measured, not temperaments).

--The authors would like the thank the reviewer for their perspective.  However, the entire analytical approach was based on Eysenck’s original work.  We feel it is important to include this information for context prior to introducing the methodological approach that considers the 4 quadrants.

 

3. Not all variables of interest (i.e., disability and fear of movement) are well justified (lines 76-77).

--The authors agree and appreciate the reviewer’s perspective.  We have included expanded discussion of the variables of interest in an attempt to justify evaluation of the different constructs.

 

4. The description of the outcome measures should be improved (lines 84-88). The 11-points NPRS is a tool to rate pain intensity, the ODI to measure disability, the TSK and the PCS to measure pain-related psychological factors (in this case, fear of movement and catastrophism), and the EPI to assess personality. It is also necessary to add cite reference 18 in the list of outcome measures. 

--Thank you for this input.  We have expanded the description for each of the outcome measures and included the Gabry’s citation.

 

5. In the introductory section, the authors highlight anxiety and depression as contextual factors in pain perception, but do not mention anything about catastrophizing. Instead, in their study they assess only catastrophizing (and not anxiety and depression). Why?

---Thank you for your comment, depression and anxiety have all been shown to impact the pain experience, especially as it relates to the development of chronic pain and therefore, we mentioned it in the introduction however evaluation of these conditions was beyond the purpose of the manuscript.

 

 

6. The authors only provide data on the reliability and validity of the EPI (lines 97-99). Why?

--The authors decided the report the reliability and validity of the EPI because that was the primary tool utilized in the study to evaluate personality.  It was our intention to report the EPI as a valid and reliable assessment.

 

7. Authors should provide more details about inclusion and exclusion criteria (lines 105-106).

--Thank you for the suggestion.  We have edited this section to reflect the inclusion and exclusion criteria as reported in the IRB documents and adhered to for subject recruitment.

 

8. The authors should explain why “a Mann-Whitney U test” is run to determined differences only “in the PCS, and TSK by neurotic and stable personality types” (lines 118-120).

--Thank you for pointing this out.  We have included additional information to justify the reason for using a Mann-Whitney U test in our analysis of these particular variables.

 

9. The authors should explain why they converted continuous data into binary variables (line 121).

--Thank for the reviewer inquiry.  We have edited this section to improve readability and clarity for how the authors manipulated the variables.

 

10. In lines 144-145, the authors mention that “Current, best, and worst pain intensity were averaged to determine the average NPRS for each participant”. This information should be indicated when describing the tool used to measure pain intensity (lines 86-87).

--Thank you for pointing this out.  We have included the additional information as requested for clarity.

 

11. In lines 175-176, the authors mention that “Collectively, neuroticism has a significant positive association with catastrophizing, disability, and kinesiophobia”. This information should be before “…A Mann-Whitney U test,…” (line 172).

--Thank you.  This sentence has been moved as requested.

 

12. What does it mean “…pain intensity relative to disability” (lines 184-185 and lines 186-187)?

--Thank you for the inquiry.  We have edited this section to improve clarity.

 

13. Table 1 should be improved: adding more relevant information (sex, race, years of pain), standardizing the use of decimals, including the statistics used, the significance (i.e., p value) and the effect size.

--Thank you for the suggestion.  We have improved the information presented including additional demographic information and standardization of significance.  The intention of this table was only to report baseline demographic information and therefore we did not include p values or effect sizes in this particular table however, some of this information is included in the other tables.

 

14. Table 2 should also be improved, including information about the statistic used and placing the Spearman’s rank correlation value before the p value. The authors might consider changing the title of this table (“Results of the correlation analysis between personality, clinical factors (pain and disability) and pain-related psychological factors (fear of movement and catastrophism)”.

--Thank you for the suggestion.  We have edited the table as requested. 

 

15. To run the cluster analysis, the authors averaged “current, best, and worst pain intensity…”. Why is the average of pain intensity not used to perform the Spearman’s rank correlation analysis? Why in Table 2 is only “NPRS-W and NPRS-B” presented? Why is current pain intensity data not included?

--Thank you for this comment.  Our intention with the cluster analysis was to have a broad understanding of our participants fell in their respective subgroups.  While this was helpful for evaluating subgroups we felt report the Worst and Best would provide the reader broader context when it comes to the range of pain participants reported.

 

16. Table 3 should also be improved, including information about the statistic used and placing the mean and SD values before the p values and effect size. The authors might also consider changing the title of this table. The authors should specify how the effect size was calculated.

--Thank you for the suggested.  Table 3 has been edited as recommended.

 

17. Table 4 should also be improved, normalizing the way in which mean values and SD values are presented in all tables, adding more relevant information (years of pain).

--Thank you.  We have edited the table to standardize the way means and SD are presented.  Unfortunately, we did not collect years of pain and therefore unable to include this information in the table.

 

 

 

18. The authors should adapt Figure 1 and Figure 2 to Eysenck's PEN model.

--Thank you for the suggestion.  However, the entire analytical approach used Eysenck’s original work and therefore we believe it is important to include figures that related to the way in which the data was analyzed. 

 

19. The section is disorganized. The authors begin by discussing EPI-N correlation results (lines 243-251), then discuss the EPI-E correlation results (lines 252-260) and the return to EPI-N.

--Thank you for pointing this out.  We have edited this section to improve clarity and flow by aligning discussion of EPI-N and EPI-E

 

20.    The authors include areas of future research. I strongly recommend that the authors rely on Eysenck's PEN model. What type of low back pain was included in the present study?

--This point is well taken.  We do plan to utilize Eysenck’s PEN model in our future research endeavors. 

 

 

Reviewer 2 Report

 

The paper presents a study investigating the relationship between personality type and pain experience in patients experiencing chronic low back pain.  The research involved one hundred twenty four participants completed the Eysenck Personality Inventory,  the Oswestry Disability Index (ODI), Tampa Scale for Kinesiophobia (TSK), Numeric Pain Rating Scale (NPRS), and Pain Catastrophizing Scale (PCS).   Despite discussing the limitations of some related studies, there are a number of related papers that need to be discussed and compared against, such as:   Ibrahim, M. E., Weber, K., Courvoisier, D. S., & Genevay, S. (2020). Big five personality traits and disabling chronic low back pain: association with fear-avoidance, anxious and depressive moods. Journal of pain research, 13, 745.   Sivik, T. M. (1991). Personality traits in patients with acute low-back pain. Psychotherapy and psychosomatics, 56(3), 135-140.
  Clark, J. R., Nijs, J., Yeowell, G., Holmes, P., & Goodwin, P. C. (2019). Trait sensitivity, anxiety, and personality are predictive of central sensitization symptoms in patients with chronic low back pain. Pain Practice, 19(8), 800-810.
  THe authors provided a detailed explanation of their methodology and study design decisions.  However, there are points that were not appropriately justified. For instance, why did the authors only considered the BMI threshold of 25?   The authors should also provide justification as to why they decided to use EPI as the personality inventory to base their analyses.  Why were other inventories not appropriate for this study?   The paper did not include keywords in its current version.   et al -> et al.

Author Response

Despite discussing the limitations of some related studies, there are a number of related papers that need to be discussed and compared against, such as:   

>Ibrahim, M. E., Weber, K., Courvoisier, D. S., & Genevay, S. (2020). Big five personality traits and disabling chronic low back pain: association with fear-avoidance, anxious and depressive moods. Journal of pain research, 13, 745.   

>Sivik, T. M. (1991). Personality traits in patients with acute low-back pain. Psychotherapy and psychosomatics, 56(3), 135-140.

>Clark, J. R., Nijs, J., Yeowell, G., Holmes, P., & Goodwin, P. C. (2019). Trait sensitivity, anxiety, and personality are predictive of central sensitization symptoms in patients with chronic low back pain. Pain Practice, 19(8), 800-810.

---Thank you for the recommendation.   We have included the these studies as part of our discussion section.

 

For instance, why did the authors only consider the BMI threshold of 25?   

---This is an interesting question and one the authors discussed at length.  Ultimately we chose 25 as it is the cut off for the upper end of the health range.   We did conduct further analysis with the variable and therefore did not report any other values.

 

The authors should also provide justification as to why they decided to use EPI as the personality inventory to base their analyses.  Why were other inventories not appropriate for this study?  

---Thank you for the comment.  We chose to use the EPI for several reasons.  The primary being that it is a relatively short assessment.  Our hope is the results of this work could, in time, be utilized as a part of clinical orthopedic practice. However, if the personality assessment is too long it will likely be a burden for clinicians and underutilized.  We understand the EPI has it’s limitations but may be of value the purpose of identifying individual’s that may be more fearful of movement which may prompt an alternate treatment approach.  We included this point in our discussion section.

 

The paper did not include keywords in its current version.  

---Thank you for pointing this out.  We have included keywords.

Round 2

Reviewer 1 Report

The authors addressed most of the suggestions (i.e., the introduction section is more organized, the description of the outcome measures is improved, more details about inclusion and exclusion criteria are included, information about when a Mann-Whitney U test was performed and “current, best, and worst pain intensity” are detailed in the Methods section, and all tables are also improved). However, a minor revision is required:

1.     The tools described in lines 120-122 are not only useful for assessing pain and disability.

2.     The acronyms of the tools described in lines 134-142 can be already used, because they have been previously defined (lines 120-122).

3.     The effect size should be also added in Table 4.

 

Author Response

We would like to thank the reviewers and editorial staff for their time and effort in review of our manuscript.  The comments were very insightful and we believe will make for a stronger manuscript.  We have addressed each of the comments below.  Please let me know if there are further questions or if additional revisions are required.

 

1. The tools described in lines 120-122 are not only useful for assessing pain and disability.

--Thank you, we have updated reference to the include a broader range of constructs evaluated.

 

2. The acronyms of the tools described in lines 134-142 can be already used, because they have been previously defined (lines 120-122).

--Thank you for the comment.  We have made edits to include acronyms for previously defined terms.

 

3. The effect size should be also added in Table 4.

--Thank you, we have calculated and included effect size in Table 4 as requested.