Comment on "evaluation of the prevalence of overactive bladder syndrome in patients with rheumatoid arthritis"

Abstract

We read with interest the study by Cirakoglu and Yuce on overactive bladder (OAB) and central sensitization (CS) in rheumatoid arthritis (RA) patients. While the study provides valuable hypothesis-generating insights, several methodological limitations warrant consideration. The OAB-V8 cut-off of 8 was validated in a primary care population, not in RA patients, and dichotomising this ordinal scale discards information and reduces statistical power. The regression model included correlated variables (HAQ-DI, VAS-pain, DAS-28) without testing for multicollinearity, which can inflate standard errors and produce unreliable p-values. Additionally, the authors did not verify key logistic regression assumptions, including linearity of continuous variables with the logit and absence of influential outliers. These limitations suggest that the findings should be interpreted as hypothesis-generating rather than definitive. Future studies should address these methodological issues to clarify the relationship between CS and OAB in RA.

Keywords
Rheumatoid arthritis; Overactive bladder; Central sensitization; Logistic regression; Multicollinearity; Dichotomisation

Dear Editor

We read with great interest the article by Cirakoglu and Yuce entitled "Evaluation of the prevalence of overactive bladder syndrome in patients with rheumatoid arthritis" [¹]. The study provides important insights into the association between overactive bladder (OAB) and central sensitization (CS) in rheumatoid arthritis (RA) patients. However, several methodological issues warrant discussion as they may influence the interpretation of the findings.

First, the authors used an OAB-V8 cut-off of 8 based on validation in a primary care population [²]. However, this cut-off has not been validated in patients with RA, a population with distinct symptom profiles, medication regimens, and inflammatory burden. The OAB-V8 is a screening tool, not a diagnostic instrument, and its performance characteristics may differ substantially in RA patients. Moreover, dichotomising an ordinal scale discards information, reduces statistical power, and may increase the risk of false-positive findings [³]. The authors themselves reported a borderline p-value for age (p = 0.054), and the OR for CS is very close to 1.0 (1.048). The additional loss of power from dichotomisation may have further compromised the study's ability to detect true associations. An ordinal logistic regression would have preserved the full score range and could have revealed whether the association with CS is consistent across the entire symptom severity spectrum.

Second, the regression model included HAQ-DI, VAS-pain, and DAS-28 simultaneously. These variables are conceptually and statistically correlated. Indeed, a study in RA patients demonstrated that the effect of DAS-28 on functional disability overlaps substantially with the effect of pain VAS [⁴]. Entering all three together without testing for multicollinearity is problematic. Multicollinearity inflates standard errors, produces unreliable p-values and confidence intervals, and can make individual predictors appear non-significant while the overall model remains significant [⁵]. The authors did not report variance inflation factor or correlation matrices, leaving the extent of multicollinearity unexamined and the stability of the reported coefficient estimates uncertain.

Lastly, logistic regression relies on several assumptions that the authors did not verify. The linearity assumption requires that continuous independent variables have a linear relationship with the logit of OAB. Unassessed non-linearity can lead to model misspecification and biased coefficient estimates. Additionally, the authors did not report diagnostic checks for influential outliers or for complete or quasi-complete separation. These omissions leave the robustness of the reported associations uncertain.

In conclusion, the study provides a valuable hypothesis-generating contribution to the understanding of OAB and CS in RA patients. However, the methodological limitations outlined above, unvalidated cut-off, dichotomisation of an ordinal scale, unexamined multicollinearity, and unverified regression assumptions, warrant caution in overinterpreting the findings as definitive evidence. Future studies should address these methodological issues to clarify the true relationship between CS and OAB in RA.

• Funding
  Not applicable.
• Declarations
  Ethics approval and consent to participate
  Not applicable.
• Consent for publication
  The final manuscript has been revised critically and approved by all the authors and they have given the necessary attention to ensure the integrity of the work.
• Publisher's note
  Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations
• OAB  Overactive bladder
• CS  Central sensitization
• RA  Rheumatoid arthritis
• VAS  Visual Analog Scale
• HAQ-DI  Health Assessment Questionnaire Disability Index
• DAS-28  Disease Activity Score in 28 joints
• OAB-V8  Overactive Bladder Questionnaire – Version 8
• OR  Odds ratio
• CI  Confidence interval

Acknowledgements

N/A.

References

• ¹ Cirakoglu D, Yuce A. Evaluation of the prevalence of overactive bladder syndrome in patients with rheumatoid arthritis. Adv Rheumatol. 2026.
• ² Coyne KS, Zyczynski T, Margolis MK, Elinoff V, Roberts RG. Validation of an overactive bladder awareness tool for use in primary care settings. Adv Ther. 2005;22:381–94.
• ³ Altman DG, Royston P. The cost of dichotomising continuous variables. BMJ. 2006;332:1080.
• ⁴ Yoshii I, Chijiwa T, Sawada N. Influence of pain score measured by a visual analog scale (PS-VAS) on the Health Assessment Questionnaire Disability Index and 28‐joint Disease Activity Index with C‐reactive protein in rheumatoid arthritis patients. Int J Rheum Dis. 2018;21:1955–61.
• ⁵ Kim JH. Multicollinearity and misleading statistical results. Korean J Anesthesiol. 2019;72:558–69.

Edited by

• Responsible editor
  Marcos Renato de Assis

Publication Dates

• Publication in this collection
  27 July 2026
• Date of issue
  2026

History

• Received
  01 May 2026
• Accepted
  10 June 2026
• Published
  16 June 2026