Methodological and Statistical Considerations in Assessing the Association Between Cumulative Blood Pressure and Cognitive Decline Authors: Etika Rana Published: 2026 Dear Editor, We read with great interest the article by Hua et al. titled “Elevated Cumulative Blood Pressure is Associated With Cognitive Decline: A Nationwide Population‐Based Cohort Study” published in your esteemed journal [1]. The authors report that elevated cumulative systolic blood pressure (SBP) and pulse pressure (PP) are associated with accelerated cognitive decline in a large Chinese cohort. The topic is important and clinically relevant, and the use of a large, nationally representative dataset is commendable. However, several methodological and statistical aspects require further clarification. Cumulative blood pressure was calculated using the mean of two measurements (2011 and 2013) multiplied by the time interval. However, this approach does not capture within‐person variability or changes over time. Cumulative exposure is more appropriately estimated using repeated longitudinal measurements or trajectory‐based methods that reflect long‐term patterns of blood pressure [2, 3]. Moreover, from 17,708 participants at baseline, only 7,877 were included in the final analysis due to missing data and loss to follow‐up. Such exclusions introduce a risk of selection bias, particularly in aging populations where missingness is often related to poorer health or cognitive decline. A comparison between included and excluded participants would help clarify the extent of this bias [4]. The handling of missing data also requires attention. The analysis relies on linear mixed models without imputation or sensitivity analyses, assuming that data are missing at random. However, this assumption was not evaluated. Methods such as multiple imputation or inverse probability weighting would provide a more robust approach when missingness may be related to outcomes or exposures [5]. In addition, the statistical analysis involves multiple exposures (SBP, DBP, PP), different parameterizations, and several cognitive outcomes, along with subgroup analyses. However, no adjustment for multiple comparisons is reported. Applying appropriate correction methods would help reduce the risk of false‐positive findings [6]. The reported effect sizes presented are small, ranging from approximately −0.03 to −0.10 standard deviation units per year. While statistically significant, these values are modest. Relating these findings to clinically meaningful thresholds would improve their interpretability [7]. The choice of covariance structure in the linear mixed model also warrants clarification. A Toeplitz structure was used, which assumes equal spacing between repeated measurements. However, follow‐up intervals in CHARLS vary between two and three years. A structure suitable for unequal intervals should be considered or clearly justified [8]. Furthermore, nonlinear associations were explored using restricted cubic splines, and significant nonlinearity was reported. However, no information is provided on model validation or comparison. This is particularly relevant where nonlinear associations are observed despite non‐significant linear results, as it raises questions about model robustness [9]. The construction of the global cognitive score also needs clarification. The score was derived by averaging standardized domain scores and re‐standardizing them, which assumes an equal contribution of each domain. Providing justification or validation for this approach would improve interpretability [10]. Finally, repeated cognitive testing across waves may introduce practice effects, where performance improves due to familiarity rather than true cognitive stability. Accounting for this effect would strengthen the estimation of cognitive decline [11]. In summary, this study provides valuable observational evidence on the relationship between cumulative blood pressure and cognitive decline in a large, nationally representative cohort. The authors are to be commended for addressing an important research question using longitudinal data. However, clarification of the methodological aspects outlined above would enhance the transparency, robustness, and overall interpretability of the findings. Author Contributions The author contributed to the conceptualization, critical appraisal, drafting, and revision of the manuscript and approved the final version. Funding This study did not receive any specific grant from funding agencies in the public, commercial, or not‐for‐profit sectors. Ethics Statement Ethical approval was not required for this study as it is a Letter to the Editor based on previously published data. Consent Patient consent was not required as no individual patient data were involved. Conflicts of Interest The author declares no conflicts of interest. Permission to Reproduce Material From Other Sources No figures, tables, or copyrighted material from other sources have been reproduced in this manuscript. Data Availability Statement No new data were generated or analyzed in support of this study. References Associated Data Data Availability Statement No new data were generated or analyzed in support of this study.