Blood Pressure Control Is Associated with Moderate, but Not Necessarily High, Adherence to the DASH Diet in Older Adults

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Evaluation of manuscript nutrients-4067729

The present manuscript investigated the association between adherence to the DASH diet and blood pressure control in an older Spanish population-based cohort (SEGOVIA Study). Below, I present my comments point by point, with the aim of helping the authors improve the overall quality of the manuscript.

The current title is generic and descriptive and does not highlight the main finding of the study, namely the independent association between moderate adherence to the DASH dietary pattern and better blood pressure control in older adults. The authors should consider a more applied and results-oriented title. If helpful, I suggest the following option:
“Blood Pressure Control Is Associated with Moderate, but Not Necessarily High, Adherence to the DASH Diet in Older Adults.”

Although the abstract is well structured and clearly organized, it places excessive emphasis on the background at the expense of the study results. I suggest that the authors present the findings more explicitly and quantitatively, including: the proportion of participants with controlled hypertension according to DASH adherence levels; the magnitude of differences observed in systolic and diastolic blood pressure across adherence categories; and the main results of the logistic regression models, particularly the reduced odds of uncontrolled blood pressure associated with moderate adherence to the DASH pattern.

The introduction clearly and coherently presents the relevance of the topic, the epidemiological context of hypertension, and the knowledge gap addressed by the study. I recommend that the authors strengthen this section by including an explanation of the physiological mechanisms through which the DASH dietary pattern contributes to blood pressure reduction. An additional paragraph could describe key mechanisms such as increased intake of potassium, magnesium, and calcium and their effects on vascular tone modulation, renal sodium excretion, and angiotensin II sensitivity; reduced sodium intake and its direct impact on plasma volume and arterial stiffness; greater intake of fiber, phenolic compounds, and antioxidants, which improve endothelial function and reduce oxidative stress; as well as the effects of the DASH pattern on insulin resistance, low-grade inflammation, and sympathetic nervous system activity. The study hypothesis should also be explicitly stated at the end of the introduction.

The study is robust; however, while the exclusion criteria are clearly and objectively described, the inclusion criteria are not presented with the same level of detail. In addition, no calculation of the minimum sample size required to detect clinically relevant associations between DASH adherence and blood pressure control is provided. Although this is an observational analysis based on an existing cohort, the authors are encouraged to justify the sample size through a statistical power calculation.

Regarding the use of acronyms, the creation of the term HTN for hypertension is unnecessary. I recommend reserving acronyms for longer expressions, such as Dietary Approaches to Stop Hypertension (DASH) or Food Frequency Questionnaire (FFQ).

Although the use of tables is appropriate and helps organize the results and reduce text length, some findings presented in the tables are subsequently repeated descriptively in the text without adding further interpretation. I recommend avoiding the repetition of numerical values, means, and proportions that are already clearly displayed in the tables, and instead using the text to highlight the most relevant findings, overall trends, or comparisons of greater scientific interest.

The discussion is well structured and comprehensive, consistently addressing the clinical and epidemiological relevance of the findings. However, as noted in the introduction, the authors are encouraged to further elaborate on the physiological mechanisms that may explain why moderate adherence to the DASH dietary pattern was sufficient to be associated with better blood pressure control. In this context, the authors could explore the hypothesis of a threshold effect, whereby partial dietary modifications, such as moderate sodium reduction and increased intake of potassium and calcium from fruits, vegetables, and low-fat dairy products, are sufficient to induce meaningful changes in fluid balance, peripheral vascular resistance, and endothelial function.

Although the SEGOVIA Study is longitudinal, the analysis presented is essentially cross-sectional, which limits causal inference. This limitation is acknowledged but could be emphasized more clearly both in the formulation of the study objective and in the interpretation of the results, in order to avoid any implicit suggestion of causality.

Finally, the adjusted models do not include potentially relevant clinical variables for blood pressure control, such as the type, number, or intensity of antihypertensive treatment, nor objective measures of medication adherence. Although this limitation is recognized, a more in-depth discussion of the potential impact of these factors on the results would strengthen the manuscript, as would the inclusion of sensitivity analyses, if data are available. Another point that deserves further discussion is the finding that moderate DASH adherence was significantly associated with blood pressure control, whereas high adherence showed only a trend. This result may be related to the small size of the high-adherence group and potential limitations in statistical power, and these aspects should be discussed more explicitly to avoid misinterpretation.

Author Response

RESPONSE TO REVIEWER 1 COMMENTS

The present manuscript investigated the association between adherence to the DASH diet and blood pressure control in an older Spanish population-based cohort (SEGOVIA Study). Below, I present my comments point by point, with the aim of helping the authors improve the overall quality of the manuscript.

We thank the reviewer for your valuable and insightful comments, which have helped to improve the quality and clarity of the manuscript. Please find the detailed responses below and the corresponding corrections highlighted in red in the re-submitted files.

Comment 1: The current title is generic and descriptive and does not highlight the main finding of the study, namely the independent association between moderate adherence to the DASH dietary pattern and better blood pressure control in older adults. The authors should consider a more applied and results-oriented title. If helpful, I suggest the following option:
“Blood Pressure Control Is Associated with Moderate, but Not Necessarily High, Adherence to the DASH Diet in Older Adults.”

Response to reviewer: Thank you for your comment. We agree with the reviewer that the title of the manuscript needed to be more applied and results-oriented, and it has been changed following the reviewer’s suggestion.

Comment 2: Although the abstract is well structured and clearly organized, it places excessive emphasis on the background at the expense of the study results. I suggest that the authors present the findings more explicitly and quantitatively, including: the proportion of participants with controlled hypertension according to DASH adherence levels; the magnitude of differences observed in systolic and diastolic blood pressure across adherence categories; and the main results of the logistic regression models, particularly the reduced odds of uncontrolled blood pressure associated with moderate adherence to the DASH pattern.

Response to reviewer: We thank the reviewer for this helpful suggestion. We have revised the abstract to reduce emphasis on background information and to present the results more explicitly and quantitatively. The revised abstract now includes: the proportion of participants with controlled hypertension; the proportion of participants with controlled hypertension according to DASH adherence levels; the magnitude of systolic and diastolic blood pressure differences across DASH adherence categories; and the main findings from the multivariable logistic regression models, highlighting the reduced odds of uncontrolled blood pressure associated with moderate DASH adherence. We believe these changes improve the clarity and clinical relevance of the abstract.

Comment 3: The introduction clearly and coherently presents the relevance of the topic, the epidemiological context of hypertension, and the knowledge gap addressed by the study. I recommend that the authors strengthen this section by including an explanation of the physiological mechanisms through which the DASH dietary pattern contributes to blood pressure reduction. An additional paragraph could describe key mechanisms such as increased intake of potassium, magnesium, and calcium and their effects on vascular tone modulation, renal sodium excretion, and angiotensin II sensitivity; reduced sodium intake and its direct impact on plasma volume and arterial stiffness; greater intake of fiber, phenolic compounds, and antioxidants, which improve endothelial function and reduce oxidative stress; as well as the effects of the DASH pattern on insulin resistance, low-grade inflammation, and sympathetic nervous system activity. The study hypothesis should also be explicitly stated at the end of the introduction.

Response to reviewer: Thank you for your comment. We have strengthened the Introduction by adding a paragraph describing the main physiological mechanisms through which the DASH dietary pattern may contribute to BP reduction, including effects related to mineral intake, sodium reduction, endothelial function, metabolic regulation, inflammation, and sympathetic activity (lines 90-96). In addition, we have now explicitly stated the study hypothesis at the end of the Introduction to clearly frame the objectives of the study (lines 118-122).

Comment 4: The study is robust; however, while the exclusion criteria are clearly and objectively described, the inclusion criteria are not presented with the same level of detail. In addition, no calculation of the minimum sample size required to detect clinically relevant associations between DASH adherence and blood pressure control is provided. Although this is an observational analysis based on an existing cohort, the authors are encouraged to justify the sample size through a statistical power calculation.

Response to reviewer: Thank you for this comment. We have clarified in the Materials and Methods section that the present analysis included all participants who attended the follow-up visit as there were no other inclusion criteria (lines 132-135). Regarding sample size, we acknowledge that this is an observational analysis based on an existing cohort. Following your suggestion, we have performed post hoc statistical power calculations to justify the adequacy of the sample size. Power analyses were conducted for the fully adjusted logistic regression model using the pwr.f2.test function in R. With a sample size of n = 185 and a model including six predictors (two dummy variables for DASH adherence categories and four adjustment variables), the estimated statistical power at a significance level of α = 0.05 ranged ≥ 0.80 for small-to-moderate effect sizes (f²≥0.08), indicating sufficient power to detect relevant associations of the magnitude observed in comparable epidemiological studies. This information has now been added to the Statistical Analysis section (lines 206-213).

Comment 5: Regarding the use of acronyms, the creation of the term HTN for hypertension is unnecessary. I recommend reserving acronyms for longer expressions, such as Dietary Approaches to Stop Hypertension (DASH) or Food Frequency Questionnaire (FFQ).

Response to reviewer: Thank you for your comment. We have revised the manuscript and now the full term ‘hypertension’ is used throughout the text.

Comment 6: Although the use of tables is appropriate and helps organize the results and reduce text length, some findings presented in the tables are subsequently repeated descriptively in the text without adding further interpretation. I recommend avoiding the repetition of numerical values, means, and proportions that are already clearly displayed in the tables, and instead using the text to highlight the most relevant findings, overall trends, or comparisons of greater scientific interest.

Response to reviewer: Thank you for this comment. We have revised the Results section to reduce the repetition of numerical values already presented in the tables and to focus the text on highlighting the most relevant findings and overall trends. The limited numerical information that has been retained in the text was considered helpful to facilitate interpretation of the results for the reader.

Comment 7: The discussion is well structured and comprehensive, consistently addressing the clinical and epidemiological relevance of the findings. However, as noted in the introduction, the authors are encouraged to further elaborate on the physiological mechanisms that may explain why moderate adherence to the DASH dietary pattern was sufficient to be associated with better blood pressure control. In this context, the authors could explore the hypothesis of a threshold effect, whereby partial dietary modifications, such as moderate sodium reduction and increased intake of potassium and calcium from fruits, vegetables, and low-fat dairy products, are sufficient to induce meaningful changes in fluid balance, peripheral vascular resistance, and endothelial function.

Response to reviewer: Thank you for this insightful comment. We have expanded the Discussion to further explore potential physiological mechanisms underlying the observed association between moderate DASH adherence and better blood pressure control. In particular, we discuss the plausibility that many partial dietary modifications may be sufficient to induce meaningful changes in fluid balance, peripheral vascular resistance, and endothelial function (lines 387-392). We acknowledge the reviewer’s point that the apparent threshold effect could also be influenced by limited statistical power in the high adherence group. We have therefore interpreted this finding cautiously and avoided overemphasizing the threshold hypothesis, presenting it as a plausible explanation rather than a definitive conclusion.

Comment 8: Although the SEGOVIA Study is longitudinal, the analysis presented is essentially cross-sectional, which limits causal inference. This limitation is acknowledged but could be emphasized more clearly both in the formulation of the study objective and in the interpretation of the results, in order to avoid any implicit suggestion of causality.

Response to reviewer: Thank you for your comment. We have addressed this issue by emphasizing it in the Material and Methods section and in the Discussion.

Comment 9: Finally, the adjusted models do not include potentially relevant clinical variables for blood pressure control, such as the type, number, or intensity of antihypertensive treatment, nor objective measures of medication adherence. Although this limitation is recognized, a more in-depth discussion of the potential impact of these factors on the results would strengthen the manuscript, as would the inclusion of sensitivity analyses, if data are available. Another point that deserves further discussion is the finding that moderate DASH adherence was significantly associated with blood pressure control, whereas high adherence showed only a trend. This result may be related to the small size of the high-adherence group and potential limitations in statistical power, and these aspects should be discussed more explicitly to avoid misinterpretation.

Response to reviewer: Thank you for this important comment. Antihypertensive medication use was collected and used to define hypertension status. However, we decided not including treatment-related variables (type of drugs, number of medications, or treatment intensity) as covariates in the regression models based on: first, no significant differences in antihypertensive treatment patterns were observed between participants with controlled and uncontrolled BP, reducing the likelihood that differences in BP control were driven by treatment heterogeneity; second, pharmacological treatment lies on the causal pathway between hypertension diagnosis and BP control, and adjusting for it could lead to overadjustment and attenuation of the association of interest. Our objective was to evaluate the relationship between DASH diet adherence and BP control as a real-world outcome, reflecting routine clinical practice where dietary behavior and pharmacological treatment coexist. This point has now been clarified in the Statistical Analysis section to improve transparency (lines 200-204). In response to your comment regarding the trend observed in the high adherence group, we have expanded the Discussion to more explicitly address the potential methodological explanations, including the reduced sample size and statistical power limitations in this category.

 

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a cross-sectional analysis of the association between adherence to the Dietary Approaches to Stop Hypertension (DASH) diet and blood pressure control among older adults in the Spanish SEGOVIA cohort. The study addresses an important public health issue—hypertension management in an aging population—and leverages a well-characterized cohort with detailed dietary assessment. The main finding that medium adherence to the DASH diet is independently associated with better blood pressure control is clinically relevant and contributes to the literature on dietary interventions in real-world settings. However, several methodological limitations and presentation issues should be addressed before publication. My comments and suggestions are as follows. 

1, The cross-sectional design limits causal inference. The observed association may reflect reverse causality (e.g., individuals with better BP control may adopt healthier diets).  Acknowledge this more prominently in the Discussion. Consider describing the findings as “associations” rather than “effects.” Future longitudinal analyses are encouraged.

2, line 134 to 145, the BP was measured during a single visit, which may not capture usual BP status and could be influenced by white-coat or masked hypertension. Discuss this limitation explicitly and consider referencing guidelines that recommend multiple measurements over time for hypertension diagnosis. If available, mention whether any participants had ambulatory or home BP monitoring.

3, Medication adherence—a critical determinant of BP control—was not objectively assessed in the study. Discuss this as a key limitation and potential unmeasured confounder. Suggest that future studies incorporate direct measures of medication adherence (e.g., pill counts, pharmacy records).

4, line 49, the finding that medium (but not high) adherence was significantly associated with BP control warrants deeper exploration. Is this due to limited power in the high-adherence group (n=56), or a true threshold?

5, The cohort is from a single Spanish province (Segovia) and comprises older adults. Generalizability to younger populations or other regions may be limited. Briefly discuss the external validity of findings in the Limitations section.

6, Figure 3 is referenced as “Figure 1” in the caption; correct numbering.

7, please define all abbreviations at first use (e.g., DASH, SBP, DBP, FFQ, BMI, OR).

Author Response

RESPONSE TO REVIEWER 2 COMMENTS

This manuscript presents a cross-sectional analysis of the association between adherence to the Dietary Approaches to Stop Hypertension (DASH) diet and blood pressure control among older adults in the Spanish SEGOVIA cohort. The study addresses an important public health issue—hypertension management in an aging population—and leverages a well-characterized cohort with detailed dietary assessment. The main finding that medium adherence to the DASH diet is independently associated with better blood pressure control is clinically relevant and contributes to the literature on dietary interventions in real-world settings. However, several methodological limitations and presentation issues should be addressed before publication. My comments and suggestions are as follows. 

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding corrections highlighted in red in the re-submitted files.

Comment 1:  The cross-sectional design limits causal inference. The observed association may reflect reverse causality (e.g., individuals with better BP control may adopt healthier diets). Acknowledge this more prominently in the Discussion. Consider describing the findings as “associations” rather than “effects.” Future longitudinal analyses are encouraged.

Response to reviewer: Thank you for this comment. We agree that the cross-sectional nature of the present analysis limits causal inference and does not exclude the possibility of reverse causality. We have now emphasized this limitation more prominently in the Discussion, consistently referring to the findings as associations rather than effects, and we highlight the need for future longitudinal analyses to further explore these relationships.

Comment 2:  line 134 to 145, the BP was measured during a single visit, which may not capture usual BP status and could be influenced by white-coat or masked hypertension. Discuss this limitation explicitly and consider referencing guidelines that recommend multiple measurements over time for hypertension diagnosis. If available, mention whether any participants had ambulatory or home BP monitoring.

Response to reviewer: Thank you for this comment. We agree that BP measurements obtained during a single study visit may not fully capture usual BP levels and may be influenced by white-coat or masked hypertension. We have now explicitly discussed this limitation in the Discussion and referenced current hypertension guidelines, which recommend repeated measurements over time primarily for the diagnosis and classification of hypertension. Importantly, hypertension status in our study was not based solely on study visit measurements but was supported by participants’ clinical history, including previous medical diagnosis and antihypertensive treatment. Participants were also routinely followed by their primary care physicians, who were familiar with their clinical status. In addition, we note that requesting home BP measurements from participants could have introduced additional measurement bias related to untrained self-measurement. This point has now been clarified in the revised manuscript (lines 424-431).

Comment 3: Medication adherence—a critical determinant of BP control—was not objectively assessed in the study. Discuss this as a key limitation and potential unmeasured confounder. Suggest that future studies incorporate direct measures of medication adherence (e.g., pill counts, pharmacy records).

Response to reviewer: We thank the reviewer for this important comment. We agree that medication adherence is a relevant determinant of BP control and that it was not objectively assessed in the present study. As now clarified in the Limitations section, this may represent a source of residual confounding, as differences in antihypertensive adherence could have influenced BP control independently of dietary habits. However, no significant differences in antihypertensive treatment patterns were observed between participants with controlled and uncontrolled BP, and our analyses focused on associations rather than causal inference. We have expanded the discussion to acknowledge this limitation and to suggest that future studies should incorporate objective measures of medication adherence, such as pharmacy refill data or pill counts, to further disentangle the relative contributions of diet and pharmacological treatment.

Comment 4:  line 49, the finding that medium (but not high) adherence was significantly associated with BP control warrants deeper exploration. Is this due to limited power in the high-adherence group (n=56), or a true threshold?

Response to reviewer: Thank you for your comment.  We have expanded the discussion of the threshold effect observed in the high adherence group in lines 385-393 to more explicitly address the potential methodological explanations, including the reduced sample size and statistical power limitations in this category to avoid misinterpretation.

Comment 5:  The cohort is from a single Spanish province (Segovia) and comprises older adults. Generalizability to younger populations or other regions may be limited. Briefly discuss the external validity of findings in the Limitations section.

Response to reviewer: Thank you for raising this important point regarding generalizability. We have explicitly addressed the external validity of our findings in the limitations (lines 436-438). We clarify that our results, derived from a population-based cohort of older Spanish adults in a single province, are primarily applicable to older populations in Mediterranean settings with similar dietary patterns and cardiovascular risk profiles. We further emphasize that extrapolation to younger age groups or non-Mediterranean populations should be approached with caution.

Comment 6:  Figure 3 is referenced as “Figure 1” in the caption; correct numbering.

Response to reviewer: Thank you for pointing this out. The mistake has been corrected, and Figure 3 is now correctly referenced in the caption.

Comment 7:  please define all abbreviations at first use (e.g., DASH, SBP, DBP, FFQ, BMI, OR).

Response to reviewer: Thank you for this comment. All abbreviations have now been defined at first use throughout the manuscript, including the Abstract.

 

Reviewer 3 Report

Comments and Suggestions for Authors

 Luengo-Dilla et al present a cross-sectional sub-study of a larger cohort, focused on DASH adherence. The effort is evident but major concerns raise and should be taken care of, before publication.

1. Authors should tighten the introduction to avoid diluting the study’s focus. Although explicit, this section is too long, and the goals appears late.

2. In the same section, authors should include one or two sentences acknowledging barriers to DASH adherence, especially in older Spanish adults. Additionally, they could avoid phrasing that implies DASH is unequivocally superior without nuance.

3. The 406 out of 632 (64%) eligibility. Analysed 371 out of 900 of the original cohort (41%)! What a small percentage: there is no comparison between participants lost to follow-up and those retained, raising concerns about selection bias and survivorship bias. The phrase “900 participants completed the study” is misleading, as the cohort is ongoing and only a subset completed follow-up. Authors should consider rephrasing “completed the study” to avoid implying full longitudinal completion.
4. The design is effectively cross-sectional within a cohort, yet this is not acknowledged. Explicitly state that the analysis is cross-sectional using follow-up data from a longitudinal cohort.
5. Authors are suggested to justify the choice of BP control thresholds or acknowledge guideline evolution.
6. Clarification needed: Timing of medication intake relative to BP measurement and white-coat effect or home BP monitoring. In addition, Combining self-reported diagnosis and medication use with measured BP may introduce classification heterogeneity.
7. The rationale for restricting the current analysis to older adults (mean age 55 ± 12) is not explicitly stated, especially given the original recruitment age range (35–65 years).
8. Acknowledge FFQ limitations and sodium misclassification explicitly in methods section. Add this to the limitations section as well.
9. In the statitical analysis section, include a brief statement on model validation and missing data handling.
10. Figure 3 not Figure 1. SBP (A) and DBP (B) across DASH diet adherence levels, stratified by BP control status. 
11. Since R software was used, data included in Figure 3 should be presented in violin plots (that are more informative). 
12. Confidence intervals should be presented as follows: [lowerbound, upperbound]
13. In the discussion section: authors are advised to use more cautious language (e.g., “may contribute” rather than “may play a central role”) and narrow the novelty claim to avoid overreach. The discussion implies stability of uncontrolled HTN prevalence over time but does not consider differences in BP targets, age distribution, or measurement protocols across studies.
14. There is no discussion of survivorship bias, which is especially relevant in a 20-year follow-up cohort. Healthier individuals are more likely to survive and attend follow-up: this has not been mentioned at all.
15. The discussion implicitly equates observational associations with intervention effects, which is problematic. The reported BP differences (4–5 mmHg SBP) are described as “clinically relevant” without acknowledging that: they are modest at the individual level and/or may reflect confounding by healthier lifestyles. authors should explicitly distinguish between efficacy (RCTs) and associations (observational data) and avoid direct numerical comparisons with intervention trials.
16. The comparison with Epstein et al. (very large BP reductions) is not appropriate, as that was an intensive intervention trial.
17. The threshold effect should be interpreted cautiously. It may be an artifact of categorization, not a biological phenomenon.
18. The lack of a statistically significant association for high adherence (if applicable) is not critically examined.
19. Language that downplays well-known FFQ limitations may be avoided.

20. Important limitations are missing or underplayed: selection bias due to attrition, reverse causality (diet change after HTN diagnosis), sodium measurement limitations. Some limitations are minimized (“partially mitigated”), which may appear defensive. 

21. Medication adherence is mentioned later as a limitation but not integrated into interpretation.

Author Response

RESPONSE TO REVIEWER 3 COMMENTS

Luengo-Dilla et al present a cross-sectional sub-study of a larger cohort, focused on DASH adherence. The effort is evident but major concerns raise and should be taken care of, before publication.

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding corrections highlighted in red in the re-submitted files

Comment 1: Authors should tighten the introduction to avoid diluting the study’s focus. Although explicit, this section is too long, and the goals appears late.

Response to reviewer: Thank you for this comment. We have revised and streamlined the Introduction to improve focus and conciseness, and the study objective has been modified to enhance clarity.

Comment 2: In the same section, authors should include one or two sentences acknowledging barriers to DASH adherence, especially in older Spanish adults. Additionally, they could avoid phrasing that implies DASH is unequivocally superior without nuance.

Response to reviewer: Thank you for this comment. We have addressed this point in the Discussion, where barriers to DASH adherence in older Spanish adults are considered in light of our findings. In addition, we have revised the text to avoid implying unequivocal superiority of the DASH diet.

Comment 3: The 406 out of 632 (64%) eligibility. Analysed 371 out of 900 of the original cohort (41%)! What a small percentage: there is no comparison between participants lost to follow-up and those retained, raising concerns about selection bias and survivorship bias. The phrase “900 participants completed the study” is misleading, as the cohort is ongoing and only a subset completed follow-up. Authors should consider rephrasing “completed the study” to avoid implying full longitudinal completion.

Response to reviewer: Thank you for highlighting this important limitation. We thank the reviewer for this comment. We agree that the wording “completed the study” could be misleading in the context of an ongoing cohort with long-term follow-up. We have therefore revised the text to clarify that the SEGOVIA cohort is ongoing and that the present analysis includes participants who attended the 20-year follow-up visit, rather than implying full longitudinal completion of the original cohort. Regarding attrition, we acknowledge that a substantial proportion of the original cohort was not available for follow-up, which is inherent to long-term population-based studies, particularly in aging populations. Importantly, loss to follow-up in this cohort was largely driven by all-cause mortality over the 20-year follow-up period, rather than selective dropout. As mortality was not restricted to cardiovascular causes, it is unlikely that attrition was systematically related to baseline DASH diet adherence or blood pressure control. While survivorship and selection bias cannot be completely excluded, our objective was to examine cross-sectional associations between dietary adherence and BP control among participants attending follow-up, rather than to estimate incidence or longitudinal risk. Therefore, although attrition may limit generalizability, its impact on the observed associations is expected to be limited. We have clarified these points in the revised manuscript.

Comment 4: The design is effectively cross-sectional within a cohort, yet this is not acknowledged. Explicitly state that the analysis is cross-sectional using follow-up data from a longitudinal cohort.

Response to reviewer: Thank you for this important clarification. We have explicitly stated in the study design section that although the SEGOVIA Study is longitudinal, the current analysis employs a cross-sectional design using data from a single follow-up visit conducted 20 years after baseline recruitment. This distinction is further emphasized in the limitations (Discussion section), where we clarify that, the cross-sectional nature of our analysis does not permit establishing causal relationships

Comment 5: Authors are suggested to justify the choice of BP control thresholds or acknowledge guideline evolution.

Response to reviewer: Thank you for this suggestion. We have justified the choice of BP control thresholds (SBP <140 mmHg and DBP <90 mmHg) in the Methods section (lines 156-159), where we explain that these thresholds are consistent with the 2022 Spanish Society of Hypertension–Spanish League for the Fight against Arterial Hypertension (SEH-LELHA) guidelines and remain widely used in clinical practice and epidemiological research. We applied these thresholds to ensure comparability with existing literature and clinical relevance for our study population. This rationale is also discussed in the Introduction (lines 107-110) where we emphasize the importance of BP control as a distinct clinical endpoint.

Comment 6: Clarification needed: Timing of medication intake relative to BP measurement and white-coat effect or home BP monitoring. In addition, combining self-reported diagnosis and medication use with measured BP may introduce classification heterogeneity.

Response to reviewer: Thank you for this important clarification request. We have addressed these methodological concerns in both the Methods and Limitations sections. We acknowledge in the limitations (Discussion section) that the timing of antihypertensive medication intake relative to BP measurement was not systematically recorded, which may have introduced variability in measured BP values. Additionally, we recognize that BP measured during a single study visit may be influenced by white-coat or masked hypertension, and that ambulatory or home BP monitoring data were not available. We discuss these limitations and their implications for BP measurement accuracy. Regarding potential classification heterogeneity, we do not believe that misclassification is a major concern in our study, as hypertension status was defined using complementary, rather than combined, criteria. Physician diagnosis, antihypertensive treatment, and measured BP values were used to capture the full spectrum of hypertension in a real-world clinical context. Any potential heterogeneity would primarily affect newly diagnosed individuals identified solely by elevated BP measurements. To minimize this, participants with de novo hypertension were excluded from the logistic regression analyses. Consequently, our main analyses focused on individuals with established hypertension, ensuring a more homogeneous study population and strengthening the validity of the observed associations.

Comment 7: The rationale for restricting the current analysis to older adults (mean age 55 ± 12) is not explicitly stated, especially given the original recruitment age range (35–65 years).

Response to reviewer: Thank you for this clarification request. We have not intentionally restricted the analysis to older adults; rather, this reflects the natural age progression of the original cohort. The SEGOVIA Study initially recruited participants aged 35–65 years between 2000 and 2003. The current analysis uses data from the follow-up visit conducted 20 years later (2021–2023), at which point all participants had naturally aged to 55 years or older. Therefore, the age range in the current study simply reflects the follow-up timepoint of the longitudinal cohort, not a deliberate inclusion criterion. We have clarified this in the Study Design and Population section.

Comment 8: Acknowledge FFQ limitations and sodium misclassification explicitly in methods section. Add this to the limitations section as well.

Response to reviewer: Thank you for this suggestion. We have explicitly acknowledged FFQ limitations and sodium misclassification in the Methods section (lines 164-166), where we clearly state that recall bias and potential misclassification of nutrient intake, particularly sodium, are inherent to FFQ-based dietary assessment methodology. We have further expanded this discussion in the limitations (Discussion section) (line 422), where we provide a more detailed acknowledgment of recall bias, misclassification of sodium intake, and underestimation of actual food intake, and discuss how these limitations may affect the accuracy of our DASH adherence estimates.

Comment 9: In the statistical analysis section, include a brief statement on model validation and missing data handling.

Response to reviewer: We have included a statement on missing data handling in Section 2.5 (Statistical Analysis) of the revised manuscript. However, regarding model validation, we respectfully note that our study presents explanatory analyses examining associations between DASH adherence and BP control rather than developing a predictive model. Therefore, formal model validation procedures are not applicable to this observational study design.

Comment 10: Figure 3 not Figure 1. SBP (A) and DBP (B) across DASH diet adherence levels, stratified by BP control status. 

Response to reviewer: Thank you for pointing this out. The mistake has been corrected, and Figure 3 is now correctly referenced in the caption.

Comment 11: Since R software was used, data included in Figure 3 should be presented in violin plots (that are more informative). 

Response to reviewer: Thank you for this suggestion. We have prepared violin plot versions of Figure 3 and are enclosing them for your review. However, we believe that the current line plot with confidence intervals is more appropriate for clearly communicating our key findings. First, the line plot format directly visualizes the trajectory of blood pressure across DASH adherence levels in both controlled and uncontrolled participants, allowing readers to immediately grasp the progression and the between-group differences. Second, the p-values for trend presented in our figure are central to our interpretation. The line plot format intuitively communicates these trend patterns, whereas violin plots would obscure them and make the clinical message less clear. Third, our format facilitates comparison with existing epidemiological literature on DASH diet efficacy. Studies examining trends across categorical exposure levels typically employ line plots, ensuring consistency with the published evidence base and making our findings more accessible to readers familiar with this standard presentation (doi: 10.1056/NEJM200101043440101; doi: 10.1016/j.jacc.2017.10.011; doi: 10.1093/ajcn/nqac067. PMID: 35285859).

 

Comment 12: Confidence intervals should be presented as follows: [lowerbound, upperbound]

Response to reviewer: Thank you for this comment. We have revised the manuscript to present all confidence intervals using the recommended format.

Comment 13: In the discussion section: authors are advised to use more cautious language (e.g., “may contribute” rather than “may play a central role”) and narrow the novelty claim to avoid overreach. The discussion implies stability of uncontrolled HTN prevalence over time but does not consider differences in BP targets, age distribution, or measurement protocols across studies.

Response to reviewer: We appreciate these suggestions. We have revised the Discussion to use more cautious language throughout (e.g., "may contribute"), narrowed the novelty claim to "among the first studies," and substantially revised the section comparing prevalence rates across studies to explicitly acknowledge methodological and geographical differences (age distribution, geographic concentration, measurement protocols, and healthcare contexts). These changes are reflected in the revised manuscript.

Comment 14: There is no discussion of survivorship bias, which is especially relevant in a 20-year follow-up cohort. Healthier individuals are more likely to survive and attend follow-up: this has not been mentioned at all.

Response to reviewer: We thank the reviewer for this insightful comment. We have revised the manuscript to clarify that the present analysis includes participants who attended the 20-year follow-up visit of the SEGOVIA cohort, rather than implying full longitudinal completion of the original cohort. While survivorship and selection bias cannot be completely excluded, our objective was to examine cross-sectional associations between dietary adherence and BP control among participants attending follow-up, rather than to estimate incidence or longitudinal risk. Therefore, although attrition may limit generalizability, its impact on the observed associations is expected to be limited. We have clarified these points in the revised manuscript.

Comment 15: The discussion implicitly equates observational associations with intervention effects, which is problematic. The reported BP differences (4-5 mmHg SBP) are described as “clinically relevant” without acknowledging that: they are modest at the individual level and/or may reflect confounding by healthier lifestyles. Authors should explicitly distinguish between efficacy (RCTs) and associations (observational data) and avoid direct numerical comparisons with intervention trials.

Response to reviewer: We thank the reviewer for this important conceptual clarification. We agree that associations observed in observational studies should not be equated with intervention effects derived from randomized controlled trials. We have revised the Discussion to more explicitly distinguish between the efficacy demonstrated in RCTs and the associations observed in real-world observational data. In particular, we have moderated the language when referring to the magnitude of BP differences, clarifying that the observed reductions are modest at the individual level and may partly reflect residual confounding by healthier lifestyle patterns. We have also avoided direct numerical comparisons with intervention trials, framing our findings instead in terms of potential clinical and public health relevance at the population level.

Comment 16: The comparison with Epstein et al. (very large BP reductions) is not appropriate, as that was an intensive intervention trial.

Response to reviewer: We thank the reviewer for this comment. We agree that direct comparison with Epstein et al., an intensive lifestyle intervention trial, may be inappropriate when interpreting results from an observational study. In the revised version of the manuscript, we have clarified this distinction and reframed the reference to Epstein et al. to emphasize differences in study design, intervention intensity, and expected magnitude of BP changes. We now avoid direct numerical comparisons with intervention trials and discuss our findings strictly in the context of observational associations observed under real-world conditions.

Comment 17: The threshold effect should be interpreted cautiously. It may be an artifact of categorization, not a biological phenomenon.

Response to reviewer: We thank the reviewer for this comment. We agree that direct comparison with Epstein et al., an intensive lifestyle intervention trial, may be inappropriate when interpreting results from an observational study. In the revised version of the manuscript, we have clarified this distinction and reframed the reference to Epstein et al. to emphasize differences in study design, intervention intensity, and expected magnitude of BP changes. We now avoid direct numerical comparisons with intervention trials and discuss our findings strictly in the context of observational associations observed under real-world conditions.

Comment 18: The lack of a statistically significant association for high adherence (if applicable) is not critically examined.

Response to reviewer: We thank the reviewer for this comment. In the revised Discussion, we have more explicitly examined the lack of statistical significance observed for the high-adherence group. We clarify that this finding is likely related to the smaller sample size and reduced statistical power in this category, rather than to an absence of benefit. We also note that the point estimates for medium and high adherence were very similar, supporting the interpretation that achieving at least moderate adherence may be sufficient for meaningful BP control. These points have now been explicitly discussed in the manuscript.

Comment 19: Language that downplays well-known FFQ limitations may be avoided.

Response to reviewer: We have revised the limitations (lines X-X) to more directly acknowledge FFQ limitations without softening language. We now state that dietary assessment using FFQ is "subject to inherent limitations, including recall bias, potential misclassification of sodium intake, and underestimation of actual food intake" without minimizing their impact.

Comment 20: Important limitations are missing or underplayed: selection bias due to attrition, reverse causality (diet change after HTN diagnosis), sodium measurement limitations. Some limitations are minimized (“partially mitigated”), which may appear defensive. 

Response to reviewer: We thank the reviewer for this comment. We acknowledge that a substantial proportion of the original cohort was not available for follow-up, which is inherent to long-term population-based studies, particularly in aging populations. Importantly, loss to follow-up in this cohort was largely driven by all-cause mortality over the 20-year follow-up period, rather than selective dropout. As mortality was not restricted to cardiovascular causes, it is unlikely that attrition was systematically related to baseline DASH diet adherence or blood pressure control. While survivorship and selection bias cannot be completely excluded, our objective was to examine cross-sectional associations between dietary adherence and BP control among participants attending follow-up, rather than to estimate incidence or longitudinal risk. Therefore, although attrition may limit generalizability, its impact on the observed associations is expected to be limited. We have clarified these points in the revised manuscript.

We agree that the cross-sectional nature of the present analysis limits causal inference and does not exclude the possibility of reverse causality. We have now emphasized this limitation more prominently in the Discussion, consistently referring to the findings as associations rather than effects, and we highlight the need for future longitudinal analyses to further explore these relationships.

We have explicitly acknowledged FFQ limitations and sodium misclassification in the Methods section, where we clearly state that recall bias and potential misclassification of nutrient intake, particularly sodium, are inherent to FFQ-based dietary assessment methodology. We have further expanded this discussion in the limitations (Discussion section),where we provide a more detailed acknowledgment of recall bias, misclassification of sodium intake, and underestimation of actual food intake, and discuss how these limitations may affect the accuracy of our DASH adherence estimates.

Comment 21: Medication adherence is mentioned later as a limitation but not integrated into interpretation.

Response to reviewer: Thank you for this comment. We have revised the Discussion to better integrate medication adherence into the interpretation of the findings, acknowledging its potential influence on blood pressure control and clarifying how this limitation may have affected the observed associations.

 

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript entitled “Blood Pressure Control and Adherence to the DASH Diet in the Spanish Adult Population” addresses an important and timely public health issue: the role of dietary patterns in controlling hypertension among older adults. Focusing on the Spanish population and using data from the SEGOVIA cohort adds regional relevance to existing literature on the DASH diet.

I have made some suggestions on how you could improve your work. This doesn't mean that you have to agree with them or rewrite your work in the same way. They are just suggestions to help you see things from a different perspective.

1) While the "Introduction" section provides a thorough discussion of the BP lowering effects of the DASH diet, it would benefit from a brief clarification that BP control (i.e. achieving guideline-recommended targets) is a distinct clinical endpoint from mean BP reduction. This would better justify the study’s focus on blood pressure control rather than absolute blood pressure changes.

2) As the study focuses on adults aged 55 years and over, it would be useful to explicitly state that evidence on DASH adherence and blood pressure control in older adults is limited, particularly in Mediterranean or Southern European populations. This would highlight the novelty and relevance of the study.

3) A short statement acknowledging that strict adherence to the DASH diet may be difficult to achieve in real-world settings, especially among older adults, would provide context for examining intermediate adherence levels.

4) Although antihypertensive medication use is collected and used to define hypertension in the "Materials and Methods" section, it is unclear whether medication use (type, number of drugs or treatment intensity) was considered as a covariate in the regression models. Given its strong influence on blood pressure control, it would be valuable to provide a brief explanation of how medication use was handled analytically (adjusted for, stratified, or not included).

5) Physical activity is described only as "hours per week". Clarity would be improved by specifying whether this variable refers to total physical activity, leisure-time activity, or moderate-to-vigorous activity, and whether a validated questionnaire was used.

6) The methods section would benefit from a brief description of how missing data were handled in the statistical analyses, beyond the initial exclusion criteria (e.g. complete-case analysis, imputation methods or exclusion).

7) Since nutrient intakes are analysed in absolute terms and relative to energy intake, it would be helpful to explicitly state whether energy adjustment (e.g. the residual method) was applied in regression models involving dietary variables.

8) Although the DASH score categorisation follows prior literature, justifying or citing the rationale for these cut-offs in older or Spanish populations would strengthen the methodological justification.

9) In the "Results" section, please, provide context for the clinical relevance of the findings. Although BP differences across DASH adherence groups are reported, adding a brief commentary on whether these differences are clinically meaningful (e.g. mmHg reduction and associated reduction in cardiovascular risk) would improve the interpretation of the results.

10) Please, include sample sizes in all figures and tables. For example, in Figures 2 and 3, explicitly stating the number of participants in each adherence group would make the figures clearer for readers.

11) For continuous variables such as SBP and DBP across adherence levels, reporting trend analyses (e.g. p for trend) would provide statistical support for the observed dose-response patterns.

12) Please, clarify the treatment covariates in the regression models. It is unclear whether antihypertensive treatment or other medications were included in the adjusted models. A brief statement specifying whether medication use was controlled for or explaining why it was not is recommended.

13) High DASH adherence showed a p-value of 0.054-0.055 in the fully adjusted model. A brief comment acknowledging the near-significant trend in the "Results" section would help to set up the "Discussion" section without overinterpreting.

14) While the limitations section mentions the cross-sectional design, it would be helpful to explicitly state that reverse causation is possible, i.e. that participants with better blood pressure control may be more likely to adhere to a healthy diet, so that readers understand the observational nature of the findings.

15) The "Discussion" could briefly elaborate on the biological mechanisms by which DASH adherence may reduce blood pressure (e.g. increased potassium and magnesium intake, increased fibre intake, reduced sodium intake and improved endothelial function), thereby reinforcing the plausibility of the findings.

16) As medication adherence was not measured, including a brief note on how this may have influenced the observed associations (i.e. potential residual confounding) would enhance transparency.

17) The "Discussion" could include a brief note on the generalisability of the results to other Mediterranean populations or younger adults, emphasising that the findings may be most relevant to older Spanish adults.

18) While the limitations are described, adding specific suggestions for future studies, such as longitudinal designs, objective monitoring of diet and medication adherence, or intervention trials, would enhance the value of the "Discussion" section.

19) In the "Conclusions" section, it would be beneficial to include a brief statement that highlights the practical implications for dietary guidance or public health interventions in older adults. For instance, emphasising that even moderate dietary changes can yield significant BP improvements would highlight the study's practical value.  Since the study also noted associations with BMI and smoking, including a sentence acknowledging that DASH adherence should be considered alongside other lifestyle modifications could provide a more holistic perspective.

While this manuscript presents valuable findings, improving the clarity and depth of the discussion, and increasing the level of methodological detail, would further enhance its impact. I recommend accepting the manuscript, provided that some revisions are made.

Author Response

RESPONSE TO REVIEWER 4 COMMENTS

The manuscript entitled “Blood Pressure Control and Adherence to the DASH Diet in the Spanish Adult Population” addresses an important and timely public health issue: the role of dietary patterns in controlling hypertension among older adults. Focusing on the Spanish population and using data from the SEGOVIA cohort adds regional relevance to existing literature on the DASH diet.

I have made some suggestions on how you could improve your work. This doesn't mean that you have to agree with them or rewrite your work in the same way. They are just suggestions to help you see things from a different perspective.

We thank the reviewer for your valuable and insightful comments, which have helped to improve the quality and clarity of the manuscript. Please find the detailed responses below and the corresponding corrections highlighted in red in the re-submitted files.

Comment 1: While the "Introduction" section provides a thorough discussion of the BP lowering effects of the DASH diet, it would benefit from a brief clarification that BP control (i.e. achieving guideline-recommended targets) is a distinct clinical endpoint from mean BP reduction. This would better justify the study’s focus on blood pressure control rather than absolute blood pressure changes.

Response to reviewer: Thank you for your comments that help us to improve the message of the manuscript. Following the reviewer’s suggestion, we have revised the Introduction, and we have included the text: “Importantly, BP control, defined as achieving guideline-recommended targets, represents a distinct and clinically relevant endpoint beyond reductions in mean BP values and is emphasized in current hypertension guidelines because of its strong association with reduced cardiovascular morbidity and mortality (lines 114-117). In this context, evaluating the relationship between DASH diet adherence and BP control, rather than focusing solely on absolute BP reductions, provides clinically meaningful information for hypertension management in routine practice.” This addition better justifies the focus of our study on BP control rather than absolute BP changes.

Comment 2: As the study focuses on adults aged 55 years and over, it would be useful to explicitly state that evidence on DASH adherence and blood pressure control in older adults is limited, particularly in Mediterranean or Southern European populations. This would highlight the novelty and relevance of the study.

Response to reviewer: Thank you for this valuable comment. We have revised the Introduction, and we have included this text: “However, evidence on the association between adherence to the DASH dietary pattern and BP control as a clinical endpoint remains limited in older adults, particularly in Mediterranean or Southern European populations. Most studies have focused on short-term reductions in SBP and DBP rather than on BP control as a clinical endpoint (lines 107-114) and have been conducted in heterogeneous populations or non-Mediterranean settings. Even in Mediterranean cohorts, available evidence has mainly addressed BP changes and cardiometabolic risk factors, rather than BP control specifically in older adults (lines 107-114). This addition highlights the novelty and relevance of our study by emphasizing the lack of data on BP control as a clinical endpoint in older populations living in Mediterranean settings.

Comment 3: A short statement acknowledging that strict adherence to the DASH diet may be difficult to achieve in real-world settings, especially among older adults, would provide context for examining intermediate adherence levels.

Response to reviewer: Thank you for this helpful comment. We have revised the Discussion to acknowledge that strict adherence to the DASH diet may be difficult to achieve in real-world settings, particularly among older adults. We also highlight that intermediate levels of adherence may represent a more realistic and achievable target in routine clinical practice, providing important context for the interpretation of our findings.

Comment 4: Although antihypertensive medication use is collected and used to define hypertension in the "Materials and Methods" section, it is unclear whether medication use (type, number of drugs or treatment intensity) was considered as a covariate in the regression models. Given its strong influence on blood pressure control, it would be valuable to provide a brief explanation of how medication use was handled analytically (adjusted for, stratified, or not included).

Response to reviewer: Thank you for this important comment. In fact, antihypertensive medication use was collected and used to define hypertension status. However, we decided not including treatment-related variables (type of drugs, number of medications, or treatment intensity) as covariates in the regression models based on: first, no significant differences in antihypertensive treatment patterns were observed between participants with controlled and uncontrolled BP, reducing the likelihood that differences in BP control were driven by treatment heterogeneity; second, pharmacological treatment lies on the causal pathway between hypertension diagnosis and BP control, and adjusting for it could lead to overadjustment and attenuation of the association of interest. Our objective was to evaluate the relationship between DASH diet adherence and BP control as a real-world outcome, reflecting routine clinical practice where dietary behavior and pharmacological treatment coexist. This point has now been clarified in the Methods/Discussion section to improve transparency.

Comment 5: Physical activity is described only as "hours per week". Clarity would be improved by specifying whether this variable refers to total physical activity, leisure-time activity, or moderate-to-vigorous activity, and whether a validated questionnaire was used.

Response to reviewer: Thank you for this comment. Physical activity was assessed as total physical activity, expressed as hours per week, based on self-reported information collected through a structured medical questionnaire that included information on occupational activity, leisure-time exercise, sedentary time, and overall activity patterns. Although a validated physical activity questionnaire was not used, this variable was included as an adjustment factor to account for overall activity levels. We have clarified this point in the Materials and Methods section to improve transparency (lines 181-183).

Comment 6: The methods section would benefit from a brief description of how missing data were handled in the statistical analyses, beyond the initial exclusion criteria (e.g. complete-case analysis, imputation methods or exclusion).

Response to reviewer: Thank you for this comment. Missing data were handled using a complete-case analysis. Participants with missing information on key variables required for the regression models were excluded from the corresponding analyses. No imputation methods were applied. We have clarified this approach in the Materials and Methods section to improve transparency.

Comment 7: Since nutrient intakes are analysed in absolute terms and relative to energy intake, it would be helpful to explicitly state whether energy adjustment (e.g. the residual method) was applied in regression models involving dietary variables.

Response to reviewer: Thank you for this comment. Dietary variables, including nutrient intakes, were analysed descriptively and for characterization of DASH adherence, but they were not included as independent variables in the regression models. Therefore, no energy-adjustment methods (such as the residual method) were applied in the regression analyses. We have clarified this point in the Materials and Methods section to avoid confusion.

Comment 8: Although the DASH score categorization follows prior literature, justifying or citing the rationale for these cut-offs in older or Spanish populations would strengthen the methodological justification.

Response to reviewer: Thank you for this comment. The DASH score cut-offs used in the present study were based on previously published literature and have been widely applied across different adult populations, including in, at least, a Spanish population study (Clin Nutr. 2021 May;40(5):2825-2836), which has been cited. Although specific validation exclusively in older Spanish adults remains limited, we have clarified this rationale to strengthen the methodological justification (lines 176-178).

Comment 9: In the "Results" section, please, provide context for the clinical relevance of the findings. Although BP differences across DASH adherence groups are reported, adding a brief commentary on whether these differences are clinically meaningful (e.g. mmHg reduction and associated reduction in cardiovascular risk) would improve the interpretation of the results.

Response to reviewer: Thank you for this comment. We have added a brief statement in the Results section to provide context regarding the clinical relevance of the observed blood pressure differences. In addition, we have expanded the Discussion to note that even modest differences in blood pressure have been associated with meaningful reductions in cardiovascular risk in large epidemiological studies, thereby facilitating a cautious interpretation of the clinical significance of our findings.

Comment 10: Please, include sample sizes in all figures and tables. For example, in Figures 2 and 3, explicitly stating the number of participants in each adherence group would make the figures clearer for readers.

Response to reviewer: Sample sizes were already presented in the tables. However, we have now revised the text to make its clinical interpretation clearer. In addition, the number of participants in each DASH adherence group has now been added to Figures 2 and 3 to improve clarity for readers, following the reviewer’s recommendations.

Comment 11: For continuous variables such as SBP and DBP across adherence levels, reporting trend analyses (e.g. p for trend) would provide statistical support for the observed dose-response patterns.

Response to reviewer: Following the reviewer’s suggestion, we have included this data in Figure 3.

Comment 12: Please, clarify the treatment covariates in the regression models. It is unclear whether antihypertensive treatment or other medications were included in the adjusted models. A brief statement specifying whether medication use was controlled for or explaining why it was not is recommended.

Response to reviewer: Thank you for this important comment. Antihypertensive medication use was collected and used to define hypertension status. However, we decided not including treatment-related variables (type of drugs, number of medications, or treatment intensity) as covariates in the regression models based on: first, no significant differences in antihypertensive treatment patterns were observed between participants with controlled and uncontrolled BP, reducing the likelihood that differences in BP control were driven by treatment heterogeneity; second, pharmacological treatment lies on the causal pathway between hypertension diagnosis and BP control, and adjusting for it could lead to overadjustment and attenuation of the association of interest. Our objective was to evaluate the relationship between DASH diet adherence and BP control as a real-world outcome, reflecting routine clinical practice where dietary behavior and pharmacological treatment coexist. This point has now been clarified in the Methods/Discussion section to improve transparency.

Comment 13: High DASH adherence showed a p-value of 0.054-0.055 in the fully adjusted model. A brief comment acknowledging the near-significant trend in the "Results" section would help to set up the "Discussion" section without overinterpreting.

Response to reviewer: Thank you for this comment. We have added a brief statement in the Results section acknowledging the near-significant association observed for high DASH adherence in the fully adjusted model. This clarification helps to appropriately frame the interpretation of the findings in the Discussion without overinterpreting the results.

Comment 14: While the limitations section mentions the cross-sectional design, it would be helpful to explicitly state that reverse causation is possible, i.e. that participants with better blood pressure control may be more likely to adhere to a healthy diet, so that readers understand the observational nature of the findings.

Response to reviewer: Thank you for this comment. We have revised the limitations section to explicitly acknowledge the possibility of reverse causation, clarifying that participants with better blood pressure control may be more likely to adhere to healthier dietary patterns. This addition reinforces the observational nature of the findings and avoids any implication of causality.

Comment 15: The "Discussion" could briefly elaborate on the biological mechanisms by which DASH adherence may reduce blood pressure (e.g. increased potassium and magnesium intake, increased fibre intake, reduced sodium intake and improved endothelial function), thereby reinforcing the plausibility of the findings.

Response to reviewer: Thank you for this comment. We have expanded the Discussion to briefly elaborate on the biological mechanisms through which adherence to the DASH dietary pattern may be associated with lower blood pressure, including reduced sodium intake, increased potassium and magnesium intake, higher fibre consumption, and improvements in endothelial function. This addition reinforces the biological plausibility of our findings.

 

Comment 16: As medication adherence was not measured, including a brief note on how this may have influenced the observed associations (i.e. potential residual confounding) would enhance transparency.

Response to reviewer: Thank you for this comment. We have now explicitly acknowledged in the Discussion that medication adherence was not measured and that this may have introduced residual confounding, potentially influencing the observed associations between DASH adherence and BP control.

Comment 17: The "Discussion" could include a brief note on the generalizability of the results to other Mediterranean populations or younger adults, emphasising that the findings may be most relevant to older Spanish adults.

Response to reviewer: Thank you for this comment. We have added a brief note in the Discussion addressing the generalizability of our findings, emphasizing that they are most directly applicable to older Spanish adults and should be extrapolated to other Mediterranean populations or younger age groups with caution.

Comment 18:  While the limitations are described, adding specific suggestions for future studies, such as longitudinal designs, objective monitoring of diet and medication adherence, or intervention trials, would enhance the value of the "Discussion" section.

Response to reviewer: Thank you for this comment. We have expanded the Discussion to include specific suggestions for future research, highlighting the need for longitudinal analyses, objective assessment of dietary and medication adherence, and intervention studies to better clarify causal relationships.

Comment 19:  In the "Conclusions" section, it would be beneficial to include a brief statement that highlights the practical implications for dietary guidance or public health interventions in older adults. For instance, emphasising that even moderate dietary changes can yield significant BP improvements would highlight the study's practical value.  Since the study also noted associations with BMI and smoking, including a sentence acknowledging that DASH adherence should be considered alongside other lifestyle modifications could provide a more holistic perspective.

Response to reviewer: Thank you for this comment. We have revised the Conclusions section to highlight the practical implications of our findings for dietary guidance and public health interventions in older adults, emphasizing that even moderate dietary changes may be beneficial for blood pressure control. We have also added a brief statement acknowledging that DASH adherence should be considered alongside other lifestyle modifications, such as weight management and smoking cessation, to provide a more holistic perspective.

While this manuscript presents valuable findings, improving the clarity and depth of the discussion, and increasing the level of methodological detail, would further enhance its impact. I recommend accepting the manuscript, provided that some revisions are made.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Evaluation of manuscript nutrients-4067729

I commend the authors and have few final observations. The inclusion of a post hoc power calculation is welcome. However, considering that post hoc power analyses can be viewed critically by some readers, the authors may emphasize that this analysis is intended to provide contextual support, and not confirmatory evidence of statistical adequacy.

The decision not to adjust the regression models for antihypertensive treatment is well justified and transparently explained. The rationale related to overadjustment, and causal pathway considerations is scientifically sound and appropriately framed. To further strengthen this section, the authors may consider explicitly stating that the analytical approach was designed to reflect real-world blood pressure control, where dietary behavior and pharmacological treatment coexist.

Overall, the manuscript is robust, requiring only minor adjustments to enhance clarity and methodological transparency.

Author Response

Comment 1: 

I commend the authors and have few final observations. The inclusion of a post hoc power calculation is welcome. However, considering that post hoc power analyses can be viewed critically by some readers, the authors may emphasize that this analysis is intended to provide contextual support, and not confirmatory evidence of statistical adequacy.

The decision not to adjust the regression models for antihypertensive treatment is well justified and transparently explained. The rationale related to overadjustment, and causal pathway considerations is scientifically sound and appropriately framed. To further strengthen this section, the authors may consider explicitly stating that the analytical approach was designed to reflect real-world blood pressure control, where dietary behavior and pharmacological treatment coexist.

Overall, the manuscript is robust, requiring only minor adjustments to enhance clarity and methodological transparency.

 

Author response 1: 

We appreciate these constructive final observations. We have implemented both suggestions in the revised Statistical Analysis section (Section 2.5):

1. Post hoc power analysis: We now explicitly clarify that this analysis "is intended to provide contextual support for understanding the study's statistical capacity rather than as confirmatory evidence of statistical adequacy," directly addressing the reviewer's concern about appropriate interpretation of post hoc power calculations.
2. Antihypertensive treatment rationale: We have strengthened the justification by stating that the analytical approach "was intentionally designed to reflect real-world BP control, were dietary behavior functions independently from pharmacological treatment, each as a distinct but complementary strategy." This clarifies that the exclusion of treatment as a covariate allows for examination of the independent dietary association while acknowledging that both interventions coexist in clinical practice.

 

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have revised the manuscript accordingly. 

Author Response

Comment 1:

The authors have revised the manuscript accordingly. 

Author response 1:

We sincerely appreciate the thorough review and constructive feedback throughout the revision process. Thank you for your contribution to improving the quality of this work.

Reviewer 3 Report

Comments and Suggestions for Authors

Authors addressed the most important of my concerns. The mscr may be published

Author Response

Comment 1: 

Authors addressed the most important of my concerns. The mscr may be published

Author response 1:

We sincerely appreciate the thorough review and constructive feedback throughout the revision process. Thank you for your contribution to improving the quality of this work.