Uncovering Gaps: Dietary Influence and Machine Learning in Hypertension and Comorbidities

Dear Editor,

We read with great interest the recent article by Zhang et al., “Diabetes Mellitus and Hyperlipidemia Status Among Hypertensive Patients in the Community and Influencing Factors Analysis of Blood Pressure Control,” which provides valuable insights into the prevalence of type 2 diabetes mellitus (T2DM) and hyperlipidemia among hypertensive patients and investigates their impact on blood pressure (BP) control using a large dataset from the National Basic Public Health Service Program in Guangzhou. The authors successfully underscore key risk factors influencing BP control, including obesity, alcohol use, physical inactivity, and poor medication adherence [1]. Although the study presents significant findings, several methodological aspects warrant further discussion.

First, the study appropriately recognizes the increased prevalence of comorbid T2DM and hyperlipidemia among hypertensive patients and its association with poorer BP control. However, one major constraint is the lack of stratification based on the severity of hyperlipidemia and diabetes. Given the varied nature of these conditions, a more detailed subgroup analysis considering glycemic control levels (e.g., HbA1c categories) and lipid profiles (e.g., LDL/HDL ratios) would provide deeper insights into their precise impact on BP regulation.

Second, while the study underlines key lifestyle factors affecting BP control, the role of dietary patterns is not sufficiently addressed. Prior studies have shown that dietary sodium intake, fat composition, and overall macronutrient distribution significantly affect BP levels in hypertensive patients with metabolic comorbidities [2]. Incorporating dietary data into the analysis would strengthen the study's conclusions and provide actionable recommendations for community-based hypertension management.

Third, although the study investigates important influencing factors on BP control, it does not sufficiently address the role of medication adherence beyond a general assessment. Hypertensive patients with comorbidities like T2DM and hyperlipidemia often require complex polypharmacy, and adherence patterns can significantly impact BP control outcomes. Previous researches have shown that factors such as medication burden, side effects, and patient perceptions of treatment efficacy affect adherence rates [3]. Including a more comprehensive evaluation of medication adherence, such as pill burden or self-reported adherence scales, would provide more in-depth insights into its effect on BP regulation.

Moreover, the study employs logistic regression models to identify factors associated with BP regulation, but does not include machine learning techniques, which have been increasingly used in cardiovascular research for predictive modeling. Advanced statistical approaches, such as the random forest model, could increase risk stratification and improve predictive accuracy in identifying high-risk hypertensive patients requiring intensified management [4, 5].

In conclusion, Zhang et al. provide an imperative contribution in understanding the effect of diabetes and hyperlipidemia on BP regulation in hypertensive patients. Though, further research is required to improve risk stratification, investigate causal pathways, and evaluate the efficiency of personalized lifestyle and pharmacological interventions. We commend the authors for their contribution and encourage further investigations into optimizing BP control strategies in hypertensive populations.

As this is a commentary on a published study and no new data were collected or analyzed, ethics approval was not required.

The authors declare no conflicts of interest.