The author team conducted a retrospective analysis on diabetes and stroke, which is a meaningful article. However, the summary and discussion still lack depth. So I suggest that it still needs to be further deepened to improve the quality of the article. At present, the incidence rates of diabetes and stroke are both high, and it is of great significance to study their existence mechanisms.

Thank you for the opportunity to review your manuscript, "Trends and disparities in mortality from comorbid non-insulin-dependent diabetes mellitus and stroke (1999-2022): A retrospective analysis." This is a timely and important study that addresses a significant public health issue. The analysis is comprehensive, covering a long time period and multiple demographic and geographic strata. Below are my comments based on the provided evaluation criteria. 1. Ethics: The study appropriately utilizes a publicly available,de-identified database (CDC WONDER). The authors have correctly stated that Institutional Review Board approval and informed consent were not required for this secondary data analysis. The ethical approach is sound and meets standard requirements for this type of research. 2. Methods: The experimental method is effective and well-described.The use of the CDC WONDER Multiple Cause of Death database is a robust choice for this nationwide epidemiological study. The application of specific ICD-10 codes to capture comorbid conditions, the use of age-adjusted mortality rates (AAMRs) standardized to the 2000 US population, and the employment of Joinpoint regression analysis to identify temporal trends are all statistically sound and established methodologies. The exclusion of T1DM is well-justified. The methods are described with sufficient detail to be repeatable by fellow researchers. 3. Results: The results appear true and authentic,derived directly from a reputable national database. The key findings—the overall shift from a decreasing to a sharply increasing mortality trend post-2014, and the pronounced disparities across sex, race, urbanization level, census region, age, and state—are clearly presented and supported by the data. The theory that mortality from this comorbidity is increasing and exhibits significant health inequities is strongly validated by your results. 4. Figures and tables: The figures provided(Figures 1, 2A-C, and 3) are generally clear and well-constructed, effectively visualizing the complex trends and disparities across different strata. The annotations with APC values and confidence intervals are helpful. A minor suggestion for improvement would be to ensure all axis labels and legends are perfectly clear (e.g., the y-axis label "Age adjusted mortality rate per 100000" in Figure 2 could be consistent in size and clarity across all subfigures). The reference to supplementary tables is appropriate for a main text; it is assumed those tables are well-structured and annotated. 5. Biostatistics: The biostatistical analysis appears perfect and rigorous.The use of Joinpoint regression is the correct choice for analyzing trends over time. The presentation of Annual Percentage Change (APC) with 95% confidence intervals for each segment, along with measures of statistical significance (P < 0.05), is exemplary. The data in the results section and figures are consistent. No mistakes in the presented data were identified. 6. References: The references are generally scientific,reasonable, and relevant to the study's context, particularly in the Introduction and Discussion sections. They support statements about the prevalence of diabetes and stroke, previous trends, and potential explanations for disparities. However, some references are slightly dated (e.g., References 14 from 2006, 15 from 2011). While these may be classics in the field, consider incorporating a few more recent reviews or studies (from the last 3-5 years) on racial and rural health disparities in diabetes and stroke outcomes to further strengthen the context. There is no evident pattern of excessive self-citation or omission of key works. 7. Language: The language is of high quality.It correctly, clearly, and concisely expresses the scientific information. The manuscript is well-written and professional. Only very minor grammatical issues were noted (e.g., Page 9, "The cause of this decline is not established yet" could be phrased as "The cause of this initial decline is not established."). These are insignificant and do not detract from the overall clarity. 8. Caveats or drawbacks: The authors have appropriately acknowledged the key limitations of their study in the Discussion section,which is commendable. The main caveats are: · Reliance on Death Certificate Data: This is the most significant limitation, as underreporting of NIDDM or non-fatal stroke on death certificates is a known source of bias, potentially leading to an underestimation of true mortality rates. · Broad Racial/Ethnic Categories: As noted, the use of broad categories (e.g., "Asian or Pacific Islander") masks important heterogeneity within these groups. · COVID-19 Confounding: The authors correctly note that the sharp increases observed in 2018-2020 may be confounded by the COVID-19 pandemic, either through direct mortality or indirect disruptions to healthcare. This is an important caveat for interpreting the most recent trends. · Ecological Fallacy: The findings identify associations at the population level, which may not directly reflect individual-level risk factors. Overall, this is a well-executed, valuable, and clinically relevant study. The findings highlight critical and widening disparities that demand public health attention. The manuscript is strong and merits publication. The suggestions above are minor and intended to further polish an already excellent paper. I recommend acceptance with minor revisions.