Special Interview with Reviews in Cardiovascular Medicine Author Prof. Paolo Emilio Puddu: Revisiting the Seven Countries Study and Homogeneity of Serum Cholesterol Coefficients Across Populations
20 April 2026
Professor Paolo Emilio Puddu is a cardiologist with extensive experience in clinical cardiology, cardiovascular epidemiology, and biostatistics. He received his MD from Sapienza University of Rome and later completed advanced training and academic qualifications in Canada and France. Throughout his career, he has contributed to numerous international studies and served as a reviewer for over 50 journals.
In this interview, we are honored to speak with Prof. Puddu about his recent publication ( Volume 27, Issue 2 (2026)) in Reviews in Cardiovascular Medicine, which revisits the classic Seven Countries Study and explores the cross-population consistency of serum cholesterol coefficients in predicting mortality in coronary heart disease.
Paolo Emilio Puddu, MD, PhD
1 Association for Cardiac Research, 00182 Rome, Italy
2 EA 4650, Signalisation, Électrophysiologie et Imagerie des Lésions D’Ischémie Reperfusion Myocardique, Normandie Université, UNICAEN, 14000 Caen, France
Interests: cardiovascular epidemiology; clinical cardiology; cardiac electrophysiology; biostatistics; cross-population risk prediction
1. Could you briefly introduce yourself and your research team, and summarize the central research question and key findings of your article in a few sentences?
I am Prof. Paolo Emilio Puddu from the University of Rome and am now retired. As a cardiologist with a PhD, my research interests initially focused on electrophysiology, then shifted to clinical research, and recently, I have been deeply engaged in cardiovascular epidemiology. There is no large research team for this study, and the core is to conduct a review and analysis of serum cholesterol based on the Seven Countries Study with collaborators. I chose to submit to your journal because it is a high-quality review journal in the cardiovascular field, which is consistent with the goals of this study to sort out and summarize a large number of results of the Seven Countries Study. The core of this paper is to explore the similarity of multivariate serum cholesterol coefficients across populations. The key finding is that the slope of the correlation between cholesterol and cardiovascular mortality is very similar among different populations.
2. Your article revisits the well-known Seven Countries Study. What inspired you to re-examine this classic study and analyze it from the perspective of the similarity of multivariate serum cholesterol coefficients across populations?
This is a very important question. The Seven Countries Study was a classic in cardiovascular epidemiology, initiated in the 1950s by Professor Keys from the University of Minnesota. The central idea was to compare different populations from Japan to Finland, because diet might influence blood cholesterol, and blood cholesterol might be the central cause of coronary artery disease. That has been extensively demonstrated.
What we showed is that the slope relating blood cholesterol to cardiovascular mortality is very similar across different populations. This is fundamental because the Seven Countries Study used exactly the same methodology across all populations: the same age range (all men aged 40–59), the same sex (male only), the same method for measuring cholesterol, and stable measurement protocols. This differs from meta-analyses that mix different methods, ages and sexes.
The key message from our figure is that although absolute risk varies between low-risk and high-risk populations (due to different average cholesterol levels), the relative risk for a given difference in cholesterol (e.g., from 200 to 220 mg/dL) is identical because the slopes are parallel. Therefore, absolute risk differs, but relative risk is the same.
3. Your paper highlights that the multivariate coefficients for serum cholesterol were not significantly heterogeneous across populations. Was this finding surprising given the wide variation in CHD mortality rates? What are your thoughts on this?
This finding is not surprising. The core reason for the differences in CHD mortality rates among different populations is the different average blood cholesterol levels of each population. Populations in Southern Europe, such as Greece and Italy, have lower average blood cholesterol levels, and the incidence and mortality of CHD are also lower accordingly; while populations in Northern Europe, such as Finland and the Netherlands, have higher average blood cholesterol levels, and the corresponding risk of CHD onset and death is also higher. In short, the differences in CHD mortality rates among populations are caused by the differences in the average levels of blood cholesterol, not the differences in the coefficients of the correlation between cholesterol and CHD. Therefore, the result that the multivariate coefficients of serum cholesterol have no significant heterogeneity across populations is not contradictory to the cross-population differences in CHD mortality rates, and is a conclusion consistent with research logic.
4. You specifically acknowledged Professor Henry Blackburn from the University of Minnesota for his review of this manuscript in the paper. As one of the pioneers in the field of cardiovascular epidemiology, could you talk about Professor Blackburn's contributions to this article?
Prof. Blackburn was an intimate collaborator of Prof. Keys, the initiator of the Seven Countries Study, and took over the relevant research position in cardiovascular epidemiology after Prof. Keys retired. Prof. Keys lived to be 102 years old and continued to carry out research and assist collaborators until his later years. Prof. Blackburn himself is 101 years old and once served as the Editor-in-Chief of Circulation, being an important scholar in the field of cardiovascular epidemiology. The manuscript of this study is the latest paper he reviewed and provided professional consultation on in his later academic career. He not only put forward many valuable revisions to improve the quality of the manuscript, but more importantly, he has always continued the research context of the Seven Countries Study and provided continuous academic support for the subsequent analysis of this classic study. Although he did not become a co-author of this paper due to his advanced age, his professional guidance was crucial to the completion of this paper, which is why we sincerely acknowledged him in the paper.
5. You mentioned in the article that preliminary tests suggest that the multivariate coefficients of blood pressure and smoking habits may also show non-heterogeneity. In light of this, does your team plan to further investigate and write a more comprehensive review that covers additional cardiovascular risk factors?
Our team does plan to carry out this research, and the relevant research is nearly completed. In fact, we have published two papers in 2017 and 2018, analyzing whether there are statistical differences in the coefficients of standard cardiovascular risk factors such as age, cholesterol, blood pressure, and smoking among different populations in the Seven Countries Study, and also discussing the recalibration of risk prediction models. We found that there are no significant statistical differences in the coefficients of core risk factors among different populations, and applying the coefficients of one population to another can achieve effective model calibration. This follow-up study will continue the research on serum cholesterol, focusing on analyzing the absolute and relative risk characteristics of risk factors such as blood pressure and smoking across populations, and exploring the cross-population consistency of their coefficients. At present, this research has been basically completed, and my collaborators and I are also considering submitting the findings of this current research to your journal.
6. In recent years, the rapid development of artificial intelligence, big data, and digital health technologies has greatly advanced medical research. Did these developments influence the research and writing of your article to some extent?
This is a difficult question. Artificial intelligence is a big evolution, but it does not replace the need for careful research. I lived through the era of data analysis from the 1980s to 2020, where we derived truth by comparing results from different studies. AI might give you an answer, but it may not necessarily be true. What we have clearly shown is that if you do not have an absolutely uniform methodology, you cannot mix results – you cannot mix apples and pears. AI is essentially mixing apples and giving you a result. That result, if interpreted as a sum, might be helpful, but if interpreted as a truly integrated result, it could be dangerous. My current evaluation of AI is like that of translation tools: if you are not a native Chinese speaker, it would be very difficult to trust an AI translation entirely. AI can help, but if used as a substitute, it is very dangerous.
7. How did you first learn about Reviews in Cardiovascular Medicine (RCM)? What aspects of the journal attracted you to submit this important article to RCM?
I have been paying attention to various journals in the cardiovascular field, and also learned that your journal is a high-quality review journal focusing on cardiovascular medicine. There are several core reasons for choosing to submit to your journal: first, your journal places a strong emphasis on review articles, while also welcoming original research with clear clinical relevance. This study is a re-sorting and analysis of the Seven Countries Study, which is in line with the manuscripts accepted by your journal; second, your journal has a good academic influence and adheres to an open running concept, with a high degree of article dissemination; third, as retired scholars, my collaborators and I have no research funding support, and your journal has provided us with the opportunity to publish without publication fees, solving our practical problems. I am very satisfied with the efficiency of the review and the journal’s professionalism.
8. Currently, RCM publishes review papers, research articles, short communications, and other types of articles on cardiovascular medicine. Do you think the type of the current article meets the needs of most scholars? In the future, what type of content would you like to see more of in our journal?
My suggestion is to invite commentators – reputed scholars – to write public commentaries on important review or research articles. They could give a frank opinion that counterbalances the conclusions of the original paper. These commentators would not necessarily be involved in the peer review process. You would have an expert article, and then ask some scholars to make a signed comment, which would be public and scientific. Since they have to sign it, their comments would be more measured and less likely to include trivial or nasty remarks. This would be a valuable addition to the journal.
This interview highlights the importance of revisiting classical epidemiological studies with modern perspectives. Prof. Puddu’s work provides valuable insights into the consistency of cardiovascular risk relationships across populations.
We sincerely thank Professor Puddu for sharing his time and perspectives with us. We look forward to more of his contributions in Reviews in Cardiovascular Medicine.
Journal Homepage: Reviews in Cardiovascular Medicine