Earlier this month, the Academy of Nutrition and Dietetics released it's commentary on the scientific report of the Dietary Guidelines for Americans' Committee. You can find these comments here. (Edit: (re-edit: it appears to be back up) AND's official commentary appears to have been taken down. You can still find it here: http://www.eatrightidaho.org/wp-content/uploads/Academy-Comments-re-DGAC-Scientific-Report2-1.pdf ) For any members of AND, I highly encourage that you reach out to Sonja Connor (president@eatright.org) and let her know your thoughts on this commentary, good or bad. If this blog post leaves you concerned about this commentary, I would especially urge you to reach out to the president.
In short, the report comments on the evidence-base used by the Dietary Guidelines Committee, their thoughts on dietary patterns, makes nice comments on the struggle to determine the amount of phosphorus in foods from food labels, supports the use of the DGAC's social ecological model to make recommendations, and makes very vague comments on the sustainability issue. The report also touches on "cross-cutting topics of public health importance", highlighting 2 nutrients, sodium and saturated fats.When the report touches on this latter part, however, we stumble onto some scientific commentaries that rely on questionable evidence, or interpretations there of.
For sodium, AND comments that the report relies on literature looking only at blood pressure as an outcome, acknowledging topics I've discussed on this blog regarding the complicated research problem child that sodium can be. It was disappointing to see a lack of a critical perspective taken by AND on the lower quality data that has led to 'J-shaped' curves (see my previous posts on sodium regarding this) and wish the comments would've given more of a shoutout to emphasizing the potassium to sodium ratio, instead of setting sodium recommendations in isolation. I certainly acknowledge that there is a rather complex sodium controversy, which can be viewed through many lenses, but I also acknowledge that much of the controversy is driven by poor quality data (1), using less than gold standard methodologies for determining regular sodium intake. A recent read on the methods used to determine sodium intake was recently published in Annual Reviews of Nutrition here that I would suggest reading for a discussion of the limitations of many of the methods widely used to estimate sodium intake.
AND's comments on saturated fat are of particular concern; the authors used simple equations to illustrate that carbohydrates have a greater effect on CVD if they replace SFAs, and cite Jakobsen's (2) pooled analysis of 11 cohorts to support this claim. This evidence does not come from RCTs, so we can't causally back up these claims. There are no RCTs looking at replacement of SFA with CHO that demonstrate increase/decrease CVD events; the best data we have to go on to look at this is observational data. In the 2 models used in Jakobsen's meta analysis, CHO replacing SFA has a hazard ratio of 1.06 and 1.07 for coronary events (not increased risk for morality), but as the authors note, the glycemic index of the carbohydrate was not considered here (fiber was included in the analyses). Further analyses by Jakobsen (3) that used a pseudo-glycemic index(see their methods) stratification of CHO quality see a non-significant inverse association with replacing low-GI carbohydrates for saturated fats (for myocardial infarction) - though this notion of low-GI carbohydrates providing some cardiovascular benefit is not supported by clinical trials (14-16). Jakobsen's (2) lack of a strong effect of replacing SFA with CHO is likely due to heterogeneity in the included cohorts; an analysis of the NHS by Hu et al in the NEJM in 1997 (13), which contained overlapping participants within the Jakobsen's pooled cohort analysis, suggested that each increase of 5 percent of energy intake from saturated fat, as compared with equivalent energy intake from carbohydrates, was associated with a non-significant 17 percent increase in the risk of coronary disease . Clearly, there is conflicting data regarding CHO, SFAs and their impact on cardiovascular outcomes. It is unclear based on the wording of this commentary whether AND means that PUFAs should replace all carbohydrates, or added sugars, as their recommendation statement is quite ambiguous in its wording, and is followed by a statement regarding added sugars. It is imperative that AND use more clear, direct language as this has already been misinterpreted, across social media, to be AND coming out in support of low carbohydrate diets. As it is interpreted now, the model employed by this AND commentary is overly simplistic, and one that is not supported by stronger clinical evidence. The trials that we have looking at CVD end points replace SFAs (and sometimes TFAs) with mixed PUFAs, not CHO with PUFAs (4-8) - these largely back up the recommendation to replace SFA with PUFAs, not food frequency questionnaire based epidemiological associations. Penn Kris-Etherton, PhD,RD and distinguished professor of nutrition at Penn State recently released an Advances in Nutrition Review (9) on the topic of fatty acids, and re-iterates the need for more research looking at the type of carbohydrate's effect on CVD risk, seeing as carbohydrates will likely serve as some of the replacement macronutrient for those trying to reduce SFAs and meet the AHA recommendations of 5-6% of calories (this being derived from dietary patterns like DASH and DELTA). I'm enthusiastic that the field is starting to acknowledge that not all SFAs are created equal, and there's controversial, interesting data for dairy fats and chocolate/tropical fats, but I'm hesitant to jump on any 'saturated fat is good/better for you train' until strong clinical evidence says so - as of now, trials that replaced saturated fat sources with PUFAs did so rather indiscriminantly. It would seem reasonable that AND highlight that dietary patterns, and recent (imperfect) clinical trials employing mediterranean style diets, all emphasize lower saturated fat diets ( less than 10% kcals), and that this focus on saturated fat is quite limiting; indeed, evidence might question their 'harm', but evidence of causal, health-promoting properties is quite sparse.
The AND commentary also appears to misconstrue an IOM document (10) evaluating biomarkers and surrogate endpoints in chronic disease to make strong claims about the use of lipoprotein endpoints . AND states that "While the body of research linking saturated fat intake to the modulation of LDL and other circulating lipoprotein concentrations is significant, this evidence is essentially irrelevant to the question of the relationship between diet and risk for cardiovascular disease" and " this body of evidence that uses lipoproteins as surrogate endpoints for cardiovascular disease must be excluded from considerations of the impact of diet on cardiovascular health". The actual IOM document doesn't quite say these things. The section on LDL in this document discusses some of the complexities of LDL-C, the better usage of LDL-particle number/non-HDL cholesterol as markers, the pleiotropic effects of drug/statin and estrogen interventions, and the general limitations of LDL-C alone as a biomarker of disease. It did not conclude that lipoprotein changes are irrelevant to cardiovascular health - indeed, those with genetically low/high LDL would likely disagree with that statement. The document largely focused on statins trials to note that just a reduction in LDL-C doesn't simply equate to better cardiovascular outcomes - this should make sense, as interventions using statins don't just lower LDL-C but have multiple other effects. The document doesn't state that LDL is useless but rather that the LDL-C is a general measurement of a heterogeneous lipoprotein, and better measurements can capture its complexity; it highlights more accurate predictors like LDL particle or ApoB number (one might note that nutrition research has kept up with these limitations and begun to look at LDL particle number (12)). I find that the AND commentary's use of this IOM report is misleading and overly reduces the complexity of what the report was trying to say; it concludes: "lowering LDL-C can never be considered a "perfect" indicator across all population groups. That said, there is high probability that lowering LDL for several interventions decreases risk of cardiovascular disease". It does not, as the AND commentary says, make 'irrelevant' the effects of saturated fats on LDL and lipoproteins. If anything, the document should be used to state that the observed beneficial effects of replacing saturated fats with mixed PUFAs cannot only be attributed to just their cholesterol lowering effects; indeed, this was argued by Mozaffarien in his meta-analysis, given that the CHD risk reductions from replacing SFA with PUFA could not be accounted for entirely by lipoprotein changes alone (this was also noted and discussed in the DGAC). Edit: Reijo Laatikainen reminded me w/ comments on twitter that if we used ApoB as an outcome, as suggested by the IOM, there is a reduction in ApoB from replacing SFAs with PUFAs, and not much of a change if CHO are replaced with SFAs, but again, there is heterogeneity likely related to the type of carbohydrate (17, 18).
UPDATE: As I've thought about this further, I've realized that one of the most confusing things about this commentary is how vague the recommendations they are, and how little they fit into the context of what the DGAC is charged with. The dietary guidelines are supposed to be based off of the IOM's recommendations. For macronutrients, one of the ways that these are set are with the acceptable macronutrient distribution ranges (AMDRs). These set ranges for individual macronutrients. Generally, nutrition recommendations focuses within a macronutrient, and gives recommendations on how to improve the nutrient density of the diet within a macronutrient (e.g. choose whole grains over refined, choose unsaturated over saturated fats). With this commentary, AND crossed over macronutrient recommendations, without making it clear how this fits within the context of the AMDRs. Are they saying that individuals should choose the lower end of the CHO AMDR and the higher end of the fat AMDR, replacing some of those CHO with PUFAs? (if that's true, most Americans can't replace too many carb/sugar kcals with PUFAs without quickly reaching the PUFA upper limit set by the IOM). I find the commentary quite confusing. Apparently, so have others, especially those who are using this commentary to promote very low carbohydrate diets. It's imperative that AND clear up what is meant by this recommendation, how it fits within the IOM guidelines (if it is meant to be interpreted in such a way).
I'll be the first to admit that the evidence surrounding the best macronutrient distribution for cardiovascular health remains unclear. The optimal amount of carbohydrate and fat continues to elude researchers. However, the commentary given by AND on the DGAC oversimplifies this picture and should be clarified. Above all, it would be prudent for all in the field to drive home that maintaining a health body composition and remaining physically active are the most important things for health, trumping any small differences in risk conferred by fatty acids or minimal differences in glycemic indices.
*Special thanks to Colby Vorland for reviewing this blog post and providing commentary prior to its publishing*
1. http://ajh.oxfordjournals.org/content/27/9/1143.full
2. http://ajcn.nutrition.org/content/89/5/1425.full.pdf+html
3. http://ajcn.nutrition.org/content/91/6/1764.long
4. http://www.ncbi.nlm.nih.gov/pubmed/21735388
5. http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1000252
6. http://www.ncbi.nlm.nih.gov/pubmed/21118617
7. http://www.ncbi.nlm.nih.gov/pubmed/21679479
8. http://www.ncbi.nlm.nih.gov/pubmed/21270379
9. http://advances.nutrition.org/content/6/3/326S.full.pdf+html
10. http://www.iom.edu/Reports/2010/Evaluation-of-Biomarkers-and-Surrogate-Endpoints-in-Chronic-Disease.aspx
11. http://ajcn.nutrition.org/content/100/3/765.long
12.http://jaha.ahajournals.org/content/4/1/e001355.abstract
13 http://www.nejm.org/doi/full/10.1056/NEJM199711203372102
14. http://www.ncbi.nlm.nih.gov/pubmed/25514303
15. http://jama.jamanetwork.com/article.aspx?articleid=2214081
16. http://www.ncbi.nlm.nih.gov/pubmed/23786819
17. http://www.ncbi.nlm.nih.gov/pubmed/12716665
18. http://ajcn.nutrition.org/content/77/5/1146/T1.expansion.html
In short, the report comments on the evidence-base used by the Dietary Guidelines Committee, their thoughts on dietary patterns, makes nice comments on the struggle to determine the amount of phosphorus in foods from food labels, supports the use of the DGAC's social ecological model to make recommendations, and makes very vague comments on the sustainability issue. The report also touches on "cross-cutting topics of public health importance", highlighting 2 nutrients, sodium and saturated fats.When the report touches on this latter part, however, we stumble onto some scientific commentaries that rely on questionable evidence, or interpretations there of.
For sodium, AND comments that the report relies on literature looking only at blood pressure as an outcome, acknowledging topics I've discussed on this blog regarding the complicated research problem child that sodium can be. It was disappointing to see a lack of a critical perspective taken by AND on the lower quality data that has led to 'J-shaped' curves (see my previous posts on sodium regarding this) and wish the comments would've given more of a shoutout to emphasizing the potassium to sodium ratio, instead of setting sodium recommendations in isolation. I certainly acknowledge that there is a rather complex sodium controversy, which can be viewed through many lenses, but I also acknowledge that much of the controversy is driven by poor quality data (1), using less than gold standard methodologies for determining regular sodium intake. A recent read on the methods used to determine sodium intake was recently published in Annual Reviews of Nutrition here that I would suggest reading for a discussion of the limitations of many of the methods widely used to estimate sodium intake.
 |
| AND's equations for why carbohydrates are worse than SFAs. |
AND's comments on saturated fat are of particular concern; the authors used simple equations to illustrate that carbohydrates have a greater effect on CVD if they replace SFAs, and cite Jakobsen's (2) pooled analysis of 11 cohorts to support this claim. This evidence does not come from RCTs, so we can't causally back up these claims. There are no RCTs looking at replacement of SFA with CHO that demonstrate increase/decrease CVD events; the best data we have to go on to look at this is observational data. In the 2 models used in Jakobsen's meta analysis, CHO replacing SFA has a hazard ratio of 1.06 and 1.07 for coronary events (not increased risk for morality), but as the authors note, the glycemic index of the carbohydrate was not considered here (fiber was included in the analyses). Further analyses by Jakobsen (3) that used a pseudo-glycemic index(see their methods) stratification of CHO quality see a non-significant inverse association with replacing low-GI carbohydrates for saturated fats (for myocardial infarction) - though this notion of low-GI carbohydrates providing some cardiovascular benefit is not supported by clinical trials (14-16). Jakobsen's (2) lack of a strong effect of replacing SFA with CHO is likely due to heterogeneity in the included cohorts; an analysis of the NHS by Hu et al in the NEJM in 1997 (13), which contained overlapping participants within the Jakobsen's pooled cohort analysis, suggested that each increase of 5 percent of energy intake from saturated fat, as compared with equivalent energy intake from carbohydrates, was associated with a non-significant 17 percent increase in the risk of coronary disease . Clearly, there is conflicting data regarding CHO, SFAs and their impact on cardiovascular outcomes. It is unclear based on the wording of this commentary whether AND means that PUFAs should replace all carbohydrates, or added sugars, as their recommendation statement is quite ambiguous in its wording, and is followed by a statement regarding added sugars. It is imperative that AND use more clear, direct language as this has already been misinterpreted, across social media, to be AND coming out in support of low carbohydrate diets. As it is interpreted now, the model employed by this AND commentary is overly simplistic, and one that is not supported by stronger clinical evidence. The trials that we have looking at CVD end points replace SFAs (and sometimes TFAs) with mixed PUFAs, not CHO with PUFAs (4-8) - these largely back up the recommendation to replace SFA with PUFAs, not food frequency questionnaire based epidemiological associations. Penn Kris-Etherton, PhD,RD and distinguished professor of nutrition at Penn State recently released an Advances in Nutrition Review (9) on the topic of fatty acids, and re-iterates the need for more research looking at the type of carbohydrate's effect on CVD risk, seeing as carbohydrates will likely serve as some of the replacement macronutrient for those trying to reduce SFAs and meet the AHA recommendations of 5-6% of calories (this being derived from dietary patterns like DASH and DELTA). I'm enthusiastic that the field is starting to acknowledge that not all SFAs are created equal, and there's controversial, interesting data for dairy fats and chocolate/tropical fats, but I'm hesitant to jump on any 'saturated fat is good/better for you train' until strong clinical evidence says so - as of now, trials that replaced saturated fat sources with PUFAs did so rather indiscriminantly. It would seem reasonable that AND highlight that dietary patterns, and recent (imperfect) clinical trials employing mediterranean style diets, all emphasize lower saturated fat diets ( less than 10% kcals), and that this focus on saturated fat is quite limiting; indeed, evidence might question their 'harm', but evidence of causal, health-promoting properties is quite sparse.
The AND commentary also appears to misconstrue an IOM document (10) evaluating biomarkers and surrogate endpoints in chronic disease to make strong claims about the use of lipoprotein endpoints . AND states that "While the body of research linking saturated fat intake to the modulation of LDL and other circulating lipoprotein concentrations is significant, this evidence is essentially irrelevant to the question of the relationship between diet and risk for cardiovascular disease" and " this body of evidence that uses lipoproteins as surrogate endpoints for cardiovascular disease must be excluded from considerations of the impact of diet on cardiovascular health". The actual IOM document doesn't quite say these things. The section on LDL in this document discusses some of the complexities of LDL-C, the better usage of LDL-particle number/non-HDL cholesterol as markers, the pleiotropic effects of drug/statin and estrogen interventions, and the general limitations of LDL-C alone as a biomarker of disease. It did not conclude that lipoprotein changes are irrelevant to cardiovascular health - indeed, those with genetically low/high LDL would likely disagree with that statement. The document largely focused on statins trials to note that just a reduction in LDL-C doesn't simply equate to better cardiovascular outcomes - this should make sense, as interventions using statins don't just lower LDL-C but have multiple other effects. The document doesn't state that LDL is useless but rather that the LDL-C is a general measurement of a heterogeneous lipoprotein, and better measurements can capture its complexity; it highlights more accurate predictors like LDL particle or ApoB number (one might note that nutrition research has kept up with these limitations and begun to look at LDL particle number (12)). I find that the AND commentary's use of this IOM report is misleading and overly reduces the complexity of what the report was trying to say; it concludes: "lowering LDL-C can never be considered a "perfect" indicator across all population groups. That said, there is high probability that lowering LDL for several interventions decreases risk of cardiovascular disease". It does not, as the AND commentary says, make 'irrelevant' the effects of saturated fats on LDL and lipoproteins. If anything, the document should be used to state that the observed beneficial effects of replacing saturated fats with mixed PUFAs cannot only be attributed to just their cholesterol lowering effects; indeed, this was argued by Mozaffarien in his meta-analysis, given that the CHD risk reductions from replacing SFA with PUFA could not be accounted for entirely by lipoprotein changes alone (this was also noted and discussed in the DGAC). Edit: Reijo Laatikainen reminded me w/ comments on twitter that if we used ApoB as an outcome, as suggested by the IOM, there is a reduction in ApoB from replacing SFAs with PUFAs, and not much of a change if CHO are replaced with SFAs, but again, there is heterogeneity likely related to the type of carbohydrate (17, 18).
UPDATE: As I've thought about this further, I've realized that one of the most confusing things about this commentary is how vague the recommendations they are, and how little they fit into the context of what the DGAC is charged with. The dietary guidelines are supposed to be based off of the IOM's recommendations. For macronutrients, one of the ways that these are set are with the acceptable macronutrient distribution ranges (AMDRs). These set ranges for individual macronutrients. Generally, nutrition recommendations focuses within a macronutrient, and gives recommendations on how to improve the nutrient density of the diet within a macronutrient (e.g. choose whole grains over refined, choose unsaturated over saturated fats). With this commentary, AND crossed over macronutrient recommendations, without making it clear how this fits within the context of the AMDRs. Are they saying that individuals should choose the lower end of the CHO AMDR and the higher end of the fat AMDR, replacing some of those CHO with PUFAs? (if that's true, most Americans can't replace too many carb/sugar kcals with PUFAs without quickly reaching the PUFA upper limit set by the IOM). I find the commentary quite confusing. Apparently, so have others, especially those who are using this commentary to promote very low carbohydrate diets. It's imperative that AND clear up what is meant by this recommendation, how it fits within the IOM guidelines (if it is meant to be interpreted in such a way).
I'll be the first to admit that the evidence surrounding the best macronutrient distribution for cardiovascular health remains unclear. The optimal amount of carbohydrate and fat continues to elude researchers. However, the commentary given by AND on the DGAC oversimplifies this picture and should be clarified. Above all, it would be prudent for all in the field to drive home that maintaining a health body composition and remaining physically active are the most important things for health, trumping any small differences in risk conferred by fatty acids or minimal differences in glycemic indices.
*Special thanks to Colby Vorland for reviewing this blog post and providing commentary prior to its publishing*
1. http://ajh.oxfordjournals.org/content/27/9/1143.full
2. http://ajcn.nutrition.org/content/89/5/1425.full.pdf+html
3. http://ajcn.nutrition.org/content/91/6/1764.long
4. http://www.ncbi.nlm.nih.gov/pubmed/21735388
5. http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1000252
6. http://www.ncbi.nlm.nih.gov/pubmed/21118617
7. http://www.ncbi.nlm.nih.gov/pubmed/21679479
8. http://www.ncbi.nlm.nih.gov/pubmed/21270379
9. http://advances.nutrition.org/content/6/3/326S.full.pdf+html
10. http://www.iom.edu/Reports/2010/Evaluation-of-Biomarkers-and-Surrogate-Endpoints-in-Chronic-Disease.aspx
11. http://ajcn.nutrition.org/content/100/3/765.long
12.http://jaha.ahajournals.org/content/4/1/e001355.abstract
13 http://www.nejm.org/doi/full/10.1056/NEJM199711203372102
14. http://www.ncbi.nlm.nih.gov/pubmed/25514303
15. http://jama.jamanetwork.com/article.aspx?articleid=2214081
16. http://www.ncbi.nlm.nih.gov/pubmed/23786819
17. http://www.ncbi.nlm.nih.gov/pubmed/12716665
18. http://ajcn.nutrition.org/content/77/5/1146/T1.expansion.html
I have a few questions. I'll readily admit that my information is cursory in this field. And correct me if I'm wrong, but only taking into account RCT and controlled feeding studies, although very important in particular scientific endeavours, may not be the most helpful perspective.
ReplyDeleteGiven how many societies with minimal coronary artery disease have subsided on primarily starch/CHO, and how diets like that also has shown reversal of CAD in patients, why is CHO continuously being questioned and low-fat diets "debunked". I have heard of quite a few examples of reversal of CAD with low-fat diets, but not with 30% fat diets. Hasn't also controlled feeding studies with chimps also shown that PUFAs are atherogenic? Not much reason to think it's different for people.
Some of the controlled feeding studies I've looked at "incriminating" CHO have used liquid diets, and using sugar for at least some of the CHO, and possibly also overfed them. Some of them seemed to match fibre intake between the various diets, putting higher CHO diets at an unfair/unnatural disadvantage, as true low-fat diets generally aren't fibre-deplete, junk-food diets.
I've seen many studies on plant-based low-fat diets which improve pretty much all relevant lipid biomarkers, with no increases in triglycerides, also. People on traditional high-CHO don't have high triglycerides, either.
Thanks for this work.
Hi Bard,
DeleteYou bring up points that many utilize to challenge the simplistic notion that carbohydrates are atherogenic. The low-fat diets you mention I believe are referring to Ornish and Esseltyn's studies, which have some problems of their own. I think 'regression' is more an appropriate term, and not reversal of CAD. Ornish employed multiple lifestyle changes, not just diet, and didn't have other intervention groups to really tease apart the effect of diet alone - we can't really say that those effects wouldn't have occurred with a different fat percentage.
I've only seen some chimpanzee literature on PUFAs and CVD - could you pass along the study you're thinking of? In evidence based medicine/nutrition, we tend to only use human work, and pull in animal data for mechanistic insights. The notion that PUFAs are atherogenic is contentious, but it's generally not supported by prospective cohort data (http://www.ncbi.nlm.nih.gov/pubmed/25161045), and Ramsden's analysis of Sydney has been highly contentious.
Unfortunately, observational evidence from other societies doesn't apply super well to our population. Much of the Harvard folk seem to think that CHO's (especially simple carbs) are less detrimental for more lean, physically active individuals - I don't find this to be well supported by RCTs, but it's still a popular opinion amongst some researchers. It would be easy to simply look at other population who eat carbohydrates and think that carbohydrates are fine; alas, that's not how the evidence-based game works.
The continuous problem with nutrition is considering what kind of evidence we choose to go off of - many don't like RCTs because they're too short or reductionist, and many don't like associative data, even prospective cohorts, because of confounding. Ideally, they would all align and make one nice, solid recommendation. Alas, the debates continue.
My personal opinion on the matter is that we're debating minutia, like macronutrient distributions, when maintaining a healthy body composition and remaining physically active are overwhelming contributing to our cardiovascular health, much more than debates about whether 11% of our diets from SFA are worse than 7%.
This comment has been removed by the author.
ReplyDeleteThanks very much for your reply.
ReplyDeleteThe low-fat diets I think of would typically be those of rural Asia or Africa where consistently many diseases, but maybe most noticeably CAD-incidence has been low. Referring to documented results with CAD in clinical settings, Esselstyn and Ornish for sure, but people further back, like Pritikin or Lester Morrison have shown equivalent results, I believe.
chimpanzees - I had this from a secondary source, so I may not have been accurate. Here's what I found, on monkeys:
http://ajcn.nutrition.org/content/62/2/463S.abstract
They do say PUFA reduces atherosclerosis, but looking at the results (Table 8) in some arms with high PUFA (and high cholesterol), there's substantial intimal thickening. This is to say that PUFA substituting SFA is better for reducing atherosclerosis buildup, but may in some cases not protect against it.
Macronutrient distribution - sure, it's just one thing, and things like fibre, salt and some other things may be more important than this in and of themselves. But I wouldn't say it's minutia. I believe cholesterol matters a whole lot to minimize CAD. I would think it's difficult to get a cholesterol of 150(say) on most diets at 30% fat. (Yes, cholesterol is but one factor, but personally I think it's an important and instrumental one. I think I could summon many experts' agreement here.)
"when maintaining a healthy body composition and remaining physically active are overwhelming contributing to our cardiovascular health, "
I kind of wouldn't want to disagree with the point about being fit and trim, but I feel this is too narrow and in fact also partly wrong. CAD has in fact been going down while obesity and diabetes (and inactivity as well?) has risen dramatically. Go figure, right? I don't think this can be all due to advances in medicine. Heart disease is still the biggest cause of death in industrialized nations. The bar should be set higher than preventing obesity and getting more active. Also, typically, this puts the blame on each individual, where obviously societal forces are behind the current obesity epidemic. With healthy plant food much would follow naturally. Think of the North Karelia project for instance. Remove incentives for unhealthy foods and have a consistent message about a healthy diet.