Big data on cardiovascular disease

Investment Analysis Clubs Macro Economic Trends and Risks
1

Cardiovascular disease has Macroeconomic significance because it’s the #1 killer of Americans and also has tremendous health care costs.

Unlike many health-associated reports, new research is based on big data ( 27,939 female health care providers living in the United States studied over 30 years beginning at age 55). The impact of 3 easily-measured factors is huge.

When all three measures — LDL cholesterol, Lp(a), and C-reactive protein (a measure of inflammation) — were assessed together, participants with the highest levels had more than a 1.5-times increased associated risk for stroke and more than a three-times increased associated risk for coronary heart disease compared to women with the lowest levels.

https://www.nejm.org/doi/full/10.1056/NEJMoa2405182

This clearly points to a need for everyone to have these three factors measured at age 55 and treated if necessary.

High LDL cholesterol can be treated with statins. Chronic inflammation is associated with many diseases, not just cardiovascular disease. Inflammation can be prevented with healthy lifestyle choices.

Statins are inexpensive and lifestyle choices are free. Prevention is much cheaper than cure.

Wendy

4 Likes
2

1.5-times increased associated risk

1.5 times what?

2 Likes
3

more than a 1.5-times increased associated risk for stroke and more than a three-times increased associated risk for coronary heart disease compared to women with the lowest levels.

4

Although this particular study had age 55 as the lower limit, this would be way late for someone with elevated Lp(a) to be initiating therapy or lifestyle interventions aimed at primary prevention. Even 40…which, for some reason, is the lower limit on the ASCVD risk calculator…would have someone who’s at high risk in the early stages of disease already.

To use a financial analogy, ASCVD is a progressive disease that harnesses the power of compound interest very effectively. The earlier the process begins (and childhood marks the start time if genetic predisposition is strong enough) the more rampant the disease.

1 Like
5

So, as I mentioned in the “breadcrumbs” thread, this study has been doing the rounds of the various health related sites I follow. So far, though, I’ve noticed very little discussion on the timeline under study (30 years) vs what’s seen as a reasonable timescale for assessment using the parameters of the standard ASCVD risk calculator (10 years…which is no time at all in the continuum from healthy homeostasis to symptomatic disease/death). This is in spite of the suggestion for a change here in the study conclusions.

At risk of repetition, here’s a take on that subject from a few years ago …

Attia is a bit wordier than I on the subject … but the power of compound interest seems to definitely be a good analogy.

2 Likes
6

What about compared to women with average levels?

7

What about compared to women with average levels?

I included the link to the research.
Wendy

8

From the link -

Women’s Health Study ClinicalTrials.gov number, NCT00000479.

I don’t quite understand it. The research appears to be about a large trial of various combinations of vitamin C/vitamin E/placebo over a period of some 10 years. But the NEJM abstract is talking about CRP, LDL, and LP(a). Could the NEJM have included the wrong link to the research data? Or maybe they are linking to a previous shorter study? But where is the data from this study?

9

@MarkR the link worked for me.

https://www.nejm.org/doi/full/10.1056/NEJMoa2405182

Inflammation, Cholesterol, Lipoprotein(a), and 30-Year Cardiovascular Outcomes in Women

Authors: Paul M Ridker, M.D. ORCID, M. Vinayaga Moorthy, Ph.D., Nancy R. Cook, Sc.D., Nader Rifai, Ph.D., I-Min Lee, Sc.D., and Julie E. Buring, Sc.D.Author Info & Affiliations

Published August 31, 2024

Background

High-sensitivity C-reactive protein (CRP), low-density lipoprotein (LDL) cholesterol, and lipoprotein(a) levels contribute to 5-year and 10-year predictions of cardiovascular risk and represent distinct pathways for pharmacologic intervention. More information about the usefulness of these biomarkers for predicting cardiovascular risk over longer periods of time in women is needed because early-life intervention represents an important risk-reduction method.

Methods

We measured high-sensitivity CRP, LDL cholesterol, and lipoprotein(a) levels at baseline in 27,939 initially healthy U.S. women who were subsequently followed for 30 years…

(Funded by the National Institutes of Health; Women’s Health Study ClinicalTrials.gov number, NCT00000479.)

ClinicalTrials.gov – The purpose of this study is to evaluate the effects of low-dose aspirin and vitamin E in primary prevention of cardiovascular disease and cancer in apparently healthy women.

I think you are correct – the link is wrong.

Abstract

Background

High-sensitivity C-reactive protein (CRP), low-density lipoprotein (LDL) cholesterol, and lipoprotein(a) levels contribute to 5-year and 10-year predictions of cardiovascular risk and represent distinct pathways for pharmacologic intervention. More information about the usefulness of these biomarkers for predicting cardiovascular risk over longer periods of time in women is needed because early-life intervention represents an important risk-reduction method.

Methods

We measured high-sensitivity CRP, LDL cholesterol, and lipoprotein(a) levels at baseline in 27,939 initially healthy U.S. women who were subsequently followed for 30 years. The primary end point was a first major adverse cardiovascular event, which was a composite of myocardial infarction, coronary revascularization, stroke, or death from cardiovascular causes. We calculated the adjusted hazard ratios and 95% confidence intervals across quintiles of each biomarker, along with 30-year cumulative incidence curves adjusted for age and competing risks.

Results

The mean age of the participants at baseline was 54.7 years. During the 30-year follow-up, 3662 first major cardiovascular events occurred. Quintiles of increasing baseline levels of high-sensitivity CRP, LDL cholesterol, and lipoprotein(a) all predicted 30-year risks. Covariable-adjusted hazard ratios for the primary end point in a comparison of the top with the bottom quintile were 1.70 (95% confidence interval [CI], 1.52 to 1.90) for high-sensitivity CRP, 1.36 (95% CI, 1.23 to 1.52) for LDL cholesterol, and 1.33 (95% CI, 1.21 to 1.47) for lipoprotein(a). Findings for coronary heart disease and stroke appeared to be consistent with those for the primary end point. Each biomarker showed independent contributions to overall risk. The greatest spread for risk was obtained in models that incorporated all three biomarkers.

Conclusions

A single combined measure of high-sensitivity CRP, LDL cholesterol, and lipoprotein(a) levels among initially healthy U.S. women was predictive of incident cardiovascular events during a 30-year period. These data support efforts to extend strategies for the primary prevention of atherosclerotic events beyond traditional 10-year estimates of risk. (Funded by the National Institutes of Health; Women’s Health Study ClinicalTrials.gov number, NCT00000479.)

Notes

This article was published on August 31, 2024, at NEJM.org.

A data sharing statement provided by the authors is available with the full text of this article at NEJM.org.

I tried to download the data but wasn’t able to. It’s probably restricted to paid subscribers to the NEJM.

It’s very common for summaries to compare the highest with the lowest participants. Obviously, it’s a sliding scale so the average will be somewhere in between.

Wendy

1 Like
10

I don’t think the link was wrong so much as the wrong link was used to try to access the primary document (which, like you suggested, is behind a paywall). The link cited was actually to what looks like a different study that had NIH funding…the Women’s Health Study…and this Lp(a) etc paper was actually a post hoc analysis on the biomarkers measured in the stored blood samples. Something I assumed anyway the first time I read the abstract as hs-CRP measurement wasn’t developed until the early aughts so had to be a retroactive measurement.

I clicked on a few of the links you provided and, sure enough, the quintiles from the lowest to highest values of LDL-C in mg/dL were available. No “average” per se but a measured range.

1 Like
11

It’s interesting now the novelty of seeing this study…and specifically the mention of Lp(a) in yet another context…I’ve started to think of it in generalities. And wondering.

Nearly 30,000 women starting in their 50s. Mentioned as being healthy. I can’t be bothered to download the full data load as I suspect the statistical analysis of the biomarkers and their impact is a big feature and I can tell from some of what I’ve read so far, it’ll be beyond me. However, in addition to some of the figures I found in the links on @WendyBG 's last post (LDL-C levels) I also noticed a few other things that make me wonder.

Apparently, within this allegedly healthy cohort, about 25% had hypertension and a smaller number had T2D. Now, I know that those figures would probably be way higher nowadays (30 years on, given the rise in obesity since) but srsly…who on earth would consider themselves “healthy” if they had these diseases in their 50s (regardless of whether they’re managed to within normal limits by medication)?

12

I thought this factor which was popular in the mid 1990s is now seen as just something off to the side that is not very determinant.

I am no expert in that. But that factor in the 90s was suspect even then. Doctors quickly put very little stock in it.

The result is exactly what should be expected otherwise. It is also why statins are so frequently used in general with the aging process.

13

Nope, the test for hs-CRP became available for use early 2000s

14

Sometimes I see a questionnaire that asks “How is your health? Excellent, good, fair, poor.” I wonder if they are asking “Without meds or with meds?”

With meds I can do an hour of Zumba. Isn’t that pretty healthy? But without meds my asthma would interfere.

I guess the nurses in the study see plenty of sick people who have really poor health and can’t function even with the best medical care. So they think that their health is good since they can work as nurses as long as they follow medical care.

I really can’t answer that question now since I’m heading toward open-heart surgery. Is my health good because I still work out 5 days a week? Or is it poor because I’ll die sooner if my heart isn’t fixed?

Wendy

2 Likes
15

Well, I guess I really meant folk like the study authors rather than public in general (they are folk you’d expect to be somewhat discriminating about disease states that are quite strongly associated with cardiovascular disease, after all)…although I would still imagine high BP and T2D to be conditions that imply, at least, sub optimal health for anyone concerned about their health, wellbeing and longevity. Although there are probably plenty of studies to show there are still significant numbers of folk walking around and unaware that they have them?

16

Lipoprotein(a) is mainly a genetic factor that affects about 20% of the population and is a strong independent predictor of risk of cardiovascular disease-it’s not affected much by lifestyle but rather measures how at risk you are for lifestyle factors such as diet to give you heart disease/kill you. In my opinion, it’s worth knowing even earlier than 55 because you don’t want to wait until you have arterial build up to act. I went on statins early because of my Lp(a) score even though i eat a pretty healthy diet (am a nutrition PhD). And then I told everyone in my family to get an Lp(a) test- the two that did were also at heightened risk- it really runs in families! I found a company that will analyse it cheaply ($24) for those whose doctors/insurance balk at it (not all dr’s are educated on this). It’s called “Precision Health Reports”. Haven’t used them yet though.

3 Likes
17

@WendyBG …Two years ago, I definitely thought I was still healthy. I’d had my Lp(a) measured and high, my CAC scan and high, started on statins and my LDL-C was down to something like 95 mg/dL (which I thought was good)…but hadn’t yet had my first visit with my intervention cardiologist (andCT Angiogram) to inform me that I was way past my primary prevention date!. Heck, I’d even run a decent time on the Golden Gallop 10k after a 3 months stint of high intensity training as a supercompensation tactic for my upcoming lapiplasty. I definitely don’t consider myself as healthy any longer.

Interestingly, a series of “memories” have come up on my FB page reminding me of where I was two years ago. At my BIL’s house in England and he was being a richard about a medical check up…especially since he’s older than my husband (his brother) and might well have the same aortopathy he had his bio-Bentall for. Also saw it as something chortle-worthy that I was concerned about a condition that was asymptomatic. I have to mention, he tends to offer opinions on medical matters quite confidently as he’s a PhD (biochemistry) and sees no incongruity in the notion of waiting for symptoms to occur…when the very first symptom could be Sudden Cardiac Death🙄

Here’s what reminded me (it’s Facebook so might not be viewable)

https://www.facebook.com/share/p/XuecTXD6Bb9C4Hbu/?mibextid=oFDknk

FYI, Quorn village center hasn’t been photoshopped to remove the crowds…this was taken quite literally during HRH’s state funeral and we’d just come out of a nearly empty pub.

1 Like
18

@VitamanD …that 20% figure is used mainly in reference to the US population and maybe others with a sort of similar demographic. I got a heads-up on the variations for this within historically smaller (more isolated/heterogeneous populations…as you’d expect with a strong genetic input) listening to one of my podcasts early last year. Guest reseacher was Dutch and FH is prevalent in the Netherlands.

I was tooling around looking for an idea of just what that prevalence might be and spotted this among the many (the headlines caught my eye)…

So it seems that there are some authorities making the case for Lp(a) measurement to be a one off standard screening test early on.

The podcast I’m referring to …

1 Like
19

Then why is it in a 30 year study?

I am on a cell but I read on. It was applied later.

It was not important

20

The study is retrospective analysis of data from stored blood from that Womens Health Study …as I mentioned upstream.

Not important to whom…you or as a useful biomarker for prediction of imcreased risk from cardiovascular disease