TL;DR: C-reactive protein is an acute phase protein your liver makes in response to inflammatory signalling. The high-sensitivity assay, hs-CRP, can detect the low concentrations relevant to chronic low-grade inflammation rather than just the dramatic spikes of acute infection. People with persistently higher hs-CRP have higher rates of cardiovascular events and higher all-cause mortality across many large cohorts, and that association is robust. What is far less clear is whether the inflammation is causing the damage or reporting on it. Mendelian randomisation work has generally failed to support CRP itself as a causal agent, which suggests CRP is a messenger rather than the message. That distinction matters enormously: it means lowering CRP for its own sake is probably pointless, while addressing what is driving it may not be. The CANTOS trial, which targeted inflammation directly with an anti-inflammatory antibody, did reduce cardiovascular events, providing genuine support for the inflammatory hypothesis even though CRP itself is likely just a marker. Practically: hs-CRP is cheap, easy to add to a blood panel, and useful as a rough gauge, but a single reading tells you very little because it moves wildly with any infection.
What is hs-CRP and how is it different from regular CRP?
Same molecule, different assay sensitivity. C-reactive protein is produced by the liver largely in response to interleukin-6 signalling, and its concentration in blood can rise more than a thousandfold during acute inflammation. Standard CRP assays were designed to detect that dramatic range, because their clinical purpose was identifying and monitoring infection and inflammatory disease. They are not precise at the low end.
The high-sensitivity assay measures the same protein but resolves the low concentrations, roughly the range from under 1 mg/L up to about 10 mg/L, that turn out to matter for cardiovascular risk prediction. Nothing else is different. If you have rheumatoid arthritis or a chest infection, your doctor orders CRP. If you are trying to gauge chronic low-grade inflammation in an otherwise well person, you want hs-CRP.
The word "inflammation" does a lot of unhelpful work in wellness content, so it is worth being concrete. Acute inflammation is a fast, coordinated response to injury or infection, and it is protective. Chronic low-grade inflammation is something else: a persistent, modest elevation in inflammatory signalling without an obvious acute trigger, associated with ageing, adiposity, smoking, poor sleep, and metabolic dysfunction. It is this second thing that hs-CRP is being used to detect, and the term "inflammageing" is sometimes used in the ageing research literature to describe the pattern.
Does high hs-CRP predict death?
Yes, consistently, and the association survives adjustment for the obvious confounders. This is one of the better-replicated findings in cardiovascular epidemiology. Analyses from large prospective cohorts, including work from the Emerging Risk Factors Collaboration which pooled data across many studies, have found that higher CRP is associated with higher rates of coronary heart disease, stroke, and mortality. Framingham, MONICA, and numerous other named cohorts have reported the same direction of effect.
The relationship also extends beyond cardiovascular disease. Higher CRP has been associated with cancer mortality, with all-cause mortality, and with outcomes like frailty and cognitive decline. That breadth is itself informative, and it hints at why interpreting CRP is tricky: a marker that predicts everything is usually reporting on general physiological deterioration rather than pointing at a specific mechanism.
Now the crucial caveat. Prediction is not causation, and in the case of CRP the distinction is unusually well studied. Mendelian randomisation studies, which examine whether genetic variants that raise lifelong CRP levels also raise disease risk, have generally not found that they do. People genetically predisposed to higher CRP do not appear to have correspondingly higher coronary risk. That is fairly strong evidence that CRP itself is not doing the damage. It is a byproduct of an inflammatory process, and the process, or whatever is driving it, is the thing that matters.
Compare that with apoB or LDL cholesterol, where Mendelian randomisation strongly supports causation. The genetic evidence points in opposite directions for the two markers, and that is the single most important thing to understand about CRP.
If CRP is not causal, is inflammation still worth worrying about?
Probably yes, and the evidence for that comes from a different direction entirely.
CANTOS was a large randomised trial testing canakinumab, a monoclonal antibody targeting interleukin-1 beta, in people who had already had a heart attack and had raised hs-CRP. It found a reduction in recurrent cardiovascular events compared with placebo, independent of any lipid lowering. That result is important because it was the first reasonably clean demonstration that targeting inflammation directly, without touching cholesterol, reduces cardiovascular events. It supported the inflammatory hypothesis of atherosclerosis as more than a correlation.
It also came with a serious caveat that gets left out of enthusiastic summaries: canakinumab increased fatal infections, which offset some of the benefit, and the drug was not approved for this indication. There has also been trial work with colchicine, an old and cheap anti-inflammatory, in coronary disease, with results that have generally been encouraging for event reduction while raising their own questions about adverse effects and about which patients benefit.
So the honest reading is: inflammation is plausibly part of the causal story in atherosclerosis. CRP is how we watch it, not how it works. Suppressing the immune system carries real costs, which is why "reduce inflammation" is not the simple win that supplement marketing implies. The immune system is not a dial you want turned down indiscriminately.
What is a normal hs-CRP level?
The conventional cardiovascular risk stratification, which came out of work associated with the Reynolds Risk Score and related research programmes, divides hs-CRP into three broad bands: under 1 mg/L as lower risk, 1 to 3 mg/L as intermediate, and above 3 mg/L as higher risk. AHA and CDC guidance has used these cut points.
Above roughly 10 mg/L, the number is generally taken to reflect an acute process, an infection, injury, or a flare of inflammatory disease, rather than chronic low-grade inflammation. The standard advice is to disregard such a reading for risk-stratification purposes and repeat it once well.
These bands are useful and also somewhat arbitrary. They are population cut points imposed on a continuous variable, and the underlying relationship with risk is graded rather than stepped. There is nothing magical that happens as you cross 3 mg/L.
The far more important point about interpretation is variability. Your hs-CRP can change by a large factor within days. A cold, a dental infection, a hard unaccustomed workout, a bad night, a recent vaccination, all of these move it. One reading is close to meaningless for characterising your chronic state. If you want a usable number, you need at least two readings taken a couple of weeks apart while you are well, and ideally more. Anyone selling you a single hs-CRP as a window into your inflammatory status is overselling a noisy measurement.
Several other things shift the number in ways that have nothing to do with cardiovascular risk. Adiposity is the biggest one: adipose tissue produces interleukin-6, so CRP correlates strongly with body fat, and much of CRP's apparent predictive power may simply be reporting on fat mass. Oestrogen-containing contraceptives and hormone therapy raise CRP. Statins lower it. Chronic inflammatory conditions such as rheumatoid arthritis, inflammatory bowel disease, and psoriasis raise it substantially, and in those people hs-CRP is not interpretable as a cardiovascular marker in the usual way.
What actually lowers hs-CRP?
The interventions with the best evidence are the ones you would expect, and they are worth doing regardless of what they do to CRP.
Losing excess fat mass, particularly visceral fat, produces the most reliable reductions. Given the adipose-IL-6-CRP link, this is mechanistically unsurprising, and the effect is consistent across weight loss trials.
Regular physical activity lowers CRP, though the effect is smaller once you account for the accompanying fat loss. There is a wrinkle worth knowing: a single hard or unaccustomed exercise session transiently raises CRP for a day or two as part of normal tissue repair. Do not test the morning after a marathon. The chronic effect of training is downward; the acute effect of a session is upward. Our exercise for longevity protocol and the walking evidence both cover the activity side.
Stopping smoking lowers CRP, and smoking cessation is the single highest-yield health behaviour available to anyone who smokes, by a very wide margin.
Dietary patterns matter modestly. Mediterranean-style eating patterns have been associated with lower inflammatory markers in trials and cohorts, and the general direction, more plants, more fibre, more unsaturated fat, fewer ultra-processed foods, is consistent with lower CRP. Our pieces on the MIND diet and fermented foods and gut longevity are relevant. The effect sizes from diet alone, independent of weight change, are real but not dramatic.
Sleep and stress are genuinely part of this picture. Sleep deprivation and chronic psychological stress both raise inflammatory signalling, and the mechanisms are reasonably well characterised. Our post on chronic stress and longevity covers overlapping ground. Prolonged sedentary time is independently associated with inflammatory markers, as discussed in sitting time and mortality.
On supplements, and this is where honesty is required: many things lower CRP a little in trials, including omega-3 fatty acids, curcumin, and vitamin D. Almost none of them have shown that this translates into fewer events or longer life. Vitamin D supplementation trials, despite promising observational associations, have not demonstrated the mortality benefits people hoped for. That is the pattern to expect. Given that CRP is probably not causal, lowering it with a supplement is a bit like taking the batteries out of a smoke alarm. The number goes down. Whether anything real changed is a separate question, and one the trials have not answered affirmatively.
Should you get hs-CRP tested?
It has a place, but a narrower one than the biohacking discourse suggests.
The reasonable case: you are at intermediate cardiovascular risk on a standard risk calculator, and the decision about whether to start preventive treatment could go either way. In that situation hs-CRP is recognised in both American and European guidance as a risk-enhancing factor that can shift a borderline call. That is a real, guideline-supported use.
The weaker case: you are well, low risk, and curious. You will get a number that bounces around, that mostly reflects your body fat and whether you have a cold, and that will not change anything you should do. The actions suggested by a high CRP, lose fat, exercise, stop smoking, sleep, are things you should do anyway.
The case where it actively misleads: you have an inflammatory condition, you are on oestrogen-containing medication, or you tested during an illness. Here the number means something different from what the risk bands assume.
If you are going to test, test twice, at least two weeks apart, while well, and take the lower or the average rather than reacting to a spike. And treat it as one input among many rather than a verdict. Composite approaches to ageing assessment, discussed in the science of biological ageing, are more robust than any single inflammatory marker.
Frequently Asked Questions
What if my hs-CRP is high but everything else looks fine? First, repeat it while well and at least two weeks from any illness, injury, vaccination, or hard unaccustomed exercise, because a single elevated reading most often reflects something transient. If it is persistently above 3 mg/L across repeated tests, it is worth raising with a clinician, who will want to consider whether there is an undiagnosed inflammatory condition, look at your body composition, and factor it into your overall cardiovascular risk assessment. A persistently high reading with no explanation deserves investigation rather than a supplement.
Do anti-inflammatory supplements reduce my risk of dying? There is no good evidence that they do. Several supplements can lower CRP modestly in trials, but since CRP appears to be a marker rather than a cause, lowering the marker without changing the underlying process is unlikely to help. Large randomised trials of supplements have generally not shown mortality benefits, and the interventions that did reduce events in trials, such as canakinumab in CANTOS, were potent prescription drugs with real adverse effects, not over-the-counter products.
Why does my hs-CRP change so much between tests? Because that is what CRP does. It is an acute phase protein designed to respond rapidly and dramatically to inflammatory signals, and its concentration can change by a large factor within days. Minor infections you barely notice, dental problems, recent vaccination, a hard workout, and poor sleep all move it. This variability is intrinsic to the molecule and is the main reason a single measurement is a poor guide to your chronic inflammatory state.
Is hs-CRP better than a cholesterol test for predicting heart disease? No, and it is not really competing. ApoB and LDL cholesterol have strong causal evidence behind them from genetic studies and from trials showing that lowering them reduces events. CRP has predictive value but the genetic evidence argues against it being causal. In practice, hs-CRP is used to refine a risk estimate at the margins, not to replace lipid measurement. If you can only have one, have the lipid panel.
Does obesity explain most of a high hs-CRP? It explains a great deal of it in many people. Adipose tissue, especially visceral fat, secretes interleukin-6, which drives hepatic CRP production, so CRP correlates strongly with fat mass. This is one reason interpreting CRP as an independent risk marker is difficult: it may partly be measuring adiposity by a more expensive route. It is also why fat loss is the most reliable way to lower it.
Medical Disclaimer
This article is for general information only and does not constitute medical advice, diagnosis, or treatment. It is not a substitute for consultation with a qualified healthcare professional who knows your medical history. Biomarker results, including hs-CRP, cannot be interpreted meaningfully in isolation and require clinical context, repeated measurement, and consideration of your full health picture to be useful. Do not start, stop, or change any medication or supplement on the basis of this content, and do not act on a laboratory result without discussing it with a clinician. This is particularly important if you have an inflammatory or autoimmune condition, cardiovascular disease, an active infection, or if you are pregnant, breastfeeding, or taking hormonal medication, all of which affect how an inflammatory marker should be interpreted. If you have symptoms that concern you, seek medical attention rather than relying on information found online.