Counting Cardiovascular Risk Factors — Six Ways It Goes Wrong Before You Do Any Arithmetic

A reader wrote in last week with three numbers and a question. Total cholesterol 225, HDL-C 68, LDL-C 135. His LDL was at or above the threshold, so that was a risk factor. But his HDL was at or above 60, which subtracts one. Did the high HDL cancel the high LDL, leaving him at zero? Or did the subtraction come off the running total, leaving him at minus one? He had asked a language model first. It told him minus one. He could tell the number was wrong without being able to prove it, which is a miserable place to be three weeks out from an exam.

The answer is neither of his two options, and the reason is worth more than the answer. He was treating this as an arithmetic problem — add the positives, subtract the negative, mind the order of operations. It is not an arithmetic problem. The addition is the only easy part. Every place this count goes wrong is a place where somebody read the criteria slightly faster than the criteria deserved.

There are six of those places. None of them is obscure. Two of them are stated outright, in plain sentences, in the textbook most candidates study from — and candidates miss them anyway, because a stated rule you read past is functionally identical to a rule that was never written down.

Why bother, when the count no longer decides anything?

First, an objection that deserves an answer, because it is a good one.

The cardiovascular risk factor count was removed from the preparticipation screening procedure in 2015. It does not decide who needs medical clearance. That decision now runs on three inputs — whether the person currently exercises regularly, whether they have diagnosed cardiovascular, metabolic or renal disease, and whether they have signs or symptoms suggestive of it. If you learned to sort clients into low, moderate and high risk by tallying risk factors, you learned a system that has been retired for a decade. (That is a whole article of its own, and it is here.)

So why is anyone still counting?

Because the Guidelines still ask for it, explicitly, and for reasons that have nothing to do with clearance. The recommendation is that you run the assessment and then put it in front of two people: the client, and whoever manages their care. The count is not a gate any more. It is a communication — a compact summary of what is working against this person’s cardiovascular future, handed to them and to the clinician who manages them. It also shapes what you program, and it remains examinable.

Which means the stakes moved rather than disappeared. A count inflated by double-counting, or deflated by a blank line on an intake form, does not wrongly refer someone to a physician any more. It misinforms a real conversation with a real client. That is arguably worse, because nothing downstream catches it.

First: a category is one line, however many ways you qualify

The most common inflation error is to count evidence instead of counting categories.

The criteria are grouped by domain of risk, not by test. Blood lipids form one domain, and there are several distinct ways to land inside it: an elevated LDL-C, a low HDL-C, an elevated non-HDL-C, or simply being on a lipid-lowering medication. Satisfy one of those and the lipid category is positive. Satisfy all four and the lipid category is still positive — once. The same holds for blood glucose, which can be reached by a fasting value, by a two-hour oral glucose tolerance test value, or by a glycated haemoglobin value. It holds for blood pressure, where a high systolic reading and a high diastolic reading together produce one positive factor, not two.

The design reason is worth holding onto, because it makes the rule impossible to forget. If categories counted per test rather than per domain, a client who happened to arrive with a full laboratory panel would score higher than an identical client who arrived with a partial one — not because they carry more risk, but because someone ordered more blood work. The count would measure the thoroughness of the workup rather than the health of the person. Grouping by domain is what keeps the number meaning something.

The third edition of Resources for the Exercise Physiologist says this outright, in a sentence about blood pressure that most readers glide over on their way to the list. It is not hidden. It is just easy to skip.

Second: the total cholesterol figure is a fallback, not a criterion

This is the one that caught our reader, and it is the single most instructive error in the set, because he got the right answer for the wrong reason — which means he will get the wrong answer the moment the numbers shift.

His reasoning was that his total cholesterol of 225 was above the 200 threshold and would have counted, except that his LDL had already used up the category’s one slot. That is a plausible story. It is not what the criteria say.

The twelfth edition wording is specific: use the 200 mg/dL total cholesterol figure if total serum cholesterol is all that is available. It is not a criterion competing for a slot. It is a substitute you reach for when the better measurements are missing. Our reader had an LDL-C and an HDL-C in hand, so his total cholesterol was not used at all. Not crowded out — simply not part of the calculation.

Same answer today. Different answer on a different panel. Consider a client whose total cholesterol is 210, whose LDL-C is 120, whose HDL-C is unremarkable, and who takes no lipid-lowering medication. Counted our reader’s way, the total cholesterol is at or above 200 and the lipid category is positive. Counted the way the criteria actually work, the total cholesterol figure never enters the reasoning, the LDL-C is below its threshold, and the lipid category is negative. One risk factor, appearing or disappearing, entirely on the strength of how one sentence was read.

There is a reason this particular sentence is treacherous, and it is not the reader’s fault. The two books do not word it the same way. The third edition of Resources for the Exercise Physiologist tells you to use the 200 figure “only if the total serum cholesterol measurement is available” — a phrasing that can be read as whenever you have it. The twelfth edition of the Guidelines says “if total serum cholesterol is all that is available,” which closes that reading. If you studied the first phrasing and never met the second, you were taught a rule that quietly invites the mistake.

Third: the negative risk factor comes off the sum, not off the category

An HDL-C of 60 mg/dL or above is cardioprotective, and it is handled as a negative risk factor. It does not sit inside the lipid category, arguing with the LDL value. It sits on its own, and the note the Guidelines attach to it is precise about where it applies: “one positive risk factor is subtracted from the sum of positive risk factors.”

From the sum. At the end. Not from the lipid category, and not conditional on what the lipid category did.

So both of our reader’s candidate answers were built on a false premise. There is no priority between the two determinations and no cancellation between them; they are simply independent. You decide whether the lipid category is positive using the lipid criteria. Separately, you decide whether the high-HDL subtraction applies. Then you total everything and apply the subtraction once.

For his numbers: the LDL-C of 135 is at or above 130, so the lipid category is positive — plus one. The HDL-C of 68 is at or above 60, so one comes off the total — minus one. The lipid panel’s net contribution is zero. The minus one he was given had simply dropped his LDL, which is what a confident-sounding wrong answer usually turns out to be on inspection: the right shape with a term missing.

One loose end, since candidates ask and the honest answer is short. Neither edition states a floor at zero. Whether a client with no positive risk factors and a high HDL-C sits at zero or at minus one is not addressed, and the worked example in the Guidelines never needs it to be — it runs a client with a single positive factor and a high HDL-C, and lands on zero. If you are asked, say what the source says and what it does not. And remember that since the count no longer routes clearance, nothing downstream turns on the difference.

Fourth: a medication is a criterion in its own right

Being on a lipid-lowering medication makes the lipid category positive. Being on an antihypertensive makes the blood pressure category positive. In both cases, regardless of what the actual measurements say.

This is stated explicitly in both texts, and it is still the trap that produces the most confident wrong answers, because it inverts the intuition that treatment is good news. A client whose LDL-C sits at 94 because a statin is holding it at 94 is a client whose lipid category is positive. The number on the page is reassuring and irrelevant. The category is not asking whether this person’s lipids are currently controlled; it is asking whether this person has a lipid problem, and a prescription is direct evidence that somebody qualified concluded they do.

In a vignette, this arrives as a set of normal-looking laboratory values followed, several lines later, by a medication list. The two are placed far apart on purpose.

Fifth: missing data counts against the client, not for them

The default human reading of an incomplete intake form is that a blank line means no. No family history mentioned, therefore no family history.

The criteria say the opposite. If the presence or absence of a risk factor is not disclosed or is not available, it is counted as a risk factor until it can be verified.

This is the quietest of the six, and the only one that changes a total without anything on the page looking wrong. A vignette that omits a family history is not being careless with its word count. It is asking whether you know the rule, and the wrong answer leaves no trace — the candidate produces a clean, confident number that is one too low, and has no reason to go back.

It is also the rule that best explains what the count is for. A risk factor assessment is not a verdict; it is a provisional picture, assembled from whatever the client could tell you, handed onward to somebody with better instruments. Counting an unknown as a risk is what that document should do. It flags the gap instead of burying it.

Sixth: two categories are not asking the same question in the book you studied, and a third gained criteria

Most candidates prepare from Resources for the Exercise Physiologist, third edition, and sit an exam keyed to the twelfth edition of the Guidelines. Set the risk factor criteria in that textbook beside the current ones and two categories turn out not to be asking the same question, while a third has gained criteria the textbook does not list.

One clarification before the details, because it is the distinction that made this article necessary. What follows is a comparison between a textbook and the current Guidelines. It is not a claim about when the criteria moved, or about which edition moved them — that is a separate question and this article does not answer it. It matters because the document candidates reach for when they want that answer, the official edition-to-edition crosswalk, does not contain it either: for the chapter housing these criteria it lists a retitle and a new figure, which I know because I went through it line by line while writing the edition comparison. The practical question is not when the gap opened. It is that the book in front of you and the exam you are sitting do not agree, and you are the one holding both.

In the textbook, the glucose category is the diagnosis of diabetes — fasting glucose at 126 mg/dL, a two-hour tolerance test at 200, a glycated haemoglobin at 6.5%. In the current criteria it is the prediabetes range: 100, 140 and 5.7% respectively. That is not a refinement. It is a different question, and it moves a large number of clients from negative to positive on that line.

In the textbook, the activity category is a schedule — thirty minutes of moderate activity, three days a week, for three months. In the current criteria it is a weekly volume of moderate-to-vigorous activity. And the current lipid category carries a sex-specific low-HDL-C threshold and a non-HDL-C option that the textbook’s list does not contain. Non-HDL-C is worth computing while you are in there: it is total cholesterol minus HDL-C, and it shares the 130 cut point with LDL-C. For our reader, that is 157 — so his lipid category was satisfied twice over, and by the first rule in this article, still counted once. (The edition comparison linked above works through the delta in full, including the accumulated flexibility volume, where the crosswalk and the Guidelines themselves point in opposite directions — quoted there from both documents.)

The labels moved too, and this catches people searching rather than reading. Where the older list named categories for the conditions — Hypertension, Obesity, Diabetes, Dyslipidemia — the current criteria name them for what is actually measured: Blood pressure, Body mass index/waist circumference, Blood glucose, Lipids. That change is not cosmetic, and it is the same signal as the shifted thresholds: what is being counted is no longer the disease, it is the risk marker. Worth knowing that the twelfth edition’s own worked answers still use the retired labels, so if you go looking for a “dyslipidemia” line in the current criteria, you will not find one under that name.

Running one all the way through

Here is a client built, deliberately, to fire four of the six. Read his file before you read the count.

He is 47. He smoked for years and quit four years ago; nobody smokes in his home. His BMI is 27 and his waist measures 96 centimetres. His resting blood pressure, averaged across two visits, is 124/78, and he takes nothing for it. He runs forty-five minutes on four mornings a week at an effort he describes as somewhere between moderate and hard. He takes a statin. His lipid panel reads: total cholesterol 205, LDL-C 94, HDL-C 63, triglycerides 240. His fasting glucose is 108. On the family history section of his intake form, he wrote nothing at all.

Working through it: he is a man of 47, which is above the age criterion — one. His smoking stopped four years ago, well outside the six-month window, so that line is negative despite reading like a yes. His BMI and waist are both under their cut points.

His activity comes to 180 minutes a week, which clears the weekly volume on any reading of that criterion, so no risk factor there. Note that we could only say that because he told us; had the file been silent about his activity, the fifth rule would have made it a risk factor. His blood pressure is under both thresholds and he takes no antihypertensive, so nothing there either.

His fasting glucose of 108 is at or above 100. Two. A candidate revising from the older list would score that line zero, because 108 is nowhere near a diagnosis of diabetes.

The lipid category is positive — not because of a number, but because he is on a statin. Three — and everything else in that panel is commentary. His LDL-C of 94 is below its threshold and does not change anything. His total cholesterol of 205 is at or above 200 and is not used, because we have his LDL-C and HDL-C. His non-HDL-C is 142, at or above the 130 cut point, which would have made the category positive on its own and still adds nothing. And his triglycerides of 240 are high enough to be classified as high in the Guidelines' own lipid classification, but triglycerides are not one of the counted categories at all — a real distinction, and one that is easy to lose when every number on a page feels like it must be worth something.

His family history is blank, so it counts. Four.

Then the subtraction: his HDL-C of 63 is at or above 60, so one comes off the sum. Three.

Now the hurried version. Age, one. Lipids look controlled, so nothing there. Glucose is not diabetic, so nothing there. Family history was not mentioned, so nothing there. High HDL, minus one. Total: zero.

Three against zero, on the same client, with no arithmetic error anywhere in either count. That gap is the entire subject of this article.

What this means for the exam — and for the person in front of you

On the exam, this is a reading task under time pressure, not a recall task. The criteria are not hard to memorise and memorising them is not what earns the point. A well-built item hands you a panel with one value that is not used, a medication line placed far from the laboratory values, a blank field that means something, and a threshold that changed between editions. The candidate who has memorised the list and the candidate who has understood the procedure will produce different numbers from identical information.

In practice, the count is the thing you hand over — to the client, and to the clinician who manages them. It no longer opens or closes a door. It describes a person. Getting it right is a different kind of obligation than passing an item, and a less forgiving one, because there is no answer key waiting at the end of it.

Frequently asked questions

Does a high HDL-C cancel out a high LDL-C? No. They are separate determinations. An LDL-C at or above 130 makes the lipid category positive. An HDL-C at or above 60 subtracts one from the sum of positive risk factors, at the end, whether or not the lipid category came out positive. On a panel with both, the two contributions happen to net to zero — but by two independent steps, not by cancellation.

Can one category ever count as two risk factors? No. Each category contributes at most one, no matter how many of its criteria are met. A client with an elevated LDL-C, a low HDL-C and a lipid-lowering prescription contributes one lipid risk factor, exactly like a client who meets only one of the three.

If I have LDL-C and HDL-C, does the total cholesterol number matter? Not for this count. The 200 mg/dL total cholesterol figure is a fallback for when total cholesterol is all you have. With the component measurements in hand, it is not used. (It is still worth computing non-HDL-C from it, which is a criterion in the twelfth edition.)

Can the total come out negative? Neither edition states a floor at zero, and the worked example in the Guidelines never needs one — it runs a client with one positive factor and a high HDL-C and lands on zero. Say what the source says. Since the count no longer drives the clearance decision, nothing downstream depends on the answer.

Does the risk factor count decide whether my client needs medical clearance? No. Risk-factor-based stratification was removed from the screening procedure in 2015. Clearance is decided by current activity, known cardiovascular, metabolic or renal disease, and signs or symptoms. The Guidelines still recommend conducting the risk factor assessment and sharing it with the client and their health care provider — assess with it, don’t screen with it. The algorithm is covered in detail here.

Key takeaways

The count no longer decides medical clearance, and ACSM still asks you to run it — because it is shared with the client and their clinician, and because it shapes what you program.
Each category contributes at most one risk factor, however many of its criteria are met. The categories count domains of risk, not tests performed.
The 200 mg/dL total cholesterol figure is a fallback used only when total cholesterol is all that is available, not a criterion competing for the lipid slot. The older phrasing invites the opposite reading.
The high-HDL-C subtraction applies to the sum of positive risk factors, independently of the lipid category.
A lipid-lowering or antihypertensive prescription makes its category positive regardless of the measurements.
Information that is not disclosed or not available counts as a risk factor until verified.
Two categories ask a different question in the textbook most candidates study than in the current Guidelines, a third gained criteria the textbook does not list, and the category labels changed from conditions to measurements. This article compares a textbook with the current criteria; it makes no claim about which edition moved them.