Browse this category by subcategory.Health
Blood Pressure
102
Contraception
27
Creatinine
80
Health
Three collections, 209 articles, and one uncomfortable fact about all of them
This section covers blood pressure, kidney function and emergency contraception. They have almost nothing in common clinically. What they share is that every number in them comes from a document somebody wrote, and all three of those documents have been rewritten in the last four years.
That is the thing worth knowing before you read anything else about these subjects, here or elsewhere. Health writing goes stale in a specific way. The number often stays the same while the document behind it is replaced, so a page can be quoting a correct figure from a superseded source and look completely fine. And a great deal of health content on the internet was written once, several years ago, and has never been revisited.
| Collection | Articles | The document behind it | Last rewritten |
|---|---|---|---|
| Blood Pressure | 102 | ACC/AHA high blood pressure guideline | August 2025 |
| Creatinine | 80 | KDIGO chronic kidney disease guideline | March 2024 |
| eGFR, within the same collection | — | The CKD-EPI estimating equation | 2021 |
| Contraception | 27 | The FDA-approved product label | December 2022 |
Article counts are live as of 18 August 2026. Each collection has its own hub page covering the clinical detail. This page covers what sits underneath all three: where medical numbers come from, who is entitled to set them, and how to tell whether the one in front of you is current.
Nothing here is advice, and that is not a formality
This section exists to help you understand numbers you have already been given, and to help you ask better questions of somebody qualified to answer them. It does not exist to replace that person. Two of the three subjects here involve decisions where being wrong has real consequences, and the third involves a decision with a clock running on it. Where something is urgent, the page says so plainly rather than burying it.
It also tries to be honest about the boundary. There are questions in this section that a page cannot answer for you no matter how carefully it is written, because the answer depends on your history, your other results, your medicines and your build. Those are marked as such rather than answered anyway, which is a smaller category than most health writing implies but a real one.
Four rewrites, and what each one actually altered
None of these were quiet technical revisions. Each changed something a reader would care about, and each left a large volume of existing writing describing a document that no longer governs.
Blood pressure: the guideline was replaced in August 2025
The 2017 ACC/AHA high blood pressure guideline was the reference for eight years and is cited in an enormous amount of material. It was replaced on 14 August 2025 by a new guideline written by thirteen sponsoring organizations. Anything describing the 2017 document as current guidance is now describing a superseded one.
The useful nuance is that the numeric categories did not change. Normal is still below 120 and below 80. Elevated is still 120 to 129 with a diastolic under 80. Stage 1 is still 130 to 139 systolic or 80 to 89 diastolic. Stage 2 is still 140 or above, or 90 or above. So a chart of thresholds copied from 2017 is still correct, which is exactly why so few people noticed the guideline underneath it had been replaced.
What did change:
- The category list is now four, not five. "Hypertensive crisis" is no longer used as a category name. The current framework instead defines severe hypertension as above 180 over 120 in nonpregnant adults, and separates that from a hypertensive emergency, which is the same reading accompanied by acute organ damage. Severe hypertension without organ damage is described as something to evaluate and treat in an outpatient setting. A five-row chart ending in "hypertensive crisis, seek care immediately" is the old framework.
- The risk calculation changed. Cardiovascular risk estimation moved from the Pooled Cohort Equations to the PREVENT equations. This is invisible in a threshold table and changes who gets offered treatment.
- A treatment-timing rule was added. Adults averaging 130 over 80 or above, with a PREVENT ten-year risk under 7.5 percent, start medication if blood pressure is still at or above 130 over 80 after a three to six month trial of lifestyle change.
- A warning about wearables. The guideline states that reliance on cuffless devices, including smartwatches, should be avoided until they demonstrate greater precision and reliability. That is new, and it is aimed squarely at how people increasingly measure themselves.
Kidney disease: KDIGO 2012 became KDIGO 2024
The chronic kidney disease guideline that most existing content rests on was published in 2012 and superseded in March 2024. One detail worth getting right, because it is quoted constantly: the definition requires abnormalities of kidney structure or function present for a minimum of three months, with implications for health. The 2012 wording was "more than three months." If a page says "more than 3 months" and cites the 2024 guideline, it is quoting the older document.
That duration requirement is the whole reason a single abnormal result is not a diagnosis of chronic kidney disease. Chronicity has to be established, not assumed.
eGFR: the equation was rewritten in 2021 to remove race
The estimating equation that converts a creatinine result into an eGFR was refit in 2021 to remove the race coefficient that earlier versions used. A joint task force recommended immediate adoption across all laboratories, and the current kidney guideline states that the use of race in computing eGFR should be avoided.
A trap worth knowing about. Older calculators running the 2009 version of the equation are still online, including on government sites, and they explicitly take race as an input. They load, they look official, and they are running a superseded equation. If a kidney calculator asks you for your race, that is the tell.
And a fifth, quietly: the contraception eligibility criteria
Alongside the label change, the CDC criteria that clinicians use to decide who can safely use which contraceptive method were reissued in 2024. Those criteria are where questions like the effect of body weight on emergency contraception are actually answered, and the current edition answers them differently from how the internet does. It places emergency contraceptive pills in the category meaning benefits generally outweigh risks at a BMI of 30 or above, notes they might be less effective there than below a BMI of 25, and states plainly that no safety concern exists. It sets no weight limit in pounds or kilograms, because none exists.
Emergency contraception: the label changed in December 2022
The FDA approved the deletion of a sentence from the levonorgestrel label stating that the product may prevent implantation, on the basis that the evidence did not support it. The current label describes the mechanism as stopping or delaying the release of an egg. Very large numbers of pages still describe the old mechanism, because removing a sentence from a label does not generate an announcement.
Four different kinds of medical authority, and they are not interchangeable
People treat every health figure as though it came from the same place. They do not, and the differences decide how much weight a number carries, who can overrule it, and how often it changes.
| Kind | Who writes it | What it binds | Example in this section |
|---|---|---|---|
| Clinical guideline | Professional societies, by committee | Nobody legally. It guides practice | The blood pressure categories |
| Product label | The manufacturer, approved by the FDA | What may be claimed and how it is sold | The emergency contraception timing window |
| Laboratory reference range | The individual laboratory | Only the results that laboratory issues | The creatinine range on your report |
| Federal regulation | Congress and agencies | Legally binding on laboratories | How reference ranges must be reported |
A guideline is a recommendation, and that is a feature
Clinical guidelines are written by committees of specialists who review the evidence and agree on what should usually be done. They are not law. A clinician can depart from one for a good reason, and guidelines themselves usually say so. They also disagree with each other across countries, which is why a threshold you read in one place may not match what you are told in another.
The practical consequence is that a guideline number tells you where you sit in a framework, not what will happen to you. Being one point over a threshold does not change your body on the day the guideline is published.
A label is narrower than people assume
An approved label describes the product as it was studied and approved. It is the authority on what the product is licensed to do and how it should be used, and it is the reason a timing window is a hard number rather than an opinion. But a label does not describe every use a clinician might reasonably make of a product, and use outside the label is common, legal and often supported by good evidence. When something is off-label, this site says so, because "the label does not cover this" and "this does not work" are completely different statements.
What happens when two of them disagree
They do disagree, and the disagreements are usually informative rather than a sign that somebody has made a mistake.
A label and a guideline can point different ways. A guideline may recommend something a label does not cover, because guidelines follow the evidence and labels follow what was submitted for approval. The copper device used as emergency contraception is a clean example: the criteria clinicians use describe a five day window for it, while the product label covers ongoing contraception and carries no emergency indication at all. That is not a contradiction and it is not a warning. It is a licensed product being used in a way the evidence supports and the licence never addressed.
A lab range and a guideline threshold measure different things. A reference range describes where most results from a reference population fall at that laboratory. A guideline threshold describes where a committee decided the risk changes enough to act. There is no reason those two should coincide, and treating a result just outside a lab range as though it had crossed a clinical threshold is one of the most common ways people frighten themselves with a normal test.
Two guidelines can disagree across borders. Different countries convene different committees who read the same evidence and set thresholds in different places, usually for defensible reasons about their own populations and health systems. A figure that is authoritative in one country is not automatically wrong somewhere else, and it is not automatically applicable either. This site writes for United States guidance and says so, which matters more than it sounds if you are reading from anywhere else.
A reference range belongs to one laboratory
This is the one that surprises people most, and it is the subject of the next section.
The normal range on your report belongs to the lab that ran the test, and federal law says so
A blood test result arrives with a range beside it, and almost everyone reads that range as a fact about human biology. It is not. It is a property of that laboratory, using that method, for that population, and United States regulation treats it that way explicitly.
The federal rule governing what a laboratory report must contain requires that pertinent reference intervals or normal values be available, as determined by the laboratory performing the tests. That phrase is doing a lot of work. Laboratories are also required, before reporting results on an unmodified test system, to verify that the manufacturer reference intervals are appropriate for their own patient population, and to establish their own intervals for methods they have modified.
The National Library of Medicine puts the consumer version plainly: to know how your result compares with the normal range, check the reference range on your own report, because you cannot compare your results to ranges you find elsewhere, and you cannot compare test results from different labs.
What this means in practice
- The same blood, split between two laboratories, can produce two results with two ranges. Neither lab is wrong. They may be using different methods, and their ranges reflect their own methods and populations.
- Tracking a value over time works best at one laboratory. A change between two labs may be a change in you or a change in the assay, and the report will not tell you which.
- Ranges differ by group. Creatinine ranges differ by sex, for the straightforward reason that creatinine production tracks muscle mass. The National Library of Medicine gives roughly 0.7 to 1.3 mg per deciliter for men and 0.5 to 0.95 for women, while noting in the same breath that ranges vary between laboratories.
- Units differ too. Creatinine is reported in milligrams per deciliter in the United States and in micromoles per liter in most of the rest of the world. The conversion factor is 88.4. A result of 1.0 mg/dL is 88.4 micromoles per liter. Reading a foreign reference range against a US result, or the reverse, produces an answer that is wrong by a factor of about ninety.
Why the method matters as much as the sample
Creatinine is the clearest example on this site. The older manual method and the modern enzymatic methods do not produce identical numbers, and published reference ranges have historically differed depending on which was used. Certain conditions interfere with particular methods: elevated glucose and ketones can produce falsely raised creatinine on one common assay. None of this is visible on a report that simply prints a number and a range.
The question to ask. If a result matters, the useful question is not "is this normal" but "is this normal for this laboratory, on this method, for someone like me, and how does it compare with my last one from the same place." That is a different question and it usually has a more useful answer.
A single measurement is not a diagnosis, and the guidelines are explicit about it
This is the most consequential idea in the section, and it applies to all three collections in different forms. People take one number, treat it as a verdict, and then either panic or relax. Both reactions are usually premature, and the documents themselves say so.
Blood pressure needs several readings on several occasions
The American Heart Association states that an average based on two or more readings taken on two or more occasions by a health care professional is what a proper diagnosis requires. Its patient guidance says outright that a single high reading is not an immediate cause for alarm. Guidance for measurement sessions is to take at least two readings a minute or two apart and average them, rather than acting on the first.
Beyond that, out-of-office measurement is recommended to confirm what an office reading suggests. Two well-described patterns are the reason: some people read high in a clinical setting and normal elsewhere, and some read normal in the clinic and high at home. Neither is detectable from a single office measurement, which is the point.
On smartwatches. The current guideline advises against relying on cuffless devices, including smartwatches, until they demonstrate greater precision and reliability. A trend line from a wrist device is not a substitute for a validated cuff, and it is not what any diagnosis is based on.
Kidney disease requires three months by definition
Chronic kidney disease is defined as abnormalities of kidney structure or function present for a minimum of three months. The duration is not a clinical nicety, it is part of the definition. One creatinine result above a reference range establishes nothing about chronicity on its own, and a great deal of alarm comes from treating a single value as though it did.
There is also a long list of things that raise creatinine without any kidney injury at all. A large meal of cooked meat does it, because cooking converts creatine in the meat to creatinine. Creatine supplements do it. High muscle mass does it, which is why a very muscular person can sit outside a reference range built on a general population. Recent hard exercise does it. And several medicines raise measured creatinine by blocking the way the kidney secretes it, without changing kidney function at all.
Emergency contraception is the exception, and the clock runs the other way
The other two subjects reward waiting and repeating. This one does not. Effectiveness depends on how quickly the method is used, the licensed windows are measured in days rather than weeks, and the correct response to uncertainty is to act and then confirm, not to research first. That inversion is worth naming explicitly, because a habit of caution that serves someone well on a blood pressure reading serves them badly here.
What all three have in common
In each case the number is an input to a judgment rather than the judgment itself. Somebody who knows your history, your medicines, your build and your other results will read the same figure differently from a page on the internet, and that is not a disclaimer, it is the actual reason these figures exist. The purpose of this section is to let you arrive at that conversation understanding what you are holding.
Not every source that sounds medical carries the same weight
Health writing quotes findings constantly, and almost never says what kind of finding it is. The difference between a guideline, a systematic review, one trial and one case is the difference between something you can plan around and something that is merely interesting. Roughly, in descending order of how much a single result should move you:
| Kind of evidence | What it is | How much one of them should move you |
|---|---|---|
| Clinical guideline | A committee reading all of the above and agreeing what to do | A great deal. It is the synthesis, already done |
| Systematic review | Every study on a question, gathered and assessed together | A great deal, and more than any single study in it |
| Randomized trial | People assigned by chance to one thing or another | Quite a lot, depending on size and who was studied |
| Observational study | Watching what happens without assigning anything | Something, but it cannot settle cause on its own |
| Case report | What happened to one person | Almost nothing about you. It generates questions |
Why this matters more in health than anywhere else
Two reasons. The first is that a single striking study is far more publishable, more shareable and more quotable than the guideline that already weighed it against forty others and decided it did not change anything. So the thing you read is systematically more likely to be the outlier than the consensus.
The second is that health findings are usually about populations, and you are one person. A trial reporting that a group given one thing did better on average than a group given another does not tell you what will happen to you, and it is not meant to. That translation is exactly the work a clinician does with your history in front of them.
The specific trap: an effect size with no interval
Numbers in health writing frequently arrive without any indication of how certain they are. A figure quoted as a range from expert judgment and a figure measured in a randomized trial with a confidence interval look identical on a page, and they are not the same kind of claim at all. Where this site quotes an effect it tries to say which it is, because "estimated to be somewhere between five and twenty" and "measured at 4.8, with the true value very likely between 3.0 and 6.6" support completely different sentences.
A quick test. When a health page gives you a number, ask what would have to be true for it to be wrong. If the answer is "the committee that wrote the guideline would have to have misread the literature," that is a strong number. If the answer is "the one small study it came from would have to not replicate," that is a weak one. Both can be worth reading. Only one is worth planning around.
How to tell in two minutes whether what you are reading has been superseded
Health pages rarely announce that they are out of date. They have no expiry, no version number, and often no date at all. But the documents behind them do, and that makes the check quick once you know what to look for.
- Find the document, not the number. A page telling you a threshold without naming what publishes it has already failed the test. If it names a guideline, a label or a set of criteria, you have something to check.
- Find the edition year. Every one of these documents carries one. The guideline that governs blood pressure in the United States is the 2025 edition. Chronic kidney disease runs on the 2024 KDIGO guideline. Contraceptive eligibility runs on the 2024 US criteria. A page citing 2017 or 2012 for these is citing a superseded edition, whether or not the number it quotes is still right.
- Check the publisher, not a search engine. The organization that issues a guideline keeps a page for it, and that page says which edition is current. Search results and even government sites frequently surface older material that still loads perfectly.
- For a lab result, check the report. Reference ranges come from the laboratory that ran your test, so the authoritative range is the one printed next to your own number, not one on any page anywhere.
- For a medicine, check the label. Labels are revised, and the revision date is on them. What a product is licensed to do is a question with a documentary answer.
The failure mode this catches
The most common problem in health content is not a wrong number, it is a right number attached to a dead source. Blood pressure is the perfect example: the category thresholds are exactly what they were in 2017, so a page reproducing them is numerically correct, while everything around those numbers, the risk model, the treatment timing, the handling of very high readings and the advice about wearables, was rewritten underneath it. A reader checking only the number would conclude the page was fine.
The second most common problem is a tool that still runs. Software outlives the guidance it was built on, and a calculator implementing a retired equation gives a confident, precise, wrong answer with no indication that anything is amiss. The kidney example is the clearest: if it asks for race, it is running an equation that was retired in 2021.
What this site tries to do. Name the document, give its edition and date, quote it where the wording matters, and say plainly when something is a convention, an estimate, or off-label rather than established. That does not make a page right forever. It makes it checkable, which is the most any page about a moving target can honestly offer.
Three collections, and which one you need
Each has its own hub covering the clinical detail. Counts are current as of 18 August 2026.
Blood Pressure
102 articles. The largest collection in the section, covering what a reading means, what moves it, and why the machine at home disagrees with the one at the clinic.
- Start with the Blood Pressure hubWhere a reading sits, what actually moves the number, and the vocabulary in one place.
- Mean Arterial Pressure CalculatorMAP from a reading, with where that reading sits.
Creatinine and kidney function
80 articles on what a creatinine result means, what moves it, and how it becomes an eGFR.
- Start with the Creatinine hubStaging on two axes, the 2021 equation rewrite, and why a result inside the range is not the whole story.
- Creatinine Clearance CalculatorClearance from a creatinine result.
- eGFR CalculatorEstimated glomerular filtration rate.
Contraception
27 articles on emergency contraception, timing and effectiveness, with the label quoted rather than paraphrased.
- Start with the Contraception hubEvery answer with the document behind it, quoted.
- Emergency Contraception CalculatorOptions and timing windows.
The other eleven sections are on the blog hub, and the health tools are in health calculators.
Questions this section gets, with the document named
Is the 2017 blood pressure guideline still current?
No. It was replaced on 14 August 2025 by a new ACC/AHA guideline written with thirteen sponsoring organizations. The numeric categories are unchanged, which is why the replacement went widely unnoticed, but the risk model, the treatment-timing rules and the handling of very high readings all changed.
What are the current blood pressure categories?
Normal is below 120 and below 80. Elevated is 120 to 129 with diastolic below 80. Stage 1 is 130 to 139 systolic or 80 to 89 diastolic. Stage 2 is 140 or above systolic or 90 or above diastolic. Above 180 over 120 is described as severe hypertension, and as a hypertensive emergency where there is acute organ damage.
Is one high blood pressure reading a problem?
Not by itself. The American Heart Association says a single high reading is not an immediate cause for alarm, and that diagnosis rests on an average of two or more readings taken on two or more occasions. Measurement outside the clinic is recommended to confirm what an office reading suggests.
Can I use my smartwatch to track blood pressure?
The current guideline advises against relying on cuffless devices, including smartwatches, until they demonstrate greater precision and reliability.
Why is my creatinine range different from the one I read online?
Because reference ranges belong to the laboratory that ran the test. Federal regulation requires reference intervals to be reported as determined by the performing laboratory, and the National Library of Medicine advises that you cannot compare your results to ranges found elsewhere or to results from a different lab.
Does one high creatinine mean kidney disease?
No. Chronic kidney disease is defined as abnormalities present for a minimum of three months, so a single result cannot establish it. Cooked meat, creatine supplements, high muscle mass, recent hard exercise and several common medicines all raise creatinine without kidney injury.
Should a kidney calculator ask for my race?
No. The estimating equation was refit in 2021 to remove the race coefficient, and current guidance is that race should not be used in computing eGFR. Older calculators running the 2009 equation are still online and still ask.
What are the units for creatinine and how do I convert them?
Milligrams per deciliter in the United States, micromoles per liter in most other countries. Multiply mg/dL by 88.4 to get micromoles per liter, or divide the other way.
The documents behind everything above
Each claim on this page is tied to the body that publishes it, with the edition and date, because the whole point of the page is that editions change. Checked 18 August 2026.
- The current blood pressure guideline. The 2025 AHA/ACC guideline for the prevention, detection, evaluation and management of high blood pressure in adults, published 14 August 2025 and replacing the 2017 version. AHA guideline hub, Top Things to Know, and the PubMed record.
- The categories and the severe hypertension wording. AHA, understanding blood pressure readings, reviewed 14 August 2025, and CDC on high blood pressure.
- Two readings on two occasions, and a single high reading. AHA, how high blood pressure is diagnosed and AHA on home monitoring. Measurement session guidance from CDC.
- The kidney disease definition. KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease, which supersedes the 2012 edition and defines CKD as abnormalities present for a minimum of three months. The KDIGO guideline page holds the executive summary.
- The 2021 eGFR equation. Inker et al., new creatinine and cystatin C based equations to estimate GFR without race, and the NKF-ASN task force report recommending immediate implementation. Implementation detail from NIDDK.
- What raises creatinine without kidney injury. National Kidney Foundation on creatinine for cooked meat, creatine supplements, muscle mass, recent exercise and the medicines involved, and NIDDK on factors affecting eGFR accuracy for body size, muscle mass, diet and assay interference.
- Reference ranges belong to the laboratory. 42 CFR 493.1291, which requires reference intervals as determined by the laboratory performing the tests, and 42 CFR 493.1253 on verifying and establishing intervals. The consumer version is at MedlinePlus, how to understand your lab results.
- Creatinine ranges and units. MedlinePlus, creatinine blood test, which gives sex-specific ranges in both units and states that ranges vary between laboratories. The 88.4 conversion factor is in the KDIGO 2024 guideline conversion table.
- The emergency contraception label change. FDA information on Plan B One-Step, current as of 23 December 2022, and the CDC US Medical Eligibility Criteria for emergency contraception.
- Noted and not published. Several widely circulated figures for how much a measurement error changes a blood pressure reading come from expert estimate rather than measurement, and the published estimates disagree with each other. Where this section quotes an effect size it says which document it came from. The blood pressure collection covers this in detail.
Scope note. This page describes documents and definitions as they stood on 18 August 2026. Clinical guidelines are replaced without notice to readers, and two of the four described here changed within the last two years. Nothing on this page or anywhere on this site is medical advice, a diagnosis, or a substitute for care from a qualified clinician who can see your history and your other results. If a symptom is severe, sudden or frightening, contact a medical professional or emergency services rather than reading. If you are outside the United States, the guidelines, labels and reference ranges that apply to you may be different ones.