Every week brings a new health headline: a food that prevents cancer, a supplement that boosts focus, a habit that adds years to your life. Some of these are solid. Many are exaggerated. A few are dangerous. The difference is rarely obvious from the headline.
Health literacy is the skill of finding, understanding, and using health information to make good decisions. It isn't medical training. It's knowing which sources to trust, how to read a study or a news story critically, and how to work with clinicians as an informed partner. It is the first step in our Health & Wellbeing learning path, and it borrows heavily from the same thinking as our guide to analyzing data.
What is health literacy?
Health literacy has two sides:
- Personal health literacy: your ability to find, understand, and use information to make health decisions for yourself and others.
- Organizational health literacy: how well hospitals, clinics, and agencies make information clear and usable.
You control the first one. Low health literacy is common and has nothing to do with intelligence; medical language, confusing forms, and conflicting headlines trip up highly educated people too. The good news is that it's a learnable skill, and the core of it fits on one page.
Step 1: Start with trustworthy sources
Not all health websites are equal. A good default order:
| Source type | Examples | Why trust it |
|---|---|---|
| Medical libraries and public health agencies | MedlinePlus (US National Library of Medicine), CDC, NIH, the UK's NHS, the World Health Organization | Written or reviewed by experts, regularly updated, no products to sell |
| Systematic reviews | Cochrane reviews | Combine all good-quality studies on a question and rate the strength of evidence |
| Professional medical organizations | Specialty societies and their patient pages | Reflect clinical guidelines, though check for industry funding |
| Academic medical centers | Patient education pages from major hospitals and universities | Clinically reviewed; may favor their own services |
| News outlets | Health reporting from established outlets | Useful for awareness, but often simplified or exaggerated; trace back to the study |
| Social media, forums, and influencers | Videos, posts, community groups | Good for lived experience and support; poor for facts unless they cite sources you can check |
Before trusting any page, check four things:
- Who wrote or reviewed it, and what are their qualifications?
- When was it last updated? Medical guidance changes.
- Who pays for the site? Is it selling a product, test, or supplement?
- Does it cite sources you can follow back to research or official guidance?
Step 2: Know what kind of study you're looking at
Health claims rest on research, and research comes in very different strengths. From weakest to strongest, roughly:
STRONGER EVIDENCE
^
| Systematic reviews and meta-analyses of good trials
| Randomized controlled trials (RCTs)
| Cohort studies (follow groups over time)
| Case-control studies (compare people with and without a condition)
| Case reports, expert opinion
| Lab and animal studies (early clues, not proof in people)
| Anecdotes and testimonials
WEAKER EVIDENCE
- Randomized controlled trials (RCTs) randomly assign people to a treatment or a comparison (such as a placebo). Randomization balances other differences between the groups, which is what lets an RCT test cause and effect.
- Observational studies (cohort and case-control) watch what people already do. They're essential for questions you can't ethically test, such as smoking, but they show associations, not proven causes.
- Lab and animal studies are early steps. A compound that kills cancer cells in a dish often does nothing useful in people.
- Systematic reviews and meta-analyses pull together many studies, so they're less likely to be fooled by one unusual result.
A few more questions to ask about any study:
- How many people were in it? Small studies swing wildly by chance.
- Who were they? Results in young men may not apply to older women, and results in mice may not apply to anyone.
- What was measured? A change in a blood marker isn't the same as fewer heart attacks.
- Was it published in a peer-reviewed journal, or is it a preprint or press release? Preprints haven't been reviewed yet.
- Who funded it? Funding doesn't make a study wrong, but it's worth knowing.
- Has anyone repeated it? Findings that replicate deserve more confidence.
Step 3: Read risk numbers correctly
This is the single most useful skill in health literacy, and it's simple once you see it.
- Relative risk compares two groups: "50% higher risk."
- Absolute risk tells you how many people are actually affected: "3 in 1,000 instead of 2 in 1,000."
The same result can sound alarming or trivial depending on which you report:
| Without the exposure | With the exposure | How it's often reported | |
|---|---|---|---|
| Example A | 2 in 1,000 | 3 in 1,000 | "50% higher risk!" (relative) |
| Same data | 0.2% | 0.3% | "1 extra case per 1,000 people" (absolute) |
| Example B | 20 in 100 | 30 in 100 | "50% higher risk!" (relative) |
| Same data | 20% | 30% | "10 extra cases per 100 people" (absolute) |
Both examples are "50% higher," but Example B affects 100 times as many people. When a headline gives only a relative number, look for the absolute one before reacting.
The same idea applies to benefits. A treatment that "cuts risk in half" might move risk from 2% to 1%. Clinicians sometimes express this as the number needed to treat (NNT): how many people need the treatment for one to benefit. Here it's 100. That can still be very worthwhile, depending on the condition, the cost, and the side effects, but it's a different picture from "cuts risk in half."
Step 4: Correlation is not causation, in health too
Our data analysis guide covers this in general. In health, it shows up constantly:
- People who take vitamins may be healthier because they also exercise more, smoke less, and have better access to care. That's confounding.
- People who feel unwell may change their diet, so a diet looks linked to illness when illness actually caused the diet change. That's reverse causation.
- Studies that find exciting results are more likely to be published and covered than studies that find nothing. That's publication bias, and it makes the overall picture look more dramatic than it is.
When a headline says a food, habit, or supplement "is linked to" an outcome, read it as "is associated with, for reasons we don't fully know yet."
Step 5: Spot misinformation and health scams
Red flags:
- One product cures many unrelated conditions.
- "Doctors don't want you to know" or other secret-knowledge framing.
- Testimonials instead of studies, or before-and-after photos.
- Urgency and scarcity: "limited supply," "act now."
- "Natural, so it's safe." Many natural substances are harmful, and supplements can interact with medicines.
- The person making the claim profits from it: selling the supplement, the test, the course, or the device.
- Big claims from a single study, especially a lab or animal study.
In the US, dietary supplements don't go through the FDA approval process for safety and effectiveness that prescription drugs do before they're sold. Tell your clinician and pharmacist about every supplement you take.
Health scams use the same manipulation playbook as other scams: urgency, authority, and fear. Our guide to spotting phishing and social engineering covers those tactics, and they transfer directly.
Step 6: Use AI chatbots carefully
AI assistants can be genuinely helpful for health literacy: explaining a term in plain English, summarizing what a type of test measures, or helping you write questions for an appointment. They also have real limits:
- They can be confidently wrong, including about doses, interactions, and guidelines.
- They don't know your history, medications, or exam findings.
- They may not reflect the latest guidance.
- What you type may be stored; check the tool's privacy settings before sharing personal health details.
Use them to prepare and understand, then confirm with trusted sources and a clinician. Never use them to diagnose yourself, change a medication, or decide whether something is an emergency. Our Claude vs ChatGPT vs Gemini comparison covers how the major assistants differ.
Step 7: Get more out of appointments
Health literacy matters most in the room with a clinician. Before you go:
- Write down your main concern and when it started, plus what makes it better or worse.
- Bring a full list of medicines and supplements, with doses.
- Bring your questions, with the most important first.
The Ask Me 3 program, developed to improve patient-clinician communication, suggests three questions:
- What is my main problem?
- What do I need to do?
- Why is it important for me to do this?
During the visit:
- Use teach-back: repeat the plan in your own words ("So I take this twice a day with food, for ten days?"). Clinicians use this technique to check understanding, and you can start it yourself.
- Ask about options: "What are my choices, including waiting? What are the benefits and risks of each?"
- Ask for numbers: "Out of 100 people like me, how many does this help?"
- Ask what to watch for, and when to call back.
- Bring someone if it's a big appointment; a second set of ears helps.
Step 8: Understand the money side
Health decisions are also financial ones. A few terms worth knowing in US health insurance:
| Term | What it means |
|---|---|
| Premium | What you pay each month to have coverage |
| Deductible | What you pay for covered care each year before the plan starts sharing costs |
| Copay | A fixed amount for a service, such as a doctor visit |
| Coinsurance | Your percentage share of costs after the deductible |
| Out-of-pocket maximum | The most you'll pay for covered care in a year; after that, the plan pays 100% of covered services |
| In-network vs out-of-network | Providers with a contract with your plan cost you less |
Our budgeting and emergency fund guide covers planning for medical costs, and how to afford a career change covers keeping coverage between jobs.
A one-minute checklist for any health claim
- Who's saying it, and do they profit from it?
- What's the evidence: anecdote, lab study, observational study, trial, or systematic review?
- How big is the effect in absolute terms?
- Who was studied, and are they like you?
- What do trusted sources (public health agencies, medical libraries, systematic reviews) say?
- What would your clinician say? Ask.
Frequently asked questions
What are the most reliable health websites?
Public health agencies and medical libraries are the best starting points: MedlinePlus, the CDC, and the NIH in the US; the NHS in the UK; and the World Health Organization globally. Cochrane publishes systematic reviews that summarize the evidence on specific questions.
How can I tell if a health study is good?
Check the study type (randomized trials and systematic reviews are strongest), the number and kind of people studied, what was actually measured, whether it was peer reviewed, who funded it, and whether other studies found the same thing.
What's the difference between relative and absolute risk?
Relative risk compares groups ("50% higher"). Absolute risk tells you the actual chance ("3 in 1,000 instead of 2 in 1,000"). Always look for the absolute number before deciding how worried to be.
Can I trust AI chatbots for health questions?
Use them to understand terms and prepare questions, then confirm with trusted sources and a clinician. They can be wrong and don't know your history, so don't use them to diagnose or change treatment.
Check your understanding
Answer each question from memory before opening it. Recalling, not rereading, is what makes learning stick (here's why).
1 A headline says a food 'raises cancer risk by 50%.' What is the first number you should look for?
The absolute risk: how many people out of 100 (or 1,000) get the disease with and without the food. A 50% relative increase from 2 in 1,000 to 3 in 1,000 is one extra case per 1,000 people.
2 An observational study finds that people who drink more tea live longer. Why can't it prove that tea causes longer life?
Tea drinkers may differ in other ways (income, diet, exercise, smoking) that affect lifespan. Observational studies show associations; randomized controlled trials are needed to test cause and effect.
3 What kind of evidence usually deserves the most weight on a health question?
Systematic reviews and meta-analyses of good randomized controlled trials, because they combine many studies instead of relying on one.
4 Name three red flags that a health claim may be misinformation.
Any three of: a miracle cure for many conditions, 'doctors don't want you to know,' testimonials instead of studies, pressure to buy now, claims that a product is 'natural so it's safe,' or a seller who also profits from the claim.
5 What are the three Ask Me 3 questions to bring to an appointment?
What is my main problem? What do I need to do? Why is it important for me to do this?
Where to go next
You've finished step 1 of the Health & Wellbeing path. Next:
- 7 types of meditation backed by research: put these skills to work on a popular wellness practice, with step-by-step instructions.
- Positive Psychiatry and Mental Health review: a well-regarded introductory course on how mental health works and what helps.
- How to analyze data for any project: the same critical thinking, applied to any dataset.
- Learning How to Learn review: the science of studying, including why the self-check above works.
Sources
- MedlinePlus: health information from the US National Library of Medicine, including how to evaluate health information online.
- Centers for Disease Control and Prevention: Health Literacy: definitions of personal and organizational health literacy.
- National Institutes of Health: research summaries and health information.
- Cochrane: systematic reviews of health evidence.
- NHS: health information from the UK National Health Service.
- World Health Organization: global health guidance.
- US Food and Drug Administration: Dietary Supplements: how supplements are regulated.
- Institute for Healthcare Improvement: the Ask Me 3 program.
- HealthCare.gov glossary: definitions of US health insurance terms.
- 988 Suicide & Crisis Lifeline: 24/7 crisis support in the US.