Open the cabinet where you keep the bottles. Count the ones that are more than half full. Then try to remember what each one was supposed to do, and whether it did it.
Most people can’t answer the second question. That isn’t a character flaw. It’s what happens when you buy something without first deciding what would count as it working.
The money involved is not small. An analysis of the 2012 National Health Interview Survey, published by the National Center for Health Statistics in 2016, found that about 59 million people aged 4 and over had at least one out-of-pocket expense for a complementary health approach that year, totaling $30.2 billion. Visits to complementary practitioners accounted for $14.7 billion of it, natural product supplements for $12.8 billion, and self-care approaches for $2.7 billion.
Per person, the averages ran $433 for practitioner visits, $368 for supplements and $257 for self-care. The income pattern is worth sitting with. People in families earning under $25,000 a year averaged $435 per user, and people in families earning $100,000 or more averaged $590. This is not a wealthy-household expense. It’s close to universal, and it lands hardest where there’s least room for it.
What follows is a decision procedure. Six questions in order, ending with the one almost nobody asks.
Buying something is a way of doing something
Any honest framework has to start here.
When you feel bad and the system feels slow, a purchase is the fastest action available. The specialist has nothing until March. The bottle arrives Thursday. You did a thing. The relief of having moved is real, and dismissing it is both unkind and inaccurate.
It’s also the exact feeling that wellness marketing is built to find. The gap between how urgent a problem feels and how long the careful path takes is where most of this spending happens.
So name the motive instead of arguing with it. If part of why you’re buying is that you’re tired of waiting, say so to yourself. It changes what you should expect from the purchase, and it changes what you’d accept as evidence it helped. A purchase made to feel less stuck will feel like it worked for about a week no matter what’s inside the bottle.
The six questions
What problem am I solving, and would I recognize it solved?
Write the problem in one sentence, using words you could measure. “I want more energy” is not measurable. “I fall asleep at my desk between two and three most afternoons” is.
If you can’t finish the sentence, that’s the finding. You’re not buying a solution. You’re buying the feeling of addressing something, and that’s worth knowing before you spend.
What tier is the evidence in?
Four tiers, and the differences matter more than people assume: established, preliminary, insufficient, and disproven or unsafe.
The US Preventive Services Task Force shows how far apart those can be inside a single product category. In its 2022 recommendation statement in JAMA on vitamin and mineral supplements for preventing cardiovascular disease and cancer, the Task Force recommended against beta carotene and vitamin E, concluding with moderate certainty that the harms of beta carotene supplementation outweigh the benefits and that vitamin E shows no net benefit. Multivitamins landed somewhere else entirely, as did most other single and paired nutrients. Each drew an I statement, meaning there isn’t enough evidence to weigh benefits against harms.
Read that carefully. “Insufficient” means nobody knows. “No net benefit” means it was studied and didn’t help. “Harms outweigh benefits” means something worse. Those are three different purchases.
One more thing about Task Force grades, which matters in an article about money: the USPSTF does not consider cost when it assigns them. A grade tells you about benefit and harm. Whether the price is justified is a separate question the grade never touches.
What does this cost per year, not per bottle?
Per-unit pricing is designed to feel small. Convert it.
A product at $29 a month is $348 a year. Three of them is $1,044 a year, roughly triple the $368 average that supplement users spent out of pocket in the federal survey above. A two-capsule serving in a 60-capsule bottle is a 30-day supply, not a 60-day one, and that doubles your real monthly cost without changing the shelf price.
There’s one place this arithmetic simply stops working. Under the FDA’s dietary supplement labeling guide, a Supplement Facts panel that uses a proprietary blend discloses only the total weight of the blend, with the ingredients listed beneath it in descending order and no individual amounts. You can calculate cost per capsule. You cannot calculate cost per milligram of the thing you’re buying. Our articles on reading the Supplement Facts panel and on calculating cost per serving go through the math.
What does the same money buy instead?
This is the question that reframes everything, and it has to use real alternatives rather than abstractions.
A 2015 study in The Journal of Pain, led by a researcher at the National Center for Complementary and Integrative Health, estimated annual per-person out-of-pocket spending on complementary approaches for specific painful conditions at between $568 for regular headaches and $895 for fibromyalgia. Set those numbers next to a few sessions with a physical therapist, a clinician visit with a plan attached, or a year of a better grocery budget.
On groceries, the trade is often smaller than people think. US Department of Agriculture Economic Research Service analysis published in May 2016, using 2010 retail scanner data, put organic price premiums above 20% for nearly every one of the 17 products examined, from 7% for fresh spinach up to 60% for salad mix and 54% for canned beans. Our article on organic premiums covers where it’s worth paying. The general rule is the same one that governs this whole article: if a budget forces a choice between a premium version of a food and simply more of that food, more food usually wins.
What does being wrong cost?
Add three things, not one.
Money is the obvious part. Delay is the dangerous part. A product bought to handle a symptom can push a real evaluation back by months, and some conditions get much harder to treat in that time.
Then there’s direct harm. The Food and Drug Administration has repeatedly found hidden pharmaceutical ingredients in products sold for sexual improvement, energy, weight loss, pain, bodybuilding and sleep. Reported effects in its case material include sudden weight gain, gastrointestinal damage including bleeding, raised blood sugar, liver toxicity and death. The agency keeps a public list of tainted products and says plainly that it “covers only a small fraction” of them, so a product’s absence from that list is not a clean bill of health.
Diagnostic purchases carry a version of this too. The FDA notes that direct-to-consumer tests “are not a substitute for a traditional health care evaluation,” and a flagged value on a large panel can generate a specialist visit, a repeat test and an imaging study that were never in the advertised price. Our article on expensive wellness panels covers that math.
MedlinePlus, the consumer service of the National Library of Medicine, makes the underlying point about results of any kind. A value outside the reference range does not automatically mean a health problem, and a value inside it does not rule one out. Labs use different methods and different ranges, so numbers do not travel cleanly between them. Buying more measurement is not the same as buying more information.
How will I decide to stop?
Almost nobody has an answer. This is the whole game, so it gets its own section.
The stopping rule, which is the part nobody has
Before you pay, write down three things: the outcome you’ll track, the threshold that counts as working, and the date you’ll check.
That’s it. It takes ninety seconds and it is the single highest-value habit in this entire subject.
Why pre-committing beats judging later
Researchers who study treatments in a single person do exactly this. In an n-of-1 trial, a person cycles through alternating periods of a treatment and a comparator, with outcomes measured repeatedly. A 2026 analysis in the journal Epilepsia by a team at University Medical Center Utrecht re-examined several such trials and judged response against a threshold set in advance, a 30% reduction in seizure frequency, treated as the smallest change that would matter.
The clinical setting isn’t yours. The design principle is. The threshold goes on paper before the data exists, because once you’ve spent the money and felt hopeful, almost anything can look like improvement.
Pick something you can count
“More energy” cannot be assessed. Pick a number.
Nights you woke after 3 a.m. Days last week with pain above a 4. Flights of stairs before you stop. Afternoons you needed a second coffee. Count it for one week before you start, so you have a baseline that isn’t a memory.
Pick a date that fits the claim
Match the window to what’s being promised. Something that’s supposed to act within hours gets two weeks. Something plausibly slow gets eight to twelve. Anything described as needing six months to show results should get a harder look, because that’s long enough for almost any condition to change on its own.
Write the date on the bottle in marker. Put it in your calendar. The date is doing the work, not your memory.
Change one thing at a time
If you start three products in one week, you’ve bought a result you can’t attribute. You’ll keep all three forever.
What a filled-in rule looks like
Say you’re buying a sleep product because you wake at 3 a.m. most nights. Your outcome is the number of nights per week you wake after 3 a.m. Your baseline is one week of counting, which comes out at five. Your threshold is two or fewer, because that’s a change you would feel. Your date is six weeks from today, written on the bottle in marker.
On that date you look at the count, not at how you feel about having bought it. If you’re still at four, you stop. You’ve spent six weeks and one bottle to get an answer, instead of two years to get a habit.
Remember when you bought it
You almost always buy at your worst moment. Many conditions swing, and the bad stretch that sent you shopping often eases on its own. That’s exactly why a single before-and-after comparison in one person is weak evidence, and why trials in a single person alternate periods instead of measuring once.
Where wellness spending tends to pay off
The pattern, stated as a judgment rather than a trial result: money does better on durable things with strong evidence than on consumables with weak evidence.
Preventive care is the clearest case, and much of it is already paid for. Under federal rules, most health plans including Marketplace plans must cover a set of preventive services with no cost sharing when you use an in-network provider, which usually means no copay for recommended screenings and immunizations even before you meet a deductible. HealthCare.gov does note that $0 isn’t guaranteed in every case, so check your plan. Still, an unused free screening is the most expensive thing in this article.
The scale explains the priority. The Centers for Disease Control and Prevention reports that about 3 in 4 US adults have at least one chronic condition and more than half have two or more, and that chronic diseases are major drivers of $5.3 trillion in annual US health care spending.
Equipment that lowers the barrier to something you already know works also tends to earn out: shoes that let you walk, a bike that replaces a drive, a decent pan. These are boring purchases. Boring is often the tell.
There’s a parallel habit on the clinical side worth borrowing. The ABIM Foundation ran an 11-year campaign starting in 2012, with more than 80 specialty society partners, that produced over 700 recommendations identifying tests and treatments that are overused or unnecessary. The campaign has ended and the foundation has moved to other work, but the recommendations remain a reference, and the underlying move transfers directly to the checkout page: ask what this will change before you agree to it.
The half-used bottle is not a reason to continue
Money already spent is gone. It’s gone whether you finish the bottle or not.
The only live question is whether the next 30 days of cost and effort are worth the next 30 days of expected benefit. A half-used supply changes nothing about that calculation, though it feels like it should.
Automatic renewals make this harder, because the default is to continue and nobody has to decide anything. The Federal Trade Commission reported in March 2026 that it has received more than 100,000 complaints in the past five years about negative option marketing and related practices. Our article on wellness subscriptions covers what protections currently exist, which is a genuinely unsettled question right now.
Routines also accumulate quietly. Products get added for a reason and never removed, because removal requires a decision and addition only required a moment. An annual review of everything you’re taking, with the bottles on the table, catches more waste than any single purchase decision. Our article on evaluating supplement costs goes deeper on that.
When the answer is a clinician, not a purchase
Some situations should end the shopping entirely.
Stop and dial 911 if weakness or numbness hits one side of the body without warning, if speech suddenly won’t come out or won’t make sense, if vision changes in an instant, if balance goes, or if a severe headache arrives with no known cause. The CDC notes the most effective stroke treatments require diagnosis within three hours of first symptoms, and says not to drive yourself and not to have someone else drive you.
Make an appointment, don’t make a purchase, if you have unexplained weight loss, a symptom that’s new and getting worse, or a problem you’ve been treating with products for more than a few weeks without anyone examining you.
Talk to a pharmacist or prescriber before buying anything if you take a blood thinner, have surgery scheduled, are pregnant or nursing, or already take several supplements. The NIH Office of Dietary Supplements notes that risk rises with high doses, with substituting a supplement for a prescribed medicine, and with combining many products at once.
And if you’re buying something to replace a treatment you’ve been prescribed, that’s the one purchase to stop and discuss first.
The card to write before you pay
One index card, six lines, filled in before the money moves:
- The problem, in a sentence with a number in it.
- The evidence tier, honestly labeled: established, preliminary, insufficient, or disproven.
- The cost per year, not per bottle.
- What else that amount buys.
- The outcome I’ll count, and my baseline for one week.
- The threshold and the date. If I’m not at X by this date, I stop.
If you only fill in one line, fill in line 6. Deciding in advance what failure looks like is what turns a purchase into a question you can answer. Skip it and you get the other outcome, which is a cabinet.
Sources
- PubMed. “Nahin RL, Barnes PM, Stussman BJ. "Expenditures on Complementary Health Approaches: United States, 2012." National Health Statistics Reports No. 95, 2016. PubMed. PMID: 27352222.” pubmed.ncbi.nlm.nih.gov
- PubMed. “Mangione CM, Barry MJ, Nicholson WK, et al. "Vitamin, Mineral, and Multivitamin Supplementation to Prevent Cardiovascular Disease and Cancer: US Preventive Services Task Force Recommendation Statement." JAMA, 2022. PubMed. DOI: 10.1001/jama.2022.8970.” doi.org
- PubMed. “Nahin RL, Stussman BJ, Herman PM. "Out-Of-Pocket Expenditures on Complementary Health Approaches Associated With Painful Health Conditions in a Nationally Representative Adult Sample." The Journal of Pain, 2015. PubMed. DOI: 10.1016/j.jpain.2015.07.013.” doi.org
- PubMed. “Defelippe V, Bartoš F, Wagenmakers EJ, Braun KPJ, Jansen FE, Otte WM. "Continuous outcome estimation in N-of-1 trials for accelerated decision-making." Epilepsia, 2026. PubMed. DOI: 10.1002/epi.70119.” doi.org
- US Preventive Services Task Force. “About the USPSTF.” uspreventiveservicestaskforce.org
- US Food and Drug Administration. “Avoiding Products Contaminated with Hidden Ingredients.” fda.gov
- US Food and Drug Administration. “Direct-to-Consumer Tests.” fda.gov
- MedlinePlus, US National Library of Medicine. “How to Understand Your Lab Results.” medlineplus.gov
- US Department of Agriculture, Economic Research Service. “Investigating Retail Price Premiums for Organic Foods.” ers.usda.gov
- Federal Trade Commission. “FTC Seeks Public Comment in Response to Advance Notice of Proposed Rulemaking Regarding Negative Option Marketing Practices.” ftc.gov
- Centers for Disease Control and Prevention. “About Chronic Diseases.” cdc.gov
- Centers for Disease Control and Prevention. “Stroke Signs and Symptoms.” cdc.gov
- HealthCare.gov. “Preventive care benefits for adults.” healthcare.gov
- ABIM Foundation. “Choosing Wisely: Patient Resources.”
- National Institutes of Health, Office of Dietary Supplements. “Dietary Supplements: What You Need to Know.” ods.od.nih.gov
- US Food and Drug Administration. “Dietary Supplement Labeling Guide: Chapter IV, Nutrition Labeling.” fda.gov