Fisetin-Rich Foods: Why Strawberries Top the List (USDA FoodData Central Analysis)

While building HealthSavvyGuide on top of the USDA FoodData Central API, I ran into a small puzzle that turned into this article. I wanted to add a "flavonoid" view to the aggregator, so I queried the standard nutrient panel for strawberries (FDC ID 167762, "Strawberries, raw" in the SR Legacy dataset). The panel had vitamin C, folate, potassium, fiber — everything you would expect. But the compound that strawberries are quietly famous for in longevity research, fisetin, was nowhere in that table.

It turns out fisetin doesn't live in FoodData Central's core nutrient panel at all. It sits in a separate, specialized dataset. That gap is the whole story of how nutrition data is structured, and it is a useful lens for understanding why a compound that's been studied in aging labs since the 2010s is still missing from most food labels. This article walks through what the USDA data actually says, what the research does and doesn't claim, and how I handled it as a data engineering problem.

This article is for informational purposes only and is not medical advice. Fisetin is an area of early research, much of it in animal models. Consult a qualified healthcare provider before changing your diet or taking any supplement, especially if you are pregnant, nursing, managing a chronic condition, or taking medication.

What fisetin is, in plain terms

Fisetin is a flavonol — a subclass of flavonoids, the plant pigments that also include quercetin and kaempferol. Chemically it is close to quercetin (the two differ by a single hydroxyl group), which is why you often see them discussed together. In plants, fisetin acts partly as a yellow-orange pigment and a defense compound. In humans, it is studied as a dietary antioxidant, and more recently as a candidate "senolytic" — a molecule that may help the body clear senescent cells.

Senescent cells are cells that have stopped dividing but refuse to die. They accumulate with age and secrete inflammatory signals (a phenomenon researchers call the senescence-associated secretory phenotype, or SASP). The idea behind senolytics is that selectively clearing these cells could reduce age-related inflammation. The U.S. National Institute on Aging funds research in this area, and it is genuinely promising — but it is still early, and most of the strongest results come from mice, not people.

The USDA data problem: two databases, one strawberry

Here is the engineering detail that most consumer articles skip. FoodData Central is not one flat table. It is a collection of datasets stitched together: SR Legacy (the old Standard Reference data), Foundation Foods, the FNDDS survey data, and branded products. The everyday nutrients — calories, protein, vitamin C, minerals — are in those panels.

Individual flavonoids like fisetin were historically tracked in a different USDA product entirely: the USDA Database for Flavonoid Content of Selected Foods (the last public release, 3.3, dates to 2018). To build a single "how much fisetin is in this fruit" view, an aggregator has to join the base food record from FoodData Central with flavonoid figures sourced from that separate database and the peer-reviewed literature it draws on. They don't share one clean API endpoint, which is exactly why fisetin "disappears" if you only query the standard panel.

For anyone building with this data, that's the practical takeaway: the absence of a compound in the FDC nutrient panel does not mean the food lacks it. It often means the compound was measured in a specialized dataset that never got merged into the main API.

Which foods actually contain fisetin

Pulling the published flavonoid figures together (the values widely cited from Arai and colleagues' 2000 measurements and later compilations), strawberries are the standout source. The commonly reported concentrations, expressed in micrograms of fisetin per gram of food, look roughly like this:

The first thing that jumps out, even from a data-cleaning standpoint, is the scale of the gap. Strawberries sit at roughly six times the level of the next food on the list and more than 25 times that of onions. When one row in a dataset is an order of magnitude above the rest, it usually tells you something real — in this case that if fisetin matters to you, strawberries are doing most of the work.

The second thing worth flagging: these are concentrations per gram, not per serving, and they are averages across the studies. Real fruit varies by cultivar, ripeness, growing conditions, and storage. A useful data point I keep coming back to is that a standard 1-cup serving of raw strawberries is about 152 grams (per the FoodData Central serving data), which at ~160 µg/g implies very roughly 24 mg of fisetin per cup — a rough estimate, not a guaranteed dose.

The dose gap nobody mentions

This is the most important number in the whole topic, and it is the one that gets lost in headlines. The senolytic studies that generated the excitement used doses far higher than food provides. The widely cited 2018 mouse study by Yousefzadeh and colleagues (published in EBioMedicine), titled "Fisetin is a senotherapeutic that extends health and lifespan," used concentrated fisetin at levels equivalent to hundreds of milligrams per kilogram of body weight in mice.

Human senolytic trials, including the Mayo Clinic-led work, have tested fisetin in the range of 20 mg/kg/day given in short intermittent bursts — which for an adult lands in the hundreds-to-thousands of milligrams per day. Compare that to the rough ~24 mg in a cup of strawberries. You would have to eat an unrealistic quantity of fruit to approach the experimental doses. So the honest framing is: eating strawberries gives you fisetin among many other beneficial compounds, but it is not the same intervention as the high-dose research, and you should not assume the lab results transfer to a bowl of fruit.

What the research supports — and what it doesn't

What is reasonably well established: flavonoid-rich diets, including berries, are associated with better cardiovascular and metabolic markers in large observational studies. Strawberries specifically deliver vitamin C, manganese, folate, fiber, and a mix of polyphenols. Those are solid reasons to eat them regularly, and they don't depend on the senolytic hypothesis being true.

What is still open: whether fisetin meaningfully clears senescent cells in humans, whether that translates to healthier aging, and what a safe and effective dose would be. The Mayo Clinic has run early-stage human trials (for example, the AFFIRM-LITE trial, registered as NCT03675724, looking at fisetin in older adults). These are small, exploratory, and not yet the kind of large outcome studies that change medical guidance. As of this writing, no major health authority recommends fisetin supplements for longevity, and the long-term safety of high-dose supplementation isn't well characterized.

How I treat fisetin in the aggregator

From a data engineering perspective, fisetin is a good case study in honest labeling. When I surface it in HealthSavvyGuide, I tag it with three pieces of metadata that I think every nutrition aggregator should carry but most don't:

• Source dataset — flagging that the value comes from the flavonoid database and literature, not the FDC core panel, so the provenance is transparent.
• Measurement variance — marking the figure as an average with wide real-world spread, rather than presenting a single number as gospel.
• Evidence stage — noting that the health claims attached to the compound are preclinical or early-human, not established.

That third tag is the one I care about most. It is trivially easy to build a "superfood ranking" that sorts foods by some trendy compound and implies a health outcome. It is much more useful, and more honest, to show the number alongside how strong the evidence behind it actually is. The data should never out-run the science.

A data note on why the numbers vary

One reason I'm cautious about presenting a single fisetin figure is that the published measurements were taken under different conditions, on different cultivars, using different analytical methods (mostly HPLC after hydrolysis). When you aggregate values from multiple studies into one cell of a database, you are quietly averaging across all of that variance. A wild strawberry and a supermarket strawberry bred for size and shelf life are not the same fruit chemically, and ripeness shifts flavonoid content too. So when you see "strawberries: ~160 µg/g," read the tilde as doing real work — it could plausibly be half or double that in a given basket of fruit.

This is also why I avoid stacking precise-looking decimals onto these values in the aggregator. Reporting "160.0 µg/g" implies a precision the underlying data simply doesn't have. Rounding to a sensible significant figure and pairing it with a variance flag is more honest than a falsely exact number — a principle that applies to a lot of nutrition data, not just fisetin.

Frequently asked questions

Is fisetin the same as quercetin? No, but they are close relatives in the flavonol family and are often studied together. They differ by a single hydroxyl group, and both are present in many of the same fruits and vegetables.

Do I get enough fisetin from food? "Enough" isn't well defined, because there is no established dietary requirement for fisetin — it is not an essential nutrient. A varied diet with berries and colorful produce gives you a reasonable amount as part of a broader flavonoid mix.

Should I take a fisetin supplement? That is a question for a healthcare provider. The high-dose senolytic research is preclinical or early-stage, long-term safety isn't well established, and no major health authority currently recommends it for longevity.

Does cooking destroy fisetin? Flavonoids can degrade with prolonged heat and certain processing, which is one more reason fresh fruit figures don't map neatly onto cooked or processed foods. Most strawberries are eaten raw, which helps.

Practical, non-medical takeaways

If you enjoy strawberries, they are a genuinely nutrient-dense fruit and a sensible part of a varied diet — the fisetin is a bonus, not the headline. Eating a range of colorful fruits and vegetables remains the best-supported way to get a broad mix of flavonoids, including quercetin, anthocyanins, and fisetin together. And if you are considering a concentrated fisetin supplement because of something you read about longevity, that is exactly the kind of decision to bring to a healthcare provider rather than to a search engine — the research is interesting, but it has not yet earned a recommendation.

Sources and further reading

• USDA FoodData Central — base nutrient data for strawberries and other foods (fdc.nal.usda.gov)
• USDA Database for Flavonoid Content of Selected Foods, Release 3.3 (2018)
• National Institute on Aging — research on cellular senescence and aging (nia.nih.gov)
• Yousefzadeh et al., "Fisetin is a senotherapeutic that extends health and lifespan," EBioMedicine, 2018
• Mayo Clinic — senolytic research and the AFFIRM-LITE trial (ClinicalTrials.gov NCT03675724)

Reminder: This article reflects an engineering perspective on public nutrition data and is for general information only. It is not medical or dietary advice, and nothing here should be used to diagnose, treat, or prevent any condition. Always consult a qualified healthcare professional for guidance specific to you.