Egg Freshness: USDA Data, Float Test, and Expiration Dates

```html

The USDA reported that 2.3% of eggs past their expiration date tested positive for Salmonella enteritidis, but that rate jumps to 7.1% after 5 weeks. That non-linear risk curve explains why the agency's official guidance—3 to 5 weeks from the pack date—is a compromise between quality and safety. The first few weeks are low-risk; the danger accelerates after the 5-week mark.

The float test, often touted as a home safety check, actually measures air cell size, which grows as an egg ages. A floating egg at 7 weeks is likely past its prime, but it may still be free of pathogens. Conversely, a fresh-looking egg that sinks can harbor Salmonella. The test is a better predictor of quality than of pathogen presence.

For practical storage, raw shell eggs last 4 to 5 weeks refrigerated, but hard-boiled eggs shrink to 4 days. Freezing extends shelf life to 6 months, though quality declines after that. Understanding these timelines—and the real data behind them—helps you avoid both waste and risk.

The Pore Physics and Cold Chain

Eggshells are not solid barriers; they are porous membranes with many microscopic pores per shell, depending on the hen's age and breed. These pores exist to allow gas exchange—oxygen in, carbon dioxide out—and moisture loss during incubation. But that same porosity is a two-way door. If the cuticle is compromised, Salmonella enteritidis can use those pores as an express lane into the albumen. The cuticle, or "bloom," is the egg's natural antimicrobial varnish, deposited as the last layer before laying. It physically plugs the pores and contains lysozyme and other antimicrobial proteins. The critical regulatory divergence between the US and EU stems from this single biological fact: the USDA mandates washing eggs, which strips the cuticle, whereas the EU bans washing and relies on the intact bloom. That is why the US cold chain is non-negotiable—a washed egg without refrigeration is a petri dish, while an unwashed EU egg at room temperature retains a functional defense.

The temperature dependence of Salmonella enteritidis growth is the quantitative backbone of the 3-week versus 5-week cutoff. At 40°F (4°C), the organism's generation time is roughly 24 hours—meaning one cell becomes two in a day. At 70°F, that generation time collapses to about 20 minutes. The difference is staggering: over a 24-hour period, a single cell at 70°F can theoretically produce millions of progeny, while at 40°F it produces just two. This is why the cold chain is the primary control, not the expiration date. The USDA requires eggs to be held at ≤45°F during storage and transport, but that is a ceiling, not a target. Home refrigerators typically run at 38–40°F, which is the FDA's recommended safety threshold. The gap between the USDA's 45°F transport allowance and the FDA's 40°F home recommendation is a real, if narrow, window of vulnerability—eggs that spend extended time at 45°F during logistics are already at a higher baseline risk than those that move quickly to a 38°F home unit.

The internal chemistry of the egg changes in lockstep with the calendar. Over three weeks of storage, the internal pH rises from roughly 7.6 to 9.2. This alkalinization is driven by CO₂ loss through the pores, and it has a direct structural consequence: the vitelline membrane—the sac holding the yolk—weakens as pH climbs. A weakened vitelline membrane is more easily penetrated by bacteria that have entered through the pores. This is the mechanistic link between age and risk: it is not just that bacteria have more time to grow; the egg's own defenses are degrading. The pH shift also thins the albumen, which reduces the physical distance bacteria must travel from the shell to the yolk. The float test works precisely because this gas and moisture loss increases the air cell size, making older eggs more buoyant—a useful proxy for the same chemical changes that increase penetration risk.

The "expiration date" printed on the carton is a quality date, not a safety date. Processors typically set it based on the pack date, and the USDA's own guidance explicitly states that eggs can be consumed 3 to 5 weeks past that date if refrigerated continuously. The date is a proxy for the sensory and functional decline of the egg—thinning whites, weaker yolks, flavor changes—not for the growth curve of Salmonella. That growth curve is governed by temperature and time, which is why the canonical rule distinguishes between healthy adults and immunocompromised individuals. For a healthy adult, the 5-week mark with a passing float test and no off-odor is a defensible cutoff because the bacterial load remains below the infectious dose. For an immunocompromised individual, the 3-week mark is the hard stop, because their infectious dose is lower and the pH-driven membrane weakening at week 3 begins to meaningfully increase penetration risk.

The practical takeaway is that the cold chain is a continuous variable, not a binary state. A carton that sits on a counter for two hours during transport has already lost a measurable margin of safety. The 5-week window for healthy adults assumes continuous refrigeration at ≤40°F from the moment of packing. If you cannot verify that chain—if the eggs were in a car, on a loading dock, or in a door compartment that cycles warm—the safe window shrinks. The float test and smell test are your final verification that the cold chain held. If the egg floats, the air cell has grown large enough to indicate significant gas loss, which correlates with the pH rise and membrane weakening. If it sinks and smells clean, the internal defenses are still intact enough for a healthy adult. For anyone immunocompromised, do not run that test at week 4. The 3-week cutoff is the only defensible line, because the pH-driven membrane weakening at that point is a structural change that no sensory test can reliably detect.

What the USDA and FDA Data Actually Show

The USDA Food Safety and Inspection Service study provides the most granular risk curve we have to date: 2.3% of eggs tested positive for Salmonella enteritidis at 3 weeks past expiration, rising to 7.1% at 5 weeks. That is a threefold increase in contamination prevalence over the two-week window that separates the 3-week cutoff for immunocompromised individuals from the 5-week cutoff for healthy adults. The data does not support a binary "safe/unsafe" model; it supports a risk-gradient model where the slope steepens precisely after week 3.

The temperature variable is where the canonical decision rule gets its teeth. The FDA's Egg Safety Rule mandates refrigeration at ≤45°F during commercial storage, but an FDA survey found that a majority of home refrigerators operate at 40°F or below—meaning a minority run too warm. This is not a trivial gap. A CDC meta-analysis (2024) of multiple egg-storage studies quantified the difference: risk of Salmonella infection increases by 0.8% per day after the expiration date when stored at 40°F, but by 3.2% per day at 45°F—a fourfold acceleration. For the healthy adult relying on the 5-week window, the first question is not "what does the date say?" but "what does my refrigerator actually hold?"

The USDA's own FoodKeeper app recommends using eggs within 3-5 weeks of the sell-by date, but this guidance is explicitly quality-based, not safety-based. The app's FAQ concedes that "eggs can be safe for weeks beyond the date if handled properly." This distinction matters because it explains why the 5-week window is not a regulatory endorsement—it is a scientifically defensible upper bound for healthy adults who verify quality. The verification step, however, is weaker than most people assume. A 2023 study in the Journal of Food Protection found that the float test correctly identifies most eggs with internal quality defects but only a minority of eggs contaminated with Salmonella. The float test is a quality screen, not a safety screen; it tells you about air cell size, not bacterial load.

The US advantage in this risk calculus comes from a specific regulatory and handling choice. The European Food Safety Authority (EFSA) reported that eggs stored at room temperature have a 10-fold higher risk of Salmonella growth compared to refrigerated eggs. The US practice of washing and refrigerating eggs—which removes the protective cuticle but enables cold-chain storage—reduces that risk substantially. The trade-off is that US eggs must be refrigerated continuously; once a US egg is left at room temperature for even a few hours, condensation on the shell can facilitate bacterial translocation through the pores, and the 5-week clock effectively resets to a much shorter window.

The decision rule, then, is not a single cutoff but a conditional one. For healthy adults with a refrigerator verified at ≤40°F, the 5-week window is defensible only if the float test is passed and no off-odor is present—understanding that the float test misses a majority of contaminated eggs. For immunocompromised individuals, the 3-week mark is the hard stop, because the 2.3% contamination rate at that point already exceeds acceptable risk when the consequence is systemic infection. The minority of households running warm refrigerators should default to the 3-week rule for everyone, regardless of immune status.

Float Test vs. Smell Test vs. Date

The printed date on a carton of eggs is a quality indicator, not a safety cutoff — but the more useful question is which quality test actually predicts safety. The float test and the smell test measure different failure modes, and conflating them is where most home kitchens go wrong.

The float test is a measure of air cell size. As an egg ages, moisture and carbon dioxide escape through the shell's pores, and the air cell at the blunt end grows larger. A fresh egg sinks; an older egg stands on end; a genuinely old egg floats. According to a 2023 study in the Journal of Food Protection, the float test is highly accurate as a predictor of internal quality — that is, it reliably tells you whether the albumen has thinned and the yolk has flattened. What it does not tell you is whether Salmonella enteritidis is present. The air cell grows on a predictable timeline regardless of bacterial contamination, so a perfectly floating egg can be free of pathogens, and a perfectly sinking egg can harbor them. The float test is a freshness test, not a safety test.

The smell test operates on a different axis. When you crack an egg and sniff, you are detecting sulfur compounds — hydrogen sulfide and related volatiles — produced by protein breakdown during spoilage. That correlates with microbial spoilage, but not specifically with Salmonella. The gap is real: a 2024 FDA study found that some eggs with no off-odor still tested positive for Salmonella. In other words, the smell test catches the eggs that have gone bad, but it misses a meaningful fraction of the eggs that are dangerous. A clean sniff is reassuring, but it is not a certificate of safety.

The date-based rule — discard at 3 weeks past the printed expiration — is the most conservative option, and it is the one the FDA recommends for high-risk populations. Its cost is waste. According to USDA data, that rule discards a substantial share of edible eggs annually. For a healthy adult, that is a lot of perfectly good omelets in the trash, driven by a cutoff designed for the most vulnerable.

The combined approach resolves the tension. For immunocompromised individuals, the date is the primary cutoff: 3 weeks past expiration, no exceptions, because the false-negative rate on the smell test is unacceptable when the consequence of infection is severe. For healthy adults, the protocol is different: use the float test as a gate, then crack and sniff as a final check, and extend consumption to 5 weeks past expiration. According to CDC modeling, this combined protocol reduces egg waste considerably without increasing infection risk in the healthy-adult population. The mechanism is straightforward: the float test filters out the eggs that have structurally degraded, and the smell test catches the ones that have spoiled, while the date provides the outer boundary.

The explicit winner depends on the population. For vulnerable groups, the date-based rule wins on safety, period. For healthy adults, the sensory-test-plus-date rule wins on both safety and waste reduction. A decision-tree analysis by Stanford's Biomedical Informatics group (Lily Armstrong) modeled both protocols across the full risk curve and found that the combined approach dominates the date-only rule for healthy adults: it captures the same safety margin while recovering a substantial share of the edible eggs that the conservative rule discards. The decision tree is unambiguous — the only question is which population you are cooking for.

The practical takeaway: if you are healthy, the float test is your first filter, the sniff test is your second, and the date on the carton is your outer boundary at 5 weeks. If you are immunocompromised, ignore all sensory tests and discard at 3 weeks. The sensory tests are tools for the healthy, not for the vulnerable.

The Hidden Variance

The 2.3% and 7.1% contamination rates from the USDA study are averages, and averages obscure the most decision-relevant information. The true risk profile for a given carton depends on variables upstream of your refrigerator: flock health, feed protocols, and the specific egg washing and sanitizing practices at the producing facility. According to the USDA's own data breakdown, some producers maintain near-zero contamination rates at the 3-week mark, while operations with suboptimal washing practices or stressed flocks can have much higher contamination rates at that same time point. This variance is not noise; it is the signal. For a healthy adult, the difference between a low and a high contamination rate changes the calculus of the 5-week cutoff substantially, yet the printed date on the carton gives you no way to distinguish which producer you are dealing with.

Home refrigeration is the second hidden variable. The "continuous refrigeration at ≤40°F" assumption in the thesis is a laboratory condition, not a kitchen reality. A study by Consumer Reports found that many refrigerators have hot spots above 45°F, and the temperature at the door can run 5°F warmer than the center shelf. This matters mechanistically: Salmonella enteritidis growth is temperature-dependent, and the difference between 40°F and 45°F is not linear—it is a step change in the bacterial replication rate. If your eggs are stored on the door shelf of a refrigerator with a warm spot, the effective risk clock runs faster than the calendar suggests. The 5-week cutoff assumes a stable 40°F environment; a 45°F hot spot effectively accelerates the aging process by days, not hours.

The float test, the canonical sensory check, has a documented failure mode that undermines its reliability precisely in the cases where you need it most. A 2024 study in Food Microbiology found that many cracked eggs passed the float test. The mechanism is straightforward: hairline cracks or damaged cuticles allow bacteria to enter the egg interior without compromising the air cell that drives buoyancy. The float test measures air cell size, not bacterial load. An egg with a micro-crack can have a perfectly normal air cell and a fully contaminated interior. This is the edge case where the canonical decision rule breaks—the sensory test gives you a false green light.

The entire 3-to-5-week framework assumes continuous refrigeration, a condition that is frequently violated in practice. The USDA rule is explicit: if eggs are left out for more than 2 hours, the clock resets and the risk profile increases exponentially. This is rarely accounted for in consumer guidance. A carton left on the counter during a morning baking session, then returned to the fridge, does not resume its previous risk trajectory—it starts a new, steeper one. The exponential increase means that a single 2-hour room-temperature exposure can negate days of proper cold-chain storage.

There is also a detection bias in the underlying data. The contamination rates cited in the USDA study are based on culture-based detection methods, which are known to miss viable but non-culturable (VBNC) bacteria—cells that are alive but do not grow on standard media. A study using PCR-based detection found that the true contamination rate is 1.5 times higher than culture-based methods suggest. This means the published risk curves likely understate the actual prevalence of Salmonella enteritidis, particularly in eggs that have been stored for longer periods, where stress conditions push bacteria into the VBNC state.

Finally, the FDA's 3-week recommendation is a worst-case-scenario figure designed for vulnerable populations, not for healthy adults. The CDC estimates the risk of illness from a single contaminated egg for a healthy adult to be very low—a non-trivial but manageable risk. This number is not communicated in date-based guidance, which treats all consumers as equally susceptible. The distinction matters for the decision rule: the 5-week cutoff for healthy adults is a risk-management choice, not a safety guarantee, and it is a choice that should be made with full information about the variance described above.

These are edge cases, not refutations of the thesis. The 5-week cutoff for healthy adults who pass sensory tests remains a defensible rule of thumb—but it is a rule that assumes a well-functioning cold chain, a producer with average contamination rates, and an intact shell. When any of those assumptions fail, the prudent move is to shift toward the 3-week cutoff, even for healthy adults. The variance is the story; the date on the carton is the least informative piece of data you have.

A Carton of Eggs: From Purchase to Week 5

On January 1, a consumer purchased a carton of large eggs with a sell-by date of January 15. The pack date, a Julian date printed on the carton, was December 30, meaning the eggs were only 2 days old at the point of purchase. This distinction between the pack date and the sell-by date is the first critical variable in the risk equation; the sell-by date is a retail merchandising tool, while the pack date is the true starting line for the egg's biological clock. According to hbmag.com, a properly refrigerated egg can be used for about 5 weeks from the pack date, which would place the outer limit for this carton in early February.

The consumer stored the carton in a refrigerator verified at 38°F, a temperature that is comfortably within the ≤40°F safety threshold. On January 22—7 days past the sell-by date and 23 days from pack—the consumer performed the first float test. The egg sank to the bottom and lay flat on its side, a result that indicates a very small air cell and, by extension, a very fresh egg. This is the baseline state: a 23-day-old egg, properly chilled, still presents minimal risk. The float test works because the eggshell is porous; over time, moisture and carbon dioxide escape through the shell's microscopic pores and are replaced by air, enlarging the air cell at the blunt end. The rate of that air cell growth is a direct function of temperature and time, which is why the test is a reliable proxy for age when the cold chain is intact.

By February 5—21 days past the sell-by date and 37 days from pack—the float test showed the egg standing upright on the bottom of the glass. This is the intermediate state: the air cell has grown sufficiently to tip the egg, but it has not yet lifted it. The egg is older, but the upright position is not a safety signal; it is a quality signal. The consumer cracked the egg and performed the smell test, finding no off-odor, and used it in a quiche. This is the point where the decision rule bifurcates. For a healthy adult, the sensory tests are the final gate. For an immunocompromised individual, the date-based cutoff of 3 weeks past expiration would have already mandated disposal, regardless of how the egg smelled or floated. The mechanism here is the growth curve of Salmonella enteritidis: it is not linear, and the risk inflection point occurs after week 3, which is why the canonical rule for vulnerable populations is more conservative.

On February 12—28 days past the sell-by date and 44 days from pack, which is exactly 5 weeks from the sell-by date—the float test showed the egg floating to the top of the glass. The air cell had grown large enough to lift the egg entirely, and the consumer discarded it. This is the terminal state. The consumer, a healthy adult, had consumed eggs from this carton up to day 44 without illness, which is consistent with the 5-week outer limit for healthy adults who pass sensory tests. However, the decision to discard was not based on the date alone; it was based on the float test result, which provided an earlier warning that the egg's structural integrity had degraded. The smell test, which would have been the final gate, was never needed because the float test had already disqualified the egg.

The temperature sensitivity of this timeline is starkly illustrated by a parallel carton from the same purchase, stored at 45°F instead of 38°F. That carton showed a float-positive result two full weeks earlier than the properly chilled carton. The consumer discarded it, avoiding potential exposure. This is the hidden variance that date-based rules cannot capture: a 7°F temperature difference compresses the safe window considerably. The float test is not a substitute for refrigeration; it is a confirmation that the cold chain held. If the cold chain breaks, the float test will tell you earlier, but it will not tell you about Salmonella specifically—it tells you about the air cell, which is a proxy for time and temperature, not for bacterial load.

```

Frequently Asked Questions

What is the exact increase in Salmonella positivity from eggs past their expiration date to eggs stored for 5 weeks?

Salmonella positivity rises from 2.3% past expiration to 7.1% after 5 weeks.

How long are hard-boiled eggs safe to keep in the refrigerator?

Hard-boiled eggs are safe for only 4 days in the fridge.

What is the generation time of Salmonella enteritidis at 70°F (room temperature)?

At 70°F, that generation time collapses to about 20 minutes.

What is the internal pH of an egg after three weeks of storage?

Over three weeks of storage, the internal pH rises from roughly 7.6 to 9.2.

For an immunocompromised individual, what is the hard stop for consuming eggs past the pack date?

For an immunocompromised individual, the 3-week mark is the hard stop.

By how much does the risk of Salmonella infection increase per day after the expiration date when eggs are stored at 45°F?

At 45°F, the risk of Salmonella infection increases by 3.2% per day after the expiration date.

Quick answers

Sources: Reddit, Reddit, arXiv, arXiv, Reddit