Auburn study targets food waste by refining ‘sell by' dates

Scientists map microbial changes in ground beef

The average American wastes more than 1,000 pounds of food each year. The resources used to generate that wasted food are staggering — emissions equivalent to 42 coal-fired power plants; enough water and energy for 50 million homes; and cultivated land equivalent to California and New York combined, according to an EPA report.

Researchers at Auburn University are taking a bite out of the food waste problem by studying how meat spoils at the microbial level. If scientists develop a better method of predicting spoilage, meat packers could affix more accurate “best by” or “sell by” dates, leading to less meat being thrown out. About 10% of all meat is discarded at the retail or consumer levels, typically due to reaching a “sell by” date, according to a U.N. report.

“Food waste is a growing issue that impacts consumers, producers, and environmental sustainability,” said Isabella Gafanha, a master’s student at the Auburn College of Agriculture. “Part of that elevated food waste has to do with conservative ‘sell by’ dates.”

When producers package beef, they affix a “sell by” date of, typically, four days after the package date. This date is often overly conservative and based on when the beef will start to lose its bright pink color, rather than when it becomes unsafe to eat. As a result, many consumers associate a change in color with meat being unsafe to eat.

“Color degradation has more to do with food quality than food safety,” said Gafanha. “If a consumer sees that on day three their meat is brown, and it's around the ‘sell by’ date, they might think it’s spoiled, but in fact it’s just quality degradation. It’s still fine to eat.”

However, consumers tend to go by sight and “sell by” label when cleaning out their refrigerators, and they often waste perfectly safe meat. Better “sell by” dates might be the best method for ensuring consumers waste less meat.

Before scientists can begin to refine “sell by” dates, however, they must first determine if it is even possible to reliably predict spoilage based on microbial activity. Gafanha’s research sought to test and predict how different microbial communities caused ground beef to spoil over time.

All meat contains some bacteria when packaged. Those bacteria break down the tissue and produce chemical compounds that cause discoloration, off-smelling odors, gross flavors and a slimy texture. The purpose of Gafanha’s study was to identify changes in the microbial community composition over the shelf life of ground beef and associate those changes with key changes in the meat quality and indicators of spoilage.

Gafanha tracked the microbial communities over 14 days, regularly feeding data into a machine learning program that could both analyze thousands of data points and predict how those thousands of data points would change over time. Gafanha collected and analyzed data from three samples of ground beef while testing them against the machine learning program’s predications.

“It’s so complex that we had to use machine learning. The human mind can’t do it,” Gafanha said.

Using this method, the researchers were able to record how the microbial community changed over time. More than simply growing and metabolizing, the bacteria — Rhodobacteraceae, Enterobacterales, Pseudomonadaceae, Carnobacteriaceae and others — spread, retreated, migrated and competed for scarce resources inside the tiny confines of a half-pound package of ground beef.

“That little half pound of ground beef is the whole world to them,” Gafanha said.

Oxygen-loving bacteria multiplied first, depleting the packaged air. Then, anerobic bacteria surged in the oxygen-depleted environment. The microscopic give and take between bacteria in the microbial community is what makes predicting food spoilage so difficult.

However, Gafanha was able to determine that spoilage patterns could be predicted, demonstrating the concept can work.

”Trends followed spoilage to a T — exactly what we wanted to see,” she said.

The study was part of a larger initiative led by Aeriel Belk, assistant professor of animal sciences at Auburn University. The research is funded by a $10,000 grant from the Alabama Beef Checkoff program. The program assesses $1 on all cattle sold in Alabama and uses the proceeds to fund promotional, educational and research initiatives.

Belk plans to replicate the study with additional samples and sessions, as well as look at other markers of spoilage. In time, the research team’s model will be a useful tool that results in more accurate “sell by” dates and fewer pounds of meat unnecessarily discarded each year.

Also on this research project were Dianna Bourassa and Amit Morey of the Department of Poultry Science and Barney Wilborn of the Department of Animal Sciences.