A week before harvest started, a private auditor gave a Colorado farm 96 points out of 100. Two months later, cantaloupes from that farm had killed 33 people.
The 2011 listeriosis outbreak took more lives than any other American foodborne outbreak of the previous two decades. It infected 147 people in 28 states, and 143 of the 145 patients with available records were hospitalized.
The part worth sitting with is that the auditor saw some of what later got listed as causes, wrote it down, and still gave the farm a 96.
What happened in the fall of 2011
On September 2, the Colorado Department of Public Health and Environment told the CDC about seven listeriosis cases reported since August 28. For scale, Colorado averages two cases in an entire August. By September 6, all seven interviewed patients had eaten cantaloupe, and three of them named melons sold as Rocky Ford.
On September 14, Jensen Farms announced a voluntary recall of its whole cantaloupes. Processors who had cut those melons into fruit trays followed with recalls of their own.
The CDC’s final count: 147 people in 28 states infected with one of five outbreak-associated subtypes of Listeria, with Colorado hardest hit at 40 cases. Illness onsets ran from July 31 through October 27. The median age was 78, and 143 of 145 patients were hospitalized. Of the 144 who could recall what they had eaten, 134 had eaten cantaloupe in the month before falling ill. Seven illnesses were pregnancy-related, three in newborns and four in pregnant women, and one pregnancy ended in miscarriage.
Thirty-three people died. Ten additional deaths among people infected with an outbreak subtype were reviewed against death certificates and not attributed to listeriosis. In December 2011 a fifth subtype turned up in a cantaloupe sample collected during the investigation, matched a patient already on the books, and pushed the final tally from 146 cases to 147.
In the outbreak review published in the New England Journal of Medicine, 127 of the 147 patients, or 86 percent, were 60 or older. Listeria hits a narrow group hard, which is how 147 cases produced 33 deaths.
What investigators found in the packing shed
On September 22 and 23, an FDA team working with Colorado state agencies ran an environmental assessment at the farm. Of ten cantaloupes taken from the walk-in cooler, five carried Listeria matching the outbreak. Of 39 environmental swabs from the processing equipment and the area around it, 13 came back positive, carrying three different outbreak patterns. The inspection was classified OAI because of the number of positive samples.
What follows reads better as a list of decisions than as a list of defects.
In May 2011 the brothers replaced their washing system. The line they bought was used, and it had been built to wash potatoes. It included a catch pan that a chlorine spray could feed into. The spray was never turned on.
The rollers on the drying equipment attached to their bars in a way that left deep crevices at every joint. Equipment like that cannot be cleaned: a brush does not reach in there, and Listeria does. How bacteria anchor themselves in places like that, and why water does not lift them out, is covered in our piece on biofilms on produce.
A condensate drain line from the old cooler emptied onto unsealed concrete next to the processing line and ran into a slit trench cut directly under the conveyor, where water pooled. The FDA treats condensate drain lines as a known harborage for Listeria.
The melons went into cold storage without pre-cooling. Field heat stayed in the fruit, condensation formed on the rind, and the cooler became a place where Listeria grows rather than a place where melons keep.
A truck used to haul culled cantaloupes to a cattle operation was parked right next to the packing facility.
The crop itself contributed too. A cantaloupe’s netted rind is full of crevices, and in its final growth phase the fruit sits in close contact with the soil. What a shopper can and cannot do about that at home is covered separately in our piece on washing melons, bananas and avocados.
The 96 percent audit
The packing shed was audited in July 2011, a week before harvest began, by Bio Food Safety, a subcontractor working under Primus Labs. The auditor, James DiIorio, scored the operation at 96 out of 100.
That same report lists major deficiencies, among them no hot water at the hand washing stations. It also records, as a separate line, that the farm was using no antimicrobial in its cantaloupe wash water. A year earlier, in August 2010, the farm had scored 95 and earned a “superior” rating, and chlorine was still in the wash water then.
A House Energy and Commerce Committee investigation released on January 10, 2012, added two details. The decision to buy that used equipment was based in part on advice from the private auditor. And a food safety representative from Frontera Produce, the farm’s distributor, came out to help prepare for the audit and was present throughout it.
The auditors’ explanations to the committee are worth quoting accurately. Robert Stovicek, president of Primus Labs, said his firm’s audits assess whether a client’s operations comply with current baseline industry standards, and that pushing clients toward best practices is not what the firm sets out to do. Jerry Walzel of Bio Food Safety put the distinction between a regulation and a guidance this way: guidelines are opinions, regulations are law.
Here is the hinge. At the time there were no FDA regulations covering cantaloupe processing at all, only draft guidance issued in July 2009. Guidance stayed out of the score because it is the agency’s opinion rather than law. So a farm could score 96 while running a potato line with no chlorine in it.
The criminal case and the sentence
On September 24, 2013, prosecutors filed a six-count Information, and on September 26 Eric and Ryan Jensen turned themselves in to US Marshals. The charge was introducing adulterated food into interstate commerce, one count for each of six shipments of Listeria-contaminated cantaloupe.
The brothers pleaded guilty to all six counts. On January 28, 2014, the court sentenced each of them to five years of probation with the first six months in home detention, 100 hours of community service, and $150,000 in restitution, $25,000 per count, payable to the victims.
There was no prison time, and probation was what the government itself recommended. The US Attorney tied that to the brothers’ cooperation, including their willingness to meet with Congress and to be confronted by the victims. For contrast, the owner of a peanut company convicted on related conduct drew 336 months, and the difference between those outcomes turns on proven intent rather than on the body count. Three American cases that make the mechanism visible are in our piece on executives who went to prison over food safety.
What the rules changed afterward
On November 27, 2015, the FDA issued the produce safety rule, 21 CFR 112. It applies to farms selling more than $25,000 of produce a year on average, and it put requirements on the exact things Jensen Farms had done in reverse. That rule grew out of FSMA, the law signed in 2011, and what FSMA changed and what it did not is covered separately.
- Equipment and tools must be of a design, construction and workmanship that lets them be adequately cleaned and maintained (112.123).
- That covers grading belts, sizing and palletizing equipment, cooling equipment including hydrocoolers, dump tanks and flumes (112.121).
- Equipment used to transport produce must be adequately clean before use (112.125).
- Water that touches produce during or after harvest, including wash and cooling water, must have no detectable generic E. coli in 100 mL, and untreated surface water cannot be used for it (112.44).
- If water is treated, the treatment method has to be effective and delivered so the water stays fit for use consistently (112.46).
What the rule left alone deserves naming too. It sets no requirement for an antimicrobial in cantaloupe wash water; the requirement runs through water quality instead. And the FSMA accreditation program for third-party auditors covers imports, so domestic American audits remain a private contract between a farm and its buyer. The underlying logic, where you identify the hazard, put a control on it and prove the control works, is laid out in our HACCP guide for microgreens.
What a grower should take from this
First, and least fair: a whole region pays for one farm. The Jensens grew melons 90 miles from Rocky Ford and sold them under that region’s name, so warnings about Rocky Ford cantaloupe landed on every neighbor. The following season, roughly a third of the previous year’s acreage went back into melons, and growers with decades behind them left the crop. The ones who stayed trademarked the Rocky Ford name themselves, spent $800,000 on upgrades, hired a full-time food safety manager, built a shared packing shed with washing, cooling and traceability codes, and moved seasonal pickers from piece rate to hourly pay.
Second: equipment you cannot clean eats the money you saved buying it. A used potato line cost less than a new one, and the story ended in bankruptcy and lawsuits from the people it harmed.
Third: water with no sanitizer in a recirculating system spreads contamination through the whole lot. The chlorine spray on this line was designed in and never once switched on.
Fourth: an audit is a snapshot, taken on one date, against the standard someone ordered. Ninety-six points meant compliance with baseline industry standards, and that is precisely what the auditor checked. If you are buying an audit for a customer, the question to ask the auditor is which standard the score runs against and what sits outside it. The rest of what safety looks like from the grower’s side is collected in our food safety hub.
