Fail: Hi, I’m from government, here to help; but shopper cards may actually help food safety efforts

If you have a warehouse membership card in your wallet or a supermarket shopper tag on your key chain, you might regard it as a good way to save money. But public health officials say it may be an even better way to save lives.

JoNel Aleccia of NBC News reports that more local health departments — along with state and federal investigators — are relying on the detailed 20100610_124707_bz10shopcard_200information about what went in consumers’ shopping carts to track down outbreaks of foodborne illness, experts say.

Identifying exactly which products were purchased by victims of food poisoning has become a standard tool for public health investigators, said officials with the Centers for Disease Control and Prevention.

“We are definitely supportive of the use of shopper cards during these outbreak investigations,” said Casey Barton Behravesh, deputy chief of the CDC’s outbreak and prevention branch of the division of foodborne, waterborne and environmental diseases. “The product, the flavor, the lot code, the best by date: That is all tracked with these purchases.”

Store cards are a rich trove for epidemiologists, who often are trying to track down suspect food a month or so after it was consumed because of the lag between when an illness strikes and when it gets reported, said Bill Keene, a senior epidemiologist with the Oregon Public Health Division. His state has been a leader in using shopper card data, along with Minnesota, but others are joining in, Keene said. 

“We rely on people’s memories, which are quite fallible, and on our interviews, which are quite fallible,” Keene told NBC News. “Shopper club cards are a good source of finding out what people ate.”

Costco has been notifying consumers about food and other products recalls for safety reasons since the late 1990s, said Craig Wilson, the company’s vice president for food safety and quality assurance. But now, they’re being called on by public health officials at every level.

“It happens a couple times a week,” said Wilson. “It’s getting to be more of a norm.”

But it’s not always easy, Keene says. Stores provide data only with the permission — usually written consent — of the consumer and a verified shopper card or membership number. And disclosure rules vary from state to state, making some information more difficult to obtain.

“We won’t just release data,” said Wilson.

Health officials like Keene say they safeguard the data carefully and use it only as a tool to keep more people from getting sick.

“We are the government, but we aren’t that part of the government,” he said. “We’re the good guys.”

337 sick from salmonella in sausage, France, 2011

Those supermarket loyalty cards helped pin down an outbreak of salmonella in sausage in France last year.

Researchers reported in Eurosurveillance last week that an outbreak of the monophasic variant of Salmonella enterica serotype 4,[5],12:i:- occurred in November and December 2011 in France. Epidemiological investigation and food investigation with the help of supermarket loyalty cards suggested dried pork sausage from one producer as the most likely source of the outbreak. Despite the absence of positive food samples, control measures including withdrawal and recall were implemented.

Between 31 October and 18 December (week 44 to week 50), a total of 337 cases of Salmonella enterica serotype 4,[5],12:i:- were identified. The median age was 10 years (range: 0–90 years) with about 30% of children under five. A majority of women were affected (female to male sex ratio: 1.22). Cases were reported throughout France.

An epidemic of Salmonella enterica 4,[5],12:i:- was already observed about three months prior to this outbreak. Between 1 August and 9 October, 682 cases were reported (Figure 1), of whom 100 cases were interviewed at the time but no common vehicle of infection could be identified. In comparison, 212 cases with this serotype had been isolated during the same period in 2010.

Epidemiological investigations pointed to a dried pork sausage purchased principally at supermarket chain A and consumed after week 44, 2011. Therefore purchases of pork delicatessen at supermarkets A and B up to four weeks prior to symptom onset were investigated by the DGAL using data recorded through supermarket loyalty cards.

The use of the loyalty card from supermarket chain A was important to identify the vehicle of infection and the local producer involved in this outbreak. These cards are used more and more and prove helpful in the investigation of food-related outbreaks. Nevertheless we should keep in mind that they do not necessarily reflect the consumption of cases perfectly. For instance, the card may not be used systematically, the household can purchase foods in additional shops and markets for which they have no loyalty cards, many food products are consumed outside the household and not recorded on the card, and the central database of the supermarket does not always contain data on all foods sold such as foods directly purchased by the retailers. For these reasons the data have to be interpreted together with the results from epidemiological and microbiological investigations.

That the producer and microbiological analysis did not find Salmonella does not exclude contamination. The limited number of samples and the processing of the food (especially salting and drying) reduce the likelihood of isolating the bacteria. Implementing checks earlier in the process (before salting and drying) and using additional methods of testing such as polymerase chain reaction (PCR) should be considered.

This is the second described outbreak in France involving dried pork sausage, and indicates that this food item might be a likely vehicle of infection and further outbreaks in humans may be expected.

Given the limitations to detect Salmonella in dried sausages, the ability of the standard reference method to detect of monophasic variant strains in dried sausages is questionable. Additional methods should be explored in order to improve monitoring protocols.

The complete report is available at

‘I guess I did eat cantaloupe’ Grocery loyalty cards help trace foodborne illness source

Elizabeth Weise writes in USA Today an outbreak of salmonella in five Eastern states has sickened 42 people so far this year, with two hospitalizations. Dozens more might have been struck down were it not for a strikingly successful new tool used by public health officials to quickly figure out what was making all those people sick: the lowly shopper-loyalty card.

Food safety officials are increasingly finding value in plumbing shoppers’ food buying habits through these loyalty cards when they’re faced with foodborne illness outbreaks across communities and even states that seem to have no obvious links.

"It’s very helpful because it’s very hard for people to remember what they ate a couple of days before, not to mention a couple of weeks ago," says Casey Barton Behravesh, with the Centers for Disease Control and Prevention.

Supermarket loyalty-card programs were introduced in 1987. By the 1990s, they were widely used. In return for discounts on some items, they allow companies to track shopping habits. For epidemiologists, who study disease outbreaks, they’re a complete record of everything shoppers bought at the store going back for years.

Privacy is a huge concern in using cards to track food-borne illness outbreaks, officials say. All health departments are required to get permission to use them, Hammond says. "This is voluntary: People are not required to consent to having the grocery chain release their shopper-card history," he says.

In the salmonella outbreak among Eastern states, New York state and local health officials noticed an increased number of salmonella cases and started conducting routine interviews.

When they realized that all the patients shopped at Wegmans, a local supermarket chain, it was a "red flag," Behravesh says. Given permission by patients to check their shopper club card data, officials found "a lot of these people were buying bulk Turkish pine nuts," or foods that contained them, Behravesh says.

Other recent cases include:
•An outbreak of E. coli O157:H7 that sickened 33 people and led to 15 hospitalizations in five Western states in 2010 was quickly traced to raw milk Gouda cheese produced by Bravo Farms in Traver, Calif., using Costco purchasing data.

•A puzzling outbreak of salmonella Montevideo that sickened 272 people in 44 states in 2009 was finally cracked when health officials examined shopper records from Costco and saw that almost everyone who had gotten sick had purchased salami from Daniele Inc. Testing showed it was not the sausage but the black and red pepper it was coated in that carried the bacteria.

The shopper loyalty cards also can help public health workers when consumers misremember what they ate. "One person swore she didn’t eat cantaloupe, she only ate honeydew melons," says William Keene, a senior epidemiologist with Oregon Public Health Services. "When we pulled her records, we found that she only bought cantaloupe, not honeydew. When we showed her that, she said ‘Oh, I guess I did eat cantaloupe.’ "