Tracking Dinner

[dragon wench]

I just came across this...

Tracking Avian Flu and Mad Cow: Is RFID Bringing Safer Food to a Store Near You?
Date: Jun 23, 2006 By Laurie Rowell.
With threats of Mad Cow and bird flu rattling our food chain, can Radio Frequency Identification (RFID) ensure a meal without epidemiological regrets? With rather a lot riding on these little transceivers, Laurie Rowell posits that it might be wise to take a good, hard look at this "tech du jour" that's supposed to save our supper.

Outbreaks of animal-borne diseases like Mad Cow and avian flu are serious business. They can spread rapidly, causing the destruction of whole flocks or herds and threatening the food chain. Even when threats to humans are minimal, consumers shy away from purchasing the at-risk items. After all, nothing puts people off their food as quickly as the suggestion that the entrée might be tainted. That kind of news is a real plate-pusher.

In the hope of short-circuiting future epidemics, the United States Department of Agriculture (USDA) has come up with a plan for tracking all livestock. The hope is that a national registry would allow rapid identification of and notification about exposed animals so that a quick quarantine could save widespread trouble. One key technology in these tracking trials is Radio Frequency Identification (RFID).

So how well is RFID working for this purpose? And just how do you chip a chicken—without later having to spoon the RFID device out of your soup?
Just-in-Time Tech

The RFID tag—the part you attach to the bird or cow or any other item you mean to identify—is a transponder, a silicon chip that can be small enough to sit on the head of a pin. The antenna that it requires can be printed in special ink so that the whole thing is discreet enough to ignore on food packaging, library books, clothes, people, pets, or credit cards. What makes RFID tags so much better than the bar codes that they’re replacing is that RFID tags don’t require line-of-sight scanning; they just need to be close enough to be in range of an appropriate reading device. Pretty nifty if your range is a couple of meters and you want to be sure a shipment is complete without having to unload the truck to scan each bar code. That’s why several large companies, Wal-Mart and Target among them, insist that their suppliers ship with RFID tags.

RFID systems vary in function and design, but most applications use passive tags—that is, the tag itself has no internal battery. Identification using a typical RFID system of this type begins with a reading device that’s either embedded in a stationary object (say, the gate on a cattle pen) or handheld like a wand. An electromagnetic signal from the RFID reader activates any RFID tag within range, which in turn transmits its data to the reader. The reader decodes the data from the tag, sending it on to a computer running appropriate application software (middleware), which filters the data and passes it to enterprise applications.

Articles in InfoWorld, cite two key improvements in RFID technology within the past year. New software has made these systems more functional at the same time that the new Gen2 protocols for hardware, now the standard for Electronic Product Code (EPC) transmissions, have ensured wider interoperability. These new, better functioning Gen2 devices didn’t hit the market until 2005, so plenty of people were already implementing hardware using earlier protocols. This means that early adopters may have money invested in older technology.

In this atmosphere of change, the farm and ranching industry finds itself at sixes and sevens, scrambling to meet the demands of government-mandated animal tracking by 2009.
Government Hands

The USDA launched the National Animal Identification System (NAIS) in 2004 to devise a means for tracing farm animals in the United States and to monitor diseases in animal populations. The initiative is supposed to provide damage control with an eye to alerting producers, distributors, and retailers before animal carriers can spread disease. The goal is to have 48-hour trace-back ability by 2009, with a system that notifies owners when their animals have been in contact with suspected infections.

The NAIS plan involves three parts:
Registering the premises (farm, ranch, household, wholesale lot, etc.)
Registering the animal
Registering each event (animal transported to the county fair, animal sold, animal dead, etc.)

Obviously, this is an expensive operation. According to Dr. John Wiemers, APHIS Senior Staff Veterinarian at USDA, premises registration is currently ongoing in all fifty states, paid for by USDA. But for the other two pieces of the project—registering the animal and registering each event—there still seems to be some confusion. While the NAIS Q&A site says that costs will be shared (presumably by consumers, government, and industry), the site is vague about who will pay for how much of which activities.

Until 2009, all registration is voluntary, with individual animal ID being implemented currently through pilot projects. Meanwhile, small holders worry about how much equipment will be required and how much it will cost.
Assessing the Risks

How serious are the risks of avian influenza (commonly known as "avian flu" or just "bird flu") and bovine spongiform encephalopathy (which we call "Mad Cow")? Given that all of us will eventually succumb to something, the odds of it being either of these diseases at the moment would appear to be miniscule. In fact, when the risks are weighed against the costs and complications of tracking, you have to wonder if an enormous government-managed database that depends on reporting from the field about every animal movement is really the best way to handle the problem.

Let’s look at the stats.
The Chicken Stakes

At the time of this writing, the World Health Organization (WHO) cites more than 200 cases of humans who have contracted avian influenza (H5N1), resulting in more than 100 deaths in nine countries. Since it was identified in 2003, the disease has spread through the movement of domestic and wild birds far from the initial cases found in Vietnam and Thailand to an ever-wider range that includes Denmark, Sweden, Germany, and Scotland. While some experts have predicted the arrival of avian flu on U.S. shores within a year, Dr. Wiemers takes a cautious approach, claiming "no crystal ball" and maintaining that the USDA is "hoping for the best, preparing for the worst."

As epidemic threats to humans go, the risk has so far remained small, limited primarily to those who work in the poultry industry. According to the WHO site, at the time of this writing there have been no confirmed cases of human-to-human transmission of avian flu (which would signal the risk of worldwide pandemic).

The danger to birds, however, is dramatic, including about 150 million so far that have died or been killed in an effort to stem this scourge, but controlling the spread is expected to take years. For chickens, this strain of avian flu is lethal within 48 hours, suggesting that the 48-hour trace-back target proposed by NAIS might not be fast enough for this particular pathogen. Certainly by the time an individual farmer learns that some of the birds just purchased might be infected, all of them could have the disease, and many a hen might be lying toes-up.

But Wiemers insists that the 48-hour trace-back should be sufficient for damage control. In that time, the USDA should completely understand the spread of the infection, which animals were exposed, which premises were contaminated, which chickens are at risk. In fact, he sees poultry producers as responding far more quickly: "Most of them can give us the information within 48 minutes."

In this regard, RFID might just offer an additional answer. In December 2005, Digital Angel, a manufacturer of RFID tags for animals, proposed using a biothermal chip for diagnosis. Usually, before it’s clear that an animal is ill, someone handling the poultry has to see a chicken sneeze, notice a discolored beak, or—more likely—observe that one of the broilers has dropped over dead in the chicken coop. Clearly what’s needed is an early-warning device. The biothermal chip can monitor and transmit information about body temperature. If a threshold number of birds register a fever, it’s time to check for an epidemic. Keven McGrath, President of Digital Angel, has suggested tagging one chicken in every 250 with a biothermal device.

So far, though, no one seems to be slapping RFID tags on their layers for ID purposes. "In fact, there’s very little RFID being used with poultry," says Wiemers, adding that chickens are generally tracked through "visual identification" methods that "don’t employ automatic data capture." Pilot projects for chickens rely on labels with the ID number attached by leg bands, metal clips, glue, or those plastic T-fasteners that attach price tags to clothing.

While some independent producers with modest holdings may need to identify birds individually, commercial poultry birds might only be marked with a lot ID number if they’re raised and shipped together, and don’t ever commingle with other birds.

Is this sort of thing enough to safeguard our food supply? Maybe.

[dragon wench]

First, our food is not currently at risk from bird flu. According to the WHO FAQ on avian influenza, risks to food supply exist only in countries with a confirmed outbreak. For the rest of us (for now, at least), eating chicken is as safe as it ever was. Better yet, even where bird flu outbreaks have been confirmed, cooking destroys the virus, so cooked poultry and eggs are safe. Just remember that wherever there have been confirmed cases of avian flu, even the yolks of eggs must be fully cooked; so if the epidemic reaches us next year, as expected, don’t order your eggs "over easy."
The New Brand on Cattle

The risks to humans from bovine spongiform encephalopathy (BSE) would appear to be very tiny indeed. According to the Centers for Disease Control (CDC), the U.S. sees fewer than 300 confirmed cases per year of Creutzfeldt-Jakob Disease (CJD), the neurodegenerative disorder related to BSE that humans can contract. Moreover, virtually none of these are the variant Creutzfeldt-Jakob Disease (vCJE) that comes from eating BSE-infected beef. In fact, according to WHO, no cases of vCJE to date can be traced to eating beef in the U.S., the single case occurring here being attributable to that individual’s having "lived in the UK during the BSE epidemic."

So far, the Food and Drug Administration (FDA) and USDA cite three confirmed cases of BSE in the United States (December 2003 and June 2005, March 2006). All animals were apparently born before the 1997 regulations that prohibited the recycling of cattle parts into cattle feed, a practice that has been tied to the spread of the disease. Meat from both the 2005 and 2006 cases was stopped at the slaughterhouse because regulations prohibit distribution of meat from a "downer" cow (one that’s unable to walk).

The 2003 BSE case, the first in the US, was more problematic. While the FDA insists that the organs that carry the infectious proteins (called prions) were discarded at slaughter, the meat from that cow did enter the food chain. Ground beef from that cow was mixed with meat from several others and subsequently sold in stores in the state of Washington. While 10,000 pounds of beef were recalled, the recall came too late for some who took it home to cook tacos and spaghetti. Unfortunately, cooking doesn’t make beef safe from this risk.

NOTE

I’m not completely impartial on this subject. The ground beef entered the food chains in stores where I shop.

But, that said, risks should be negligible in cases where organ and nerve tissue have been discarded, as was done in the case cited above. Muscle tissue doesn’t carry the risk of infectivity, according to a WHO fact sheet on BSE—and, just for the record, neither does milk.

For those who are still concerned by the small percentage of cattle tested for BSE in the U.S., there’s always the option to limit consumption to organic beef, which has never carried the risk of recycled cattle feed.
Marking the Beasts

The NAIS initiative has sparked plenty of debate in the cattle industry. Charlie Gaskins, professor and animal scientist at Washington State University, says most beef producers see NAIS regulations as inevitable, although many are taking a wait-and-see attitude toward how things will shake out. "I think most producers, in fact maybe all, believe it’s going to happen sooner or later. The questions would be how it’s going to be implemented, what the exact regulations will be, and when. And of course there are arguments about who is going to pay for it."

Probably because the health risks from Mad Cow emerged first, much more has been done on the NAIS initiative to track cattle than the one to monitor chickens. From a technological standpoint, NAIS has been coy about how livestock tracking of any kind will be done. They don’t specify RFID as the technology for this project, claiming in their FAQ that "USDA is maintaining a technology-neutral position with regard to the technologies that will be used to identify animals."

While RFID isn’t the only player in the field, it’s an aggressive front-runner. Pilot projects for NAIS are currently in progress to track U.S. cattle destined for the plate, and the great majority are testing RFID tracking. Jack Field, executive director of the Washington Cattlemen’s Association, is involved with the seven-state Northwest Pilot Project. The goal, says Field, is "to test as many different ID systems as possible at the local and regional level to ensure they meet the USDA 48-hour standard for trace-back." These methods of ID include traditional branding, visual dangle tags that are hand-numbered or imprinted, retinal scanning, and RFID buttons as well as combinations of these methods.

The thumbnail-size RFID button is often combined with a tag, with the whole thing fastened to an animal’s ear. While there are other methods of securing RFID tags, including embedding under the skin, ear buttons seem to be favored by most people in the industry. While they might tear loose if the cow scratches hard enough against a post or tree, the ear buttons are at least big enough that they shouldn’t find their way into the food supply. The chips will transmit a unique 15-digit number for each animal that can be matched with a database file containing information on the animal’s owner, premises, and current location.

This kind of low-frequency ear tag is estimated by Field as costing $1.75–2.50, depending on quality and read-range, with higher-performance tags clustering in the neighborhood of $2.20–2.50. Right now, in the Northwest Pilot Project, the participating producers are footing the bill for the tags. "Nobody’s happy to hear this, but it looks like a cost that will be borne by the cow/calf producers," says Field.

Readers, whether secured to posts or handheld, run $650–1,000 each, depending on whether they’re can be secured to posts or handheld. Producers might only need to buy the tags, however, leaving the costs of scanners and computing equipment to those who manage the transport, feedlot, and processing of cattle.

Unfortunately, these particular passive RFID tags can’t be read from one to three feet away. That can be a disappointment if you’re hoping to hold up a reader or slap it into a gate and scan the entire herd as they hurtle past. This limitation is being addressed by some RFID manufacturers. Advanced ID, for example, is using ultra-high-frequency (UHF) tags to improve the range and accuracy of scanning to between one and three meters. These devices allow more data storage so that more information about the animal can be transmitted. Another company, ZigBeef, is developing active RFID tags for reading at still greater ranges.

Again, RFID tags are not the only technology that promises to meet NAIS standards for animal tracking. Companies such as Optibrand are promoting retinal scanning of cattle for identification (based on the unique patterns of blood vessels in the retina). While this technique solves problems such as cattle losing their ear tags, it also requires that the animal be restrained so the scanning device can be held about a finger’s width from its eye.

The current pilot projects requires only minimal data to be kept for each animal: date, animal ID, premise number, and event code. There are genuine privacy concerns in the cattle industry that are triggered by this tracking initiative, not just because many producers are keen on privacy as a principle, but because the tracking of information about a producer’s herds can offer just the kind of market edge to a competitor that no one likes to give away. In Washington state, representatives of the cattle industry recently were successful in convincing lawmakers to pass legislation ensuring that any information submitted by an individual will be exempt from public disclosure. How this situation will play out in other states is not yet clear.
All in the Packaging

The good news is that tracking was adequate to identify tainted beef for recall even back in 2003, when the first case of Mad Cow was identified. Better yet, one grocery store chain (Albertson’s) used customer discount cards to trace people who had purchased the beef and provide a courtesy alert to the recall. In other words, technology was able to draw a clear line of data to those who had purchased the dodgy product. Meanwhile, other stores in the area hesitated to use the same technology because of concerns about customer privacy.

This is in part because consumers have, from the first, raised justifiable questions about these kinds of technologies that allow retailers to target people more effectively for marketing—and pose some genuine security risks. Some of the fiercest arguments arise because items scanned by the RFID reader at the checkout can be linked to a customer discount card, and the information that was gathered when this card was issued. If you pay with a credit card, the potential exists for retailers to track items purchased in other stores.

The potential for use and abuse are high because anyone might have a reader. As critics point out, a bar code is visible and the information revealed by your credit or discount card at least requires you to swipe it. RFID, on the other hand, is frequently implemented without customer participation or recognition.

Yet even those familiar bar codes in your local grocery store allow a company to compile a lot of data about what you buy and link it to things like your address and telephone number. The one bit of good news about all this stored information is that if the chicken you bought on Saturday morning came from a facility where birds were found to have avian flu, at least the grocery store should be able to track the purchase, pull up your phone number, and give you a call before you cook Sunday dinner.