It’s All Antibiotic Free, Baby!

After all of the recent Panera and Chipotle hype about antibiotic free production, I decided to look at the data. This is also a follow up to my previous blog about antibiotic free (ABF) meat; I am going to present some data to back up my claim that there is very little difference between conventional and ABF – in other words, it’s all antibiotic free, baby! #ItsAllABF!

Due to farmers following appropriate withdrawal times, there are very few violations. In fact in the last three years of USDA testing no broiler chickens have been found with violative residues for the scheduled (random) sampling. For beef only 2 violations out of 1,600 samples were found and only 3 out of 2,200 from market hogs.  Note that antibiotics are not toxins, there are useful and very safe products used by us all.

The Bottom Line

The residue detection levels in the 3 classifications that I analyzed (beef cattle, market hogs, and broilers) are extremely small and well below the levels that would cause adverse effects to a human eating the meat. In addition, if an animal tests positive for residues, it does not enter the food supply.

Meat from an ABF farm would supposedly have zero levels of residues – but, if you aren’t going to get sick or be affected by the perfectly healthy, wholesome conventional meat, why should you pay more for something that potentially carries more foodborne illness?

From a veterinary perspective, I am concerned with the internal struggle that the ABF farmer must face. Most farmers get some premium for raising ABF meat, so if the animals get sick does the farmer treat and lose the financial benefits of ABF or wait a day or two? Waiting can increase mortality and spread of infectious disease significantly. What about the veterinarian, who has taken an oath to prevent animal suffering, but management will only let him treat a small percentage of the barns? Can these restaurateurs really argue their ABF meat provides a better “conscience choice,” if it comes at the cost of additional mortality and animal suffering?

For further clarification of the terms used above, here are some explanations:

The Food Safety Inspection Service (FSIS), the inspection arm of the USDA, does sampling for a variety of residues in meat through the National Residue Program (NRP). They sample for many types of residues, including antibiotics as well as pesticides and environmental toxins. I will be discussing the combined NRP data from 2009, 2010, and 2011.

The Basic Definitions

Residue: A residue indicates that traces of a substance are present in meat. Residue does not mean that the substance is harmful, and it can be one of many classes of compounds. For antibiotics in particular, if a residue is present, it is likely due to the producer not waiting long enough after the animal was given antibiotics to send it to market (not following proper withdrawal guidelines).

Resistance: The presence of a residue in meat does not indicate antibacterial resistance. The two are separate issues. If resistance is detected, this means that there are bacteria on the meat that have tested resistant to one or more antibiotics. Resistance is measured and reported through the National Antimicrobial Resistance Monitoring System (NARMS). If resistance is detected, that does not mean there are residues; likewise, if a residue is found, that does not mean that there are resistant bacteria to that antibiotic.

Violative vs. Non-Violative Residues

Below, you will find 2 charts that list the percentage of both violative and non-violative residues in beef cows, market hogs, and broilers. There are 3 main levels for regulating residues in meat. The first is called acceptable daily intake, which is the highest level of a residue at which it is safe to consume the meat.  The second is called acceptable tolerance, which is the level at which a residue becomes “violative” if exceeded. The third is the detectable limit, which is significantly less than the acceptable tolerance and is the lowest level that the lab test can detect.

For example, here are the levels for tetracycline:

  • Acceptable daily intake: 25 micrograms/kg of body weight
  • Acceptable tolerance level in meat: 2 parts per million
  • Detectable level: 0.05 parts per million

A 165 lb person could consume up to 33 ounces of steak that contains the maximum allowable tetracycline residue (2 ppm) per day for the rest of their life and not feel any adverse effects. This is virtually impossible, because less than 1% of beef tested even comes close to this violation level.

In order for a meat sample to be called violative, it must test higher at levels than acceptable tolerance – it is essential to note, here, that even though this residue may be considered violative, it is still significantly below the level at which a person may begin to feel adverse effects.  A meat sample that is considered non-violative is one which tests positive for residues but is lower than the acceptable tolerance. Even though a residue may be detected, that by no means indicates that it will transfer to sickness or increased resistance in people. There are many factors at play here and the FDA has done extensive testing in order to clarify and regulate what is a safe food supply.

Testing Methods

Both of the charts below list violative and non-violative residues, but each represents a different sampling method. Table 1 represents a method called scheduled sampling, in which random meat samples are taken and tested for residues. This is an unbiased approach and is generally considered representative of the greater population. Table 2 represents another method called targeted sampling, in which only animals that appeared sick or abnormal are tested. This method is more biased, and generally more likely to have positive results because the animals are displaying abnormalities to begin with.

Table 1: Scheduled Sampling Antibiotic Residue Percentages by Classification

Antibiotic residues from scheduled sampling 2009-2011*

Production Class

Total sampled

Non-violative positives

% Non-violative positives


% Violations

Beef Cows






Market Hogs












Table 2: Targeted Sampling Antibiotic Residue Percentages by Classification

Antibiotic residues from targeted sampling 2009-2011*

Production Class

Total Sampled

Non-violative positives

% Non-violative positives


% Violations

Beef Cows






Market Hogs






*Includes data from the following compound classes: Antibiotics, Chloramphenicol, Florfenicol, Nitrofurans, Nitroimidazole, Sulfonamides

How many antibiotics are actually used in livestock?

In our last post entitled Antibiotic Sales Data Released by FDA, I failed to answer a simple question: How many antibiotics are actually used in livestock? The amount of antibiotics sold  is easy to find if you read the FDA report and add up the columns, but who wants to read government reports?   The short answer, to my original question how many antimicrobials are used, cannot be easily answered. FDA does provide a report on how much was sold, but we cannot tell how much was used and exactly for what it was used. We don’t know how much is still sitting on a shelf. Therefore, looking at year to year variations can be risky.

FDA told us that about 13 million kg of antimicrobials were sold; however, since the concern is really about antimicrobials which might also be used in humans, we must deduct the 4.1 mil kg which were ionophores (ionophores are not used in humans). Therefore, the total amount of “concern” is 9.4 million kg.

Is that a lot? Is that too much?  It is only too much if it is “excessive,” which can only be defined as those sales which are not needed.  Farmers and veterinarians agree that antimicrobial abuse and excessive use is wrong and not prudent, but if animals need these medicines to provide a healthy food supply it is not too much.

Antibiotic Sales Data Released by FDA

Welcome to our new blog site.  It would be nice to say “Hi” and talk about friendly stuff, but critical animal health food safety issues never seem to stop pouring through the media.

Today FDA released some valuable reports about antimicrobial sales in livestock. The links from FDA noted below explain from where the data are derived and offer some good thoughts about interpretation.  I will highlight a few points.

1. There appears to have been an increase  in antibiotic sales from 2010 to 2011. If you remove ionophores, which are never used in humans, nor are their cousins,there was actually 0.01 percent decrease.  Additionally, if you compare those data to meat production which increased by at least 0.2%, producers have apparently reduced antimicrobial use quite a bit.

FDA notes this is only the 3rd year these data have been reported, so don’t get too excited about slight changes in the volumes.  They say it more professionally of course (see below).  Note also that since these are sales data it could easily reflect changes in inventory on hand.  NOTE, they do not reflect actual usage in animals!!

“It is not unexpected to see transient increases or decreases in sales of certain drugs in a given year due to various factors (e.g., occurrence of disease outbreaks, fluctuations in animal populations). Therefore, given that only three years of sales data have been compiled so far, it is difficult at this time to draw conclusions about whether any changes in sales data represent meaningful trends in overall volume of sales or the resulting effect on antimicrobial resistance.”

2. FDA is very clear these data cannot be compared to other sources on animal and human use. Some folks try to make those comparisons. FDA says, “Don’t go there!” – again, more professionally of course.

“The Section 105 summary report is not directly comparable to other previously reported estimates of antimicrobial drug sales or distribution. The Section 105 summary report includes data on the quantity of antimicrobial drugs sold or distributed for use in all food-producing animal species for all purposes, both therapeutic and non-therapeutic, and includes all dosage forms. Previously reported estimates included different categories of antimicrobials and used different sources of information and thus cannot be directly compared to the Section 105 summary report.”

How do the data collected for animal use compare to the amount of antimicrobial drugs used in humans?

“It is very difficult to compare the animal and human reports in any meaningful way. Before making comparisons between human and animal sales and distribution data, there are a number of differences in the circumstances of use of antibacterial drugs in human and veterinary medicine that must be carefully considered.”

3. FDA mentions these data are valuable for evaluation resistance risk, etc:

“Antimicrobial drug data collection helps FDA evaluate antimicrobial resistance trends. The data also help the agency in its analyses of other issues that may arise relating to the safety and effectiveness of antimicrobial drugs approved for use in food-producing animals.”

However, as we point out in our recent Food and Drug Law Institute (FDLI) paper aimed toward the lawyers, data on usage is only marginally informative. Some comparison or outcome measure is also needed: “conclusions based on one half of the equation (sales or usage data only) cannot be substantiated.” In other words, if on-farm antimicrobial use does not increase public health harm due to treatment failures from resistant infections, then sales data are not that interesting.  It is true that treatment failures occur in human hospitals, but those are usually from bacteria and resistance genomes not closely related to the farm, e.g staphylococcus and enterococcus, not Salmonella and Campylobacter.  Therefore, the removal of on-farm antimicrobials may make little to no improvement in public health, as discovered in Denmark. More on that later.

For more information, see: