Ogden Mills Phipps:
Thank you, Denis. Denis has shown real leadership in Ireland, and we appreciate him coming here today.
We have heard from representatives McKinsey & Company at the Round Table on several occasions. In 1991, they recommended steps needed to build a world-class drug detection system. In 2011, they delivered a set of recommendations to build sustainable growth in our industry.
Several weeks ago, The Jockey Club reached out to them again to conduct a study of current practices in state‑based medication regulation testing the enforcement of how to compare to best practices internationally and in other sports.
Dan Singer has worked with major sports leagues around the world. He oversaw the 2011 study we commissioned. I can tell you that he and his team have a deep understanding of our business, and he’s here today with some eye‑opening research findings about drug testing and enforcement. Dan…
Our topic today is raising the standard of testing and enforcement in U.S. Thoroughbred racing. As we discussed in our 2011 report here in this room to the Round Table, raising the standards of equine medication regulation is a critical element of the growth strategy for Thoroughbred racing. It removes a barrier to new fan development, it addresses a significant better concern, and it protects the value of US bloodstock internationally.
The goals of equine medication testing and enforcement are clear.
First, to preserve the integrity of competition.
Second, to provide a level playing field.
Third, to promote animal welfare, and finally to protect the rights of owners, trainers, and racing fans.
What is less obvious is what the standard should be since each racing jurisdiction acts independently.
Now earlier this year, The Jockey Club asked McKinsey to study the practices in equine medication regulation in U.S. racing, both state by state and internationally, to identify the best practices, and the gaps of best practice in order to give racing a road map to a uniform high standard.
We also compared racing practices to anti‑doping practices in various human sports. This presentation that I’m about to show you summarizes our findings and recommendations on how racing can achieve that uniform high standard.
Now we were very fortunate to have as sources the leading experts on equine medication, both the science and the medicine, and the regulation. We had 12 state commissioned executive directors and Ed Martin of RCI.
We had five science advisors, including Drs. Rick Sams, Scott Stanley, Dionne Benson, Mary Scollay and Rick Arthur. We had international lab directors from Hong Kong, France, the UK and Australia. Eighteen states responded to our request for data. We also analyzed a ruling database of 25,000 medication rulings from 2001 to 2014.
We’re extremely grateful to these experts for their timing and guidance. Through their work through RMTC, TOBA, and other bodies, racing has already made very significant progress in medication regulation and testing.
Finally, we’d like to acknowledge Travis Tygart and his team at USADA.
Now, one caveat before we go into the findings. We did not ask these advisors to endorse our findings or recommendations. Those are ours. We did rely on them for data and perspectives on the underlying science of medication, and to the extent that we are making recommendations that are in any way controversial, those are ours.
Don’t expect that we had endorsements from every single source that we had.
We looked across collection testing investigation, enforcement, research and other equine medication activities; there is no single source for what racing is spending today, but based on all the research we did we estimate that racing across all breeds is currently spending a total of $44 million on medication regulation.
That equates to $60 approximately per start. This total is 3% less than the industry spent in 1991, but there has been a 50% drop in Thoroughbred starts since that time.
So that would imply a much higher cost per start. Actually, that cost per start has pretty much gone up in line with inflation, in real terms racing is spending about the same per start as it spent in 1991.
Let’s start with sample collection, which is the first area of practices that we looked at. The primary difference between the U.S. and international racing jurisdictions -- and also compared to human sports -- is an out‑of‑competition testing. Post-race collection practices are largely consistent across the U.S. and with international racing jurisdictions, but the level of out‑of‑competition testing is much lower in the U.S. as we’ll see in a minute.
There are three reasons why out‑of‑competition testing is a powerful tool for medication oversight.
The first is detection of performance‑enhancing substances that can be used in training but leave the horses system by race day. Examples might be EPO or corticosteroids. With the risk of detection, out‑of‑competition testing acts as a deterrent when trainers and vets know that horses can be tested anytime it deters them from using medications in excess or prohibited substances because they know they can be detected out‑of‑competition.
And finally out‑of‑competition testing generates longitudinal data on horses in training, which allows measurement of the level allowed medications versus what is prescribed for the particular horse.
Let’s look at the differences in out‑of‑competition testing frequency between the U.S. and other sports and jurisdictions here. What you can see is in Hong Kong 11% of the out of the tests for equine medication are taken out‑of‑competition.
In France, that number is 10%; in Victoria, Australia, it’s 21%; in the U.S., it’s only 1%, and we’ll go into that in a second.
We also looked at other sports. For testing for the Olympics in Sochi, 52% of the samples were taken out‑of‑competition. In swimming it’s 59%; in baseball, it’s 25% and in cycling 63%. USADA collects 67%, takes two‑thirds out‑of‑competition, and WADA, which is the world anti-doping agency, takes 50%.
In almost every case more than half the samples are taken out‑of‑competition.
A few states have started collecting out‑of‑competition samples in recent years but the total is only 1% in the U.S. so far.
New York, California, Indiana, Delaware, Kentucky and New Jersey are all doing out‑of‑competition sampling for racing. They recognize the value of out‑of‑competition testing, but face a number of obstacles to reaching the levels these other countries have achieved for racing. Let’s see what those are.
The first is that many states need to update their regulations to give their racing commissions the authority to test for the full range of regulated substances out‑of‑competition.
Secondly, racing needs a system for tracking the location of each Thoroughbred in training. Many other sports require athletes to report their location to an online database so they can be tested on a short notice. For example, a skater has to report her location at all times and could be tested on an hour’s notice anywhere in the world. That is common in human sports.
Racing already tracks the location of each recorded workout, so it’s really just a short step to a common tracking database for Thoroughbred racing.
Racing also needs reciprocity agreements between the states to facilitate out‑of‑competition sample collection for horses training in a different state from where they race. For example, if a horse is going to race in Kentucky but is training in Ohio, we need a reciprocity agreement that would allow Ohio or require Ohio to collect on behalf of Kentucky and vice versa.
Racing needs to set a standard for the frequency of out‑of‑competition sample collection. You saw the frequencies in other international jurisdictions, we’d suggest at least 10% be done out‑of‑competition. It could be more, but that would bring us in line with international best practices.
Now to do that, racing will need additional collectors to manage the increase in out‑of‑competition samples. Since it’s more time consuming to collect a sample out‑of‑competition or go to where the horse is training or presented than to do it at the track. That would require more labor, but it’s not that significant of a cost, and we’ve projected that cost.
We’d recommend that urine and hair samples be collected out‑of‑competition in addition to blood samples. In most cases, where U.S. states are collecting samples out‑of‑competition, they’re taking only blood.
For at least a portion of out‑of‑competition tests, it would make sense to take paired samples. It’s more expensive, but would allow for many greater detection and deterrence if we took urine and hair as well.
Finally, we’ll need additional research for effective testing methods for out‑of‑competition samples. For example, to identify gene doping medications or protein sources in blood we’ll need additional research funding.
Now let’s talk about testing practices. You’ll see here that we found a really large variance in testing costs across the states. Here you’re going to see the cost per paired sample that state racing commissions are paying to 11 different U.S. labs…And the cost for paired sample varies from the low of $55 all the way up to $230, depending on the state and depending on the lab.
You could ask yourself, why should the cost be so different? This is, by the way, a fundamental importance: that we detect the therapeutic medications and prohibited substances in the samples. This is where the rubber hits the road, so to speak.
So why should the cost be so different? Part of the answer is we believe differences in testing methods. For example, one acceptable method is to use ELISA kits, the other is to use mass spectrometry for screening, and there is a big difference in the efficiency and to some degree the sensitivity of those methods.
Some are appropriate for some and others for others...There can be differences from state to state, which drugs are required for screening, and we think that is a major difference in the cost you’re seeing there. In the chart there in France and Melbourne, the cost to prepare a sample is significantly higher.
There are a number of reasons for that difference. That is a really big difference. Part of it is cost factor, and we believe labor costs in those countries could be higher and that could explain some of the variance. And also those labs include research in costs per test, which most U.S. labs do a much, much lower extent. Another driver that is critical is the number of drugs tested in each screen.
So in the U.S. in those numbers, they could be as few as 15 drugs tested, it could be as high as 1,800, and in both of those international samples, it’s over 2,000 drugs per screen.
We think the main reason for the differences in methods and costs across U.S. labs is the way the states procure testing services.
This is a really important issue in driving these differences. We looked at requests for proposals from a number of different states. The request for proposals that go to the labs, the labs respond in order to get the contracts for testing. Some of those were excellent, but many of them were completely silent on what they require from the labs, and one or more of these four areas.
The first is a specific drug to test in each sample. The second is the threshold levels that would constitute a positive. Third, the testing methods, so, for example, the instrumentation that the state requires and finally accreditation of the lab.
Racing has made progress through the RMTC on accreditation through the years. They use a process that is based on the WADA process. RMTC accreditation ensures that labs have the proper equipment and practices in place to perform tests at high standards, but in practice the only way to verify that the labs are using those through practices is through double blind efficiency testing, and we aren’t there yet.
Labs can currently differentiate the proficiency test [PT] samples from other samples based on the packaging of the sample, based on the substances within the sample, and sometimes even by word of mouth.
There is currently no certified source for double‑blind PT samples for equine testing, and that is key to best practice in lab accreditation.
Overall, in all the areas we talked about, we need a uniform rigorous set of specifications from the states for the test labs.
Now, let’s turn to regulations. We see significant variation in medication regulations across the states. Currently, there are nine states that have fully adopted phase one of the national uniform medication program which establishes threshold levels for 26 therapeutic medications. The NUMP has passed commission or in the adoption process in is it more states and in nine states they’re either under discussion or not committed yet to the NUMP.
To show you how different the thresholds can be, here’s a comparison for four different drugs between the NUMP standard which is in green there, that is the thresholds in blood of per grams per milliliter, and state one, which we’re disguising the name of the state, but it’s one that’s either passed commission or is in the adoption process, and state two is one of the nine where it’s either under discussion or noncommitted.
Take, for example, Dantrolene, which is a muscle relaxant. The level in state two for a positive is 1000 times higher than the NUMP standard. For Dexamethasone, which is a corticosteroid, state 1 is 200 times the NUMP level, and state 2 is 300 times the NUMP standard. That is just an extremely high difference in thresholds, which we need to get to a common standard.
We also see a wide variation across the states and enforcement outcomes. In 2013, the number of medication related rulings ranged from a high of 74 per 10,0000 starts, down to a low of one medication related ruling for 10,000 starts. This is across 19 states that we analyzed.
It is difficult to say how much of a variance in the number of rulings per 10,000 starts is because of differences in rules across the states, difference in testing procedures across the states or difference in investigation enforcement or whether in the end it’s a difference of the underlying level of medication misuse.
But we can certainly say that with uniform standards across the states, we would not see a 74:1 ratio from the highest to lowest persist over time.
Penalties for medication violations are also inconsistent across the states. This is an example of Clenbuterol. The average suspension of a trainer for a Clenbuterol violation ranged from a low of 0 to a high of 83 days across the states we studied here. The average fine ranged from $500 to $4,421 across the same states. So the penalties for a violation can be radically different from state to state.
The final area we’re going to talk about is research. There is less funding for research on racing medication and testing in the U.S. than in the international jurisdictions that we studied, and the research is less tightly coordinated.
There are only a few states that provide equine research funding to labs. The RMTC had $200,000 in funding for research last year, and there are other pockets of research fund page. But when you consider the cost to develop a new test, the research to develop a new test can be anywhere from $100,000 to $200,000, and as we increase out‑of‑competition testing, we’ll need to be able to identify more substances in blood than we can today and increase our testing proficiency and accuracy in many other ways.
Our conclusion is that racing will need to increase research funding and take more advantage of research done in other racing jurisdictions and other sports.
Here’s the benefits of greater coordination of research in our view. First, we could have a coordinated research agenda which would direct funding to the highest priority initiatives such as tests for gene‑doping agents and protein‑based substances.
We could have a uniform national highly expert scientific review board to review and evaluate grants and research proposals.
We could aggregate financial resources and funding for research to conduct larger scale studies more efficiently with less overlap. And we’d add that greater coordination with international jurisdictions in human sports is just a key to getting where racing needs to get.
There is never going to be enough money for racing on its own to do all the research we’d like to do, so we need to take advantage of the findings of other bodies.
We should recognize the advances racing has made in medication testing and enforcement, including the work done by the states and the RMTC on medication thresholds and research into new drug types.
States such as California and Indiana are expanding their out‑of‑competition testing programs. Many states as you saw are adopting the NUMP standards.
To complete the journey towards uniform best practices equine medication, we believe racing should take action in five areas, which you’ll see here.
The first is racing needs to update the out‑of‑competition testing regulations to allow more extensive and rigorous out‑of‑competition testing. And within the out‑of‑competition testing, we need to create a system to track the location of horses in training, implement reciprocity agreements among the states, collect at least 10% of samples, and add more collectors.
We need to raise the level of out‑of‑competition samples to at least 50% and ideally more, and we need to add more collectors to go collect those out‑of‑competition samples.
The second area of our recommendation would be to standardize testing specifications for the labs across the states. That would include the list of drugs to test and the threshold of those tests to get to a uniform high standard of instrumentation and process.
The third is we need to achieve true double‑blind proficiency testing for lab accreditation and auditing. It’s difficult to do but it’s a fundamental piece of ensuring the quality and consistency of lab results.
The fourth is moving to uniform medication thresholds and penalties for violations.
And, finally, [we recommend] an increase in research funding and particularly the coordination of research investments in partnership with other jurisdictions in sports.
I’d like to thank The Jockey Club for the opportunity to share these findings with you today, and thanks also for your kind patience.