Chicken Litter -- Prohibited?

Help Support CattleToday:

Status
Not open for further replies.

HerefordSire

Well-known member
Joined
Aug 2, 2006
Messages
5,212
Reaction score
0
Location
Arkansas
I hope this new thread clears this issue up. I placed this thread besides the other thread for convenience.

There was some doubt about chicken litter being legal or not. During a drought several years ago in Arkansas, in late winter, many had no hay including me. Hay was running 400% higher than the normal market. Ranchers we paying big money for corn stock hay. One of my fellow local ranchers has many chicken houses. We mixed chicken litter, rice bran, and corn and I fed some cattle for about one month. Prior to feeding this mix, I research on the internet thoroughly. About the only thing I located that referenced litter was if fed longer than short term, it could lower milk production.

Later, I found out Arkansas farmers have been feeding litter to cattle for years and years.


Here is one comment by hillsdown. Thank-you for your help hillsdown.

hillsdown":2xakn4hn said:
DO NOT EVER FEED CHICKEN LITTER, TURKEY LITTER ,EGGS ETC...THAT I HAVE READ BEING POSTED LATELY ........ :mad: THERE IS A FRICKEN REASON FOR IT AND I SUGGEST YOU ALL DO SOME RESEARCH OR YOU HAVE NO F'ing BUSINESS AT ALL TO BE RAISING CATTLE..

Feeding animal protein to cattle is prohibited and it p@sses me off to no end when I read of people advocating it..

Sorry to jack your thread Jeanne.. 8)

Very interesting comment hillsdown. Could you provide some references to your information? I have been in research and developement for over a decade. I specialize in finding information. Perhaps the information was right in front of my face?
 
It's obviously not prohibited here at CT or you wouldn't be allowed to continue posting, HS! :tiphat:

My "one-liner" for the day!

George
 
Here's a link to Mississippi State University concerning the use of chicken litter as feed and fertilizer:

http://www.msstate.edu/dept/poultry/pub1998.htm

Oh and by the way....NOT ALL animal protein is prohibited for use in cattle feeds. Ruminant Meat and Bone meal is certainly one of them but we are talking about chicken. litter. At one time one company was mixing chicken litter, corn and bakery waste and making a 16% ration to be fed to cattle.

In 2004, the FDA considered a ban after the first case of Mad Cow disease was discovered in the U.S. but later rejected the proposal. (16,17) A year later the FDA proposed a rule to ban 90 percent of 'specified risk material' (bovine brain and spinal cord material) from all feed consumed by food animals.(18) The agency has delayed publication of the final rule indefinitely, thereby allowing the use of poultry litter in animal feed.
 
Very clever George. :mrgreen: :mrgreen: :mrgreen:

Here is one reference on the "pro" side not the "con" side.

"Recent studies have shown that broiler litter possesses substantial nutritional value. Since the 1950s, broiler litter has become a common component for cattle diets in areas where broilers are produced. Cattle requirements for calcium, phosphorus, and other minerals are usually met when broiler litter comprises 20% or more of the diet. Broiler litter may have a value of $80-$120 per ton when formulated as feed to supply protein to growing and finishing cattle (see University of Maryland's Department of Animal Science Fact Sheet B1303). This feed is best suited for ruminant animals like cattle and sheep, because ruminants digest their feed twice and can extract nutrients left after the first pass.

Ensiled broiler litter has a fermented smell and is very acceptable in cattle diets. When stored in deep stacks, the heating process not only eliminates potentially toxic bacteria, but also replaces the characteristic ammonia smell of manure with a sweeter aroma. Dry litter retains crude protein and other nutrients better than wet litter.

A survey of Virginia producers using broiler litter as a feedstuff indicated that they considered it to have a feed value five times greater than its cost. Beef cattle feeds containing broiler litter have effectively been utilized in maintenance, growing and finishing rations. However broiler litter rations should not be fed to cattle producing milk for human consumption and should be discontinued as feed for beef cattle 15 days prior to slaughter.

No feeding problems attributed to manure feed have occurred, nor have there been any problems with cattle grade, carcass quality, or marketing.

http://www.ehso.com/ehshome/chickenpoop.htm
 
My apologies I was wrong.

I got this from Dr Anderson this morning....

"There was a movement to ban feeding chicken litter indeed after the 1st
BSE case. However, the ban was not enacted into law and it is indeed
okay to feed chicken litter to cattle. The poultry folks also decided
not to put any ruminant derived feedstuffs into their diets anymore - so
there should be no way a cow eating litter should contract BSE.

So.... you too can go back to feeding litter and corn if you would
like!

Lisa Kriese-Anderson
Associate Professor
Extension Animal Scientist
210 Upchurch Hall
Auburn University, AL 36849-5625 "

I am sorry if I spread any misinformation on the subject.
 
Here is a "con" argument and not a "pro" argument. For now, I will reserve opinion in regards to the author.

Cut out the crap
Published: Friday, January 30, 2004
The federal government, responding to the mad cow scare, has prohibited the use of byproduct animal tissue, blood and chicken litter in cattle feed, as well as prohibiting the use of "downer" cattle.

All I can say is, it's about time!

"Downer" cattle are those too weak or too sick to stay on their feet. Wouldn't you have thought the meat industry would've been smart enough to have turned those critters into dog food or something instead of putting them on our plates and hamburger buns?

Not so; the bottom line is the thing, and if they can get us to eat sick cows' meat at steak prices, they're going to do it. Or at least, they have been.

Cattle that have eaten feed containing animal parts are the cows most likely to develop "mad cow" disease. Cattle that have the disease are capable of passing it on to humans, and the disease takes years to develop. So, a cow that looks OK and acts OK can have the disease, undetected, for years before it shows the classic signs of "mad cow."

I'm not trying to scare you off beef here. I've eaten at least a couple of hamburgers since the single mad cow was discovered in the U.S. But I wonder why a bigger effort wasn't made to protect consumers before now.

And the chicken litter thing - yes, you are correct, chicken litter is chicken s..., I mean, chicken crap. Excrement. Manure. Dung!

And I see a loophole here. What the government is prohibiting is the use of chicken crap in commercial feed, and commercial feed only. Plenty of cattle, especially in our home state with its multitude of poultry farms, are going to still be eating chicken crap.

That's because farm experts have been encouraging chicken farmers who also raise cattle - and they virtually all do - to feed chicken s..., I mean, manure, litter, crap, whatever you want to call it, to their cattle to save money on feed. I sat in on one of those meetings at the Poughkeepsie School cafeteria a few years ago.

I learned a lot. Like, some cattle that are pastured near chicken houses go over by themselves, without any encouragement, and eat from the chicken crap pile that's just been cleaned from the poultry houses. (That's the dumb cattle.)

They don't all do that. But all that's required to get most of the other cattle to eat chicken crap is to keep it dry and fairly free of foreign objects, and add a little grain to it. Hey, the experts say chicken crap is full of protein.

A few cattle (the smart ones) will refuse to eat the stuff, even with the suggested amount of grain included. If I remember correctly, the experts recommended getting rid of those smart cattle after their first growing season.

The suggestion was even made that Texas cattle ranches might eventually turn into a market for our state's dry chicken crap.

In the days after that meeting, I thought a lot about this whole chicken-crap-to-hamburger thing. Cows eat grass and turn it into hamburger, and it's OK. They eat grain and turn it into hamburger, and it's OK. So, I reasoned, cattle can eat chicken crap and turn it into hamburger, and it must be ... OK? I just wasn't sure. Something just didn't seem quite right about the whole thing.

On the other hand, what about people who raise chickens and cattle? They have to look at the bottom line, I suppose, and if feeding crap to cattle keeps them from paying money for other feed, they almost have to do it.

And there's the loophole - some cattle going to market are going to have been fed chicken crap, one of the ingredients the federal government is prohibiting in commercial feed.

Let me put in this disclaimer: no one, anywhere, has confirmed that chicken crap ingested by cattle has led to any problem for either the cattle or people who eat meat from those cattle. It's just on the list of stuff that's now banned in commercial feed for cattle.

And while I still have the occasional hamburger, I eat more fish and chicken than in the past. And I've also noticed that when I buy hot dogs, I buy the ones made with chicken and pork.

After all, the prions that cause mad cow disease are, if I understand correctly, a form of protein that is running amok, sometimes referred to in news articles as a "rogue protein."

All I know is that after attending that meeting at Poughkeepsie, I could never look at a hamburger the same way as before.


http://www.guardonline.com/?q=node/24079
 
In my and Hillsdown's defense, the livestock press gave a whole lot more press to FDA writing their litter ban rule, than the livestock press has given to FDA's decision not to make a decision reguarding it's final passage.
 
HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM..." Beef its what for dinner."........................followed by brought to you by the chicken growers of America...Anyone see where I am going with this......Consumer confidence in our products is not great as it is........Whats next slogan....."Eat Beef...Chicken Sh.ts Finest"
 
houstoncutter":3teahx9h said:
HMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM..." Beef its what for dinner."........................followed by brought to you by the chicken growers of America...Anyone see where I am going with this......Consumer confidence in our products is not great as it is........Whats next slogan....."Eat Beef...Chicken Sh.ts Finest"

lol, ever see a chicken $hit and then turn right around and eat it?? Nothing nastier than chickens. They'll even cannibalize their own, not to even mention hogs. They're gotta be second only to vultures as scavengers. Truth is we don't know where a lot of beef comes from, what it was fed nor why it taste so bad (or good).
 
Use the chicken litter in the fields, not in the cows. :dunce:
 
TexasBred":3ifvor2i said:
Here's a link to Mississippi State University concerning the use of chicken litter as feed and fertilizer:

http://www.msstate.edu/dept/poultry/pub1998.htm

Oh and by the way....NOT ALL animal protein is prohibited for use in cattle feeds. Ruminant Meat and Bone meal is certainly one of them but we are talking about chicken. litter. At one time one company was mixing chicken litter, corn and bakery waste and making a 16% ration to be fed to cattle.

In 2004, the FDA considered a ban after the first case of Mad Cow disease was discovered in the U.S. but later rejected the proposal. (16,17) A year later the FDA proposed a rule to ban 90 percent of 'specified risk material' (bovine brain and spinal cord material) from all feed consumed by food animals.(18) The agency has delayed publication of the final rule indefinitely, thereby allowing the use of poultry litter in animal feed.

TexasBred....I think you pegged the issue. Did you happen to have a link for that refererence? What about Canada since hillsdown shows to be from there?
 
You're not eating chicken litter any more than you're eating the bermuda grass the cow ate. :roll: Some of you sound like the anti-beef people. :devil2: I thought it was against the law to feed it to cattle, too, so I appreciate the clarification from these sources. I think it was the University of AL (?) that showed BSE prions can't survive a trip through a chicken. So, scientifically, there was no reason for chicken people to stop using bovine byproducts in their food. But perception is often more important than fact.
 
Brandonm22":o3grbhpm said:
My apologies I was wrong.

I got this from Dr Anderson this morning....

"There was a movement to ban feeding chicken litter indeed after the 1st
BSE case. However, the ban was not enacted into law and it is indeed
okay to feed chicken litter to cattle. The poultry folks also decided
not to put any ruminant derived feedstuffs into their diets anymore - so
there should be no way a cow eating litter should contract BSE.

So.... you too can go back to feeding litter and corn if you would
like!

Lisa Kriese-Anderson
Associate Professor
Extension Animal Scientist
210 Upchurch Hall
Auburn University, AL 36849-5625 "

I am sorry if I spread any misinformation on the subject.


There is nothing to apologize for Brandonm22. You were very polite and sincere in your feelings and provided professional information disagreeing with your beliefs. If anything, you helped ranchers faced with similar drought conditions, that I faced several years ago, understand the consequences of actions. You, sir, are a good man.
 
HerefordSire":19bnptjs said:
TexasBred":19bnptjs said:
Here's a link to Mississippi State University concerning the use of chicken litter as feed and fertilizer:

http://www.msstate.edu/dept/poultry/pub1998.htm

Oh and by the way....NOT ALL animal protein is prohibited for use in cattle feeds. Ruminant Meat and Bone meal is certainly one of them but we are talking about chicken. litter. At one time one company was mixing chicken litter, corn and bakery waste and making a 16% ration to be fed to cattle.

In 2004, the FDA considered a ban after the first case of Mad Cow disease was discovered in the U.S. but later rejected the proposal. (16,17) A year later the FDA proposed a rule to ban 90 percent of 'specified risk material' (bovine brain and spinal cord material) from all feed consumed by food animals.(18) The agency has delayed publication of the final rule indefinitely, thereby allowing the use of poultry litter in animal feed.

TexasBred....I think you pegged the issue. Did you happen to have a link for that refererence? What about Canada since hillsdown shows to be from there?

Herf.....Canada does have a law making it illegal to feed chicken litter to cattle in Canada.....however, you can feed it to your cows in the US and still export the meat to Canada....duh.... ;-)
 
Frankie":mxh8w8aa said:
You're not eating chicken litter any more than you're eating the bermuda grass the cow ate. :roll: Some of you sound like the anti-beef people. :devil2: I thought it was against the law to feed it to cattle, too, so I appreciate the clarification from these sources. I think it was the University of AL (?) that showed BSE prions can't survive a trip through a chicken. So, scientifically, there was no reason for chicken people to stop using bovine byproducts in their food. But perception is often more important than fact.
I think the concern is that chickens spill alot of feed, so it goes into the litter without passing through the chicken.
 
This is an interesting article, and yes to my knowledge chicken/turkey litter was included in the enhanced feed ban here in Canada . I agree TB that makes no sense at all, but is also why I buy only Alberta or Canadian beef if I have to buy any at all..

The sensitivity analysis implies that other assumptions should be investigated. For example, the risk assessment suggests investigating the use of chicken litter (which can contain chicken feed made from ruminant protein) in the production of cattle feed.
To the extent that chicken litter contains cattle protein, its use in cattle feed could represent the mathematical equivalent of mislabeling prohibited feed as nonprohibited. I will feed my cattle grass, hay and the occasional grain and you can feed sh@tzen as Bez puts it..

You also have to look at how cattle would be in the real world if you did not have them penned up where they are at the mercy of what you choose to feed them. They would choose grass, hay or grain over a pile of litter guaranteed ..There is a reason animals should not eat some things as it is not part of their natural diet. We all need to do our best that we ensure the consumer a safe and delicious product. Also feeding litter can and has led to numerous cases of botulism.



This is the html version of the file http://compepid.tuskegee.edu/RiskConfer ... 0paper.doc.
Google automatically generates html versions of documents as we crawl the web.
Criteria for Risk Assessments to Aid in Decisions to Import Agricultural Products


Joshua T. Cohen*, Ph.D., Harvard Center for Risk Analysis, Harvard School of Public Health, Boston, MA.


* Address for corresponding author and reprint requests:

Harvard Center for Risk Analysis

718 Huntington Avenue

Boston, MA 02115

Tel.: 617 432 0394; Fax.: 617 432 0190

email: [email protected]






Abstract

Recent trade agreements have increased the emphasis on sanitary and phytosanitary risk assessment as a tool to help identify cases where restrictions on the import of agricultural products are justified. Setting aside the legal issues, the role of a risk assessment in these situations is to determine if the risks associated with imports are acceptably small given the benefits (e.g., lower food costs). This paper describes criteria a risk assessment must satisfy in order to fulfill this role. In particular, I argue that 1) a risk assessment must quantitatively characterize impacts relevant to stakeholders, 2) that the underlying science must be reproducible; and that 3) the analysis must address uncertainty in an unbiased manner. Finally, this paper discusses the potential use of risk assessment to evaluate USDA's rule to allow cattle imports from Canada.




Introduction

Identification of appropriate standards for the conduct of sanitary and phytosanitary (SPS) risk assessments has become more critical in recent years as trade agreements have lead to a greater reliance on such analyses. For example, the North American Free Trade Agreement (NAFTA) and the General Agreement on Tariffs and Trade (GATT) require risk assessments to support import limits in order to prevent such measures from being used as protectionist measures (Gray 1998).


Plaintiffs have challenged the legitimacy of rules promulgated by the U.S. Department of Agriculture (USDA) to allow the import of various commodities into the U.S. Cases include Harlan Land Co. vs. USDA (186 F. Supp 2d 1076), involving the import of citrus from Argentina; Cactus Corner vs. USDA (346 F. Supp 2d 1075), involving the import of Clementines from Spain; and in Ranchers Cattlemen Action Legal Fund vs. USDA, involving the import of cattle from Canada in the wake of the discovery of cattle infected with bovine spongiform encephalopathy (BSE) in that country (Ninth Circuit, decision filed July 25, 2005). In all three cases, the plaintiffs claimed USDA's actions were arbitrary and capricious because the Department did not conduct a legitimate risk assessment and because it did not establish an acceptable level of risk.


This paper discusses the criteria for risk assessment not from a legal perspective, but from the more abstract perspective of what aids good decision making. After all, SPS risk assessments are conducted, at least ostensibly, for the purpose of deciding how to manage a risk – e.g., whether import of a particular commodity should be allowed, and if so, under what conditions. Based on the goal of promoting good decision making, I argue that that 1) a risk assessment must quantitatively characterize impacts relevant to stake holders, 2) that the underlying science must be reproducible; and that 3) the analysis must address uncertainty in an unbiased manner. Finally, I discuss the potential use of risk assessment to evaluate USDA's rule to allow cattle imports from Canada.

1 Risk Assessment Must Quantify Impacts Relevant to Stake Holders

The claim made in this section might at first glance seem to be self evident. After all, if the risk assessment does not quantify outcomes relevant to stake holders, then why would anyone be interested in its results? On the other hand, it is not immediately obvious what outcomes stakeholders are interested in, and there are certainly many outcomes that do not have value in and of themselves, although they seem relevant because they are related to impacts of interest to stakeholders.


Consider the following SPS and environmental examples: 1) a risk assessment of fruit imports might quantify the probability that a pest would survive measures aimed at its elimination (e.g., cold treatment); 2) a risk assessment of cattle imports might estimate the probability that an animal with BSE is imported into the U.S. in a given year; 3) a risk assessment of diesel buses might estimate human exposure to diesel particulates. These outcomes are certainly important, but they do not by themselves help to identify optimal actions (e.g., should fruit or cattle imports be allowed; what risk mitigation measures should be taken; should a transit authority purchase diesel powered buses or compressed natural gas powered buses) because they do not help a decision maker (or stakeholder) determine if the benefits of a risk management option justify its costs.


Knowing how good or bad the outcome is depends on characterizing it in terms of impacts that stakeholders value. Hence, quantifying the probability that a pest or BSE-infected animal is imported into the U.S. is not sufficient to inform decision making because those events in and of themselves do not have a value that can be compared to the benefits of allowing imports. For example, suppose that allowing imports of fruit from another country contributes $100 million annually to the U.S. economy. That benefit (which clearly does have a value to stakeholders) cannot be compared to the value of preventing a single pest from being imported into the country.


Technically, comparing risks and benefits for the purpose of determining whether the risks associated with an action are acceptable is part of "risk management" (National Research Council 1983). In some cases, in an effort to aid risk managers, risk assessments provide information to aid such comparisons. A common approach is to compare the identified tradeoff to other tradeoffs that are generally viewed as acceptable or not acceptable. For example, it follows that a contemplated tradeoff is acceptable if it is more favorable than other acceptable tradeoffs. Conversely, a contemplated tradeoff can be viewed as unacceptable if it is less favorable than other unacceptable tradeoffs.


Cohen and Graham (2003b) used such comparisons in a risk assessment of cell phone use while driving. Allowing the use of cell phones while driving represents a convenience that consumers place a monetary value on. However, available evidence also suggests that it increases the risk of motor vehicle crash fatalities. Cohen and Graham estimated that a ban on cell phone use while driving would result in the loss of convenience valued at $380,000 for each year of life saved. In the parlance of health economics, the intervention is said to have a cost-effectiveness of $380,000 per life-year. Assuming that the underlying analysis is valid, how would one determine if such a result supports or counters the case for banning cell phone use while driving? Cohen and Graham addressed this question by comparing this cost-effectiveness ratio to the cost-effectiveness ratio for other motor vehicle safety interventions. They noted that imposing a 55 mph speed limit on rural interstate highways achieves risk reduction at the lower cost of $82,000 per life year. The fact that the U.S. has not imposed a 55 mph speed limit on interstate highways suggests that a cost of $82,000 per life year saved is too high to be considered acceptable. In that case, it follows that the cost of $380,000 per life year saved by a ban on cell phone use while driving is also too high. Put another way, based on costs and benefits, a rationale individual could not simultaneously favor a ban on cell phone use and oppose imposition of a 55 mph speed limit on interstate highways.


Note that such comparisons must be used judiciously. The comparisons are most useful when the risks and benefits are both comparable. The example in the Cohen and Graham paper is compelling because the costs associated with the two interventions are a loss of either time or convenience. Likewise, the risks associated with each intervention are similar, if not identical – i.e., motor vehicle crash fatalities.


On the other hand, comparing dissimilar risks is problematic. An extensive literature on risk perception suggests that the urgency people place on a risk depends not just on its magnitude but also on other factors, such as whether it is incurred voluntarily, its immediacy, a "dread" factor, its potential for catastrophe, how much knowledge there is about the risk, and its distribution, among other factors (Wilson 2001). In one effort to describe how people rank risks, Slovic (1987) noted that people place greater weight on avoiding risks that are less well understood and on risks that inspire greater dread. Figure 1 illustrates some of the examples identified by Slovic. In another paper, Slovic (1998) showed that people are less tolerant of risks that are perceived to be involuntary than they are of more voluntary risks.

2 Risk Assessment Science Must Be Reproducible

A risk assessment organizes science so that it can be used by a decision maker (e.g., a regulatory body) to act on behalf of the citizenry (Gray 1998). The key characteristic of scientific reasoning is reproducibility. That is, presented with the same data and scientific theories, a suitably trained third party can understand how the risk assessor reached a conclusion. The court in Harlan vs.. USDA quoted Gray et al. (1998) as stating (p. 779), "Any kind of analysis requires a complete and transparent documentation of data considered and used, models and their assumptions and results, sources and justification of parameter values and all other relevant facts… In all cases, the standard should be that anyone could understand and reproduce an analysis based on its documentation."


Unless a risk assessment is reproducible, it is not possible to determine if its results are based on data and scientific theory, rather than on subjective judgments or the desire of a decision maker to support one course of action over another. One way to help ensure that a risk assessment is reproducible is to subject it to independent scientific peer review. Peer review involves gathering comments on the risk assessment from scientists not involved in its development, asking the risk assessment authors to revise the assessment in response to these comments or to explain why they believe revision is not necessary, and having a third party independent of the authors and reviewers arbitrate the process.


The Office of Management and Budget (OMB) (2003) notes that the peer review process has been used by scholarly journals to ensure the scientific quality of articles and by federal agencies for major scientific analyses. Such a tool would help ensure the reproducibility (and credibility) of significant risk assessments. I identify four principles for peer review based in part on requirements specified in an OMB guideline (pp. 10-11 in (Office of Management and Budget 2003)).


The review should be conducted by individuals who do not have a particular interest in its conclusions. If it is not possible to identify individuals who are both knowledgeable and impartial, a balanced reviewer group should be identified;
The reviewers should be charged with addressing specific contentious assumptions, but should also be given a sufficiently broad mandate that permits them to question other aspects of the risk assessment;
The reviewers should be given access to scientific information sufficient to allow them to reproduce the risk assessment;
Peer review comments should be made available to the public, along with an explanation of how the risk assessment authors responded to those comments.
3 Risk Assessment Must Address Uncertainty Without Bias

Invariably, risk assessment involves the use of uncertain assumptions, models, and data. In many cases, the uncertainty is so substantial that it threatens to paralyze efforts to conduct the assessment. The precautionary principle, which strives to avoid uncertain risks by demanding a demonstration of safety before permitting an action, has been advanced as one approach to address such situations (Applegate 2000; Hammitt 2000).


The problem with a strict interpretation of the precautionary principle is that it imposes an artificial asymmetry on a decision that favors the status quo. For example, an adherent to the precautionary principle one hundred years ago might have favored prohibiting the introduction of drinking water disinfection because the risks of the resulting chemical exposures were (and are) uncertain. Such a position ignores the potential benefits of drinking water disinfection. Ironically, it is possible that in the present, a strict adherent to the precautionary principle might oppose proposals to eliminate or otherwise alter drinking water disinfection technologies until the safety of doing so is demonstrated.


In the case of agricultural imports, the asymmetry of the precautionary principle is likewise problematic. For example, if the U.S. currently imports no fruit and vegetables from a given country, the precautionary principle might require demonstration that allowing such imports would be "safe" (e.g., that doing so would not introduce pests or disease into this country). If the U.S. currently does import fruit and vegetables from that country, then the precautionary principle might require demonstration that stopping these imports would impose no risk (e.g., by driving up the price of fruits and vegetables, which could shift consumption towards less healthy foods among members of the population).


An iterative approach to addressing uncertainty in an unbiased manner breaks this asymmetry. The key is to start with a relatively simple analysis and to increase the level of sophistication to address uncertainty only as needed to identify the optimal course of action. The first step in this approach is to conduct a "screening" assessment using bounding "conservative" (i.e., pessimistic) assumptions that tend to overstate risk. A finding that risks calculated in this manner are acceptable implies that the actual risks are acceptable (because the actual risks are smaller than the risks calculated using pessimistic assumptions).


It may be tempting to use results from a pessimistic screening analysis to establish that a risk is not acceptable. However, because actual risks are smaller than risks calculated using a screening assessment, such an inference is invalid. Use of pessimistic assumptions may nonetheless be defended on the grounds that doing so protects public health. However, the purpose of conducting the risk assessment is to determine if a risk is unacceptably large compared to some compensating benefit. A biased estimate of risk cannot be meaningfully compared to a benefit measured in an unbiased manner.


If a screening assessment suggests a risk is unacceptably large, the assessment should be refined using more realistic assumptions. These assumptions require additional information but tend to decrease the magnitude of the estimated risk. This approach can be repeated until either the screening assessment reveals that the risk is acceptable (because the conservatively estimated risk is acceptable) or until the screening assessment can be revised no further with readily available information.


If the screening assessment cannot be further refined, a "full-blown" risk assessment (i.e., one characterizing the full range of plausible risk estimates) can be conducted. The goal of such an assessment is to convey the potential upside and downside for each course of action, and to convey which uncertain assumptions contribute the most to the risk assessment's overall uncertainty.


Uncertainty can be characterized using different approaches that vary in terms of their sophistication. Approaches fall into two general categories – bounding analyses and probabilistic analyses. A bounding analysis identifies the range of plausible values for an assumption without quantifying their relative plausibility. The risk assessment describes the extent to which predictions vary when an individual parameter is assigned its bounding values while holding other parameters equal to their "base case" values. Such a one-way sensitivity analysis can be extended by varying multiple parameters simultaneously. Figure 2 illustrates the results of a sensitivity analysis from a risk assessment of cell phone use while driving (Cohen 2003b).


A probabilistic analysis describes the relative plausibility of alternative values for each assumption in terms of a probability distribution. Probabilistic analyses are often conducted using Monte Carlo simulation (Cullen 1999). Monte Carlo simulation draws values randomly from the each uncertain parameter's probability distribution, computes the risk estimate corresponding to this set of values, and then repeats the process. A large number of such trials produces a set of risk estimates that can be described as a probability distribution (see Figure 3). Note that a probabilistic analysis provides more information than a bounding analysis because it characterizes not just the range of possible risks, but also their relative likelihood. However, a probabilistic assessment requires more information than a sensitivity analysis – i.e., a probability distribution for each uncertain assumption, not just a range of plausible values.


The "full blown" stage of the risk assessment can likewise be viewed as iterative (Hammonds 1994). If the risk assessment produces results that are sufficiently precise to identify the optimal course of action with an acceptable degree of confidence, then there is no reason to further refine it. However, if the result is insufficiently precise, the decision maker can gather additional information, focusing attention on those parameters identified as contributing the most to the risk assessment's uncertainty. After these assumptions are refined (made more precise), the risk assessment is repeated and the results evaluated to see if they are sufficiently precise for decision making. Whether further investment in refining an analysis is worthwhile depends on four factors:


The consequences being wrong: A decision where a wrong choice has a substantial downside (large potential risks or large foregone benefits) tends to favor investing in an improved analysis.
The likelihood of being wrong: A situation where it is not clear whether the risk is acceptable or not tends to favor investing in an improved analysis. If the answer is already clear, then it is unlikely that a new analysis would provide new insight.
The likelihood that new information would improve the risk assessment's precision: Gathering information makes more sense if that information tends to reduce uncertainty substantially.
The cost of new information: Low cost information tends to favor investing in an improved analysis.


Figure 4 illustrates how the analyst can decide when it is appropriate to further refine the screening assessment, when it is appropriate to move on to a full-blown assessment, and when it makes sense to gather additional information to refine the full-blown assessment. In short, the screening assessment is conducted until it can be shown the risk is acceptably small or that information is not available to definitively make such a determination. In the latter case, a full blown assessment is conducted. That assessment is refined until either the likelihood of getting the "wrong answer" is reduced sufficiently, or until the cost of gathering new information (relative to its value) gets too high.


It is possible that this iterative process will arrive at a point where the risk estimates remain uncomfortably uncertain for the purpose of decision making and yet important assumptions cannot be sufficiently refined, at least not in a timely manner. It is important to recognize that such a situation is an inherent property of the decision at hand, rather than reflecting a problem with risk assessment as a methodology. In these situations, it simply is not possible to know what the best course of action is with a sufficiently comfortable level of certainty. Other criteria can be used to make the decision in these situations. However, these alternatives will not improve the likelihood that the optimal course of action will be identified.

Example: Risk Assessment of Canadian Cattle Imports

In response to the discovery of a dairy cow infected with bovine spongiform encephalopathy (BSE) in Canada in May, 2003, USDA immediately banned the import of cattle from that country. The decision reflected long-standing policy that classified countries as "(1) A region free of BSE; (2) a region in which BSE is known to exist; or (3) a region that presents an undue risk of BSE" (p. 62,387 in (U.S. Department of Agriculture 2003)). This scheme effectively placed Canada, the source of four identified BSE-infected cattle, in the same category as the UK, where some 185,000 BSE-infected cattle have been identified. In January, 2005, USDA published a new rule introducing an additional category for countries, referred to as "minimal risk," and allowed for the import of some cattle from these countries (U.S. Department of Agriculture 2005).


Before the issue of BSE in Canada arose, the Harvard Center for Risk Analysis had conducted a risk assessment characterizing the impact of several hypothetical scenarios involving the introduction of BSE into the U.S. (2003a). Among other scenarios, the assessment considered the hypothetical import of 10 BSE-infected animals. Using the three criteria for risk assessment described earlier in this paper, the following text briefly discusses use of the Harvard analysis to characterize risks associated with the import of cattle from Canada.


Quantification of Risks Relevant to Stakeholders


Among the key outcomes relevant to stakeholders are the impact of imports on animal health, public health, and demand for beef. The Harvard risk assessment was developed to estimate the impact of a hypothetical BSE introduction. However, it could be tailored to address scenarios specific to the import of cattle from Canada.


Animal health: The Harvard risk assessment provides the information needed to evaluate the impact of Canadian imports on animal health. In particular, the assessment estimates how many additional animals in the U.S. would become BSE-infected in the 20 year period subsequent to the introduction of disease. It also estimates the number of infected animals that would develop clinical signs. Ultimately, the Harvard risk assessment seeks to determine if BSE would tend to spread or die off after an introduction. This question is answered by quantifying the so-called disease reproductive constant (designated R0), which is the average number of animals infected by each existing BSE case. If R0 is less than 1, then disease prevalence tends to decrease over time. If it is greater than 1, then the prevalence tends to grow. Knowing if the disease will tend to die off or if its prevalence will grow (exponentially) is of key importance to stakeholders.


Public health: The Harvard risk assessment estimates human exposure to the BSE agent, measured in terms of cattle oral ID50s. One cattle oral ID50 will infect a single bovine with 50% probability following oral intake. Because the BSE agent is thought to be less capable of infecting humans, the risk of contracting variant Creutzfeldt Jokob Disease (vCJD, the human form of BSE) is substantially less following exposure to the same quantity of BSE agent. How resistant humans are to the BSE agent, i.e., how large the so-called species barrier is, is very uncertain, with recent estimates spanning several orders of magnitude (Comer 2003).


It is because of this uncertainty that the Harvard assessment does not go as far as estimating human risk (p. 2 in (Cohen 2003a)). Better information on the magnitude of the species barrier could facilitate extension of the Harvard risk assessment for the purpose of estimating vCJD risk. Alternatively, the more extensive vCJD surveillance data from the UK could be used to estimate this risk by quantifying the number of vCJD cases per BSE-infected animal in that country. The expected number of vCJD cases in the U.S. () could then be estimated as


,


where is the number of vCJD cases ultimately resulting from the UK BSE epidemic, andand are the number of BSE cases in the UK and the U.S., respectively. The E parameter (EUK and EUS for the UK and U.S., respectively) is the average amount of BSE contamination (cattle oral ID50s) that ends up in human food per infected animal. The value of can be estimated from projections based on vCJD surveillance data and other considerations (Clarke 2005). Fergusson et al. (Ferguson 1997) have estimated the value of . The Harvard risk assessment model estimates and EUS. Although I am unaware of an estimate of EUK, that parameter could be estimated by reviewing slaughter practices and meat consumption practices (in particular, what tissues are consumed) in the UK during the past two decades.


Beef Demand: For cattle producers, the impact of BSE cases on demand for their product is of course important. This impact has two components – domestic demand and international trade. The Harvard risk assessment did not address this issue.


Reproducibility of the Science


The Harvard risk assessment provides an extensive description of the development of parameter values for the simulation model. This work was independently peer reviewed. The peer review comments are available to the public (http://www.hcra.harvard.edu/publications.html), along with the original version of the report (November, 2001), and version updated in response to those comments (October, 2003). In addition, the Harvard Center for Risk Analysis will provide the simulation source code at no charge to anyone who requests it.


Unbiased Treatment of Uncertainty


Using "base case" assumptions (i.e., best estimates of likely parameter values), the Harvard risk assessment estimated a value for R0 substantially less than 1. The sensitivity analysis evaluated the impact of setting each assumption individually to its "worst case" value, while holding all other parameters to their base case value. The sensitivity analysis also analyzed the impact of altering sets of parameters simultaneously.


This approach differs somewhat from a typical sensitivity analysis, which evaluates the impact of both best case and worst case assumptions. However, use of optimistic assumptions would only result in predictions of R0 smaller than the base case value. Because the base case analysis already indicated that the prevalence of BSE would decrease over time (i.e., R0 < 1), there was limited value in determining the impact of optimistic assumptions, which would only result in predictions that the disease would die out somewhat more quickly (p. 69 in (Cohen 2003a)).


Instead, the Harvard risk assessment focused on identifying assumptions that would result in a prediction that R0 would exceed 1, suggesting exponential prevalence growth over time. Harvard defined 17 sets of parameters and identified three as having the largest impact on the number of additional BSE cases subsequent to disease introduction into the U.S.: 1) the misfeeding rate, defined to be the proportion of properly labeled prohibited feed (feed prohibited from being fed to ruminants) incorrectly administered to cattle; 2) the proportion of prohibited feed improperly mislabeled as non-prohibited (permitted for use in cattle); and 3) the proportion of rendering facilities using various technologies, each of which eliminates the BSE agent in processed cattle tissue to a different degree.


The impact of using a pessimistic assumption for the misfeeding rate was particularly dramatic. For example, when the base case assumptions are used, the risk assessment predicts prevalence will drop to zero within 20 years with a probability exceeding 99%. Use of the worst case misfeeding rate (and base case values for all other assumptions) decreases this probability to less than 90%. The misfeeding parameter is important because its value is very uncertain (Harvard assumed that the worst case value was a factor of 10 larger than the base case value) and because relatively modest changes in this parameter's value have a large impact on the spread of disease.


The sensitivity analysis implies that other assumptions should be investigated. For example, the risk assessment suggests investigating the use of chicken litter (which can contain chicken feed made from ruminant protein) in the production of cattle feed. To the extent that chicken litter contains cattle protein, its use in cattle feed could represent the mathematical equivalent of mislabeling prohibited feed as nonprohibited.


Finally, it should be noted that while an effort was made to develop unbiased characterizations of the uncertainty for the vast majority of the assumptions used in the model, there is (at least) one assumption that should be regarded as pessimistic. In particular, the assessment assumes that the dose-response relationship for BSE is linear – i.e., it assumes that the incremental risk of disease is proportional to the incremental exposure, no matter how small that exposure is. Alternative assumptions may posit a sigmoidal dose-response (i.e., one that implies the incremental risk per unit of exposure is smaller at low exposure levels), or one that has a threshold below which the risk of infection is zero. Because of the potentially "conservative" linearity assumption, the Harvard analysis could be viewed as having some screening assessment characteristics. How important this particular assumption is remains a matter of speculation, as the information to confidently characterize the shape of the dose-response relationship for BSE remains limited.

Conclusion

Using risk assessment as a tool to help optimize decision making for stakeholders dictates its characteristics. These characteristics follow from the presumption that the "primary" risks are being estimated so that they can be compared to a cost or alternative risk that would be incurred as part of eliminating that primary risk. Failure to express risk in terms of outcomes stakeholders place a value on makes the results irrelevant to that population. Failure to make the science reproducible puts citizenry in the position of having to accept the results as a matter of faith. Failure to maintain an unbiased approach when addressing uncertainty places an implicit premium on one side of the tradeoff to be evaluated.


Risk assessment has been used extensively to evaluate both medical interventions (as part of the development of cost-effectiveness analyses – see the Harvard cost-effectiveness analysis registry at http://www.hsph.harvard.edu/cearegistry/ ) and to manage environmental hazards (National Research Council 1994). As in these domains, risk assessment can be useful for the purpose of evaluating sanitary and sanitary and phytosanitary risks, and in particular, weighing the risks associated with allowing imports of various products against the countervailing benefits of imports. Risk assessments that do adhere to the criteria described in this paper provide decision makers a tool to help improve stakeholder outcomes by ensuring that the results are relevant, that they are credible, and that the impact of subjective bias on the analysis is limited.






Figure 1: Classification of Hazards(a)






















Notes: (a) Source Slovic (1987). This figure represents a subset of the 31 hazards plotted by Slovic.





Figure 2: Illustrative Bounding Sensitivity Analysis(a)





















Notes: (a) Based in part on Cohen and Graham (2003b). Contributions to uncertainty by assumed annual driving distance and number of cell phone users are illustrative.





Figure 3: Monte Carlo Simulation





























Figure 4: Iterative Process to Address Uncertainty






















Notes: (a) Refining a risk assessment involves gathering more information. Doing so makes sense if the consequences of a suboptimal action are substantial, the probability of being wrong is high, the likelihood that new information will substantially reduce uncertainty, and the new information is not too costly (see text).




References

Applegate, J. S. (2000). The precautionary preference: An American perspective on the precautionary principle. Human and Ecological Risk Assessment 6(3): 413-444.


Clarke, P. and Ghanni, A. C. (2005). Projections of the future course of the primary vCJD epidemic in the UK: inclusion of subclinical infection and the possibility of wider genetic susceptibility. Interface 2: 19-31.


Cohen, J. T., Duggar, K., Gray, G. M. and Kreindel, S. (2003a). Evaluation of the Potential for Bovine Spongiform Encephalopathy in the United States: Report to the U.S. Department of Agriculture (revised October, 2003). Boston, MA, Harvard Center for Risk Analysis. Available at: http://www.hcra.harvard.edu/peer_reviewed_analysis.html.


Cohen, J. T. and Graham, J. D. (2003b). A revised economic analysis of restrictions on the use of cell phones while driving. Risk Analysis 23(1): 5-17.


Comer, P. J. and Huntly, P. J. (2003). Exposure of the human population to BSE infectivity over the course of the BSE epidemic in Great Britain and the impact of changes to the Over Thirty Month Rule. OTMR Review Paper, Version 3. DNV Consulting. Available at: http://www.food.gov.uk/multimedia/pdfs/otmcomer.pdf.


Cullen, A. C. and Frey, H. C. (1999). Probabilistic Techniques in Exposure Assessment: A Handbook for Dealing with Variability and Uncertainty in Models and Inputs. New York, Plenum Press.


Ferguson, N. M., Donnelly, C. A., Woolhouse, M. E. and Anderson, R. M. (1997). The epidemiology of BSE in cattle herds in Great Britain. II. Model construction and analysis of transmission dynamics. Philosophical Transactions of the Royal Society of London - Series B: Biological Sciences 352(1355): 803-838.


Gray, G. M., Allen, J. C., Burmaster, D. E., Gage, S. H., Hammitt, J. K., Kaplan, S., Keeney, R. L., Morse, J. G., North, D. W., Nyrop, J. P., Sthevitch, A. and Williams, R. (1998). Principles for conduct of pest risk analyses: Report of an expert workshop. Risk Analysis 18(6): 773-780.


Hammitt, J. K. (2000). Global climage change: Benefit-cost analysis vs. the precuationary principle. Human and Ecological Risk Assessment 6(3): 387-398.


Hammonds, J. S., Hoffman, F. O. and Bartell, S. M. (1994). An Introductory Guide to Uncertainty Analysis in Environmental and Health Risk Assessment. Report No. ES/ER/TM-35/R1. SENES Oak Ridge, Inc. Available at: http://risk.lsd.ornl.gov/homepage/tm35r1.pdf.


National Research Council (1983). Risk Assessment in the Federal Government: Managing the Process. Washington, DC, National Academy Press.


National Research Council (1994). Science and Judgment in Risk Assessment. Washington, DC, Nationatl Academy Press.


Office of Management and Budget (2003). Peer Review and Information Quality. Washington, DC, Office of Information and Regulatory Affairs. Available at: http://www.whitehouse.gov/omb/inforeg/p ... uality.pdf.


Slovic, P. (1987). Perception of risk. Science 236: 280-285.


Slovic, P. (1998). The risk game. Reliability and System Safety 59(1): 73-78.


U.S. Department of Agriculture (2003). Bovine spongiform encephalopathy; Minimal Risk Regions and importation of commodities. Federal Register 68(213): 62386-62405.


U.S. Department of Agriculture (2005). Bovine spongiform encephalopathy; Minimal Risk Regions and importation of commodities. Federal Register 70(2): 460-553.


Wilson, R. and Crouch, E. A. C. (2001). Risk-Benefit Analysis. Cambridge, MA, Harvard University Press.

- -
 
"You also have to look at how cattle would be in the real world if you did not have them penned up where they are at the mercy of what you choose to feed them. They would choose grass, hay or grain over a pile of litter guaranteed ..There is a reason animals should not eat some things as it is not part of their natural diet"

Actually I have seen a lot of cattleman dump a large pile of litter out in their field ~come first frost and they let the cattle have it free choice like the hay and the grass and cows WILL dig into the litter pile. They don't have to be forced. I don't know that that is the wisest way to feed it or not. I am not saying that they would eat it if they were standing in a field of knee high rye grass; but that's not when you would supplement cows either.
 
Thanks for providing the link hillsdown. The person that wrote the paper hillsdown referred to labeled "Criteria for Risk Assessments to Aid in Decisions to Import Agricultural Products" appears to be qualified. I am currently trying to find a date on the document. I am estimating the document was authored in 8/2005. Here is a power point presentation that is easier to read. You can navigate to the BSE reference on page 22 and a couple of screens thereafter. It appears to me the author is a mathematician. This specific post more than likely has to do with matematical risk and nothing more, that I can determine, pertinent to feeding cattle chicken litter.

http://compepid.tuskegee.edu/RiskConfer ... 8%2011.ppt

EPA Science Advisory Board Staff

Joshua Cohen

Dr. Joshua T. Cohen is a Research Associate Professor of Medicine, Tufts University School of Medicine, Institute for Clinical Research and Health Policy Studies, Center for the Evaluation of Value and Risk, at the Tufts New England Medical Center, Boston, MA. Formerly, he was a Senior Research Associate at the Center for Risk Analysis, Department of Health Policy and Management, School of Public Health, Harvard University. In that capacity, Dr. Cohen develops and conducts analyses, authors manuscripts and reports, develops health economic models and environmental risk assessments, provided expertise in Monte Carlo simulation, and is a guest lecturer on the use and design of simulation models in continuing education and graduate student courses. Prior to his position at Harvard, Dr. Cohen was a Senior Associate at Gradient Corporation from 1994-1999. Dr. Cohen earned his Master's Degree in Applied Mathematics from Harvard University in 1990. He earned his Ph.D. in Decision Sciences from Harvard in 1994. Dr. Cohen's particular areas of expertise relative to the SAB CASAC Lead Review Panel include lead exposure modeling, risk assessment and uncertainty characterization, and evaluation of economic effects of lead. Dr. Cohen has published on several modeling issues relevant to the Review Panel, including several peer-reviewed articles on lead exposure modeling. In 2001, Dr. Cohen co-authored an article on trends in childhood blood lead levels, and another article on the development of a stochastic physiologically-based pharmacokinetic model for lead. In 1998, he published an article on blood lead slope factor models for adults. In 1995, he published an article on the use of Monte Carlo simulation techniques to predict population blood lead levels. Since 2004, Dr. Cohen has served as a member of the National Academies of Sciences Committee on EPA's Exposure and Human Health Reassessment of TCDD and Related Compounds.
 
Brandonm22":159yxuxp said:
"You also have to look at how cattle would be in the real world if you did not have them penned up where they are at the mercy of what you choose to feed them. They would choose grass, hay or grain over a pile of litter guaranteed ..There is a reason animals should not eat some things as it is not part of their natural diet"

Actually I have seen a lot of cattleman dump a large pile of litter out in their field ~come first frost and they let the cattle have it free choice like the hay and the grass and cows WILL dig into the litter pile. They don't have to be forced. I don't know that that is the wisest way to feed it or not. I am not saying that they would eat it if they were standing in a field of knee high rye grass; but that's not when you would supplement cows either.


When I fed my cattle chicken litter mixed in with rice bran and corn during the bad drought we had, I emptied the mix held in an old fertilizer spreader into concrete troughs with small drain holes at the bottom so water would not accumulate. I could barely work they wanted it so bad. They loved it and they loved to see and hear my one ton coming towing the fertilizer spreader. I had to be careful not to get hurt they wanted it so bad.
 
Brandonm22":vwgwfdf5 said:
Actually I have seen a lot of cattleman dump a large pile of litter out in their field ~come first frost and they let the cattle have it free choice like the hay and the grass and cows WILL dig into the litter pile. They don't have to be forced. I don't know that that is the wisest way to feed it or not. I am not saying that they would eat it if they were standing in a field of knee high rye grass; but that's not when you would supplement cows either.
Same thing I've witnessed. Cows got real well on it. Not my cows, they were my cousin's.
 
Status
Not open for further replies.

Latest posts

Top