7
Costs of Eliminating Subtherapeutic Use of Antibiotics

Under current food-animal production practices in the United States, antibiotics are used to treat specific health problems (therapeutic use) and to improve animal performance (subtherapeutic use), as discussed in earlier chapters. Used subtherapeutically, antibiotics result in enhanced growth rates and improved feed efficiency, thereby contributing to lower costs of meat and eggs. However, this practice also is associated with the development of antibiotic-resistant strains of bacteria that contribute to the presence of drug-resistant pathogens in humans, as discussed in Chapters 3 and 6.

It is frequently suggested that, because of the resistance issue, subtherapeutic use of antibiotics should be banned. The main arguments against a ban are that it would cause an economic hardship for livestock and poultry producers and raise costs for consumers. In large part, subtherapeutic feeding of antibiotic drugs is a management tool to prevent infection and to facilitate the use of confinement housing. This practice allows larger numbers of animals to be maintained in a healthy state and at a lower cost per unit to the farmer. If subtherapeutic use of antibiotic agents were eliminated, these production advantages would be reduced or lost and consumers would pay more.

To gauge the cost to consumers of eliminating the subtherapeutic use of antibiotics, the Committee on Drug Use in Food Animals conducted an economic analysis. Under current production practices in the United States, it is difficult to quantify either the probability that subtherapeutic drug use results in human health problems, or the economic value of the current and potential stock of antibiotics. The role of economic analysis is limited to measuring the benefits of subtherapeutic drug use to producers or, alternatively, to identifying the costs



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks 7 Costs of Eliminating Subtherapeutic Use of Antibiotics Under current food-animal production practices in the United States, antibiotics are used to treat specific health problems (therapeutic use) and to improve animal performance (subtherapeutic use), as discussed in earlier chapters. Used subtherapeutically, antibiotics result in enhanced growth rates and improved feed efficiency, thereby contributing to lower costs of meat and eggs. However, this practice also is associated with the development of antibiotic-resistant strains of bacteria that contribute to the presence of drug-resistant pathogens in humans, as discussed in Chapters 3 and 6. It is frequently suggested that, because of the resistance issue, subtherapeutic use of antibiotics should be banned. The main arguments against a ban are that it would cause an economic hardship for livestock and poultry producers and raise costs for consumers. In large part, subtherapeutic feeding of antibiotic drugs is a management tool to prevent infection and to facilitate the use of confinement housing. This practice allows larger numbers of animals to be maintained in a healthy state and at a lower cost per unit to the farmer. If subtherapeutic use of antibiotic agents were eliminated, these production advantages would be reduced or lost and consumers would pay more. To gauge the cost to consumers of eliminating the subtherapeutic use of antibiotics, the Committee on Drug Use in Food Animals conducted an economic analysis. Under current production practices in the United States, it is difficult to quantify either the probability that subtherapeutic drug use results in human health problems, or the economic value of the current and potential stock of antibiotics. The role of economic analysis is limited to measuring the benefits of subtherapeutic drug use to producers or, alternatively, to identifying the costs

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks incurred if current use of subtherapeutic drugs were prohibited. Ideally, the costs associated with a ban should be compared with the benefits to consumers (valued as the benefits from reduced health problems). Because of the difficulty in measuring economic benefits, only the costs are addressed here. The estimated cost measures can be compared among themselves to elucidate the sensitivity of the results to various assumptions and to provide an understanding of the magnitude of the costs. CONSIDERATIONS IN DETERMINING THE EFFECT OF A BAN The best way to determine the economic benefits of subtherapeutic antibiotics is to examine what would occur if the U.S. Food and Drug Administration were to prohibit all forms of subtherapeutic drug use. To make this estimate, several areas must be considered. Definition of Subtherapeutic Use If regulators were to decide to limit subtherapeutic drug use, it would become essential to define the difference between therapeutic and subtherapeutic uses more accurately. For a detailed discussion of the different uses, see Chapter 2 and Hays and Black (1989). The current practice of incorporating antibiotics in beef cattle diets in the feedlot is done to prevent liver abscesses and the diseases associated with the stress of moving and commingling animals. Under current regulations, there is little incentive to determine whether such feeding is therapeutic or subtherapeutic, and an argument can be made for either definition. Such feeding is therapeutic in that the incidence of liver abscesses and stress-related diseases would be higher if the drugs were withdrawn. If the symptoms appear after drug withdrawal, the drugs can be used therapeutically. However, a strict interpretation of therapeutic use is to treat a symptom, and if antibiotics are used to prevent a symptom, they are used prophylactically. The argument over definitions is more than one of semantics. The entire beef-feeding industry would be exempt from any ban if the first definition were applied. In the analysis, the strict definition is used, because many of the benefits of subtherapeutic use in poultry and pork industries could also be described as treating symptoms before they develop (that is, before subclinical problems become clinical). This issue is important because it implies that once a subtherapeutic-use ban were in place, there would be strong incentives to restrict therapeutic use. Measurement Choice The first effect of any ban on subtherapeutic use of antibiotics would be felt in the animal health industry. In 1995, this industry generated $3.3 billion in sales. In the same year, the human health pharmaceutical industry produced

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks approximately $63 billion in sales. Approximately 62 percent of all animal health products marketed in the United States are used in food-animal production; 38 percent of those products are classified as feed additives and include antibiotics; antibacterial drugs such as sulfonamides, nitrofurans, arsenical compounds, anthelmintics, and coccidiostats; and other pharmaceutical agents such as ionophores, melengesterol acetate, antioxidants, mold inhibitors, probiotics, and nonantibacterial growth promoters (Richard Carneval, AHI, personal communication 1996). The animal health industry, if asked, could not estimate the reduction in sales it would suffer nor estimate the changes in employment and profits that would occur. An alternative economic measure would be to enumerate the consequences for farm profits and farm costs. Approximately 100 percent of chickens and turkeys, 90 percent of swine and veal calves, and 60 percent of beef cattle receive diets containing antibiotic drugs during some part of their lives (Manchanda 1994). Thus, it is obvious that most producers find these products useful. One study estimated that subtherapeutic drugs saved the U.S. hog industry approximately $2 billion in annual production costs (Wade and Barkley 1992). However, changes in production costs would not necessarily translate directly into lower profits. First, the cost of the drugs themselves must be considered. For average producers, that amounts to about 3.75 percent of total ration costs, or about 50 percent of the value of the compounds to animal producers (Beran 1987). Second, not all producers rely on these compounds to the same extent. Subtherapeutic antibiotics are most effective in animals under the stress of inadequate nutrition and suboptimal sanitation (Braude et al. 1953). That means the incentive to use these compounds decreases as management practices improve. For example, pork producers who wash hoghouses every time a group of pigs is moved and who move piglets to off-site growing facilities can reduce their reliance on antibiotics (Dial et al. 1992). Thus, producers who practice good management would not be as greatly affected by a ban as producers who do not. This raises the interesting possibility that a ban on subtherapeutic drug use would actually result in an economic incentive to improve animal care and could result in a more efficient industry in the long term. However, the process required to reach that point would be painful for those producers forced out of business. Because some producers might actually benefit from a ban on subtherapeutic drug use, the estimate of costs to a typical producer could be misleading. Examples of this can be seen in the successes of specialty producers such as Colemon Natural Beef in Colorado (NRC 1989a). Colemon beef, raised without antibiotic treatments or exogenous growth promoters, costs approximately 25 percent more than conventional beef. Colemon beef is produced for a specific niche market, and farmers pass on the increased production costs to the consumer. A more viable alternative to cost estimating would be to measure costs to consumers in terms of the higher prices they would pay for meats. This alterna-

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks tive has the advantage of being reasonably representative across all consumer groups. The dollar value expressed on a per capita or per family basis is readily understandable: It is a number anyone can put into perspective. Using a consumer measure also makes sense from an economic perspective, because all changes in production costs must eventually be passed on in output prices in a competitive industry. The effects of a ban on subtherapeutic drug use might need to be offset by technological improvements to obtain equal levels of production. Therefore, costs would increase and meat prices would be higher than they were before the ban. Viable Antibiotic Substitutes The studies referenced above assume a worst case, in which the effects of a ban on production are exactly equal to the known production benefits of subtherapeutic use. The rationale for this assumption was suggested in the CAST (1981) report: “Probably most of the economists did not know all the administrative and technical alternatives but expected that the socially optimum restriction would be less than complete elimination. The analyses reviewed did not try to design a socially optimum partial restriction.” (P. 39) A more reasonable assumption would be that the responses of drug companies and the producers cannot be predicted but that some response will occur. As mentioned elsewhere in the CAST report, the response could take the form of different management practices, new products, or even genetic selection. Thus, in a different scenario, only 50 percent of the estimated effect on production is incorporated. (The worst case contains the entire production effect.) The issue of substitutes for subtherapeutic drug use also gives rise to the question of whether some producers would purchase antibiotics legally but add them illegally to the feed or water. If that occurs on a widespread basis, there would be little effect on output or prices, because illegal use would substitute for legal use. Again, the extent of such activity cannot be predicted, but it should be noted that the incentive to violate the ban would be enormous as long as therapeutic antibiotics were sold without prescription. The issue could become critically important if widespread violations forced a ban on over-the-counter sales of antibiotics. Banning subtherapeutic use of antibiotics without regulating therapeutic use might be impossible. Any attempt to regulate therapeutic use would increase production costs for the U.S. beef and pork industries, because veterinary visits would be required for each diagnosis and prescription. The costs of a ban on over-the-counter sales of antibiotics would possibly be greater than would be the costs of a ban on subtherapeutic drug use. Total versus Partial Ban As has become clear in the debate on fluoroquinolone use in animals, any new antibiotics approved for human use will be too expensive for subtherapeutic

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks use in animals. That is an elegant example of the market at work. New antibiotics are pharmacologically valuable because human pathogens have become resistant to the old ones, and by comparison new drugs are quite expensive to manufacture and purchase. The old antibiotics are inexpensive, in part, because they are less useful to humans and also because the manufacturing chemistry is simpler. To some extent, market forces will deter the use of antibiotics in animal feed if human pathogens have not yet developed resistance to these particular drugs. If regulators accept that scenario, then the group of antibiotics in current use will not be banned, and newer, more expensive ones will. In that case, the consequences of any ban would be minimal because they would occur only as animals develop increased resistance to older antibiotics. The rate at which microorganisms in food-animal populations become resistant to antibiotics is slow because of the short lifespans and high turnover of these animal populations (Walton 1986). The food-animal industry could also be expected to take additional steps to avoid multiple-drug resistance as long as the industry knew that replacements would be difficult to obtain. This market-driven solution has much to recommend it. However, the numbers presented in the economic analysis assume that all subtherapeutic use is banned. Consumer Behavior It has been argued that the best way to measure the consequences of a subtherapeutic-use ban is at the consumer level. Because information is available at the producer level, some assumptions must be made about how the extra costs would be passed on to consumers. Assumptions also must be made on the responses of consumers to higher prices and to any improvement or reduction in the quality of the meat. Attempts have been made in previous economic analyses to allow consumers to respond to higher prices by reducing consumption and to provide potential responses from retailers and meat processors (Allen and Burbee 1972; Dworkin 1976; Mann and Paulsen 1976; CAST 1981; Wade and Barkley 1992; Gilliam et al. 1993; FSIS 1995a; Office of Technology Assessment, Washington, D.C., unpublished material). To derive the effects of these responses, estimates of consumer–demand elasticity and producer–retail markups must be made. But such estimates are subjective and can vary widely among studies. A more straightforward approach is to assume that all costs are passed on to consumers and then to measure how much consumers would need to spend to maintain consumption. This measure of consumer costs will slightly overestimate the true cost (by an amount that depends on how meat consumption is affected by prices), but it has the advantage of not depending on elasticity estimates (see Layard and Walters, 1978, p. 147, Figure 5–9). If the effect of a ban were more severe, it might make more sense to build consumer response to higher prices and a production response to lower demand. This would require use of elasticity measures and would make the final results sensitive to these mea-

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks sures. There is no real consensus on the appropriate size of these elasticities, and, given the very small price effect, the committee decided that the elasticity approach would raise more questions than it answered. A second consideration is whether consumers will pay more for meat that is produced without subtherapeutic antibiotics. In the two most recent studies on this topic (Wade and Barkley 1992; Manchanda 1994), it was assumed that consumption in the United States would increase by 5 percent in response to such a ban. This assumption seems difficult to justify, because no change can be expected in the concentrations of antibiotic residues. The incidence of drug residue violations cited earlier in this report is so low that a ban on subtherapeutic drug use would be unlikely to have any detectable effect (FSIS 1995a). Consequently, the committee concluded that the correct assumption would be that no change in consumption (positive or negative) occurs. That assumption is equivalent to assuming that the positive effect of improved meat quality exactly offsets the negative effect of higher meat prices. A final part of this question is whether the marketing system itself would pass on the higher costs in terms of cost per pound or in terms of percentage price change. The latter scenario is used in all the previous studies on this topic and implicitly assumes that meat processors and retailers increase their margins on a per-pound basis in response to increases in the prices they pay. This assumption is justified, in part, because the U.S. marketing system works on a percentage-markup basis. This convention is used in the results presented below. RESULTS OF ECONOMIC ANALYSIS AND CONCLUSIONS Based on the assumptions discussed above, the committee derived the estimated economic impact of a ban on subtherapeutic use (Table 7–1). Per capita cost is estimated as follows: Per capita Costs = %C × P × Q Where %C is %increase in annual production cost, P is retail price, and Q is annual retail quantity sold per capita. The committee’s conclusion is that the average annual per capita cost to consumers of a ban on subtherapeutic drug use is $4.84 to $9.72. The effect of the ban is lowest for poultry prices and highest for beef. That cost seems small; however, assuming a U.S. population of 260 million, the total amounts to about $1.2 billion to $2.5 billion per year. Of course, the higher per capita cost means low-income consumers would spend an even larger proportion of their income on food than would high-income consumers. To determine whether the increase in cost is justified, the amount should be compared with estimated health benefits. Additional costs not included in Table 7–1 are (1) a slight erosion in U.S. export

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks TABLE 7–1 Approximate Annual Costs of a Ban on Subtherapeutic Antibiotic Use in Four Domestic Retail Markets Meat Change in Price ($/lb) Per Capita Consumption in 1997, Retail Weight (lb) Extra Cost per Capita per Year ($) Extra Cost Family of Four per Week ($) Total National Extra Cost per Yeara (million $)   Ab Bc   A B A B A B Chicken 0.013 0.026 84 1.09 2.20 0.08 0.17 283 572 Turkey 0.015 0.031 18 0.27 0.56 0.002 0.04 70 146 Beef 0.03 0.06 67 2.01 4.02 0.15 0.31 523 1,045 Pork 0.03 0.06 49 1.47 2.94 0.11 0.23 382 764 Total NAd NA 218 4.84 9.72 0.342 0.75 1,258 2,527 aCalculated based on a population of 260 million × column 5 (for scenario A), or 260 million × column 6 (for scenario B). bA = With substitutes. Scenario A assumes that substitutes mitigate the effect by 50 percent. cB = Without substitutes. dNot applicable. Source: Columns 2 and 3: CAST 1981. Numbers are based on the quantity effects presented in Tables 24 and 25 (that is, a 2.052 percent change in production costs was used) of the CAST report; Column 4: FAPRI 1998; Columns 5–8: Calculated from Columns 2–4.

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks competitiveness; (2) the personal and financial costs of producers forced out of business; (3) the lower profits and revenues of the companies that manufacture these compounds; and (4) the additional costs that would occur in markets for eggs, dairy, and pet-food, which are not discussed here. The values in Table 7–1 also ignore the possibility that a subtherapeutic-use ban eventually would lead to restrictions on over-the-counter antibiotic sales. There has not been any previous attempt in the literature to estimate the consequences of an economy-wide ban of subtherapeutic antibiotic use on consumers. Four studies have focused on the economic effects of a ban (Burpee et al. 1978; Wade and Barkley 1992; Gilliam et. al. 1993; Machanda 1994). Those studies present estimates of production cost increases of 4 to 20 percent. Only 2 of the studies (Wade and Barkley 1992; Machanda 1994) attempted to estimate the effect of such a ban on consumers. They were specific to the pork sector and both calculated retail price increases of $0.04 per pound, which is within the range of $0.03 to $0.06 shown in Table 7–1. A more difficult task would be to estimate the effect of such a ban on the development of new animal drugs by the animal health industry. For example, the animal health industry invested $381 million in research and development: $355 million was spent for internal research and $26 million was invested in external research, primarily at universities. Seventeen percent of the total research and development investment was allocated to feed additives (Richard Carneval, AHI, 1996, personal communication). The reduction in profits and industry confidence that could occur after such a ban would cause a reduction in research and society would lose the research benefits. Although that loss might well be one of the most important consequences of such a ban, it is impossible to put a monetary value on future research, in part because no one knows what drugs would be developed or approved. Because a value cannot be placed on animal drug research, the associated costs are not discussed. That omission means the numbers provided here underestimate the true costs of a ban. APPENDIX Technical Notes for Table 7–1 Chicken Data On the basis of personal communication with Jerry Sell (Iowa State University, 1997), it was assumed that poultry feed conversion efficiency (FCE) changes from 1.85 tons of grain per ton of meat to 1.90 tons of grain per ton of meat, a 2.7 percent increase. This would represent a 1.76 percent increase in total production costs because feed represents 65 percent of producers’ total costs. To calculate the expected effect in the scenario without substitutes, this 1.76 percent was multiplied by the 1997 retail price of chicken ( $1.46 per pound) to arrive at cost

OCR for page 179
The Use of Drugs in Food Animals: Benefits and Risks of 2.6 cents per pound. The scenario with substitutes is set equal to one-half of that value. The key point of this substitute scenario is that some substitution will inevitably occur and it will diminish the effect of a ban. Because these substitutions will occur in the future, there is no accurate way to know what they or their likely magnitude will be. The committee used a value of one-half as a crude estimate of the likely effect of substitution. Turkey Data For turkeys FCE was assumed to change from 1.68 to 1.75 tons of feed per ton of meat, a 4.2 percent increase. Feed was assumed to represent 70 percent of total production costs. The total cost increase was calculated at 2.94 percent. The 1997 turkey price of $1.05 per pound was then used to calculate a 3.1-cent-per-pound increase in the scenario without substitutes. The value with substitutes was arbitrarily assumed to be one-half of that value. Beef Data Personal communications with Richard Cowman (nutrition expert at the National Cattlemen’s Beef Association, 1995) indicated that the consequence of a ban would be an increase of $0.06 per pound in the price of beef. However, this expert did not consider that these particular uses stated were subtherapeutic, because the treatments were for preventing liver abscesses and stress-related diseases and, therefore, suggested a zero value. The retail price used was $2.80 per pound to derive a no-substitutes value of 6 cents per pound. The scenario with substitutes was one-half of that value. The analysis assumes that only 60 percent of all beef animals are affected by such a ban. Pork Data The pork data are taken from the Pork Industry Handbook (1996). The data showed a change in FCE of 6.5 percent for the first 40 pounds of gain and a change of 3.18 percent for the remaining 145 pounds of gain. The ration costs and FCE for young pigs were $150 per ton and 2.04, respectively. The values for fattening were $120 and 3.0, respectively. These 4 values were used to weigh the changes in feed conversions. The weights were calculated to be 4.3:1. Thus, the 6.5 percent change in FCE in the starter ration came to approximately 19 percent of total ration costs. The 3 percent change was added to the remaining 81 percent. The total change in FCE was, therefore, calculated at 3.6 percent. Assuming that ration costs equal 70 percent of total production costs, the total change in retail prices represent a 2.5 percent increase. The retail price used for 1997 was $2.30 per pound to arrive at a 6-cent-per-pound increase in the scenario without substitutes.