A Case Study of Escherichia Coli Testing of Beef Trim

Food Control 38 (2014) 227e234

Contents lists available at ScienceDirect

Food Control

journal homepage: www.elsevier.com/locate/foodcont

Review The limits of food safety testing: A case study of Escherichia coli testing of beef trim

Peyton Ferrier*, Jean C. Buzby

Economic Research Service, USDA, Markets Trade and Economics Division, 1400 Independence Ave., SW, Mailstop 1800, Washington, DC 20520-1800, USA article info abstract

Article history: Beef trim, the primary component of ground beef, is tested by both the Food Safety and Inspection Received 15 February 2013 Service (FSIS) and meat packers for Escherichia coli O157:H7 under a zero tolerance standard. We Received in revised form compare and contrast the process control and filtering roles of this testing on costs and food safety 19 September 2013 outcomes. Both of these roles of testing have been alternatively emphasized in recent legislative pro- Accepted 19 October 2013 posals and policy statements on beef trim testing. In the process control role, test design e including the decisions to increase sample sizes, re-test lots, and change tolerance thresholds e only affects the cost of Keywords: implementing a targeted food safety standard, but does not directly affect health outcomes. In contrast, Escherichia coli O157:H7 fi fl Testing in the ltering role, test design in uences the likelihood of errors and directly affects health outcomes. Beef trim Neither role eliminates all risk from this pathogen. More broadly, we discuss the incentives food pro- Process control ducers face to increase their frequency of testing and improve food safety processes in response to Filtering positive test findings as the process control role emphasizes. FSIS Published by Elsevier Ltd.

Contents

1. Introduction ...... 227 2. The concern over E. coli O157:H7 and its presence in beef ...... 228 3. Testing for E. coli O157:H7...... 228 4. The dual role of testing: process control and filtering ...... 230 4.1. The process control role of testing ...... 230 4.2. The filtering role of testing ...... 231 5. Conclusion ...... 233 Disclaimer ...... 233 References...... 233

1. Introduction imports for adulterants relatively infrequently and a failed FDA test often leads to disruptive, increased future testing or placement on There are two broad roles of testing a product for a food safety an import alert, thereby incentivizing importers to undertake risk: process control and ﬁltering. In the United States, these two additional food safety measures. In the ﬁltering role, testing in- roles of testing are used in different current and proposed regula- tercepts products of concern, such as adulterated food that may tions and applications. In the process control role, testing measures sicken consumers if ingested. Another example is that the U.S. the effectiveness of a plant’s production process and provides in- Department of Agriculture’s Animal and Plant Health Inspection formation for adjusting the process following a failed test. For Service (USDA/APHIS) inspects most fresh agricultural imports for instance, the U.S. Food and Drug Administration (FDA) tests food potentially invasive quarantine pests. In these cases, a failed test may create only a relatively small cost to treat or recondition the commodity and have little effect on importer behavior. * Corresponding author. Tel.: þ1 202 694 5224. We extend the literature by contrasting these two roles of E-mail address: [email protected] (P. Ferrier). testing through a case study on the testing of beef trim for

0956-7135/$ e see front matter Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.foodcont.2013.10.028 228 P. Ferrier, J.C. Buzby / Food Control 38 (2014) 227e234

Escherichia coli O157:H7. This paper contributes to the literature by in spurring food safety legislation in the United States (Loader & presenting the dichotomy on how a change in the test design (e.g., Hobbs, 1999). Since 1994 when FSIS declared it an adulterant in increased testing) affects testing costs, incentives to packing plants beef trim, E. coli O157:H7 has been a primary pathogen targeted in to adopt additional food safety investments, and public health FSIS’ testing of beef trim for human consumption (Food Safety and outcomes differently, through its roles as a process control or a Inspection Service, 2011a, 2011b). This designation of E. coli filter. As we shall see later in this article, this information is O157:H7 as an adulterant only applies to non-intact beef cuts, important for policymaking. including beef trim later made into ground beef, but not to whole The characteristics of this pathogen make it a particularly severe cuts which include steaks or roasts (Alvares, Lim, & Green, 2008). food safety hazard. As part of the Federal Meat Inspection Act This designation provides FSIS with the legal authority to seize (FMIA), the USDA’s Food Safety and Inspection Service (USDA’s products in commerce shown to contain E. coli O157:H7 and later led FSIS) regularly tests beef trim (i.e., the primary component of to mandated food safety and quality control measures, such as FSIS ground beef) for E. coli O157:H7 to help ensure that the production testing and the requirement that all Federally-inspected meat process outlined in the packing plant’s Hazard Analysis and Critical slaughter and processing plants maintain HACCP plans. Control Points (HACCP) plan is under control. Using the FSIS test More recently in September 2011, FSIS announced that six protocol, meat packers may also voluntarily test beef trim to filter additional serotypes of Shiga toxin producing E. coli (STECs) (i.e., out adulterated products, which in turn helps limit their liability strains carrying the anti-bodies O26, O45, O103, O111, O121, and exposure, reduce recall costs, and meet customer requirements. In O145) were also adulterants and gave plants notice that FSIS testing either case, a lot that tests positive for the pathogen may not be sold and plant HACCP plans needed to address for the possible presence to consumers unless it undergoes an FSIS-approved process, typi- of these pathogens as well (Food Safety and Inspection Service, cally cooking, that mitigates the food safety risk. 2011a, 2011b). The focus of the remainder of this paper is primar- Because no level of E. coli O157:H7 is permissible in beef trim, ily on E. coli O157:H7. testing constitutes a “zero tolerance” standard. However, passing The Shiga toxin produced by E. coli O157:H7 can cause acute and this test does not imply that beef trim is entirely free of pathogens chronic illness when it is ingested in contaminated food or spread because: by other routes, such as person-to-person contact. Beef is a common food vehicle for this pathogen. The U.S. Centers for Disease (1) the test is not necessarily free of errors (e.g., testing some- Control and Prevention estimates that 63,153 domestically- times fails to detect the pathogen when it is present in the acquired foodborne illnesses are associated with E. coli O157:H7 sample), each year (Mead et al., 1999, Scallan et al., 2011) which imposes an (2) the test’s sample may not contain E. coli O157:H7 even when estimated $219.9 million in annual health costs according to the the lot is contaminated, or USDA’s Economic Research Service (Economic Research Service, (3) other pathogens may be present for which the test is not 2010; Frenzen, Drake, Angulo, & The Emerging Infections Program designed to detect. Foodnet Working Group, 2005; Scallan et al., 2011). Because E. coli O157:H7 lives harmlessly in the gastrointestinal tracts of While the first concern is primarily technical, the second cattle and can easily be transferred to the carcass during slaughter, concern is addressed by recent proposals that would increase it is difficult to completely eliminate it from beef trim (Naugle, Holt, sample sizes. The third concern is addressed, in part, by the recent Levine, & Eckel, 2005). Box 1 provides greater detail on E. coli classification of additional six Shiga toxin producing E. coli (STECs) O157:H7 in the slaughter process. serotypes as adulterants. In recent years, legislative proposals in the Following fabrication, FSIS or packers may test beef trim for E. coli United States, such as the proposed 2010 E. coli Traceability and O157:H7 before it is used for grinding. If trim tests positive or, as a Eradication Act, would have made testing mandatory and matter of semantics, non-negative, then FSIS requires it to be treated required packing plants and processors to increase the size of tests, or re-conditioned with a kill step to mitigate the pathogen risk so reducing the likelihood of testing errors. These recent proposals that the meat can be sold for human consumption. Given the lack of and the new classification of adulterants reflect the current and consumer acceptance of irradiation, contaminated beef trim is heightened attention paid to this issue by the scientific community typically sold to a facility that cooks and sells it for uses such as pizza and policymakers in the United States. This issue is also relevant to topping or sausage. If FSIS confirms that the product is adulterated other developed countries with similar food safety regulations and with E. coli O157:H7, it will investigate what actions on the part of the approaches by packing plants as well to countries where beef is packer may have led to the adulteration so that the production important in national consumption (e.g., Canada). process can be brought back under control. It will also perform more As background, we first describe E. coli O157:H7 and its public frequent follow-up tests following a positive test. If the process is health consequences, the avenues by which beef trim can become judged to not be in control, FSIS can request changes to the packer’s contaminated with this pathogen during processing, the beef trim HACCP plan or withdraw inspection services, which effectively shuts testing process, and the incentives faced by producers to test their down production and no further beef trim can be produced and sold. product regularly. In this case study, we summarize how both the process control and filtering roles of testing have been the focus of 3. Testing for E. coli O157:H7 economic inquiryand public policyconcern and we highlight the most relevant literature on the role of pathogen testing in the meat industry. The frequency that FSIS tests of beef trim for E. coli O157:H7 depends on the plant’s output.1 For very large plants producing 2. The concern over E. coli O157:H7 and its presence in beef more than 250,000 lbs daily, FSIS guidelines require more than one

The bacterium E. coli, abbreviated E. coli, is ubiquitous but only speciﬁc serotypes of the bacteria are harmful to humans. The 1 Plants also conduct a separate test of the external surfaces of beef carcasses for serotype E. coli O157:H7 (so named because it carries the O157 and generic E. coli and salmonella as mandated by FSIS for approval of their HACCP plan. However, these tests do not have zero tolerance and a positive test does not pre- H7 antibodies) has been the predominant focus of public policy clude the carcass from being sold for fresh consumption because generic E. coli is attention since it was linked to the 1993 Jack-in-the-Box outbreak, not necessarily pathogenic. In testing, generic E. coli serves as an indicator organism which caused over 300 illness and 4 deaths and marked a watershed for fecal contamination. P. Ferrier, J.C. Buzby / Food Control 38 (2014) 227e234 229

Box 1 E. coli O157:H7 in the slaughter process. E. coli O157:H7 prevalence on the animals that comprise the source material of beef trim. However, packers control animals for far too short of a period (typically less than a day) While a serious threat to human health, Escherichia coli to affect pathogen loads through feeding or housing O157:H7 does not affect the health of beef cattle or create practices. any visible symptoms. Living and reproducing in the ani- Importantly, none of the investments or actions previously mal’s alimentary track, the pathogen is ubiquitous in cattle, discussed are considered “kill” steps e actions that, unto although individual animals differ in terms of concentra- themselves, entirely mitigate the food safety risk. Rather, tions of the pathogen shed in fecal matter. These shed rates these sequential interventions are designed to reduce the are seasonal, being highest in late summer months, and pathogen to low or imperceptible levels so that consumer episodic, reaching very high concentration levels in some cooking mitigates the remaining hazard. Packer level kill cattle (i.e., super-shedders) (Arthur, Bosilevac, Nou, & steps include cooking or irradiating the final product. In the Koohmaraie, 2005, Arthur et al., 2010, Matthews et al., case of Lean Finely Textured Beef (LFTB), treatment with 2006). E. coli O157:H7 survives most stages of the beef gaseous ammonium hydroxide reduces, but does not supply chain, including freezing, and it can grow and eliminate, E. coli O157:H7 if it is present in the beef trim multiply slowly at temperatures as low as 44 F(Food and (Greene, 2012). Because irradiation is controversial and can Drug Administration, 1994). However, the pathogen is reduce consumer demand, it has not been adopted widely easily killed when cooked to at least 160 F. (Ferrier, 2010; Hayes, Fox, & Shogren, 2002). Alternatively, Beef slaughter and packing has been divided into distinct cooking limits the range of potential end uses and the steps which are, sequentially: loading and stunning, de- marketability of the meat. And, importantly, cook-only beef hiding, first stage decontamination, evisceration, second trim sells at a price discount to fresh beef trim. stage decontamination, chilling and fabrication. De-hiding and evisceration are the stages of slaughter with the greatest likelihood of transferring the E. coli O157:H7 test a month (Food Safety and Inspection Service, 2008). For large pathogen to meat, described in detail by FSIS (Food Safety plants producing between 50,000 and 250,000 lbs daily, one test and Inspection Service, 2001; Koohmaraie et al., 2007, per month is required. For medium plants producing between Koohmaraie et al., 2005). During de-hiding, the pathogen 1,000 pounds and 50,000 lbs per week, one test is required every can be transmitted to the carcass from hides that have been two months.2 The smallest plants are tested only once every three fouled with mud and feces. During evisceration, the months. gastrointestinal tract of the animal may become perforated, Packers, however, may voluntarily test their product more exposing the carcass to the contents of the colon. Carcass contamination is particularly problematic for beef trim as frequently, with some plants testing all lots, or in some instances, this material is often collected from exterior sections of the all combo bins containing the trim (Ollinger, Muth, Karns, & Choice, carcass, including the fat cover near the hide, rather than 2011). A lot is a production unit of beef trim presented for testing. from whole cuts. Although for small packers, a lot may include as little as 50 pounds of trim, ordinarily a lot consists of one to five combo bins, which Preventative measures and decontamination steps inhibit the growth of microorganisms which, along with testing each weigh approximately one ton. and other actions, create a hurdle process that enhances the When either FSIS or packers test beef trim, they follow the effect of individual measures (McMeekin, Olley, Ratkowsky, standard FSIS-determined protocol for sampling referred to as “n- & Ross, 2002). Packers can invest in equipment that pas- 60” testing, which takes a sample of 60 pieces of trim weighing a teurizes the carcass (e.g., with steam or hot water) or total of 375 g from the lot and then tests the sample collectively for removes spot contamination (e.g., steam vacuuming in the the presence of targeted pathogens. In practice, this involves de-hiding stage). They can also wash carcasses with manually removing the pieces of trim and cutting a sliver off it with organic acid during evisceration. Equipment can de-hide a knife or using a mechanical sampler.3 Later in a laboratory, the the animal mechanically in a less hazardous manner than sample is enriched in a growth medium for 8e12 h to make the manual de-hiding. For instance, equipment is available that creates negative pressure around the hide so that the pathogen easier to detect. N-60 testing is the most rigorous testing pathogen is not aerosolized during de-hiding and spread to protocol considered by the International Commission for the other parts of the carcass. Additionally, packers can provide Microbiological Specifications of Food (International Commission instruction and incentives for workers to maintain best for the Microbiological Specifications of Foods, 2002). practices. These include: regular sanitization of knives, Because sampling does not draw from the entire lot of beef trim, gloves, uniforms, handling equipment, and work areas; it will necessarily fail to detect E. coli O157:H7 when it is present in proper instruction on how to de-hide and eviscerate ani- a certain percentage of cases. More specifically, probability theory mals; maintenance of proper temperature controls in shows that for a particular lot of beef trim, there is a 95% probability plants; and control steps in addressing contamination when of drawing a contaminated sample when pieces of trim are it occurs. Regular testing of beef trim for E. coli O157:H7 contaminated at the 5% prevalence rate. If either the sample size is also reduces the potential for food safety risks. larger than 60 or the prevalence rate is larger than 5%, the Although FSIS recommends five practices to reduce E. coli O157:H7 prevalence in cattle herds, these practices have not been explicitly linked to contamination and food safety outcomes (Arthur et al., 2009; Food Safety and Inspection 2 In general, the beef packing industry is very concentrated, with the top four Service, 2010a, 2010b, 2010c; Koohmaraie et al., 2007; firms slaughtering more than 80% of national output (Lamb & Beshear, 1998). The Loneragan & Brashears, 2005; Sargeant, Amezcua, Rajic, processing and retailing sectors have also become more concentrated and vertically & Waddell, 2007). These practices include maintaining coordinated as noted by Barkema, Drabenstott, and Novack (2001). 3 clean water and feed, keeping living environments well- Some mechanical samplers may test more than 60 samples and provide for a drained, separating calves and heifers in housing or more uniform sampling process. They only need to be validated by FSIS to sample at least 60 samples. One mechanical sampler resembles a long drill which collects 12 reducing animal density, and maintaining biosecurity to samples at its tip when inserted into the combo bin at five separate spots. In this exclude wildlife that might be carriers of this and other way, it can reach the bottom of the combo bin in a way that is impractical with pathogens. Ideally, better cattle practices would lower manual sampling. 230 P. Ferrier, J.C. Buzby / Food Control 38 (2014) 227e234 probability of drawing, and therefore detecting, a contaminated implicated by the test and also considered ‘adulterated’. Packers piece rises. Ironically, if the rate of contamination is less than 5% may test trim regularly to limit the amount of volume they could (but still positive), the probability that sampling fails to detect the potentially be forced to sell on the lower-priced, cook-only market. pathogen rises. Critics of the n-60 protocol argue that the protocol Similarly, if E. coli O157:H7 is found in ground beef, other ground samples too little material to detect contamination (Healy, 2009; beef produced from the same production run may also be consid- Office of Inspector General, 2011; Safe Tables Our Priority, 2008). ered adulterated by FSIS. Again, regular testing limits the size of the The testing method (i.e., specific lab test or technology) used by potential recall and associated economic losses. packers, however, filters or screens for E. coli O157:H7 but is not intended to provide confirmatory evidence of its presence. Con- 4. The dual role of testing: process control and filtering firming this is expensive and firms are not legally required to do it. And, without confirmation of the pathogen’s presence, producers This section presents the key literature on the role of pathogen may have limited incentives to address food safety concerns in the testing in the meat industry. It also extends the literature by production process. Subsequently, firms may simply re-route comparing and contrasting the two roles of testing beef trim for screened positive beef trim to cook only uses. In this case, testing E. coli O157:H7. improves food safety predominately by filtering out adulterated product, rather than through changes in the production practices. 4.1. The process control role of testing In contrast to the concern that too-small sample sizes fail to detect contamination when it is present, alternative FSIS-validated Since 2000, FSIS has required federally-inspected meat and testing methods vary in their likelihood of having the opposite poultry slaughter and processing plants to have HACCP plans to error e to detect contamination when it is not present. In general, ensure that a plant’s food safety process is in control (Unnevehr & quicker less expensive methods, such as the lateral Flow (LF) Jensen, 1999). Typically, these HACCP plans include both preven- method are more likely to have a false positive than longer more tative measures to reduce the occurrence of pathogens in finished expensive methods, such as the Polymerase Chain Reaction (PCR) products and testing of intermediate and final products. Within method (see Ferrier & Buzby, 2013). FSIS recognizes that the PCR these plans, testing is intended to determine whether these food test is less prone to error than the LF test and permits product that safety processes are effective, rather than to filter or “catch” test positive for E. coli O157:H7 in an LF test to be sold as fresh if it contaminated products (Ollinger et al., 2011). subsequently passes a PCR test. From a regulatory standpoint, As previously mentioned, FSIS expanded testing of beef trim to however, both the LF and the PCR tests are screening tests (Food the six non-O157 STECs and in a 2011 Federal Register posting, Safety and Inspection Service, 2010a, 2010b, 2010c) where a still reiterated the text of a 1994 speech by then-FSIS Administrator more lengthy and rigorous test is necessary to isolate and confirm Michael R. Taylor that: the presence of the pathogen (Food Safety and Inspection Service, To clarify an important legal point, we consider raw ground beef 2010a, 2010b, 2010c). When FSIS itself conducts tests of beef trim that is contaminated with E. coli O157:H7 to be adulterated within lots, each initial positive screening test is followed by a confirma- the meaning of the [Federal Meat Inspection Act]. We are prepared tory test. Packers, however, may choose to act differently on the to use the Act’s enforcement tools, as necessary, to exclude adul- results of a positive screening test and not follow them up with terated product from commerce.We plan to conduct targeted confirmatory tests. Because screening tests do not conclusively sampling and testing of raw ground beef at plants and in the prove that E. coli O157:H7 is present, a positive screen test does not marketplace for possible contamination. We know that the ulti- obligate packers to correct potential flaws in the production process mate solution to the (E. coli) O157:H7 problem lies not in under HACCP as there is no direct evidence that E. coli O157:H7 is comprehensive end-product testing but rather in the development present.4 and implementation of science-based prevention controls, with The specific costs of testing are difficult to verify. Based on product testing to verify process controls.” communications with firms offering testing services and industry representation, Ferrier and Buzby (2013) estimate that physical costs to sampling and testing at the n-60 level in an LF test is $30 The economic literature provides some discussion of the process per test and that the less frequently performed PCR test is $60. control role of testing. In particular, Ollinger and Moore (2009) and Embedded in any testing regime are the additional costs of inter- Starbird (2005) consider how testing influences process control and rupting the supply chain, maintaining traceability of lots, and the provision of food safety in the beef industry. Ollinger et al. sorting and refrigerating lots while waiting for test results. This last (2011) argue that the potential for recalls and failed safety tests process is often referred to as “test and hold” although the actual creates reputation effects. Moreover, a significant portion of the “holding” of lots on site is not actually necessary. Instead, packers firm’s value is intangible and subject to loss if the firm becomes must simply maintain control of the lot and delay its final pro- associated with negative food safety events. For example, Food- cessing, while still shipping it to a distant destination. Lots that test maker Inc. [the parent company of Jack in the Box Inc.] lost an positive in screening tests will be held even longer if confirmatory estimated $160 million in the first 18 months after the 1993 E. coli tests are performed. O157:H7 outbreak (Roberts, Morales, Lin, Caswell, & Hooker, 1997). The current U.S. regulatory structure supports the homogeni- A processor whose product is recalled or implicated in a food safety zation of testing practices across packers. FSIS generally accepts the lawsuit may discontinue all business with the packer that supplied packer’s test as evidence that a production process is in control if the trim or make future purchases contingent upon the packer the testing and sampling method is as rigorous as the FSIS method. undertaking additional food safety investments. Testing outcomes If a lot of beef trim tests positive and is confirmed to contain E. coli play a similar role in establishing reputation effects, and some O157:H7, other lots produced in the production run may be processors may request that suppliers of beef trim test all their output. This role of processors making purchases contingent upon food safety investments may be linked to contractual ties or vertical integration ties that bind packers and processors in long-run re- 4 In projecting the costs to expanding E. coli O157:H7 testing to include the six other STECs in the future, FSIS assumed that the screen positive sample rate is 2% lationships. This view stands in contrast to popular press accounts while the confirmed positive sample rate is .5% (FSIS, 2011a, 2011b). of large packers with market power pressuring small processors to P. Ferrier, J.C. Buzby / Food Control 38 (2014) 227e234 231 not re-test beef trim before processing, presumably for fear of Table 1 immediate revenue loss (Moss, 2009). Firm tendency to undertake food safety investments and testing by plant size. In general, the beef packing industry is very concentrated, but the Plant size quintile All fi plants scale of the rm can vary dramatically from plant to plant. Large 0e19 20e39 40e59 60e79 80e99 packers can slaughter 12,000 cattle a day on the high-end; small Output (1,000 pounds of output) 28 158 508 3,329 317,592 plants may slaughter only a dozen or so animals a week. Table 1 Number of Plants Surveyed 50 52 50 51 51 254 fi shows how different sized rms undertake food safety investments Overall Index (Mean Index .32 .32 .35 .37 .54 .38 (based on their quintile distribution of annual output). In this work, Values) Ollinger et al. (2011) develop several index values for food safety Dehiding Index .43 .42 .44 .46 .62 .47 measures, ranging from the least rigorous as 0 to the most rigorous as Sanitation Index .55 .55 .65 .65 .71 .62 fi ’ Operations Index .44 .44 .46 .45 .57 .48 1, that capture the rms food safety investments in critical areas Equipment Index 10 .13 .15 .18 .51 .22 including de-hiding, sanitation, operations, testing, and equipment. Testing Index .05 .05 .05 .08 .30 .11 For example, a firm can invest in equipment that removes the hide in Note: Mean Index Values range from 0 to 1 where 1 represents the safest value of a way that reduces the chance that it contaminates the carcass. each index. Ollinger et al. (2011) find that larger firms tend to invest more in Source: Ollinger et al. (2011). measurable food safety equipment, testing, and employee training than smaller firms and offer several explanations. First, larger firms provide safety assurance. To the extent that more frequent testing face a larger liability and negative media exposure when a food safety reduces the risk faced by agents, it also lowers the contracting cost event results in a lawsuit or lost business. This larger downside risk to implement a desired food safety objective. spurs greater food safety investment. Second, scale economies make In practical applications, packers (as principals) may find it difficult 5 the per-unit costs of food safety investment cheaper. Third, to the to efficiently devise contracts that incentivize food safety actions for extent that they sell to large purchasers under contract, large plants line workers (as agents) for several reasons. First, the benefittoprin- may face greater buyer reprisal if their food is found to be unsafe, cipals of avoiding a food safety event e in terms of size, costs, duration, unlike a firm that has many alternative sales outlets. For instance, a or effect on brand reputation e may be unknown. Second, the agent’s 2001 survey by the Economic Research Service and the University of costof effort may be variable (depending on experience), idiosyncratic Washington found that 60% packers test raw or cooked product for (depending on the pathogen loads on carcasses), or jointly determined E. coli O157:H7 or Listeria over and above that which is required under (depending on several links in the supply chain). Determining a the Pathogen Reduction/HACCP Rule (Economic Research Service, compensation rate that provides an incentive for increased food safety 2004). The share of ground beef that is tested regularly is likely to be effort may be challenging in these circumstances. Third, the distinc- even higher because testing is positively correlated with a plant’ssize tion between principal and agent may be unclear. Contractual in- and whether the plant is part of a multi-plant operation (Ollinger et al., centives may come from buyers (i.e.,processorsofbeeftrim),inwhich 2011). case, packing plant managers may be considered the agents whose Conversely, economists (Starbird, 2000, Starbird, 2005; Starbird effort is their capacity to direct the actions of line workers. Moreover, if & Amanor-Boadu, 2006) have analyzed the design of firm in- processors wish to gain greater control over the production process, centives to undertake specific food safety actions in a principal- they can vertically integrate with the packing process to more directly agent framework. In this approach, the principal may be repre- influence workers’ actions. Hennessy (1996) argues that such vertical sented by the processor or the packer while the agent may be linkages remove a portion of the information asymmetry problem that represented by the line workers or plant managers where the leads firms to underinvest in food quality or safety improvements. principal contracts with an agent to exert effort that reduces the probability that a lot is contaminated. For example, agents (i.e., line 4.2. The filtering role of testing workers at a packing plant) can clean knives more frequently than others or may make more careful cuts. The principal, however, By diverting beef trim to cook-only uses, testing also performs a cannot easily observe whether the agent exerts effort, but only filtering role that improves food safety. Draft guidelines from the observe the outcomes of food safety tests after the production FSIS (2008) express concern that testing is only used for filtering ’ process occurs. The principal, therefore, makes the agent s and recommend, but do not require, producers to alter production compensation dependent on the outcome of the food safety tests. practices based on positive screen tests stating: If the agent exerts effort, the probability that a lot passes a food safety test increases. However, even when the agent exerts effort, For each positive result, there should be an investigation of its the probability that a lot passes a food safety test contains some cause. Once a possible cause is identified, then appropriate action randomness because testing may fail to uncover contamination. In should be taken to make corrections and to eliminate the cause. addition to the cost of effort, the agent may demand a risk premium Doing so will bring a steady decline in the percentage of positive which increases the cost to the principal of targeting any specific results. Feedback of sampling and testing results to the supplier of food safety objective. Contracting for desired food safety objective the source materials should be provided as a matter of good is less costly if the test for contamination in lots contains less manufacturing practice. Importantly, there should be an affirma- randomness as it increases the correlation of testing outcomes and tive following-up on each and every positive test result with the agent actions. Reducing the error associated with testing, however, supplier, whether you produced the trimmings or procured the is expensive if it requires more frequent testing, increased sam- source materials from another supplier. Without reporting, the pling, or re-testing. As Ollinger et al. (2011) shows, frequent testing sampling and testing program is merely a “test-and-divert” pro- is more common with larger producers who may have a lower gram. Test-and-divert programs will not prevent, eliminate, or average cost of testing or may face greater demand side pressure to reduce to a non-detectable level E. coli O157:H7 in raw beef. and Optimally, every production lot should be sampled and tested 5 Hooker, Nagaya and Seibert (2002) similarly find that food safety investment (in terms of the cost of implementing HACCP requirements) are more expensive on before leaving the supplier and again before use at the receiver. a per-unit basis for smaller plants. Results of the testing program should be conveyed back to supplier 232 P. Ferrier, J.C. Buzby / Food Control 38 (2014) 227e234

Table 2 Comparison of process control and ﬁltering roles in beef trim testing for E. coli O157:H7.

Process control role Filtering roles

Incentives to adopt additional Principal determined, depending on the desired Market determined, depending on recall and liability food safety investments food safety objective. costs and relative value of products. Effect on pathogen risks from Reduction in other non-O157 pathogen risk because No reduction in other non-O157 pathogens unless non-O157 pathogens. improved process controls likely affects both set of pathogens. the presence of both sets of pathogens is correlated. Effect of frequency of testing Smaller, a higher frequency of testing only has an effect Larger, a higher frequency of testing uncovers on health risk if the production process is out of control. more contaminated lots. Effects of increasing sample Indirect, smaller errors only reduce the uncertainty Direct, smaller errors lessen the chance that size and lower error on health risk associated with contracting for a food safety investment. contaminated trim reaches the market.

in order for the supplier to assess the adequacy of its slaughter and face liability and recall costs and the loss of brand reputation if the dressing program and the sampling and testing program for product is sold as a fresh product to consumers. If these costs and trimmings. the markdown on cook-only products are small, however, positive tests may create relatively small incentives for packers to make Because “n-60” testing already has a zero tolerance threshold, food safety investments.6 many proposals to increase the rigorousness of testing require Ferrier and Buzby (2013) estimate the economically optimal some combination of more tests to be performed, more sampling sampling rates for E. coli O157:H7 when it acts primarily as a within each test (i.e., increase sample size), and re-testing. The filtering mechanism, rather than a process control mechanism. In E. Coli Traceability and Eradication Act (2010), if implemented, would their model, testing via sampling improves food safety by diverting have increased testing by: requiring packers to test all lots of beef contaminated products to be re-conditioned, but does not provide trim, requiring processors to test all beef trim received (unless they incentives to invest more in food safety improvements. Increasing are the same entity as the packers), and mandating regular scien- sampling also increases testing costs. They find that, in most cases, tific review of testing standards. The proposed act, however, would the economically efficient sample size, equating the marginal also have obligated packers and processors to test beef trim and benefits and costs of further sampling, should be increased above report the results to the FSIS within a day of learning the results. the current level of 60, but not above 120, the level consistent with While both these documents call for more testing, they also re-testing. For instance, they use simulations to show that the encourage packers and processors to undertake process control economically efficient sample size is 94 when a lot weighs one measures (by reporting positive tests) rather than simply filtering metric ton, a test (of 60 samples) costs approximately $35, and the products out of the fresh supply chain. contamination rate of lots is 8%. Using a similar method but In practice, the rigor of testing may be increased in a variety of different assumptions on the modeling of risk, Powell (2013),in ways. Packers can test all lots or even all combo bins, processors can contrast, finds that current test size of 60 is efficient. Both works, test lots before grinding, or tests could require larger sample sizes. however, show that when lots are larger and expected contami- As previously mentioned, Moss (2009) reports that processors nation rates are high, the optimal sample size is higher and vice sometimes face pressure from packers not to test lots for fear that it versa. Importantly, in the filtering framework, increased testing is might increase recall costs. Relatedly, a 2011 Office of Inspector oriented towards intercepting lots contaminated at low, but still General report (2011) notes that: positive, prevalence rates. “Dirtier” lots where the prevalence of FSIS’ sampling dilemma is that given the likely low occurrence of E. contamination is high (i.e., most sampled pieces of beef trim will coli O157:H7 in U.S. beef trim, FSIS must collect more than 60 pieces contain E. coli O157:H7) are easier to detect with sampling and so before it can attain the precision that is reasonable for food safety less sampling may be needed. purposes. With just 60 sample pieces, FSIS is currently more likely FSIS’ concern that testing is simply used as a filtering process not to find E. coli O157:H7 than it is to find it in a sample of product may not be limited to the sampling error problem alone. As a with, for example, 1 percent contamination. The problems with N- filtering mechanism, testing has a narrow band problem in that it 60 occurred because FSIS, from the outset, did not design its sam- only detects the specific pathogens for which it is testing. While pling program to achieve a specific level of statistical precision. the inclusion of other pathogens within the test reduces this problem, it will also increase the costs of testing, particularly if it The report also notes that: increases the likelihood of errors in screening process. As a process Plants are also not required to inform FSIS inspectors of positive test control mechanism, testing leads to corrections of the production results, but they are required to make the test results available for process which addresses all pathogens simultaneously. If the effect the inspector’s review. FSIS inspectors are required to examine the of food safety actions on pathogen risk is correlated across path- test results at least weekly to determine whether plants took ogens, then testing for additional pathogens may have relatively appropriate corrective actions. little additional effect as a process control but a substantial effect as a filter. Outside of FSIS testing, firms are not required to test trim Ollinger (2013) illustrates how filtering may occur in his ex- themselves. However, if firms do test trim, the magnitude of the amination of the testing of beef trim for Salmonella by the National financial penalties they face for producing a product that tests School Lunch Program (NSLP). The NSLP only purchases beef trim positive for E. coli O157:H7 are unclear. Within the principal-agent from approved facilities and that meet a higher tested, safety framework associated with the process control role of testing, a standard than for otherwise commercially acceptable beef trim. principal creates incentives for workers to expend effort to improve While hard data is not available, purchases into the program are food safety. Within the filtering role of testing, incentives for food safety improvement emerge from market prices where the incentives for improving food safety are likely to be weaker because of the lack of a coordinated relationship between that particular 6 Unfortunately, there is no publically available data on markdown between fresh buyer and seller. If a lot is contaminated, packers receive lower and cook-only beef, although Powell (2013) reports that the markdown is about 14% prices if the product is sold on the cook-only market or potentially ($.33 cents on a $2.27 per lb fresh price). P. Ferrier, J.C. Buzby / Food Control 38 (2014) 227e234 233 assumed to be associated with a price premium as well as greater References penalties if a product is found to be contaminated in the test. Importantly, approved producers may forego selling some of their Alvares, C., Lim, C., & Green, K. (2008). Results of checklist and reassessment of control fi for Escherichia coli O157:H7 in beef operations. Washington, DC: Food Safety and product to the NSLP program. Ollinger nds that lots of beef trim Inspection Service. sold to the NSLP are less likely to positive than lots from the un- Arthur, T. M., Bosilevac, J. M., Nou, X., & Koohmaraie, M. (2005). Evaluation of cul- approved producers. However, lots from these same approved ture- and PCR-based detection methods for Escherichia coli O157:H7 in inocu- lated ground beef. Journal of Food Protection, 68, 1566e1574. producers that are not sold to the NSLP are more likely to test Arthur, T. M., Brichta-Harhay, D. M., Bosilevac, J. M., Kalchayanand, N., positive than those of unapproved producers. In this case, pro- Shackelford, S. D., Wheeler, T. L., et al. (2010). Super shedding of Escherichia coli ducers seem to have private knowledge of the risk of salmonella O157:H7 by cattle and the impact on beef carcass contamination. Meat Science, e contamination at the time that they choose marketing outlets. 86(1), 32 37. Arthur, T. M., Keen, J. E., Bosilevac, J. M., Brichta-Harhay, D. M., Kalchayanand, N., Moreover, the lower risk profile of lots sold to the NSLP program by Shackelford, S. D., et al. (2009). Longitudinal study of Escherichia coli O157:H7 in approved producers is explained in part by the sellers removing a beef cattle feedlot and role of high-level shedders in hide contamination. e that lots they suspect are more likely to be contaminated. Ollinger Applied and Environmental Microbiology, 75(20), 6515 6523. Barkema, A., Drabenstott, M., & Novack, N. (2001). The new U.S. meat industry. could not conclude whether the lots of approved producers are, in Kansas City Federal Reserve Bank of Kansas City. fact, less risky than those of unapproved producer’s on average (i.e., E. Coli Traceability and Eradication Act. (2010). Amendment to meat inspection act. testing improves process control). In Washington, DC: 111th U.S. congress, 2nd Session. Economic Research Service. (2010). Foodborne illness cost calculator. In USDA (Vol. 2011). Washington, DC: USDA. Economic Research Service and University of Washington. (2004). Food safety 5. Conclusion technologies and HACCP compliance survey: Meat slaughter and processing plants. Washington, DC: USDA. In short, we show that test design affects the cost of imple- Federal Register. (2011). Not applying the mark of inspection pending certain results. Washington, DC: Food Safety and Inspection Service. menting food safety investments in the process control role, Ferrier, P. (2010). Irradiation as a quarantine treatment. Food Policy, 35(6), 548e555. whereas it directly affects health outcomes in the filtering role. Ferrier, P. M., & Buzby, J. C. (2013). The economic efficiency of sampling size: the Subsequently, we show that by affecting the test error, the level of case of beef trim. Risk Analysis, 33(3), 368e384. Food and Drug Administration. (1994). FDA consumer. Washington, DC U.S: sampling directly affects health outcomes when testing is used as a Department fo Health and Human Services. filter. While both roles improve consumer health outcomes, they Food Safety and Inspection Service. (2008). Compliance guidelines for sampling beef have different implications when testing is associated with error. trimmings for Escherichia coli O157:H7 (Draft for stakeholder comment). Wash- Within the process control role, errors where tests fail to detect ington, DC: USDA. Food Safety and Inspection Service. (2010a). Detection, isolation and identification of contaminated material primarily increase the cost of incentivizing Escherichia coli O157:H7 from meat products. Washington, DC: USDA. additional food safety investment and thus only indirectly affect Food Safety and Inspection Service. (2010b). FSIS procedure for the use of Escherichia health outcomes. If firms face weak incentives to improve food coli O157:H7 screening tests. Athens, GA 30605: USDA. Food Safety and Inspection Service. (2010c). Pre-harvest management controls and safety, reduced testing error may have little effect on food safety intervention options for reducing Escherichia coli O157:h7 shedding in cattle. investment. Within the filtering role, similar errors result in Washington, DC: USDA. contaminated product reaching consumers, directly affecting Food Safety and Inspection Service. (2001). Risk assessment of the public health impact of Escherichia coli O157:H7 in ground beef. Washington, DC: USDA. health outcomes. Table 2 presents a comparison of process control Food Safety and Inspection Service. (2011a). Shiga toxin-producing Escherichia coli and filtering roles in beef trim testing for E. coli O157:H7. in certain raw beef products. Federal RegisterUSDA (Ed.). (Vol. 76; pp. 58157e Using beef trim as a case study, we examined how testing can 58165). Federal Register. Food Safety and Inspection Service. (2011b). Shiga toxin-producing Escherichia coli improve food safety outcomes, by acting both as process control in certain raw beef products. Federal RegisterUSDA (Ed.). (Vol. 76; pp. 72331e and filtering mechanism. When acting as a process control, testing 72332). Federal Register. may be infrequent but connected to strong incentive to adjust the Frenzen, P. D., Drake, A., Angulo, F. J., & The Emerging Infections Program Foodnet Working Group. (2005). Economic cost of illness due to Escherichia coli O157:H7 production process to make it safer. On the other hand, when acting infections in the United States. Journal of Food Protection, 68, 2623e2630. as a filtering mechanism, testing may still be infrequent and asso- Greene, J. L. (2012). Lean finely textured beef: The “pink slime” controversy. Wash- ciated with weak incentives to improve the process if both recall ington, DC: Congressional Research Service. costs and markdowns on cook-only products are low. We have Hayes, D. J., Fox, J. A., & Shogren, J. F. (2002). Experts and activists: how information affects the demand for food irradiation. Food Policy, 27(2), 185e193. shown that at different times, current and proposed regulations Healy, A. (2009). New investigations of school lunches and n-60 sampling sought and policy guidelines have emphasized these alternative roles. following latest E. coli recall (Vol. 51 (pp. 21e22). Washington: DC Agra Zero tolerance in testing does not imply a lack of testing error. Informa Ltd. Hennessy, D. A. (1996). Information asymmetry as a reason for food industry ver- Increased sampling, however, can lower the probability of an error tical integration. American Journal of Agricultural Economics, 78(4), 1034e1043. in testing as recent policy proposals have emphasized. Within a Hooker, N. H., Nayga, R. M., Jr., & Siebert, J. W. (2002). The impact of HACCP on costs e process control framework, increased sampling reduces the cost of and product exit. Journal of Agricultural and Applied Economics, 34(1), 165 174. International Commission for the Microbiological Specifications of Foods. (2002). implementing a given performance standard and only indirectly Microorganisms in foods 7. London: James and James. affects health outcomes. Within a filtering framework, the effect on Koohmaraie, M., Arthur, T., Bosilevac, J., Brichtaharhay, D., Kalchayanand, N., health outcomes is direct where increased sampling lowers errors Shackelford, S., et al. (2007). Interventions to reduce/eliminate Escherichia coli O157:H7 in ground beef. Meat Science, 77(1), 90e96. and directly prevents contaminated products from reaching the Koohmaraie, M., Arthur, T., Bosilevac, J., Guerini, M., Shackelford, S., & Wheeler, T. market. Institutional mechanisms concerning recall costs and lia- (2005). Post-harvest interventions to reduce/eliminate pathogens in beef. Meat bility have led some producers to test all their output using the FSIS Science, 71(1), 79e91. Lamb, R., & Beshear, M. (1998). From the plains to the plate: can the beef industry sampling protocol. However, without strong incentives, testing regain market share? Economic Review, 64(10), 1e18. outcomes may not impact the production process, even if increased Loader, R., & Hobbs, J. E. (1999). Strategic responses to food safety legislation. Food sampling strengthens the role of testing as a filter. Policy, 24(6), 685e706. Loneragan, G. H., & Brashears, M. M. (2005). Pre-harvest interventions to reduce car- riage of E. coli O157 by harvest-ready feedlot cattle. Meat Science, 71(1), 72e78. Disclaimer Matthews, L., McKendrick, I. J., Ternent, H., Gunn, G. J., Synge, B., & Woolhouse, M. E. J. (2006). Super-shedding cattle and the transmission dy- namics of Escherichia coli O157. Epidemiology and Infection, 134(1). The views expressed in this article do not reflect those of the McMeekin, T. A., Olley, J., Ratkowsky, D. A., & Ross, T. (2002). Predictive microbi- Economic Research Search or the United States Department of ology: towards the interface and beyond. International Journal of Food Micro- biology, 73(2e3), 395e407. Agriculture. 234 P. Ferrier, J.C. Buzby / Food Control 38 (2014) 227e234

Mead, P. S., Slutsker, L., Dietz, V., McCaig, L. F., Bresee, J. S., Shapiro, C., et al. (1999). Roberts, T., Morales, R. A., Lin, C.-T. J., Caswell, J. A., & Hooker, N. H. (1997). In Food-related illness and death in the United States. Emerging Infectious Diseases, L. T. Wallace, & W. R. Schroder (Eds.), Worldwide opportunities to market food 5(5), 607e625. safety. Boston: Kluwer Academic Publishers. Moss, M. (2009). E. Coli outbreak traced to company that halted testing of ground beef Safe Tables Our Priority. Re: Docket No. FSIS-2008-0011: “Shiga Toxin-producing trimmings. New York. E. coli: Addressing the Challenges, Moving Forward with Solutions”. Naugle, A. L., Holt, K. G., Levine, P., & Eckel, R. (2005). Food safety and inspection Sargeant, J. M., Amezcua, M. R., Rajic, A., & Waddell, L. (2007). Pre-harvest interventions service regulatory testing program for Escherichia coli O157:H7 in raw ground to reduce the shedding of E. coli O157 in the faeces of weaned domestic ruminants: beef. Journal of Food Protection, 68, 462e468. a systematic review. Zoonoses and Public Health, 54(6e7), 260e277. Ofﬁce of Inspector General. (2011). FSIS sampling protocol for testing beef trim for Scallan, E., Hoekstra, R., Angulo, F., Tauxe, R., Widdowson, M.-A., & Roy, S. (2011). E. coli O157:H7. Washington, D.C.: United States Department of Agriculture. Foodborne illness acquired in the United Statesdmajor pathogens. Emerging Ollinger, M. (2013). Regulatory oversight and the food safety of ground beef in the Infectious Diseases, 17(1), 7e15. national school lunch program. In Paper presented at agricultural & applied Starbird, S. A. (2000). Designing food safety regulations: the effect of inspection economics association’s 2013 & Canadian agricultural economic associations joint policy and penalties for noncompliance on food processor behavior. Journal of annual meeting. Washington, DC, August 4e6, 2013. Agricultural and Resource Economics, 25(2), 616e635. Ollinger, M., & Moore, D. (2009). The interplay of regulation and marketing incentives Starbird, S. A. (2005). Moral hazard, inspection policy, and food safety. American in providing food safety. Washington, DC: Economic Research Service. Journal of Agricultural Economics, 87(1), 15e27. Ollinger, M., Muth, M. K., Karns, S. A., & Choice, Z. (2011). Food safety audits, plant Starbird, S. A., & Amanor-Boadu, V. (2006). Do inspection and traceability provide in- characteristics, and food safety technology use in meat and poultry plants. centivesfor food safety? Journal of Agricultural and Resource Economics, 31(1),14e26. Washington, DC: Economic Research Service. Unnevehr, L. J., & Jensen, H. H. (1999). The economic implications of using HACCP as Powell, M. R. (2013). The economic efﬁciency of sampling size: the case of beef trim a food safety regulatory standard. Food Policy, 24(6), 625e635. revisited. Risk Analysis, 33(3), 385e396.