1389

Journal of Food Protection, Vol. 63, No. 10, 2000, Pages 1389±1398

Visual Color and Doneness Indicators and the Incidence of Premature Brown Color in Beef Patties Cooked to Four End Point Temperatures

B. G. LYON,1* B. W. BERRY,2 D. SODERBERG,3² AND NELSON CLINCH3

1U.S. Department of Agriculture, Agricultural Research Service, Russell Research Center, P.O. Box 5677, Athens, Georgia 30604-5677; 2U.S. Department of Agriculture, Agricultural Research Service, Meat Science Research Laboratory, Beltsville, Maryland 20705-2350; and 3U.S. Department of Agriculture, Food Safety Inspection Service, Washington, D.C., USA Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021 MS 00-35: Received 7 February 2000/Accepted 20 May 2000

ABSTRACT

An interlaboratory study was undertaken to assess the frequency that cooked color of ground beef patties appeared brown at internal temperatures of 52.7ЊC (135ЊF), 65.6ЊC (150ЊF), 71.1ЊC (160ЊF), and 79.4ЊC (175ЊF). In general, as internal cooked temperature of the patties increased, the following results were observed in the patties: (i) more brown meat color, (ii) less pink or red juice color, and (iii) more cooked texture. However, brown meat color occurred prematurely at the two lower internal temperatures (57.2ЊC/135ЊF and 65.6ЊC/150ЊF) that are insuf®cient to eliminate foodborne pathogens without holding times. The common consumer practice of freezing bulk ground beef, followed by overnight thawing in a refrigerator, led to substantial premature brown color in patties cooked from this product. In addition, at 71.1ЊC (160ЊF), recognized to be the lowest temperature for cooking ground beef safely in the home, meat color, juice color, and texture appearance were not fully apparent as doneness indicators. In fact, at no temperature studied did 100% of the patties appear done when evaluated by the criteria of no red or pink in the meat, no red or pink in the juices, or by texture appearance. Patties in this study were evaluated under a set protocol for forming the products, cooking, and viewing under the same lighting conditions. Other preparation conditions are possible and may produce different results. Thus, temperature to which patties have been cooked cannot be judged by color and appearance. This study provided the evidence to support the message to consumers regarding cooking of beef patties of ``use an accurate food thermometer and cook beef patties to 160ЊF (71.1ЊC)'' in place of messages based on consumer judgment of cooked color.

The association between foodborne illness caused by In another 1990 publication (11), the USDA also stated that Escherichia coli O157:H7 and inadequately cooked ground patties should be cooked until they are hot and steaming beef patties has stimulated research into incidence and cau- and the juices run clear, with no evidence of pink color in sation and the evolution of public health recommendations the center. for safe cooking of this product. As heating occurs and Because of the food safety concerns associated with internal temperature (IT) increases in meat products, the consuming an undercooked (Ͻ71ЊC/160ЊF) hamburger, in interior color changes from red to pink to tan or brown (2). 1994 the USDA (12) developed a nationwide mailing of With intact beef muscle cuts, cooking to various degrees of concern for schoolchildren to deliver to their parents. The visual doneness from rare to well done is assumed to pro- message of this document was to cook hamburgers (ground vide microbiologically safe food as long as safe handling beef patties) until no pink color remained. The message was has been practiced. However, there has been concern that modi®ed in 1995 (13) to state ``cook ground meat to at cooking to visual colors of pink or red might not be a re- least 160ЊF (71.1ЊC), which means there is no pink left in liable indicator of ®nal IT in ground beef patties, thus im- the middle or in the juices.'' posing a possible safety risk. In 1996, as a result of the increasing research docu- Before the concern of food safety risks with ground mentation that cooked color varies substantially in ground beef due to E. coli O157:H7, the advice to consumers from beef cooked to 160ЊF (71.1ЊC), the USDA (14) released a the Food Safety and Inspection Service (FSIS) of the U.S. technical document that highlighted a condition in which Department of Agriculture (USDA) (9) was to avoid eating the interior of cooked patties appears brown at temperatures rare ground beef and to cook patties until they were brown well below 160ЊF (71.1ЊC) (premature brown color) and or brownish pink in the center. Five years later in 1990, the another condition in which patties stay persistently pink USDA (10) voiced a food safety concern associated with when cooked well beyond 71.1ЊC (160ЊF) (persistent pink consuming hamburger that was red in the middle and ad- color). In 1997, because of the growing evidence (2±5, 16, vised that ground beef should be cooked to 71ЊC (160ЊF). 17) of premature browning (PMB), the USDA again mod- i®ed its advice to consumers to ``use a thermometer when * Author for correspondence. Tel: 706-546-3167; Fax: 706-546-3607; E-mail: [email protected]. cooking hamburger, and do not rely on the internal color ² Present address: Environmental Protection Agency, Washington, D.C. of the meat to insure food safety'' (15). 1390 LYON ET AL. J. Food Prot., Vol. 63, No. 10

PMB strongly suggests a potential food safety concern. To simulate the same myoglobin chemistry encountered by Oxidation of myoglobin appears to be a contributing factor consumers, it was necessary to obtain samples at retail. However, to the incidence of PMB (6, 18, 20). The presence of oxy- there was no practical way to randomize retailers as weighted by myoglobin (3, 6), added ingredients (16, 17), and cooking the volume of ground beef they sell or by the state of the myo- of previously frozen patties from the thawed state (17) may globin chemistry at the time of sale. Instead, the study attempted, to the extent possible, to have USDA buyers imitate consumer also increase the incidence of PMB. However, the actual buying behavior in the selection of samples. That is, each buyer incidence of PMB in beef patties made from retail-pur- was instructed to purchase the sample as though it would be for chased ground beef has remained uncertain, although esti- his or her own home use. A variety of buyers and retail areas mates (5) range from 25 to 62%. If the frequency of oc- were used. Despite the nonrandomness of sample selection, the currence is very small, safety is little compromised by use study was structured using a balanced statistical design with num- of a color message. However, if the incidence of PMB is bers of samples based on random selection. More samples would high, color cannot be used. To resolve the issue, a joint not have compensated for the nonrandomness in the sampling. study was undertaken between USDA's FSIS and USDA's A ®nal limitation affected the procedures used to cook and Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021 Agricultural Research Service (ARS) to determine the fre- evaluate the color of the ground beef patties. Consumers cook quency of ground beef patties appearing brown when ground beef patties using many different cooking styles. They cooked to temperatures equal to or less than 71.1ЊC (160ЊF) may use high or low temperatures; patties can be made thick or when some pink would be expected. Factors that had to be thin or large or small. The lighting conditions under which patties considered in determining the incidence of PMB were the are visualized range from daylight to sunset outdoors and from incandescent to ¯uorescent indoors. It was not possible to dupli- in¯uence of ground beef processing, selection and storage cate all possible cooking techniques under all possible lighting practices, and internal cooked patty temperature. This arti- conditions. Therefore, the study used a single standardized cook- cle focuses on the portions of that study that relate to the ing technique, cooking patties of a single standard size and shape frequency that browning occurs in patties at temperatures and viewing results under a single set of lighting conditions. Ad- below 71.1ЊC (160ЊF) and the possible necessity that the ditional studies are under way to address the effects of other dif- USDA revise advice to consumers on how to cook ground ferent cooking and lighting conditions. beef patties for food safety. Sampling procedures. Cooking and evaluations were done MATERIALS AND METHODS at each of ®ve USDA laboratories: ARS in Beltsville, Md., ARS Study design considerations. The study was designed to in Athens, Ga., FSIS in Athens, Ga., FSIS in St. Louis, Mo., and control as many variables as possible but had certain necessary FSIS in Alameda, Calif. Personnel from these laboratories pur- limitations. The most important limitation was control of the ox- chased samples locally (local) for same day cooking. Because two idation state of the myoglobin. Because the oxidation state of of the laboratories were located in Athens, Ga., only one labora- myoglobin affects the temperature at which browning occurs dur- tory purchased samples used by both participating laboratories. ing cooking (3, 19), the condition of the ground beef myoglobin Six other USDA of®ces also purchased samples (Distant) in Bos- at purchase and any changes between purchase and cooking were ton, Mass., Atlanta, Ga., Lawrence, Kan., Lincoln, Neb., Alameda, critical factors. This oxidation problem could not be resolved by Calif., and Bothel, Wash. The retail area for samples purchased freezing the product, because freezing also affects the temperature by a USDA laboratory in Alameda and by another USDA of®ce of browning. It was necessary to sample ground beef at retail in for the larger San Francisco area overlapped and are recorded as a way that re¯ected the oxidation states in ground beef commonly single retail area. bought by consumers, and it was critical to control the time from In total, 240 pairs of samples were scheduled for purchase purchase to cooking. Finally, because fresh ground beef is often during the 6-week period of the study (i.e., 120 Local and 120 frozen and stored at home before cooking and because freezing Distant). During the ®rst week, 25 extra ground beef packages affects browning, home freezing and thawing techniques were also were also obtained by the laboratories to be used to evaluate the included in the study design. effects of long-term freezing (4-week) storage. Thus, a total of All ground beef was bought at retail markets and selected 505 packages were scheduled for collection. In the ®nal tally, only from the normal meat displays in an unfrozen state. Product in 490 sample packages were actually obtained because of unantic- this state is referred to as ``fresh,'' with no connotation of the ipated dif®culties in sampling. Individual packages of ground beef prior condition of the meat. All samples were purchased as pairs weighed from 0.57 to 0.91 kg. of ground beef packages. One package from each pair was made Of the 490 sample packages, 240 nonfrozen packages were into patties and cooked the same day (Fresh). The second package prepared and cooked the day of purchase or receipt in the labo- was frozen (either as patties or in original packages) and later ratory (referred to as Fresh). The second package from each of thawed by one of several methods before cooking. Because the the 240 pairs was assigned to one of four freeze-and-thaw treat- two samples in each pair could not be ensured to be identical in ments. For three treatments, patties were formed, frozen for 1 myoglobin chemistry, the pairs were not considered to be exactly week, and then thawed using techniques commonly reported by matched and were analyzed separately. consumers. The thawing approaches were thawing in a microwave Sample pairs were obtained in two ways. Samples in one oven (TH-MW), thawing for 2 h at room temperature (20 to 22ЊC/ group (Local) were bought locally in the morning near the ®ve 68 to 72ЊF) (TH-RM), and thawing for 4 h in a refrigerator (TH- laboratory sites and cooked or frozen later on the same day. Sam- REF). For a fourth treatment, a set of packages was frozen in ples in the second group (Distant) were bought at several remote original packs for 1 week and thawed overnight in a refrigerator locations (see below) and shipped overnight to the laboratories in (TH-BK-7). An additional 25 packages were kept frozen in their chilled containers. The myoglobin chemistry of these latter sam- original pack for about 1 month before thawing overnight in a ples approximated that of ground beef held in a home refrigerator refrigerator (TH-BK-28). The statistical prevalence of use of these 1 day after purchase before cooking or freezing. different sample handling techniques by consumers was estimated J. Food Prot., Vol. 63, No. 10 END POINT TEMPERATURE AND VISUAL ASSESSMENTS IN COOKED BEEF PATTIES 1391 using information taken from the USDA's Economic Research most consistent surface heating of 162.8ЊC (325ЊF), with the nar- Service 1995 Continuing Survey of Food Intakes by Individuals. rowest range of temperatures (low to high) during the heating cycle. The patty was placed on the griddle surface at the prede- Laboratory procedures. Protocols for forming patties, pack- termined location at the end of a heat cycle (signal light off). aging, labeling, freezing, thawing, cooking, and evaluating the Patties were turned at 2-min intervals for the ®rst 4 min then at patties were developed by representatives from each of the par- 1-min intervals until target ITs were reached. ticipating laboratories. The steps in the protocols were videotaped The ITs of patties were measured with a copper-constantan, and sent to each laboratory to use for reference and training ad- 23-gauge thermocouple probe (Physitemp, MT-23/3) attached to ditional workers at each site. As was practical, each laboratory Barnant JTEK-115 thermometers. The probe was inserted from attempted to use the same equipment. As soon as the product the side of the burger so that the tip was at the center of the arrived at the laboratory, notes were recorded on temperature, drip thickest part of the patty and parallel to the griddle surface. Usu- in the package, and package tears, etc. Based on the study design, ally, the thickest part was the center of the patty, but sometimes a treatment (Fresh plus either TH-MW, TH-RM, TH-REF, TH- it was slightly off from the center. During the last 1 to 2 min of BK-7, or TH-BK-28) was assigned to the sample packages. All expected cooking time, the thermocouple probe was inserted and Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021 packages were refrigerated (Ϫ1to4ЊC/34 to 40ЊF) until used for kept in the burger until target IT was reached. Preliminary trials immediate cooking or preparation for storage treatment. indicated the probe should be inserted and kept in the patties after Forming patties. Each 0.57- to 0.91-kg package of ground 4 to 5, 7 to 8, 9 to 10, and 11 to 12 min, respectively, for ITs of beef was divided into four approximately equal portions by cutting 52.7, 65.6, 71.1, and 79.4ЊC (135, 150, 160, and 175ЊF, respec- vertically through the unwrapped meat block lengthwise and then tively). These times were based on fully thawed patties. TH-RM crosswise. Each cut section had approximately the same amount and TH-REF patties were colder at the start of cooking and gen- of exterior and interior tissue that was processed into a patty, one erally required more time. If the measured temperature was much at a time just before cooking, by mixing for about 15 s by hand higher than expected when ®rst checked, the probe was removed (wearing gloves), taking care not to overmix. Approximately 115- and repositioned to locate a lower-temperature section of the patty. g portions of the mixed section was placed in a stainless steel ring Patties were turned at 1-min intervals with the thermocouple in (89 mm diameter by 25 mm high) and distributed evenly inside place until the target temperature was reached. Patties were re- the ring. A Te¯on press made speci®cally for the ring was used moved with spatulas to a tared, white plate. At this time, weights to compact the meat under manual pressure to a height (thickness) and time required to reach IT were recorded. of 15 mm. The patty was removed from the ring and reweighed. Patties averaged 115 g, with dimensions of 15 mm thick by 89 Visual color of heated patties. After heating to target tem- mm diameter. peratures, patties were immediately divided into two portions by a vertical cut perpendicular to the top ¯at surface. The two interior Freezing and thawing treatments. Patties to be frozen and cut surfaces were placed on a white plate and propped between then thawed were placed into individual plastic freezer bags (ap- small blocks (white Te¯on or blond wooden segments) so that the proximately 0.95 liters, 17.8 by 20.3 cm), closed, and labeled. viewing light was perpendicular to the cut surfaces. The test lights Individual bags were stored in a single layer in an automatic de- were two Philips deluxe, warm white 3,000ЊK lights, placed to Ϫ Њ Њ frost home-style freezer at 18 C(0F). Treatments TH-BK-7 and provide 1,614 lx of illumination. With room lights off, visual eval- Ϫ TH-BK-28 were frozen in their original packages and held at 18 uations were performed by individuals (n ϭ 2 or 3) who were Ϫ Њ Њ to 8 C(0to18F) for 7 and 28 days, respectively. trained to score samples by comparing the color and appearance Before cooking, TH-MW patties were thawed individually in of test patties to color charts of cooked ground beef patties pro- household microwave ovens. Brand or manufacturer of microwave vided by Kansas State University (KSU) (8). The color photo- ovens was not standardized, but all ovens were similar-sized coun- graph charts from KSU depicted the interior appearance of ®ve tertop models, 650 to 1,000 W, equipped with a turntable. Single ground beef patties cut in half that had been cooked to 65, 68, patties were removed from the storage bag, placed on a tared 71, 74, and 77ЊC (149, 154, 160, 166, and 170ЊF), labeled ac- white plate, and microwaved at 30% power (i.e., thaw setting) for cordingly, and also numbered 1 to 5, respectively. These charts 1 min. After a standing time of 2 min, the patty was turned over, also carried a descriptive degree of doneness as medium rare, and the process was repeated once. Each laboratory modi®ed this medium, and well done for patties at 65, 71, and 77ЊC (149, 160, procedure to accommodate its particular equipment. The goal in and 170ЊF), respectively. There were some variations in color microwave thawing of a patty was to have no juices on the plate, from chart to chart in the sets obtained from KSU. For that reason, no visually cooked surface areas, an IT of approximately 1ЊC each laboratory received charts that were matched by Minolta col- (34ЊF), and a surface temperature of about 18ЊC (65ЊF). For thaw- or measurements. Use of the charts provided a visual frame of ing TH-RM patties, each patty in its individual storage bag was reference to the panelists for comparing color and appearance of placed on a counter for 2 h at room temperature (20 to 22ЊC/68 the interior of the patties. Therefore, labels describing the tem- to 72ЊF). For thawing TH-REF patties, the individually bagged perature and degrees of doneness of the patties on the KSU charts patty was placed on a refrigerator shelf (3 to 4ЊC/38 to 40ЊF) for were obscured by black tape. Panelists only saw the numerical 4 h. For TH-BK-7 and TH-BK-28 samples, bulk packages were label of 1 to 5, beginning with the 65ЊC (149ЊF), medium rare thawed overnight (approximately 12 to 18 h) in a refrigerator (3 photograph. Panelists were instructed to focus evaluations on the to 4ЊC/38 to 30ЊF). Patties were formed from the thawed meat middle horizontal section of the inside cut surface of the test patty just before cooking as previously described. Raw weights of all and to record a color score on the paper ballot. Strict comparisons patties were recorded just before cooking. between study patties and chart patties were dif®cult, because the Cooking. Each laboratory used an electric (ഠ1,500 W) study patties were formed under manual pressure, whereas the homestyle griddle (for example, WestBend Inc, West Bend, Wis.). chart patties were formed mechanically. Manually formed patties Each laboratory performed preliminary tests to map the griddle simulated home preparation variations in conformation. Panelists surface for temperature variations using an infrared thermometer were allowed to give a score at 0.5 intervals to allow for observed and to determine dial setting and surface location that gave the variations in the test patties that fell between the integer scores 1392 LYON ET AL. J. Food Prot., Vol. 63, No. 10

TABLE 1. Incidence of cooked patties scored 4.0 or higher (judged as having no pink or red color)a Final internal temperature, ЊC(ЊF)

Procedures for patties 57.2 (135) 65.6 (150) 71.1 (160) 79.4 (175)

Ground beef purchase procedures Local 3.3 14.1 34.3 54.1 Distant 3.0 17.1 32.8 52.8 Chi-square P value Ͻ0.88 Ͻ0.37 Ͻ0.74 Ͻ0.88 Product storage and handlingb Fresh (as received, nonfrozen) 1.8 12.3 36.1 56.4 TH-MW 1.8 7.0 12.3 52.6

TH-RM 1.8 7.1 14.6 30.0 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021 TH-REF 5.7 5.7 13.2 33.3 TH-BK-7 7.1 48.2 67.9 80.0 TH-BK-28 7.4 33.3 63.0 77.8 Chi-square P value Ͻ0.17 Ͻ0.001 Ͻ0.001 Ͻ0.02 a Values are percentages of scores 4.0 or higher for cooked patties within purchase procedures or product storage and handling categories. Chi-square test performed within ®nal internal temperatures. Color scores assigned on the basis of a scale of 1 (very red or pink) to 5 (no pink or red) using Kansas State University Ground Beef Cooked Color Guide (Marksberry et al., 1993) as visual reference for appearance comparison. A numerical score of 4.0 or higher indicated no red or pink color but varying degrees of tan or brown. b TH-MW, patties frozen individually for 7 days then thawed in microwave oven; TH-RM, patties frozen individually for 7 days then thawed at2hatroom temperature (20 to 22ЊC/68 to 72ЊF); TH-REF, patties frozen individually for 7 days then thawed 4 h in refrigerator; TH-BK-7, meat frozen in original package for 7 days then thawed overnight in refrigerator; TH-BK-24, meat frozen in original package for 28 days then thawed overnight in refrigerator. that corresponded to the chart patties. Thus, the scale used was into two categories, undercooked and cooked, and two situations 0.5 (lowest possible score, most red color intensity) to 5.0 (highest were considered in selecting these categories based on different possible score, most tan or brown color with no red). Furthermore, criteria for appearance of cooked color. In the ®rst situation, we because cut surfaces of the study patties often had color pattern assigned a color score of 4.0 and higher as cooked, based on a variations, panelists were instructed to score any patty that had de®nite category of tan or brown color (i.e., no red or pink color) any pink or red color no higher than 3.5. A score of 4.0 or above and a color score of 3.5 and lower as undercooked, based on a indicated that the sample had no pink or red color. Aside from de®nite category of some red or pink color. In a second situation, this stipulation, great care was taken to ensure that judgments we assigned a division in the scale at color score of 3.0 and lower were made only by visual comparison to the charts and that all as undercooked, based on a de®nite amount of some red or pink verbal descriptors of color such as pink, red, etc., were avoided. color. The score of 3.5 was considered a transition area, where In addition, panelists were asked to view the collected juice panelists observed mostly tan and brown but also observed some on the plate and mark the score sheet Yes or No in response to small areas of pink or red that prevented a de®nite score of 4.0 the question, ``Is there presence of pink/red color in the juice?'' or more, denoting tan or brown color with no red or pink color. Panelists were also asked the question, ``Can the overall texture Therefore, a score of 3.5 or more was considered cooked in this of the cut surface be described as undercooked: i.e., moist, glossy, situation. Chi-square analysis was also conducted on the responses soft, spongy, mushy?'' A response of Yes equated ``undercooked to the questions on juice color and texture appearance. Percentages appearance'' and No equated ``cooked appearance.'' were based on one color score for each burger, using the median Additional observers were trained at each of the ®ve labo- values where multiple observers evaluated patties. ratories to evaluate the appearance of cooked patties. With few exceptions, there were at least two independent observers, and RESULTS AND DISCUSSION most often there were three. However, it was dif®cult to blind The percentages of color scale responses that were 4.0 observers to the cooking process, because observers had to be or higher, according to purchase procedure within each ®nal immediately present during cooking to facilitate fast observations IT, are presented in Table 1. There were no differences be- and instrumental measurements of color on reaching temperature. Frequently, a cook was also one of the observers. As noted, in- tween the Local and Distant percentages at any of the four strumental color measurements of raw and cooked patties were temperatures. Since the duration between purchase and obtained but are part of a separate report in progress. It can be evaluation was longer and the opportunity for temperature stated here that instrumental results were consistent with the hu- ¯uctuation greater for Distant procured product, it was ex- man observations. pected that these factors would have an in¯uence on cooked patty color. Previously, overnight refrigerated storage has Data analysis. Visual color evaluations from all ®ve labo- been shown to result in more purple pigment in the interior ratories were combined and analyzed for the percentage of re- sponses to the color scale values from 0.5 to 5. Data were ana- of retail-purchased ground beef, which was associated with lyzed comparing treatments within each temperature and purchase less premature brown color than that observed in patties procedures within treatments. Chi-square tests were performed to made from ground beef purchased the same day (5). Fre- test the signi®cance of percentages within each ®nal temperature. quency of responses of 4.0 or more, indicating no pink or For the objectives of this report, the full color scale was divided red color in the patties, increased with increasing IT. Ap- J. Food Prot., Vol. 63, No. 10 END POINT TEMPERATURE AND VISUAL ASSESSMENTS IN COOKED BEEF PATTIES 1393

TABLE 2. Incidence of cooked patties scored 3.5 or higher (judged as having a few small areas of pink or red color)a Final internal temperature, ЊC(ЊF)

Procedures for patties 57.2 (135) 65.6 (150) 71.1 (160) 79.4 (175)

Ground beef purchase procedures Local 11.7 33.8 56.8 86.9 Distant 5.7 25.5 49.2 80.6 Chi-square P value Ͻ0.02 Ͻ0.05 Ͻ0.10 Ͻ0.33 Product storage and handlingb Fresh (as received, nonfrozen) 4.8 28.2 58.2 90.9 TH-MW 8.8 14.0 28.1 79.0

TH-RM 7.1 12.5 29.1 70.0 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021 TH-REF 7.6 13.2 30.2 53.3 TH-BK-7 17.9 64.3 83.9 100.0 TH-BK-28 22.2 63.0 85.2 100.0 Chi-square P value Ͻ0.004 Ͻ0.001 Ͻ0.001 Ͻ0.001 a Values are percentages of scores 3.5 or higher on cooked patties within purchase procedures or product storage and handling categories. Chi-square test performed within ®nal internal temperatures. Color scores assigned on the basis of a scale of 1 (very red or pink) to 5 (no pink or red) using Kansas State University Ground Beef Cooked Color Guide (Marksberry et al., 1993) as visual reference for appearance comparison. A score of 3.5 indicated small localized areas of pink or red color on cut surface of patties. b TH-MW, patties frozen individually for 7 days then thawed in microwave oven; TH-RM, patties frozen individually for 7 days then thawed at2hatroom temperature (20 to 22ЊC/68 to 72ЊF); TH-REF, patties frozen individually for 7 days then thawed 4 h in refrigerator; TH-BK-7, meat frozen in original package for 7 days then thawed overnight in refrigerator; TH-BK-24, meat frozen in original package for 28 days then thawed overnight in refrigerator. proximately 20% of all patties cooked to temperatures less incidence of pinkness or very small areas of redness and than 71.1ЊC (160ЊF) were rated 4.0 or more, i.e., appearing pinkness that could be missed by the consumer before serv- to be fully cooked by the most stringent color interpretation ing. criteria. Local patties cooked to 57.2ЊC (135ЊF) were scored as At ®nal ITs of 65.6 and 71.1ЊC (150 and 160ЊF), prod- 3.5 or higher in 11.7% of the responses, whereas Distant uct storage and handling procedures in¯uenced (P Ͻ 0.001) patties (i.e., at least 1-day storage before cooking) had a the frequency of scores of 4.0 or more (i.e., brown). The lower incidence (5.7%) of scores of 3.5 or higher. This higher incidence of brown color in patties made from difference was statistically signi®cant (P Ͻ 0.02), as was thawed bulk ground beef after 7 and 28 days of frozen the difference in percentages between Local and Distant storage is largely responsible for differences in scores. Al- patties cooked to 65.6ЊC (150ЊF). Decrease in brownness though many of the bulk ground beef packages were not would be expected in ground meat held overnight because completely thawed following overnight refrigerator storage, of pigment conversions to deoxymyoglobin, which is more the length of time out of frozen storage may have in¯u- likely to be pink at the lower temperatures (5, 6). Differ- enced the appearance of brown color in cooked patties, ences were not statistically signi®cant between the sources more so than product temperature. Van Laack et al. (17) of purchase for the patties cooked to ®nal ITs of 71.1 and reported that occurrence of brown color in cooked patties 79.4ЊC (160 and 175ЊF). increased as thawing time increased from 0.5 to 54 h under Within each of the ITs, differences in the percentage refrigerated conditions. In the present study, the thaw con- of responses scored at 3.5 or higher among the product ditions for patties had scores less than 4.0 (i.e., less brown storage and handling treatments were signi®cant. Patties color) more often than patties cooked from the fresh, non- made from ground beef that had been frozen in whole packs frozen state. for 7 or 28 days before being thawed, then made into patties Table 2 includes the percentages of responses that were and cooked, had signi®cantly higher proportions of color 3.5 or higher. These values would include those presented responses that were scored as 3.5 or higher compared with in Table 1 plus scores of 3.5. The 3.5 score served as a patties made from fresh, nonfrozen ground beef and then transition area between the patty references that clearly cooked to 57.2ЊC (135ЊF). The extra 21 days of frozen stor- showed some pink or red at 3.0 but none at 4.0. Cooked age for bulk product did not seem to in¯uence the occur- color patterns in our patties would more likely represent the rence of brown color. Samples that were frozen as patties color variations that would be found in hand-formed patties and then thawed by microwave, room temperature, or re- prepared by consumers. Thus, a score of 3.5 could indicate frigerator before being cooked to 57.2ЊC (135ЊF) had slight- a small area of pink or red color, or it could indicate a ly higher percentages of responses of 3.5 or higher. How- de®nite portion of the interior that had just a hint of pink ever, samples from these thawing treatments that were color. The signi®cance of combining the 3.5 score into the cooked to 65.6, 71.1, or 79.4ЊC (150, 160, or 175ЊF) had upper category of color and doneness was to capture the lower percentages scored at 3.5 or higher than the fresh, 1394 LYON ET AL. J. Food Prot., Vol. 63, No. 10 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021

FIGURE 1. Brown color in cooked patties by ®nal IT using a two- FIGURE 2. Brown color in cooked patties by ®nal IT using a two- way classi®cation criteria, where score of 3.5 or less equals un- way classi®cation criteria, where score of 3.0 or less equals un- dercooked, pink or red and 4.0 or more equals cooked, brown dercooked, pink or red and 3.5 or more equals cooked, mostly (no pink or red color). Ⅵ, score of 3.5 or less is pink or red; Ⅺ, brown, with few small areas of pink or red. Ⅵ, score of 3.0 or score of 4.0 or more is brown (no pink or red). less is pink or red; Ⅺ, score of 3.5 or more is mostly brown, with few small areas of pink or red. nonfrozen cooked samples or the samples from the frozen unsafe to consumers when, in fact, they had reached or whole packs. Rapid, short-time thawing procedures have exceeded a safe cooked IT. In the case of patties cooked to been reported to result in limited ``browning'' of patties 79.4ЊC (175ЊF), reductions in eating quality might be ex- predisposed to pink or red color when cooked from the pected. frozen state (6, 17, 19). When the criterion for cooked color was changed to The categories of color scores are summarized across incorporate the 3.5 transition score into the upper bracket temperatures according to the two color classi®cation sys- with 4.0 or higher (Figure 2), as the ®nal ITs increased, tems and presented in Figures 1 and 2. In Figure 1, visual more patties were scored as appearing mostly tan or brown, color is represented as pink or red (scores 0.5 to 3.5, inclu- with no red or pink or with only small areas of red or pink. sive) and tan or brown with no pink or red (scores 4.0 to On the other hand, the increase in ®nal temperature reduced 5.0). In Figure 2, the visual color is represented as pink or the proportion of patties that were considered to have pink red (scores 0.5 to 3.0) and mostly tan or brown (scores 4.0 or red color. Although including the color scores of 3.5 into to 5.0), with possible small areas of pink or red (score 3.5) a classi®cation of brown color permits some pink color, it included in the latter group. In both ®gures, the chi-square cannot be ascertained from this study how many consumers test indicated signi®cant differences between the categories under in-home cooking situations would consider these pat- across the ITs. These data would represent some of the risk ties as cooked or brown. However, under typical lighting associated with relying on cooked patty color versus patty and cooking conditions used by consumers and with small, temperature when the oxidation and oxygenation state of localized areas of pink color in patties, it is conceivable myoglobin and history of product handling are not known. that many of the patties would be considered brown (or In Figure 1, 3.2% of the patties cooked to 57.2ЊC (135ЊF) cooked) in color. and 15.8% of patties cooked to 65.6ЊC (150ЊF) were eval- Table 3 includes the percentage of responses to juice uated as having no pink color. These temperatures are lower color that indicated no pink or red color in the juices that than the safe temperatures recommended for destruction of collected on the white plate from the cut patties as they E. coli O157:H7. So, these patties cooked to 57.2 or 65.6ЊC were held under the Philips lights for evaluation. Presence (135 or 150ЊF), respectively, without a holding period, of red juices would indicate undenatured myoglobin, might appear tan or brown and be assumed to have reached whereas an absence of red juices would indicate that patties a safe temperature of 71.1ЊC (160ЊF) when, in fact, they were cooked to an IT high enough to denature the pigment. had not. Although these frequencies of scores of 4.0 or Purchase location (Local vs Distant) did not in¯uence the higher (i.e., brown color) are low, they con®rm the earlier presence or absence of pink or red color in juices from laboratory studies (2±6, 16, 17, 19, 20) that reported evi- cooked patties. The absence of red or pink juice color in- dence of brown, cooked patty color at temperatures less creased as IT increased. Within each temperature group, than 71.1ЊC (160ЊF). About 60% of the patties cooked to there were signi®cant differences in distribution of respons- 71.1ЊC (160ЊF) and 45% of the patties cooked to 79.4ЊC es on presence or absence of pink or red color due to prod- (175ЊF) had pink or red color and might be rejected as uct storage and handling procedures. At 57.2ЊC, all respons- J. Food Prot., Vol. 63, No. 10 END POINT TEMPERATURE AND VISUAL ASSESSMENTS IN COOKED BEEF PATTIES 1395

TABLE 3. Incidence of responses indicating an absence of pink or red color in juice from cooked pattiesa Final internal temperature, ЊC(ЊF)

Procedures for patties 57.2 (135) 65.6 (150) 71.1 (160) 79.4 (175)

Ground beef purchase procedures Local 1.4 9.4 26.5 58.3 Distant 0.8 10.0 23.8 61.9 Chi-square P value Ͻ0.49 Ͻ0.83 Ͻ0.50 Ͻ0.67 Product storage and handlingb Fresh (as received, nonfrozen) 0.0 8.4 25.4 74.1 TH-MW 1.8 1.8 8.8 27.8

TH-RM 0.0 0.0 3.6 40.0 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021 TH-REF 1.9 1.9 11.3 26.7 TH-BK-7 5.4 32.1 60.0 93.3 TH-BK-28 0.0 25.9 55.6 88.9 Chi-square P value Ͻ0.02 Ͻ0.001 Ͻ0.001 Ͻ0.001 a Values are percentages of responses indicating no pink or red color in the juice from cooked patties within purchase procedures or product storage and handling categories. Chi-square test performed within ®nal internal temperatures. b TH-MW, patties frozen individually for 7 days then thawed in microwave oven; TH-RM, patties frozen individually for 7 days then thawed at2hatroom temperature (20 to 22ЊC/68 to 72ЊF); TH-REF, patties frozen individually for 7 days then thawed 4 h in refrigerator; TH-BK-7, meat frozen in original package for 7 days then thawed overnight in refrigerator; TH-BK-24, meat frozen in original package for 28 days then thawed overnight in refrigerator. es for patty treatments Fresh, TH-RM, and TH-BK-28 in- then thawed whole packages. If we assume the percentages dicated pink or red color in the juices. The absence of pink to equate to a numerical incidence of patties appearing done or red color in juices for 57.2ЊC (135ЊF) patties were 1.8 (based on seeing no red in juice), then there was a greater and 1.9% for TH-MW and TH-REF and 5.4% for TH-BK- chance that no pink juice would be seen in patties made 7. For patties cooked to 65.6ЊC (150ЊF), pink or red color from whole packages that were frozen more than 7 days, disappeared in none of the TH-RM, less than 2.0% of the thawed, and cooked as patties. The incidence of no red or TH-MW and TH-REF, and 8.4% of the Fresh patties. How- pink juices was signi®cantly decreased by freezing the ever, for patties made from whole packaged ground beef product as patties and then thawing under the conditions frozen 7 and 28 days before thawing, there were 32.1 and reported in this study compared with cooking the patties 25.9% responses, respectively, that no pink or red color from the fresh state and frozen and thawed whole-packaged appeared in the juice. Patties cooked to 71.1ЊC (160ЊF) product. ranged from 3.6% no pink or red in juice for TH-RM treat- Figure 3 includes the distribution of the Yes and No ments to 60 and 55.6% for the patties made from frozen categories to the question of presence of pink or red color in the juice as affected by purchase and storage practices across the ®nal ITs. Even at 79.4ЊC (175ЊF), an IT above the recommended temperature (71.1ЊC/160ЊF) for destruc- tion of E. coli and above the denaturation temperature of myoglobin (71ЊC/160ЊF) (7), more than 40% of responses for juice that drained (was not pressed) from the patties were scored pink or red. Thus, use of a criteria of ``no pink or red color in juice'' as an indicator of safe food would provide consumers more protection than the use of internal color of the cooked meat. However, use of this criteria as a food safety indicator, if accurately applied by consumers, could result in considerable overcooking of patties. The end result of this practice would likely be reduced eating qual- ity, such as lack of juiciness (1). In addition, unless a suit- able contrasting background material is used to evaluate juice color, appraisal of juice color is dif®cult. While the patty is cooking, assessment of juice color through release of juice from a slit in the burger is even more dif®cult. Lightness and yellowness of expressible juice from cooked patties has been shown to increase up to 71ЊC (160ЊF), with FIGURE 3. Presence or absence of pink or red color in juice redness decreasing above 65ЊC (150ЊF) (5). Earlier studies from cooked patties according to ®nal IT. Ⅵ, yes, pink or red (2, 8, 19) have shown weak relationships between express- color in juice; Ⅺ, no, juice has no pink or red color. ible juice color and internal cooked patty color. 1396 LYON ET AL. J. Food Prot., Vol. 63, No. 10

TABLE 4. Incidence of responses indicating the appearance of undercooked/cooked texture on the cut surface of cooked pattiesa Final internal temperature, ЊC(ЊF)

Procedures for patties 57.2 (135) 65.6 (150) 71.1 (160) 79.4 (175)

Ground beef purchase procedures Local 7.5 21.1 45.0 82.0 Distant 2.7 13.7 30.9 76.1 Chi-square P value Ͻ0.01 Ͻ0.03 Ͻ0.002 Ͻ0.41 Product storage and handlingb Fresh (as received, nonfrozen) 3.1 16.8 35.7 90.7 TH-MW 3.5 14.0 35.1 57.9

TH-RM 3.6 3.6 20.0 65.0 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021 TH-REF 3.8 3.8 26.4 53.3 TH-BK-7 14.3 39.3 63.0 100.0 TH-BK-28 7.4 33.3 59.3 88.9 Chi-square P value Ͻ0.02 Ͻ0.001 Ͻ0.001 Ͻ0.001 a Values are percentages of no responses to question of whether patties had visually undercooked texture in the cut surface of cooked patties. Chi-square test performed within ®nal internal temperatures. Undercooked texture described as glossy, moist, soft, spongy, mushy appearance. b TH-MW, patties frozen individually for 7 days then thawed in microwave oven; TH-RM, patties frozen individually for 7 days then thawed at2hatroom temperature (20 to 22ЊC/68 to 72ЊF); TH-REF, patties frozen individually for 7 days then thawed 4 h in refrigerator; TH-BK-7, meat frozen in original package for 7 days then thawed overnight in refrigerator; TH-BK-24, meat frozen in original package for 28 days then thawed overnight in refrigerator.

Panelists were also asked to evaluate the appearance of cluded. The responses given in Table 4 include any con- the texture of the cut surface of the patties and to note notation of doneness or underdoneness that would support whether (Yes) or not (No) the patties would be considered an individual's assessment of texture doneness and appear- undercooked. In preliminary surveys and meetings, respon- ance. dents from a broad demographic pro®le had been asked to Chi-square results indicated that the distribution of re- describe what undercooked and cooked patties looked like. sponses regarding appearance of cooked texture within an The compiled descriptions of undercooked texture appear- IT group was signi®cantly different between Local or Dis- ance from that survey (moist, glossy surface and soft, tant purchases and signi®cantly affected by product storage spongy, mushy meat) were listed for respondents to con- or handling treatments. The occurrence of cooked appear- sider. For this report, only the Yes or No responses were ance was higher for Local purchases than Distant purchases tallied. Interpretation relative to the descriptors is not in- (Table 4). Only at 79.4ЊC (175ЊF) was the IT not signi®- cantly different. At 57.2ЊC (135ЊF), the percentages were similar for fresh patties and each category of frozen then thawed patties. At 65.6ЊC (150ЊF), TH-RM and TH-REF percentages were equal to those at 57.2ЊC (135ЊF). How- ever, there was a marked increase in cooked appearance in Fresh, TH-MW, TH-BK-7, and TH-BK-28 patties. At 71.1ЊC (160ЊF), 35% of the patties from Fresh and TH-MW appeared cooked compared with less than 27% of the TH- RM and TH-REF patties and more than 59% of the patties made from whole pack frozen for 7 or 28 days. At 79.4ЊC (175ЊF), Fresh, TH-BK-7, and TH-BK-28 patties displayed cooked appearance in more than 88% of the patties, where- as less than 65% of product frozen as patties and then thawed before cooking appeared cooked. Texture responses combined across treatments are giv- en in Figure 4 as a two-category classi®cation de®ned as Yes (texture appears undercooked) and No (texture does not appear undercooked). At 71.1ЊC (160ЊF), there were more patties that had undercooked texture appearance than ap- Њ FIGURE 4. Presence or absence of undercooked texture appear- peared cooked. Almost 23% of patties cooked to 79.4 C ance of cut surface of cooked patties according to ®nal IT. Ⅵ, (175ЊF) appeared undercooked, and these would mostly be yes, texture appears undercooked; Ⅺ, no, texture does not appear those products frozen as patties, thawed, and then cooked undercooked. (Table 4). The texture responses followed a very similar J. Food Prot., Vol. 63, No. 10 END POINT TEMPERATURE AND VISUAL ASSESSMENTS IN COOKED BEEF PATTIES 1397 trend as the results for juice color evaluation. Additionally, and time frame of the work could not have been accomplished without the ratios of cooked to undercooked texture in patties at the many dedicated people involved in the total study. We attempt here to list, in alphabetical order, those involved in some way, including ad- ®nal temperatures of 57.2, 65.6, and 71.1ЊC (135, 150, and Њ visement, planning, procurement, cooking, and evaluations. We apologize, 160 F) were similar to the brown to pink or red ratios in in advance, for omissions that have undoubtedly, but unintentionally, oc- patties cooked to these three temperatures when a score of curred. We thank Carl Davis, Judy Davis, Andra Dickens, Peggy Feldner, 4.0 was used to de®ne brown color. Undercooked or cooked Mara Fogarassy, Gene Lyon, Elizabeth Savage, Allan Savage, and Dr. texture appearance was the most dif®cult indicator of done- Robert Windham of ARS; and Estrella Agamata, Ralph Bakowski, Lily ness to evaluate, according to panelist comments. Obvious- Chounlamountry, John Dreas, Gail Graves, William Havlik, Carolyn Hen- ry, Dr. Michael Hoffman, Kathleen Holland, William Kelly, Rita Kishore, ly, assessing the condition of cooked-appearing texture in Bill Koscinski, Dr. Patrick McCaskey, Tom Phillippo, Steve Powell, Diane pink or red meat or uncooked texture in brown meat is VanLonkhuyzen, Dr. Kaye Wachsmuth, Veronica Wade, and Catalina Yee dif®cult. of FSIS. The primary question posed in this study was ``What

REFERENCES Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021 proportion of cooked ground beef patties appear brown (no red) before reaching the safe IT of 71.4ЊC (160ЊF)?'' We 1. Berry, B. W., and M. E. Bigner-George. 1999. Properties of beef examined four ITs (52.7, 65.6, 71.1, and 79.4ЊC/135, 150, patties cooked to elevated internal temperatures as a means of re- 160, and 175ЊF), representing a low- and high-temperature ducing pink color. J. Muscle Foods 10:215±230. range. Visual inspection of cooking state of burgers by ob- 2. Hague, M. A., K. E. Warren, M. C. Hunt, D. H. Kropf, C. L. Kastner, S. L. Stroda, and D. E. Johnson. 1994. Endpoint temperature, inter- servers was dif®cult to translate to data even when using a nal cooked color and expressible juice color relationships in ground color chart for reference. First, a value of 4 was de®ned as beef patties. J. Food Sci. 59:465±470. no red or pink in the burger and corresponded to the cooked 3. Hunt, M. C., O. Sùrheim, and E. Slinde. 1995. Effects of myoglobin patty reference on the KSU charts of patties that were form on internal cooked color development in ground beef. In Pro- cooked to 74ЊC (166ЊF) (with temperature and description ceedings of the 41st International Congress on Meat Science and hidden for panelists). Color scale responses that were 4.0 Technology, San Antonio, Tex. 4. Hunt, M. C., K. E. Warren, D. H. Kropf, M. A. Hague, C. L. Wald- or higher, grouped by purchase and at-home handling pro- ner, S. L. Stroda, and C. L. Kastner. 1994. Factors affecting pre- cedures, indicated that approximately 20% of all patties mature browning in cooked ground beef, S-VIB.24. In Proceedings cooked to temperatures less than 79.4ЊC (175ЊF) were rated of the 40th International Congress on Meat Science and Technology, greater than or equal to 4.0, thereby appearing to be fully The Hague, The Netherlands. cooked by the most stringent color interpretation criteria. 5. Killinger, K. M., M. C. Hunt, and R. E. Campbell. 1998. Incidence of premature browning in ground beef purchased retail, p. 185. Abstr. The percentage (approximately 30%) of patties cooked to 51st Recip. Meat Conf. less than 71.1ЊC (160ЊF) that were considered brown in- 6. Lavelle, C. L., M. C. Hunt, and D. H. Kropf. 1995. Display life and creased when a color score of 3.5 was included in the internal color of ground beef from vitamin E-supplemented steers. brown color interpretation. The score of 3.5 was a transition J. Food Sci. 60:1175±1178, 1196. color score between patties that were clearly still pink and 7. Machlik, S. M. 1965. The effect of heat on bovine myoglobin de- rivatives in model systems and in beef semitendinosus muscle. Ph.D. those that were clearly all brown. Because hand-formed dissertation. Purdue University, West Layette, Ind. patties created an irregular color surface when cut and be- 8. Marksberry, C. L. 1990. The effect of fat level, pH, carcass maturity cause the scores in this study were speci®cally attuned to and compaction on the cooked internal color ground beef patties at reporting the least traces of pink as seen under the best of ®ve end-point temperatures. M.S. thesis. Kansas State University, lighting conditions, it is a reasonable expectation that many Manhattan. patties scored at 3.5 would be considered brown by many 9. U.S. Department of Agriculture. 1985. The safe food book: your kitchen guide. FSIS Home and Garden Bull. no. 241. Food Safety consumers. By either color criteria, and especially given the Inspection Service, U.S. Department of Agriculture, Washington, major effects of freeze-and-thaw procedures, the study D.C. showed that a message to consumers to cook until the pat- 10. U.S. Department of Agriculture. 1990. A quick consumer guide to ties are brown could lead to consumption of a signi®cant safe food handling. FSIS Home and Garden Bull. no. 248. Food number of patties cooked to less than 71.1ЊC (160ЊF). For Safety Inspection Service, U.S. Department of Agriculture, Wash- both juice and texture responses, there were fewer samples ington, D.C. 11. U.S. Department of Agriculture. 1990. Preventing foodborne illness: that appeared without red or pink juice and without a a guide to safe food handling. FSIS Home and Garden Bull. no. 247. cooked appearance than would be indicated by color chart Food Safety Inspection Service, U.S. Department of Agriculture, comparisons to rate burger color. Thus, this study provided Washington, D.C. the evidence that the temperature to which patties have 12. U.S. Department of Agriculture. 1994. Recipe for a safe and deli- been cooked cannot be judged by color and appearance. cious hamburger: a post card to the parents of school children. Food Safety Inspection Service, U.S. Department of Agriculture, Wash- The results support the message to consumers regarding ington, D.C. cooking of beef patties of ``use an accurate food thermom- 13. U.S. Department of Agriculture. 1995. A quick consumer guide to eter and cook beef patties to 71.1ЊC (160ЊF)'' in place of safe food handling (revised). FSIS Home and Garden Bull. no 248. messages based on consumer judgment of cooked color and Food Safety Inspection Service, U.S. Department of Agriculture, appearance (15). Washington, D.C. 14. U.S. Department of Agriculture. 1996. Color of cooked ground beef ACKNOWLEDGMENTS and juices as it relates to doneness. FSIS Technical Information. Food Safety Inspection Service, U.S. Department of Agriculture, The authors gratefully acknowledge all persons involved in the suc- Washington, D.C. cessful planning, executing, and reporting of this study. The scope, extent, 15. U.S. Department of Agriculture. 1997. Focus on ground beef. FSIS 1398 LYON ET AL. J. Food Prot., Vol. 63, No. 10

Food Safety Focus. Food Safety Inspection Service, U.S. Department Cooked color of patties processed from various combinations of nor- of Agriculture, Washington, D.C. mal or high pH beef and lean ®nely textured beef (LFTB). J. Muscle 16. van Laack, R. L. J. M., B. W. Berry, and M. B. Solomon. 1996. Foods 8:287±299. Variations in internal color of cooked beef patties. J. Food Sci. 61: 19. Warren, K. E., M. C. Hunt, and D. H. Kropf. 1996. Myoglobin ox- 410±414. idative state affects internal cooked color development in ground 17. van Laack, R. L. J. M., B. W. Berry, and M. B. Solomon. 1996. beef patties. J. Food Sci. 61:513±515, 519. Effect of precooking conditions on color of cooked beef patties. J. 20. Warren, K. E., M. C. Hunt., D. H. Kropf, and C. L. Kastner. 1996. Food Prot. 59:976±983. Chemical properties of ground beef patties exhibiting normal and 18. van Laack, R. L. J. M., B. W. Berry, and M. B. Solomon. 1997. premature brown internal cooked color. J. Muscle Foods 7:303±314. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/63/10/1389/1686076/0362-028x-63_10_1389.pdf by guest on 26 September 2021