Evaluating the Performance of Sampling Plans to Detect Fumonisin B1 in Maize Lots Marketed in Nigeria

Abstract Fumonisins are toxic and carcinogenic compounds produced by fungi that can be readily found in maize. The establishment of maximum limits for fumonisins requires the development of scientifically based sampling plans to detect fumonisin in maize. As part of an International Atomic Energy Agency effort to assist developing countries to control mycotoxin contamination, a study was conducted to design sampling plans to detect fumonisin in maize produced and marketed in Nigeria. Eighty-six maize lots were sampled according to an experimental protocol in which an average of 17 test samples, 100 g each, were taken from each lot and analyzed for fumonisin B1 by using liquid chromatography. The total variability associated with the fumonisin test procedure was measured for each lot. Regression equations were developed to predict the total variance as a function of fumonisin concentration. The observed fumonisin distribution among the replicated-sample test results was compared with several theoretical distributions, and the negative binomial distribution was selected to model the fumonisin distribution among test results. A computer model was developed by using the variance and distribution information to predict the performance of sampling plan designs to detect fumonisin in maize shipments. The performance of several sampling plan designs was evaluated to demonstrate how to manipulate sample size and accept/reject limits to reduce misclassification of maize lots.

Download Full-text

Variability and distribution among sample test results when sampling unprocessed oat lots for ochratoxin A

World Mycotoxin Journal ◽

10.3920/wmj2014.1858 ◽

2015 ◽

Vol 8 (4) ◽

pp. 511-524 ◽

Cited By ~ 6

Author(s):

T.B. Whitaker ◽

A.B. Slate ◽

T.W. Nowicki ◽

F.G. Giesbrecht

Keyword(s):

Ochratoxin A ◽

Negative Binomial ◽

Test Procedure ◽

Regression Equations ◽

Test Results ◽

Sampling Step ◽

Test Portion ◽

Sample Test ◽

Total Variability ◽

Sampling Statistics

In 2008, Health Canada announced it was considering the establishment of maximum levels for ochratoxin A (OTA) in a number of foods, including unprocessed wheat and oats and their products. The Canada Grains Council and Canadian National Millers Association initiated a study to measure the variability and distribution among sample test results so that scientifically based sampling plans could be designed to meet regulatory and industry requirements. Twenty lots of oats naturally contaminated with OTA were identified and sampled according to a nested experimental protocol where 16-two kg laboratory samples were taken from each lot, two 100 g test portions were taken from each comminuted laboratory sample, and two aliquots of the extract from each test portion were analysed for OTA by LC. The variance associated with each step of the OTA test procedure were found to be a function of OTA concentration and regression equations were developed to predict the functional relationship. When using the above OTA test procedure on an oat lot at 5 μg/kg, the sampling, sample preparation, analytical, and total variances were 11.26, 0.10, 0.13 and 11.49, respectively. The 2 kg sampling step accounted for 98.0% (11.26/11.49) of the total variability. The observed OTA distribution among the 16 OTA sample results was found to be positively skewed and the negative binomial distribution was selected to model the OTA distribution among sample test results. The sampling statistics were incorporated into the FAO Mycotoxin Sampling Tool where operating characteristic curves were calculated to predict the chances of rejecting good lots (seller’s risk) and accepting bad lots (buyer’s risk) for various sampling plan designs.

Download Full-text

Sampling Hazelnuts for Aflatoxin: Effect of Sample Size and Accept/Reject Limit on Reducing the Risk of Misclassifying Lots

Journal of AOAC International ◽

10.1093/jaoac/90.4.1028 ◽

2007 ◽

Vol 90 (4) ◽

pp. 1028-1035 ◽

Cited By ~ 7

Author(s):

Guner Ozay ◽

Ferda Seyhan ◽

Aysun Yilmaz ◽

Thomas B Whitaker ◽

Andrew B Slate ◽

...

Keyword(s):

Sample Size ◽

Negative Binomial Distribution ◽

Binomial Distribution ◽

Negative Binomial ◽

Food Additives ◽

Sampling Plan ◽

Test Results ◽

Sampling Plans ◽

Wide Range ◽

Sample Test

Abstract About 100 countries have established regulatory limits for aflatoxin in food and feeds. Because these limits vary widely among regulating countries, the Codex Committee on Food Additives and Contaminants began work in 2004 to harmonize aflatoxin limits and sampling plans for aflatoxin in almonds, pistachios, hazelnuts, and Brazil nuts. Studies were developed to measure the uncertainty and distribution among replicated sample aflatoxin test results taken from aflatoxin-contaminated treenut lots. The uncertainty and distribution information is used to develop a model that can evaluate the performance (risk of misclassifying lots) of aflatoxin sampling plan designs for treenuts. Once the performance of aflatoxin sampling plans can be predicted, they can be designed to reduce the risks of misclassifying lots traded in either the domestic or export markets. A method was developed to evaluate the performance of sampling plans designed to detect aflatoxin in hazelnuts lots. Twenty hazelnut lots with varying levels of contamination were sampled according to an experimental protocol where 16 test samples were taken from each lot. The observed aflatoxin distribution among the 16 aflatoxin sample test results was compared to lognormal, compound gamma, and negative binomial distributions. The negative binomial distribution was selected to model aflatoxin distribution among sample test results because it gave acceptable fits to observed distributions among sample test results taken from a wide range of lot concentrations. Using the negative binomial distribution, computer models were developed to calculate operating characteristic curves for specific aflatoxin sampling plan designs. The effect of sample size and accept/reject limits on the chances of rejecting good lots (sellers' risk) and accepting bad lots (buyers' risk) was demonstrated for various sampling plan designs.

Download Full-text

Comparing the USDA/AMS Subsampling Mill to a Vertical Cutter Mixer Type Mill Used to Comminute Shelled Peanut Samples for Aflatoxin Analysis1

Peanut Science ◽

10.3146/ps11-17.1 ◽

2012 ◽

Vol 39 (1) ◽

pp. 69-81 ◽

Cited By ~ 2

Author(s):

Thomas B. Whitaker ◽

Andrew B. Slate

Keyword(s):

Operating Characteristic ◽

Test Procedure ◽

Sampling Plan ◽

Test Results ◽

Test Portion ◽

Two Samples ◽

The Us ◽

Sample Test ◽

Total Variability ◽

Very High

Abstract The US peanut industry can use up to three 21.8 kg samples per lot to determine if shelled peanut lots are acceptable or unacceptable due to aflatoxin content. If a lot is accepted by the first 21.8 kg sample (1AB≤8 ng/g) prepared with the USDA/AMS Subsampling mill (DM), then some peanut buyers request that the sheller prepare the second sample (2AB) and in some cases the 3AB sample (called special samples in the trade) with a vertical cutter mixer (VCM) type mill. These requests to specifically use the VCM instead of the DM to prepare official (1AB) and special (2AB or 3AB) samples is based in part on a perception that analytical results associated with a test portion taken from the 21.8 kg sample comminuted with a DM does not detect the full magnitude of aflatoxin in the 21.8 kg sample and that negative aflatoxin certificates (lot acceptance) are more likely to occur when samples are prepared with a DM than a VCM. Analysis of aflatoxin test results from two shellers along with Monte Carlo simulation indicate that differences between the 1AB and special sample test results are due to the use of a cut-off limit (≤ 8 ng/g associated with the 1AB) requested by the buyer as part of the acceptance criteria and not due to any bias associated with the DM. Operating characteristic curves were used to demonstrate that the performance of the USDA/AMS aflatoxin sampling plan is about the same regardless of the use of a DM or a VCM for sample preparation. The performances are similar because the DM, with an 1100 g test portion, account for only 8% of the total variability of the aflatoxin test procedure (sampling and analysis account for about 92%). A sampling plan that requires two 21.8 kg samples to test less than a limit, regardless of mill used to prepare the two samples, has a very low risk of accepting bad lots above the FDA limit of 20 ng/g, but has a very high risk of rejecting good lots, which makes for an extremely high economic burden on the sheller.

Download Full-text

Sampling dried figs for aflatoxin - Part II: effect of sampling plan design on reducing the risk of misclassifying lots

World Mycotoxin Journal ◽

10.3920/wmj2016.2127 ◽

2017 ◽

Vol 10 (2) ◽

pp. 99-109 ◽

Cited By ~ 1

Author(s):

H. Ozer ◽

H.I. Oktay Basegmez ◽

T.B. Whitaker ◽

A.B. Slate ◽

F.G. Giesbrecht

Keyword(s):

Negative Binomial Distribution ◽

Binomial Distribution ◽

Negative Binomial ◽

Sampling Plan ◽

Export Market ◽

Test Results ◽

Sampling Plans ◽

Binomial Distributions ◽

Sample Test ◽

Codex Standard

Because aflatoxin limits vary widely among regulating countries, the Codex Committee on Contaminants in Foods (CCCF) began work in 2006 to harmonise maximum levels (MLs) and sampling plans for aflatoxin in dried figs. Studies were developed to measure the variability and distribution among replicated sample aflatoxin test results taken from the same aflatoxin contaminated lot of dried figs so that a model could be developed to evaluate the risk of misclassifying lots of dried figs by aflatoxin sampling plan designs. The model was then be used by the CCCF electronic working group (eWG) to recommend MLs and aflatoxin sampling plan designs to the full CCCF membership for lots traded in the export market. Sixteen 10 kg samples were taken from each of 20 dried fig lots with varying levels of contamination. The observed aflatoxin distribution among the 16-aflatoxin sample test results was compared to the normal, lognormal, compound gamma, and negative binomial distributions. The negative binomial distribution was selected to model aflatoxin distribution among sample test results because it gave acceptable fits to observed aflatoxin distributions among sample test results taken from the same contaminated lot. Using the negative binomial distribution, a computer model was developed to show the effect of the number and size of samples and the accept/reject limits on the chances of rejecting good lots (seller's risk) and accepting bad lots (buyer's risk). The information was shared with the CCCF eWG and in March 2012, the 6th session of CCCF adopted at step 5/8 an aflatoxin sampling plan where three 10 kg samples must all test less than an ML of 10 µg/kg total aflatoxins to accept a dried fig lot. The 35th Session of the Codex Alimentarius Commission met in July 2012 and adopted the CCCF recommendations for the ML and the sampling plan as an official Codex standard.

Download Full-text

Evaluation of Sampling Plans to Detect Cry9C Protein in Corn Flour and Meal

Journal of AOAC International ◽

10.1093/jaoac/87.4.950 ◽

2004 ◽

Vol 87 (4) ◽

pp. 950-960 ◽

Cited By ~ 5

Author(s):

Thomas B Whitaker ◽

Mary W Trucksess ◽

Francis G Giesbrecht ◽

Andrew B Slate ◽

Francis S Thomas

Keyword(s):

Normal Distribution ◽

Test Procedure ◽

Enzyme Linked Immunosorbent Assay ◽

Sampling Variance ◽

Regression Equations ◽

Corn Flour ◽

Test Procedures ◽

Sampling Plans ◽

Sample Test ◽

Almost All

Abstract StarLink is a genetically modified corn that produces an insecticidal protein, Cry9C. Studies were conducted to determine the variability and Cry9C distribution among sample test results when Cry9C protein was estimated in a bulk lot of corn flour and meal. Emphasis was placed on measuring sampling and analytical variances associated with each step of the test procedure used to measure Cry9C in corn flour and meal. Two commercially available enzyme-linked immunosorbent assay kits were used: one for the determination of Cry9C protein concentration and the other for % StarLink seed. The sampling and analytical variances associated with each step of the Cry9C test procedures were determined for flour and meal. Variances were found to be functions of Cry9C concentration, and regression equations were developed to describe the relationships. Because of the larger particle size, sampling variability associated with cornmeal was about double that for corn flour. For cornmeal, the sampling variance accounted for 92.6% of the total testing variability. The observed sampling and analytical distributions were compared with the Normal distribution. In almost all comparisons, the null hypothesis that the Cry9C protein values were sampled from a Normal distribution could not be rejected at 95% confidence limits. The Normal distribution and the variance estimates were used to evaluate the performance of several Cry9C protein sampling plans for corn flour and meal. Operating characteristic curves were developed and used to demonstrate the effect of increasing sample size on reducing false positives (seller's risk) and false negatives (buyer's risk).

Download Full-text

Designing Sampling Plans to Detect Foreign Material in Bulk Lots of Shelled Peanut1

Peanut Science ◽

10.3146/ps07-115.1 ◽

2008 ◽

Vol 35 (2) ◽

pp. 159-164 ◽

Cited By ~ 4

Author(s):

T. B. Whitaker ◽

A. B. Slate ◽

F. G. Giesbrecht

Keyword(s):

Binomial Distribution ◽

Small Sample ◽

Sampling Plan ◽

Test Results ◽

Foreign Material ◽

Sampling Plans ◽

Sample Test ◽

The Mean ◽

Mean Variance ◽

Food Manufacturer

Abstract When food manufacturers specify a maximum limit for the amount of foreign material (FM) in the lot, handlers estimate the true percent FM in a commercial lot by measuring FM in a small sample taken from the lot before shipment to a food manufacturer. Because of the uncertainty (variability) in FM among samples taken from the same lot, it is difficult to obtain a precise estimate of the true FM in the lot. The objectives of this study were to (1) measure the variability and FM distribution among sample test results when estimating the true lot proportion of FM in a lot of shelled peanuts, (2) compare the measured variability and FM distribution among sample test results to that predicted by the binomial distribution, (3) develop a computer model, based upon the binomial distribution, to evaluate the performance (buyer's risk and seller's risk) of sampling plan designs used to estimate FM in a bulk lot of shelled peanuts, and (4) demonstrate with the model the effect of increasing sample size to reduce misclassification of lots. Eighty-eight samples, 9 kg (20 lb) each, were selected at random from each of six commercial lots of shelled medium runner peanuts. The percent FM (PFM), based upon number of kernels was determined for each sample. The mean, variance, and distribution among the 88 sample test results were calculated for each of the six lots. Results indicated that the variance and distribution among the 88 sample test results are very similar to that predicted by the binomial distribution. The performance of various sampling plan designs was demonstrated using the binomial distribution.

Download Full-text

Evaluating the Performance of Sampling Plans to Detect Hypoglycin A in Ackee Fruit Shipments Imported into the United States

Journal of AOAC International ◽

10.1093/jaoac/90.4.1060 ◽

2007 ◽

Vol 90 (4) ◽

pp. 1060-1072 ◽

Cited By ~ 9

Author(s):

Thomas B Whitaker ◽

Joyce J Saltsman ◽

George M Ware ◽

Andrew B Slate

Keyword(s):

United States ◽

Variance Components ◽

Monitoring Program ◽

The United States ◽

Sampling Plan ◽

Regression Equations ◽

Test Results ◽

Experimental Protocol ◽

Sampling Plans ◽

Lognormal Distributions

Abstract Hypoglycin A (HGA) is a toxic amino acid that is naturally produced in unripe ackee fruit. In 1973, the U.S. Food and Drug Administration (FDA) placed a worldwide import alert on ackee fruit, which banned the product from entering the United States. The FDA has considered establishing a regulatory limit for HGA and lifting the ban, which will require development of a monitoring program. The establishment of a regulatory limit for HGA requires the development of a scientifically based sampling plan to detect HGA in ackee fruit imported into the United States. Thirty-three lots of ackee fruit were sampled according to an experimental protocol in which 10 samples, i.e., ten 19 oz cans, were randomly taken from each lot and analyzed for HGA by using liquid chromatography. The total variance was partitioned into sampling and analytical variance components, which were found to be a function of the HGA concentration. Regression equations were developed to predict the total, sampling, and analytical variances as a function of HGA concentration. The observed HGA distribution among the test results for the 10 HGA samples was compared with the normal and lognormal distributions. A computer model based on the lognormal distribution was developed to predict the performance of sampling plan designs to detect HGA in ackee fruit shipments. The performance of several sampling plan designs was evaluated to demonstrate how to manipulate sample size and accept/reject limits to reduce misclassification of ackee fruit lots.

Download Full-text

Comparing the performance of sampling plans that use a single regulatory limit based upon total aflatoxins to sampling plans that use dual limits based upon B1 and total aflatoxins

World Mycotoxin Journal ◽

10.3920/wmj2009.1169 ◽

2010 ◽

Vol 3 (1) ◽

pp. 35-44 ◽

Cited By ~ 7

Author(s):

T. Whitaker ◽

A. Slate ◽

J. Adams ◽

T. Birmingham ◽

F. Giesbrecht

Keyword(s):

European Union ◽

Monte Carlo ◽

Test Results ◽

The European Union ◽

Sampling Plans ◽

Tree Nuts ◽

Study Results ◽

Sample Test ◽

The Mean ◽

Aflatoxin B

The European Commission (EC) aflatoxin sampling plan for ready-to-eat tree nuts such as almonds requires that each of the three 10 kg laboratory samples must all test less than 2 ng/g aflatoxin B1 (AFB1) and 4 ng/g total aflatoxins (AFT) for the lot to be accepted. Exporters have observed that the AFB1/AFT ratio varied greatly from sample to sample and the ratio appeared to average more than 50%. Because of the concern that dual limits associated with the EC aflatoxin sampling plans may reject more lots than similar sampling plans that use a single limit based upon total aflatoxins, studies were designed with the objectives to (a) measure the distribution of AFB1/AFT ratio values using sample test results associated with testing U.S. almond lots exported to the European Union; (b) use Monte Carlo methods to develop a model to compute the effects of using dual limits based upon AFB1 and AFT on the probability of accepting almond lots; and (c) compare the probability of accepting almond lots using the current Codex aflatoxin sampling plans for tree nuts when using single limits versus the use of dual limits. The study results showed that the mean and median among 3,257 AFB1/AFT ratio values was 87.6% and 91.9%, respectively, indicating that the distribution among the ratio values was negatively skewed. Only 31% of the 3,257 AFB1/AFT ratio values are less than the mean ratio of 87.6%. Codex aflatoxin sampling plans for tree nuts using a single limit based upon total aflatoxins had the highest probability of accepting lots at all lot concentrations when compared to the probability of accepting lots with dual limits. As the AFB1 limit decreased from 90 to 50% of the total limit, the probability of rejecting lots at all concentrations increased when compared to the Codex aflatoxin sampling plans with a single limit based upon total aflatoxins.

Download Full-text

Predicting the Distribution of Aflatoxin Test Results from Farmers’ Stock Peanuts

Journal of AOAC International ◽

10.1093/jaoac/77.3.659 ◽

1994 ◽

Vol 77 (3) ◽

pp. 659-666 ◽

Cited By ~ 2

Author(s):

Thomas B Whitaker ◽

Francis G Giesbrecht ◽

Jeremy Wu ◽

Winston M Hagler ◽

Floyd E Dowell

Keyword(s):

Negative Binomial ◽

Accurate Estimate ◽

Test Results ◽

Sample Sizes ◽

Significance Level ◽

Sample Test ◽

Unknown Distribution Function ◽

Acceptance Levels ◽

Ks Test ◽

Binomial Function

Abstract Suitability of the negative binomial function for use in estimating the distribution of sample aflatoxin test results associated with testing farmers1 stock peanuts for aflatoxin was studied. A 900 kg portion of peanut pods was removed from each of 40 contaminated farmers1 stock lots. The lots averaged about 4100 kg. Each 900 kg portion was divided into fifty 2.26 kg samples, fifty 4.21 kg samples, and fifty 6.91 kg samples. The aflatoxin in each sample was quantified by liquid chromatography. An observed distribution of sample aflatoxin test results consisted of 50 aflatoxin test results for each lot and each sample size. The mean aflatoxin concentration, m; the variance, s2xamong the 50 sample aflatoxin test results; and the shape parameter, k, for the negative binomial function were determined for each of the 120 observed distributions (40 lots times 3 sample sizes). Regression analysis indicated the functional relationship between k and m to be k = 0.000006425m0.8047. The 120 observed distributions of sample aflatoxin test results were compared to the negative binomial function by using the Kolmogorov–Smirnov (KS) test. The null hypothesis that the true unknown distribution function was negative binomial was not rejected at the 5% significance level for 114 of the 120 distributions. The negative binomial function failed the KS test at a sample concentration of 0 ng/g in all 6 of the distributions where the negative binomial function was rejected. The negative binomial function always predicted a smaller percentage of samples testing 0 ng/g than was actually observed. However, the negative binomial function did fit the observed distribution for sample test results at a concentration greater than 0 in 4 of the 6 cases. As a result, the negative binomial function provides an accurate estimate of the acceptance probabilities associated with accepting contaminated lots of farmers' stock peanuts for various sample sizes and various sample acceptance levels greater than 0 ng/g.

Download Full-text

Variability and distribution among sample test results when sampling unprocessed wheat lots for ochratoxin A

World Mycotoxin Journal ◽

10.3920/wmj2015.1970 ◽

2016 ◽

Vol 9 (2) ◽

pp. 163-178 ◽

Cited By ~ 3

Author(s):

T.B. Whitaker ◽

A.B. Slate ◽

T.W. Nowicki ◽

F.G. Giesbrecht

Keyword(s):

Liquid Chromatography ◽

Sample Preparation ◽

Ochratoxin A ◽

Test Procedure ◽

Test Results ◽

Laboratory Sample ◽

Sampling Step ◽

Test Portion ◽

Sample Test ◽

Sampling Statistics

In 2008, Health Canada announced it was considering the establishment of maximum levels for ochratoxin A (OTA) in unprocessed wheat, oats, and their products. The Canada Grains Council and Canadian National Millers Association initiated two studies to measure the variability and distribution among sample test results for unprocessed wheat and oats so that scientifically based OTA sampling plans could be designed to meet regulatory and industry requirements. Sampling statistics related to detecting OTA in oats has been published. 54 OTA contaminated wheat lots representing three wheat classes were identified for the sampling study. Each lot was sampled according to a nested experimental protocol where sixteen 2-kg laboratory samples were taken from each lot, multiple 5-g test portions were taken from each comminuted 2-kg laboratory sample, and multiple OTA measurements were made on each test portion using liquid chromatography. The sampling, sample preparation, and analytical variances associated with each step of the OTA test procedure were found to be a function of OTA concentration and regression equations were developed to predict the functional relationships between variance and OTA concentration. When sampling a wheat lot containing 5 µg/kg OTA with an OTA test procedure consisting of a sampling step employing a single 2-kg laboratory sample, sample preparation step employing a single 100-g test portion, and an analytical step that used liquid chromatography to quantify OTA, the sampling step accounted for 95.3% of the total variability. The observed OTA distribution among the 16 OTA sample results was found to be positively skewed and the negative binomial distribution was selected to model the OTA distribution among sample test results. The sampling statistics were incorporated into the FAO Mycotoxin Sampling Tool and the chances of rejecting good lots and accepting bad lots were calculated for various sampling plan designs.

Download Full-text