A big data approach to macrofaunal baseline assessment, monitoring and sustainable exploitation of the seabed.

In this study we produce a standardised dataset for benthic macrofauna and sediments through integration of data (33,198 samples) from 777 grab surveys. The resulting dataset is used to identify spatial and temporal patterns in faunal distribution around the UK, and the role of sediment composition and other explanatory variables in determining such patterns. We show how insight into natural variability afforded by the dataset can be used to improve the sustainability of activities which affect sediment composition, by identifying conditions which should remain favourable for faunal recolonisation. Other big data applications and uses of the dataset are discussed.

Metabolomics reveals novel biomarkers of illegal 5-nitromimidazole treatment in pigs. Further evidence of drug toxicity uncovered.

The aim of the study was to investigate the potential of a metabolomics platform to distinguish between pigs treated with ronidazole, dimetridazole and metronidazole and non-medicated animals (controls), at two withdrawal periods (day 0 and 5). Livers from each animal were biochemically profiled using UHPLC-QTof-MS in ESI+ mode of acquisition. Several Orthogonal Partial Least Squares-Discriminant Analysis models were generated from the acquired mass spectrometry data. The models classified the two groups control and treated animals. A total of 42 ions of interest explained the variation in ESI+. It was possible to find the identity of 3 of the ions and to positively classify 4 of the ionic features, which can be used as potential biomarkers of illicit 5-nitroimidazole abuse. Further evidence of the toxic mechanisms of 5-nitroimidazole drugs has been revealed, which may be of substantial importance as metronidazole is widely used in human medicine.

Mass spectrometric analysis of muscle samples to detect potential antibiotic growth promoter misuse in broiler chickens.

Mass spectrometric methods were developed and validated for the analysis in chicken muscle of a range of antibiotic growth promoters: spiramycin, tylosin, virginiamycin and bacitracin, and separately for two marker metabolites of carbadox (quinoxaline-2-carboxylic acid and 1,4-bisdesoxycarbadox), and a marker metabolite of olaquindox (3-methyl-quinoxaline-2-carboxylic acid). The use of these compounds as antibiotic growth promoters has been banned by the European Commission. This study aimed to develop methods to detect their residues in muscle samples as a means of checking for the use of these drugs during the rearing of broiler chickens. When fed growth-promoting doses for 6 days, spiramycin (31.4 µg kg(-1)), tylosin (1.0 µg kg(-1)), QCA (6.5 µg kg(-1)), DCBX (71.2 µg kg(-1)) and MQCA (0.2 µg kg(-1)) could be detected in the muscle 0 days after the withdrawal of fortified feed. Only spiramycin could consistently be detected beyond a withdrawal period of 1 day. All analytes showed stability to a commercial cooking process, therefore raw or cooked muscle could be used for monitoring purposes.

Emergency slaughter of casualty cattle increases the prevalence of anthelmintic drug residues in muscle.

The ProSafeBeef project studied the prevalence of residues of anthelmintic drugs used to control parasitic worms and fluke in beef cattle in Ireland. Injured (casualty) cattle may enter the human food chain under certain conditions, verified by an attending veterinarian and the livestock keeper. An analytical survey was conducted to determine if muscle from casualty cattle contained a higher prevalence of anthelmintic drug residues than healthy (full slaughter weight) cattle as a result of possible non-observance of complete drug withdrawal periods. A validated analytical method based on matrix solid-phase dispersive extraction (QuEChERS) and ultra-performance liquid chromatography-tandem mass spectrometry was used to quantify 37 anthelmintic drugs and metabolites in muscle (assay decision limits, CCα, 0.15-10.2 µg kg⁻¹). Of 199 control samples of beef purchased in Irish shops, 7% contained detectable anthelmintic drug residues but all were compliant with European Union Maximum Residue Limits (MRL). Of 305 muscle samples from injured cattle submitted to abattoirs in Northern Ireland, 17% contained detectable residues and 2% were non-compliant (containing either residues at concentrations above the MRL or residues of a compound unlicensed for use in cattle). Closantel and ivermectin were the most common residues, but a wider range of drugs was detected in muscle of casualty cattle than in retail beef. These data suggest that specific targeting of casualty cattle for testing for anthelmintic residues may be warranted in a manner similar to the targeted testing for antimicrobial compounds often applied in European National Residues Surveillance Schemes.

Anthelmintic drug residues in beef: UPLC-MS/MS method validation, European retail beef survey, and associated exposure and risk assessments.

Anthelmintic drugs are widely used to control parasitic infections in cattle. The ProSafeBeef project addressed the need for data on the exposure of European consumers of beef to potentially harmful drug residues. A novel analytical method based on matrix solid-phase dispersive extraction and ultra-performance liquid chromatography-tandem mass spectrometry was validated for 37 anthelmintic drugs and metabolites in muscle (assay decision limits, CCα, = 0.15-10.2 µg kg⁻¹). Seven European countries (France, Spain, Slovenia, Ireland, Italy, Belgium and Portugal) participated in a survey of retail beef purchased in local shops. Of 1061 beef samples analysed, 26 (2.45%) contained detectable residues of anthelmintic drugs (0.2-171 µg kg⁻¹), none above its European Union maximum residue limit (MRL) or action level. Residues detected included closantel, levamisole, doramectin, eprinomectin, moxidectin, ivermectin, albendazole and rafoxanide. In a risk assessment applied to mean residue concentrations across all samples, observed residues accounted for less than 0.1% of the MRL for each compound. An exposure assessment based on the consumption of meat at the 99th percentile of consumption of adults in 14 European countries demonstrated that beef accounted for less than 0.02% of the acceptable daily intake for each compound in each country. This study is the first of its kind to apply such a risk-based approach to an extensive multi-residue survey of veterinary drug residues in food. It has demonstrated that the risk of exposure of the European consumer to anthelmintic drug residues in beef is negligible, indicating that regulation and monitoring is having the desired effect of limiting residues to non-hazardous concentrations.

Implications of dredging induced changes in sediment particle size composition for the structure and function of marine benthic macrofaunal communities.

A meta-analysis approach was used to assess the effect of dredging induced changes in sediment composition, under different conditions of natural physical disturbance, for the structure and function of marine benthic macrofaunal communities. Results showed the sensitivity of macrofaunal communities increased as both the proportion of gravel increased and the level of natural physical disturbance decreased. These findings may be explained by the close association of certain taxa with the gravel fraction, and the influence of natural physical disturbance which, as it increases, tends to restrict the colonisation by these species. We conclude that maintaining the gravel content of surface sediments after dredging and, where practicable, locating extraction sites in areas of higher natural disturbance will minimise the potential for long-term negative impacts on the macrofauna.

Stability during cooking of anthelmintic veterinary drug residues in beef.

Anthelmintic drugs are widely used for treatment of parasitic worms in livestock, but little is known about the stability of their residues in food under conventional cooking conditions. As part of the European Commission-funded research project ProSafeBeef, cattle were medicated with commercially available anthelmintic preparations, comprising 11 active ingredients (corresponding to 21 marker residues). Incurred meat and liver were cooked by roasting (40 min at 190°C) or shallow frying (muscle 8-12 min, liver 14-19 min) in a domestic kitchen. Raw and cooked tissues and expressed juices were analysed using a novel multi-residue dispersive solid-phase extraction method (QuEChERS) coupled with ultra-performance liquid chromatography-tandem mass spectrometry. After correction for sample weight changes during cooking, no major losses were observed for residues of oxyclozanide, clorsulon, closantel, ivermectin, albendazole, mebendazole or fenbendazole. However, significant losses were observed for nitroxynil (78% in fried muscle, 96% in roast muscle), levamisole (11% in fried muscle, 42% in fried liver), rafoxanide (17% in fried muscle, 18% in roast muscle) and triclabendazole (23% in fried liver, 47% in roast muscle). Migration of residues from muscle into expressed cooking juices varied between drugs, constituting 0% to 17% (levamisole) of total residues remaining after cooking. With the exception of nitroxynil, residues of anthelmintic drugs were generally resistant to degradation during roasting and shallow frying. Conventional cooking cannot, therefore, be considered a safeguard against ingestion of residues of anthelmintic veterinary drugs in beef.

A 4-week randomized study of acetaminophen extended-release vs rofecoxib in knee osteoarthritis.

OBJECTIVE: To compare the safety and efficacy of acetaminophen extended-release (APAP ER) with rofecoxib for the management of pain associated with knee osteoarthritis (OA). METHODS: Four hundred and three adult patients with moderate pain secondary to knee OA were randomized to receive APAP ER 1300 mg three times daily, rofecoxib 12.5mg once daily, or rofecoxib 25mg once daily. Primary end point was change from baseline at week 4 in the Western Ontario and McMaster Universities Osteoarthritis Index pain subscale score using a visual analog scale. This 4-week study was conducted at 23 US research sites from October 1999 to October 2000. RESULTS: APAP ER was noninferior to rofecoxib 12.5mg because the upper 95% confidence limit (CL) for the least squares mean (LSM) change from baseline (35.27 mm at week 4) did not exceed the prespecified noninferiority limit of 50mm. The upper CL (57.39 mm) exceeded the noninferiority limit for APAP ER compared with rofecoxib 25mg at week 4. There were no significant differences among groups in the overall incidence of adverse events. CONCLUSION: APAP ER 3900 mg daily was noninferior to rofecoxib 12.5mg daily, but noninferiority was not established to rofecoxib 25mg daily. APAP ER was well tolerated and no safety issues were identified. Based on the results of this study, APAP ER 3900 mg daily is an alternative to nonsteroidal anti-inflammatory drugs (NSAIDs), such as rofecoxib, in treating pain associated with knee OA.

Residues of nitrofuran antibiotic parent compounds and metabolites in eyes of broiler chickens.

Nitrofuran antibiotic residues in food continue to be of international concern. The finding of sources of semicarbazide (SEM), other than through the misuse of nitrofurazone, present a challenge to the use of SEM as a definitive marker residue for this drug. Detection of intact (parent) nitrofurazone would avoid confusion over the source of SEM residues. Broiler chickens were fed sub-therapeutic nitrofuran-containing diets and their tissues were analysed for parent compounds and metabolites by liquid chromatography coupled with tandem mass spectrometry detection (LC-MS/MS). Depletion half-lives in muscle were longer for tissue-bound metabolite residues, 3.4 days --3-amino-2-oxazolidinone (AOZ), 3-amino-5-morpholinomethyl-2-oxazolidone (AMOZ) -- to 4.5 days (SEM), than total metabolite residues, 2.0 days (AOZ) to 3.2 days (SEM). Metabolite concentrations were higher in eyes than in muscle. Metabolite half-lives in eyes ranged from 8.5 days (1-aminohydantoin (AHD)) to 20.3 days (SEM). Nitrofuran parent compounds were also detected in eyes. Furaltadone was detected in single eyes after 21 days' withdrawal of a 6 mg kg(-1) furaltadone diet. When 50 eyes from broilers containing metabolites in muscle close to the 1 microg kg(-1) minimum required performance level (MRPL) were pooled into single samples, 1.2 ng of furazolidone and 31.1 ng of furaltadone were detected, but nitrofurazone was not detected due to the long depletion half-life of SEM in muscle. Further studies are required to improve LC-MS/MS nitrofurazone sensitivity and refine the sample size necessary to use nitrofurazone detection in pooled eyes as a complement to SEM detection in muscle.

Kinetics of semicarbazide and nitrofurazone in chicken eggs and egg powders.

The accumulation, depletion and partitioning of semicarbazide (SEM) and its parent compound nitrofurazone (NFZ) in eggs were studied using hens fed NFZ at therapeutic and sub-therapeutic levels. Dietary NFZ correlated strongly with NFZ and total SEM in eggs, while 28% of observed SEM was present in the form of parent NFZ. Depletion half-life in eggs was 2.4 days for SEM and 1.1 days for NFZ. NFZ accumulated preferentially in yolk (57-63%) as opposed to albumen, while 71-80% of SEM was found in yolk. In whole egg, 29% of SEM was present as tissue-bound residues compared with 80% in breast muscle. Whilst NFZ and SEM were partly degraded by pasteurization and spray drying, sufficient NFZ remained to suggest it might be detectable in egg powders when SEM is observed at low microg kg(-1) concentrations. NFZ was detectable in whole eggs during ingestion of only 0.1% of the therapeutic NFZ dose, making detection of intact NFZ in eggs a feasible means to prove conclusively the administration of this banned compound.

The identification of potential alternative biomarkers of nitrofurazone abuse in animal derived food products.

Semicarbazide (SEM) was considered to be a characteristic protein-bound side-chain metabolite of the banned veterinary drug nitrofurazone and used as a marker of nitrofurazone abuse. It was recently discovered that SEM can arise in food from sources other than nitrofurazone. This uncertainty over the source of SEM may be overcome if alternative markers specific to tissue-bound nitrofurazone residues can be determined. The structure of nitrofurazone metabolites in vivo and particular proteins to which they are bound are not known. These proteins with altered structure due to the presence of the drug metabolites can be considered as potential alternative biomarkers of nitrofurazone abuse. The proteins implicated in the in vivo binding of nitrofurazone were separated and identified. A crude mixture of proteins extracted from the liver of a rat treated with the drug was separated using a series of different techniques such as preparative isoelectric focusing and size exclusion HPLC. Multiple fractions were assayed by LC-MS/MS to detect the presence of SEM. The proteins containing SEM residues were identified by peptide mass mapping using trypsin digestion and MALDI-TOF. The first protein identified as containing high concentration of SEM was albumin. It was also shown that low molecular weight species within a protein mixture whose main constituent was glutathione S-transferase contained a high concentration of SEM. The chemical composition of these components is under investigation. Preliminary data suggest the SEM forms part of a nitrofurazone metabolite conjugated to glutathione.

Stability studies of the metabolites of nitrofuran antibiotics during storage and cooking.

Nitrofuran antibiotics cannot be used in food production within the European Union because of their potential health risks to consumers. The recent discovery of their widespread use in global food industries and the finding of semicarbazide in baby food as a result of packaging contamination have focused attention on the toxicity and stability of these drugs and their metabolites. The stability of the nitrofuran marker residues 3-amino-2-oxazolidinone (AOZ), 3-amino-5-morpholinomethyl-2-oxazolidone (AMOZ), 1-aminohydantoin (AHD) and semicarbazide (SEM) were tested. Muscle and liver of nitrofuran treated pigs were cooked by frying, grilling, roasting and microwaving. Between 67 and 100% of the residues remained after cooking, demonstrating that these metabolites are largely resistant to conventional cooking techniques and will continue to pose a health risk. The concentration of metabolites in pig muscle and liver did not drop significantly during 8 months of storage at -20 degrees C. Metabolite stock and working standard solutions in methanol were also stable for 10 months at 4 degrees C. Only a 10 ng ml(-1) solution of SEM showed a small drop in concentration over this extended storage period.

Depletion of four nitrofuran antibiotics and their tissue-bound metabolites in porcine tissues and determination using LC-MS/MS and HPLC-UV.

Depletion of the nitrofuran antibiotics furazolidone, furaltadone, nitrofurantoin and nitrofurazone and their tissue-bound metabolites AOZ, AMOZ, AHD and SEM from pig muscle, liver and kidney tissues is described. Groups of pigs were given feed medicated with one of the nitrofuran drugs at a therapeutic concentration (400?mg?kg(-1)) for ten days. Animals were slaughtered at intervals and tissue samples collected for analysis for six weeks following withdrawal of medicated feed. These samples were analysed both for parent nitrofurans (using LC-MS/MS and HPLC-UV), and for tissue-bound metabolites (using LC-MS/MS). The parent drugs were detectable only sporadically and only in pigs subjected to no withdrawal period whatsoever. This confirms the instability of the four major nitrofuran antibiotics in edible tissues. In contrast, the metabolites accumulated to high concentrations in tissues (ppm levels) and had depletion half lives of between 5.5 and 15.5 days. The metabolites of all four drugs were still readily detectable in tissues six weeks after cessation of treatment. This emphasizes the benefits of monitoring for the stable metabolites of the nitrofurans.

Maternal serum soluble fms-like tyrosine kinase 1 concentrations are not increased in early pregnancy and decrease more slowly postpartum in women who develop preeclampsia.

OBJECTIVE: We measured maternal serum soluble fms-like tyrosine kinase 1 concentrations across pregnancy and immediately postpartum in women who developed preeclampsia and normal pregnant women. STUDY DESIGN: This was a nested case control study of 113 normal pregnant women and 55 women with preeclampsia. RESULTS: Serum soluble fms-like tyrosine kinase 1 concentrations increased similarly in early pregnancy in both groups. Mean serum soluble fms-like tyrosine kinase 1 concentrations were increased in women who developed preeclampsia, compared with normal pregnant women, and this increase was most pronounced in severe preeclampsia. However, many women with preeclampsia had soluble fms-like tyrosine kinase 1 concentrations similar to normal pregnant women. Lastly, soluble fms-like tyrosine kinase 1 decreased rapidly after delivery, but this decrease was significantly slower in women with severe preeclampsia. CONCLUSION: Increased soluble fms-like tyrosine kinase 1 is not an early-pregnancy event among women who later develop preeclampsia. Increased soluble fms-like tyrosine kinase 1 is more likely to be present in women with severe preeclampsia, but it is not present in all women with preeclampsia. Soluble fms-like tyrosine kinase 1 concentrations decrease more slowly after delivery in women with preeclampsia, consistent with a decreased rate of excretion or continued production.

Detection of 3-amino-2-oxazolidinone (AOZ), a tissue-bound metabolite of the nitrofuran furazolidone, in prawn tissue by enzyme immunoassay.

Furazolidone, a nitrofuran antibiotic, is banned from use in food animal production within the European Union. Increasingly, compliance with this ban is monitored by use of analytical methods to detect a stable tissue-bound metabolite, 3-amino-2-oxazolidinone (AOZ). Widespread use of furazolidone in poultry and prawns imported into Europe highlighted the urgent need for development of nitrofuran immunoassay screening tests. The first enzyme-linked immunoabsorbant assay for detection of AOZ residues in prawns (shrimps) is now described. Prawn samples were derivatized with o-nitrobenzaldehyde, extracted into ethyl acetate, washed with hexane and applied to a competitive enzyme immunoassay based on a rabbit polyclonal antiserum. Assay limit of detection (LOD) (mean + 3 s) calculated from the analysis of 20 known negative cold and warm water prawn samples was 0.1 microg kg(-1). Intra- and interassay relative standard deviations were determined as 18.8 and 38.2%, respectively, using a negative prawn fortified at 0.7 microg kg(-1). The detection capability (CCbeta), defined as the concentration of AOZ at which 20 different fortified samples yielded results above the LOD, was achieved at fortification between 0.4 and 0.7 microg kg(-1). Incurred prawn samples (n = 8) confirmed by liquid chromatography coupled with tandem mass spectrometry detection to contain AOZ concentrations between 0.4 and 12.7 microg kg(-1) were all screened positive by this enzyme-linked immunoabsorbant assay. Further data are presented and discussed with regard to calculating assay LOD based on accepting a 5% false-positive rate with representative negative prawn samples. Such an acceptance improves the sensitivity of an ELISA and in this case permitted an LOD of 0.05 microg kg(-1) and a CCbeta of below 0.4 microg kg(-1).

Interlaboratory ring test of time-resolved fluoroimmunoassays for zeranol and alpha-zearalenol and comparison with zeranol test kits.

Many zeranol immunoassay test kits cross-react with toxins formed by naturally occurring Fusarium spp. fungi, leading to false-positive screening results. This paper describes the evaluation and application of recently published, dry reagent time-resolved fluoroimmunoassays (TR-FIA) for zeranol and the toxin alpha-zearalenol. A ring test of bovine urine fortified with zeranol and/or alpha-zearalenol in four European Union National Reference Laboratories demonstrated that the TR-FIA tests were accurate and robust. The alpha-zearalenol TR-FIA satisfactorily quantified alpha-zearalenol in urine fortified at 10-30 ng ml(-1). The specificity-enhanced zeranol TR-FIA accurately quantified zeranol in the range 2-5 ng ml(-1) and gave no false-positive results in blank urine, even in the presence of 30 ng ml(-1) alpha-zearalenol. Zeranol TR-FIA specificity was demonstrated further by analysing incurred zeranol-free urine samples containing natural Fusarium spp. toxins. The TR-FIA yielded no false-positive results in the presence of up to 22 ng ml(-1) toxins. The performance of four commercially available zeranol immunoassay test kits was more variable. Three kits produced many false-positive results. One kit produced only one potential false-positive using a protocol that was longer than that of the TR-FIA. These TR-FIAs will be valuable tools to develop inspection criteria to distinguish illegal zeranol abuse from contamination arising from in vivo metabolism of Fusarium spp. toxins.

Efficacy and safety of acetaminophen and naproxen in the treatment of tension-type headache. A randomized, double-blind, placebo-controlled trial.

The objective of this study was to evaluate and compare the efficacy and safety of single doses of acetaminophen (paracetamol) 1000 mg and naproxen 375 mg vs. placebo over a six-hour period in the treatment of tension-type headache. The treatments were compared in a randomized, double-blind, multicentre, placebo-controlled study. Efficacy was evaluated using four standard analgesic summary endpoints (the sum of pain intensity differences from baseline, the maximum pain intensity from baseline, the sum of the pain relief scores, and the maximum pain relief score). Both acetaminophen 1000 mg and naproxen 375 mg were significantly superior to placebo (Por=0.498) for these four endpoints. For example, the mean sum of pain intensity differences from baseline was 9.14+/-0.34 for acetaminophen 1000 mg and 8.81+/-0.35 for naproxen 375 mg compared with 7.42+/-0.34 for placebo. Other efficacy endpoints (percentage of responders (pain reduced to none) at two hours, onset of meaningful relief, time to use of rescue medication and subject's overall impression of study medication) showed similar trends. A significantly larger mean pain intensity difference from baseline was observed for acetaminophen 1000 mg (1.13) than for naproxen 375 mg (0.95) (P=0.036) at one hour after treatment. There was no significant difference among the treatment groups in the incidence of adverse events (P=0.730). In summary, the results of this well-controlled, double-blind study demonstrate that over-the-counter acetaminophen 1000 mg and prescription naproxen 375 mg are effective and well tolerated in the treatment of tough (moderate-to-severe) tension-type headache.

Development and validation of dry reagent time-resolved fluoroimmunoassays for zeranol and alpha-zearalenol to assist in distinguishing zeranol abuse from Fusarium spp. toxin contamination in bovine urine.

Zeranol, an oestrogenic growth promoter in food animals, is banned within the European Union (EU). However, commercially available immunoassay kits for zeranol cross-react with toxins formed by naturally occurring Fusarium spp. fungi, leading to false-positive screening results. This paper describes the validation of a specificity enhanced, rapid dry reagent time-resolved fluoroimmunoassay (TR-FIA) for zeranol (recovery 99%, limit of detection 1.3 ng ml(-1)) demonstrating that up to 150 ng ml(-1) of Fusarium spp. toxins in urine do not lead to false-positive results. This assay will assist EU Member States to implement Council Directive 96/23/EC, which requires states to monitor for potential abuses of zeranol. A similar TR-FIA for the Fusarium spp. toxin alpha-zearalenol, using the same sample extract, is also described (recovery 68%, limit of detection 5.6 ng ml(-1)). Only the addition of diluted sample extract is required to perform these dry-reagent TRFIAs, the results being available within 1h of extract application. The EU-funded project 'Natural Zeranol' (FAIR5-CT97-3443) will use these fluoroimmunoassays to screen bovine urine in four Member States to gather data on the seasonality of Fusarium spp. toxin contamination of urine and the incidence of zeranol screening test positives.

Health barriers to walking for exercise in elderly primary care.

Health status is acknowledged by nursing theory as a factor in the probability of health behavior changes. Yet few studies have addressed the health-related barriers encountered by chronically ill elderly patients in primary care clinics who are trying to increase their physical activity. This study used self- and interviewer-administered instruments to assess potential barriers to physical activity in a sample of 60- to 80-year-old patients entering a walking program. Pain, fatigue, and mobility and sensory impairments were prevalent and could be significant barriers to participation. The authors present specific suggestions for helping patients overcome these barriers.

Optimizing the balance between false positive and false negative error probabilities of confirmatory methods for the detection of veterinary drug residues.

GC-MS data on veterinary drug residues in bovine urine are used for controlling the illegal practice of fattening cattle. According to current detection criteria, peak patterns of preferably four ions should agree within 10 or 20% from a corresponding standard pattern. These criteria are rigid, rather arbitrary and do not match daily practice. A new model, based on multivariate modeling of log peak abundance ratios, provides a theoretical basis for the identification of analytes and optimizes the balance between the avoidance of false positives and false negatives. The performance of the model is demonstrated on data provided by five laboratories, each supplying GC-MS measurements on the detection of clenbuterol, dienestrol and 19 beta-nortestosterone in urine. The proposed model shows a better performance than confirmation by using the current criteria and provides a statistical basis for inspection criteria in terms of error probabilities.