Data Representativeness: Issues and Solutions

Milanzi, Elasma; Njagi, Edmund Njeru; Bruckers, Liesbeth; Molenberghs, Geert

doi:10.2903/sp.efsa.2015.en-759

Cited by 7 publications

(6 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, from a statistical point of view, in the design of future studies it would be interesting to combine aspects of the MAS protocol with aspects of survey sampling methodology (15,44). The MAS protocol could be used to define the CSLs, with survey sampling principles being used to specify a survey sampling design and to calculate the number of factories and the number of samples at each CSL in each factory that would be needed to achieve the objectives of the statistical analysis.…”

Section: Resultsmentioning

confidence: 99%

Microbial Performance of Food Safety Control and Assurance Activities in a Fresh Produce Processing Sector Measured Using a Microbial Assessment Scheme and Statistical Modeling

Njage

Sawe

Onyango

et al. 2017

Journal of Food Protection

Self Cite

View full text Add to dashboard Cite

Current approaches such as inspections, audits, and end product testing cannot detect the distribution and dynamics of microbial contamination. Despite the implementation of current food safety management systems, foodborne outbreaks linked to fresh produce continue to be reported. A microbial assessment scheme and statistical modeling were used to systematically assess the microbial performance of core control and assurance activities in five Kenyan fresh produce processing and export companies. Generalized linear mixed models and correlated random-effects joint models for multivariate clustered data followed by empirical Bayes estimates enabled the analysis of the probability of contamination across critical sampling locations (CSLs) and factories as a random effect. Salmonella spp. and Listeria monocytogenes were not detected in the final products. However, none of the processors attained the maximum safety level for environmental samples. Escherichia coli was detected in five of the six CSLs, including the final product. Among the processing-environment samples, the hand or glove swabs of personnel revealed a higher level of predicted contamination with E. coli, and 80% of the factories were E. coli positive at this CSL. End products showed higher predicted probabilities of having the lowest level of food safety compared with raw materials. The final products were E. coli positive despite the raw materials being E. coli negative for 60% of the processors. There was a higher probability of contamination with coliforms in water at the inlet than in the final rinse water. Four (80%) of the five assessed processors had poor to unacceptable counts of Enterobacteriaceae on processing surfaces. Personnel-, equipment-, and product-related hygiene measures to improve the performance of preventive and intervention measures are recommended.

show abstract

Section: Resultsmentioning

confidence: 99%

Microbial Performance of Food Safety Control and Assurance Activities in a Fresh Produce Processing Sector Measured Using a Microbial Assessment Scheme and Statistical Modeling

Njage

Sawe

Onyango

et al. 2017

Journal of Food Protection

Self Cite

View full text Add to dashboard Cite

show abstract

“…In general, a simple random sampling scheme will suffice for detection surveys, while stratified sampling may be applied in some cases. The methodology for sample size calculations in pest freedom surveys (detection and delimiting surveys) has been described by EFSA (2012), while Milanzi et al (2015) have described the methodology that can be applied for monitoring surveys. These methodologies have been developed for animal pests and maximum residue level compliance, respectively, but the underlying principles are the same for plant pests.…”

Section: Statistical Background For Sample Size Estimationmentioning

confidence: 99%

General guidelines for statistically sound and risk‐based surveys of plant pests

Lázaro¹,

Parnell²,

Civera³

et al. 2020

EFS3

View full text Add to dashboard Cite

At the request of the European Commission, EFSA prepared the general guidelines for surveys of plant pests, describing the legal, international and scientific context in which the surveys are designed, the basic principles implemented for surveillance of quarantine pests and introducing the concepts needed for the design of statistically sound and risk-based surveys. Three types of specific surveys are addressed: detection surveys for substantiation of pest freedom, delimiting surveys for determining the boundaries of an infested zone, and monitoring surveys for prevalence estimation when measuring the progress of eradication measures or for confirming a low pest prevalence area. For each survey, the survey parameters are introduced and their interactions analysed showing the importance of the assumptions that are taken for each one of them: (i) the aims of the survey are defined as the confidence of detecting a given pest prevalence (design prevalence), this reflects the trade-off between the acceptable level of the risk and availability of resources that determine the strength of the evidence to support the conclusion of the survey; (ii) the target population is addressed in terms of its structure and size, including the risk factors; and (iii) the method sensitivity is defined as the combination of the sampling effectiveness and the diagnostic sensitivity. EFSA's RiBESS+ tool is introduced for calculating the sample size using the survey parameters as input values for a statistically sound and risk-based survey design. The mathematical principles behind the tool are in line with the International Standards for Phytosanitary Measures. The survey design is flexible and can be tailored to each pest and specific situation in the Member States. Once the survey is implemented following this approach, the conclusions allow surveys to be compared across time and space, contributing to the harmonisation of surveillance activities across the EU Member States.

show abstract

“…Estimation objectives mainly involve production of quantitative and numerical descriptions (estimation) of relevant aspects of a target population, like the population mean or the population total, mean difference between two groups of the same population and proportion of the population with a trait of interest, etc. On the other hand, inferential objectives are about testing a particular hypothesis about the population of interest, examples include, testing that the population mean is greater (less) than a certain value, or that means of groups of the same population are not equal (Milanzi et al, 2015). For the baseline survey, we are seeking to estimate the likelihood that a sample unit is contaminated with NoV.…”

Section: Survey Objectivesmentioning

confidence: 99%

“…A clear definition of targeted population is important to determine the extent to which results from the survey can be generalised (Milanzi et al, 2015).…”

Section: Survey Objectivesmentioning

confidence: 99%

Technical specifications for a European baseline survey of norovirus in oysters

Publication¹

2016

EFS2

View full text Add to dashboard Cite

The European Commission requested scientific technical assistance in the preparation of a survey protocol for a European Union (EU) coordinated monitoring programme on the prevalence of norovirus (NoV) in raw oysters. The objective of the survey is to estimate the European prevalence of norovirus‐contaminated oysters at production areas and batches of oysters at dispatch centres, with a 95% level of confidence and a level of precision of 5% considering an expected prevalence of 50%. The survey protocol defines the target population, the sample size for the survey, sample collection requirements, the analytical method for the quantification of NoV copy number (genotype I and genotype II), the data reporting requirements and the plan of analysis. The sample unit in production areas is a classified production area actively growing commercial oysters (whether harvesting or not is occurring) and for dispatch centres is a quantity of live oysters which are being packed and labelled with an Identification Mark. Based on a multistage sampling scheme, 1,026 samples from 171 production areas and 1,182 samples from 197 dispatch centres should be taken annually in Europe. To reduce the probability of surveying an atypical year, the survey is to be repeated for a second year. The samples are to be analysed according to the method specification developed by the European Union Reference Laboratory (EURL) expert working group, which is compliant with ISO/DIS 15216‐1. Generalised linear models will be used to estimate proportion (with 95% confidence intervals) of sample units with NoV contamination for the following thresholds: < limit of quantification (LOQ), 100, 200, 500, 1,000, 5,000, 10,000 and > 10,000 copies/g. The necessary data to be reported by the sampler and the laboratory to support this analysis is presented in two data models. The results of the survey should be reported using the EFSA data collection framework.

show abstract

Data Representativeness: Issues and Solutions

Cited by 7 publications

References 37 publications

Microbial Performance of Food Safety Control and Assurance Activities in a Fresh Produce Processing Sector Measured Using a Microbial Assessment Scheme and Statistical Modeling

Microbial Performance of Food Safety Control and Assurance Activities in a Fresh Produce Processing Sector Measured Using a Microbial Assessment Scheme and Statistical Modeling

General guidelines for statistically sound and risk‐based surveys of plant pests

Technical specifications for a European baseline survey of norovirus in oysters

Contact Info

Product

Resources

About