Corporate Director, Quality Management System | Pharmaceutical QP (Eligible)| Biologist
EMA : Data Quality Framework (#DQF) for EU medicines regulation- Deep Dive - Part II- Data Quality Dimensions and metrics The definition of #DQdimensions and #metrics rely on the general definition of dimension, metrics, and measures: • A dimension represents one or more related aspects or features of reality • A metric represents a way to assess the value of a specific feature • A measure represents a single instance of a metric. More measures can be combined to derive more general metrics DQ metrics can be defined as indicators that when applied to a #datasource, can derive an assessment of one of more quality dimensions. A single #qualitymetric can be used as an indicator for more than one dimension as expressed below in the examples for #Coherence. For some metrics, #acceptancethresholds can be defined. Dimensions of data quality #Reliability Reliability is defined as the dimension that covers how closely the data reflect what they are directly measuring. Reliability sub-dimensions #Accuracy defined as the amount of discrepancy between data and reality. #Precision defined as the degree of approximation by which data represents reality. Strictly related to Reliability is the concept of #Plausibility, defined as the likelihood of some information being true. #Traceability refers to data presenting the knowledge of how data came to be, what source it originated from, and what processing it went through before appearing in its current form. #Extensiveness Extensiveness is defined as the dimension capturing the amount of data available. Sub-dimensions of Extensiveness #Completeness measures the amount of information available with respect to the total information that could be available given the capture process and data format. #Coverage measures the amount of information available with respect to what exists in the Real World. #Coherence Coherence (also referred to as #Consistency) is defined as the dimension that expresses how different parts of an overall dataset are consistent in their representation and meaning Sub-dimensions of Coherence #FormatCoherence: whether data are expressed in the same way throughout a dataset #Structural or #RelationalCoherence: whether the same entities are identified in the same way throughout a dataset. #SemanticCoherence: whether the same value mean the same thing throughout a dataset. #Uniqueness: Uniqueness is the property that the same information is not duplicated but appears in the dataset once. #Timeliness Timeliness is defined as the availability of data at the right time for regulatory decision making, that in turns entails that data are collected and made available within an acceptable time. Sub-dimensions of Timeliness #Currency is a specific aspect of Timelines that considers how fresh the data are. #Relevance Relevance is defined as the extent to which a dataset presents the data elements useful to answer a given research question.
