1 |
DQ-Task |
DQ-Task |
Specifications for the DQA activity |
|
2 |
DQ-Task |
DQ-Dimension |
Data error to investigate, quality properties determining how well data are fit for use, or label for grouping measurements |
Completeness [64–66, 69, 71, 74, 75, 77, 79–81, 83, 84], conformance,[64, 65, 68, 70, 71, 77, 82] plausibility [48, 65, 70, 71, 73, 77, 83], consistency [74, 75, 79, 83], accuracy [69, 84], timeliness [75], out of range [73, 83], representation completeness [78, 79], domain completeness [78, 79], domain constraints [78, 79], syntax accuracy [69], duplicate [83], domain consistency [79], precision [74], violations of logical order [83], redundancy [84], readability [84]. |
3 |
DQ-Task |
Data-Element |
An individual unit of an observation |
Data elements determined at runtime [73, 78, 81, 84] pre-defined data elements, e.g. growth measurements [48], discharge summaries [74], and emergency records [83]. |
4 |
DQ-Task |
DQ-Metric |
An aggregate measure for assessing defined DQ-Dimensions
|
Simple ratio [69, 73, 75, 79, 80], counts [66, 70, 73, 77], weighted scores [81, 84], and Boolean values[73]. |
5 |
DQ-Task |
Baseline |
A threshold for judging the level of a DQ-Dimension in a dataset for a particular use case |
Greater than 99.9% [67] and 90%[79], user-defined [73, 84], previous DQ-Metric score [65, 74]. |
6 |
DQ-Task |
Periodicity |
The type of execution and frequency supported |
On-demand [68, 81, 83] scheduled e.g. every 24 h [72], quarter [66] and other specified intervals [65, 67, 71]. |
7 |
DQ-Task |
Application area |
The point in the EHR data cycle where the DQA program or tool would be applicable |
Directly on EHRs data stores [72, 74], EHR data exchanged via health information exchange frameworks [64, 75] |
8 |
DQ-Task |
Priority |
The rationale for focusing on selected dimensions and data elements |
Data elements type supported by available measurement [71, 84], data elements are necessary for intended use cases [71, 72], dimensions prevalent in previous records and literature [65, 84], dimensions for which measurements and required data are available [65], demands of internal and external data consumers [71]. |
9 |
Target-Data |
Target-Data |
One or more tuples containing observations |
|
10 |
Target-Data |
Data-Source |
The range of sources or datasets that the program can be applied to |
Single [72, 74, 75], multiple[68–70, 78] |
11 |
Target-Data |
Data connection |
The method for accessing data sources. DQA program can support more than one type of connection |
CSV files [70, 84], database scripts or connections [70–72, 74, 80, 81], REST API [69], Health Level Seven (HL7) document [75], XML[77]. |
12 |
Target-Data |
Data-Integrator |
The method for consolidating data from different sources into a single model or view. |
Extract, transform and load (ETL) [68, 69, 71, 76, 77] |
13 |
Target-Data |
Data-Model |
Logical representation of data elements, their relationships, and constraints that is used to enable other components to operate and share the same data uniformly |
Observational Medical Outcomes Partnership (OMOP) [68, 69, 80], extended OMOP [71], Clinical Research Document [77], openEHR,[78], PCORnet [65, 66, 80] Informatics for Integrating Biology & the Bedside (i2b2) [70], Digital Imaging and Communications in Medicine (DICOM) [72, 82], National Summary Care Record Format [76], locally defined standards [69, 81]. |
14 |
Target-Data |
Data-Location |
The physical location of the Target-Data |
Users’ repository [68, 69, 77], central server [71, 76] |
15 |
Target-Data |
Size |
The amount of data the program can support or has been validated with. |
Small (0-100k) [74, 77], medium (100k to 1 M) [79, 80], large (1 M+) [68]. |
16 |
Target-Data |
Data-Transformer |
Functions for converting data from one format, structure and value to another |
Vocabulary crosswalks [71, 75] |
17 |
DQ-Measurement |
DQ-Measurement |
Criteria for measuring DQ-Dimension |
|
18 |
DQ-Measurement |
Data-Level |
This refers to the data level considered in the DQ measurement. |
Cell level [69], field level [65, 67, 70, 84], record level [74, 81, 83], table level [65, 67, 71]. |
19 |
DQ-Measurement |
Measurement-Source |
Method for creating measurements and accompanying reference items |
Domain experts [68–72, 79, 80], crowdsourcing [68, 71], data standards or dictionaries [71, 77, 78], national guidelines [76], literature review [71], statistical analysis [83, 84]. |
20 |
DQ-Measurement |
Representation |
Format for representing measurements |
Natural text [68, 72, 80], conditional logic statements [75, 78, 79], database queries [67, 69, 70, 73, 78], metadata repository[67, 69], programming language scripts [71, 73, 83], mathematical and computational models [48, 74, 81]. |
21 |
DQ-Report |
DQ-Report |
The content of reports and type of analysis |
Summary metrics [69], DQA metadata [67, 79], date and time the result was obtained [67, 71], severity warnings or comments [64, 65, 68, 82], error message to display [68, 71, 73], data profile of source data [68, 80], records returned per dataset or site [77], records returned linked to assessment metadata [67, 69, 70, 72, 73, 83, 84], aggregate results from multiple assessments or sites [66, 70, 77], results grouped by data element [66–68, 71, 83], suggestions on improvements [64], information to exclude [69]. |
22 |
DQ-Report |
Dissemination-Method |
Techniques or tools for communicating assessment methods |
Store results in a repository [66, 67, 69, 70, 80], file export [71, 76, 77], Tables [68, 70, 73], charts [66, 68, 79, 80, 84], longitudinal views [66], collaborative workspace, e.g. Github [71] |
23 |
DQ-Mechanism |
DQ-Mechanism |
The mechanism for operationalising DQA components |
Visualisation tool [84], dedicated tool [48, 68, 71, 80, 83] |
24 |
DQ-Mechanism |
Feature |
Functions that enable a DQ-Mechanism to perform satisfactorily and meet Stakeholder requirements |
See Table 3. |