Limits...
A Prototype for Executable and Portable Electronic Clinical Quality Measures Using the KNIME Analytics Platform.

Mo H, Pacheco JA, Rasmussen LV, Speltz P, Pathak J, Denny JC, Thompson WK - AMIA Jt Summits Transl Sci Proc (2015)

Bottom Line: To prototype this capability, we implemented eCQM CMS30 (titled: Statin Prescribed at Discharge) using KNIME.The implementation contains value set modules with connections to the National Library of Medicine's Value Set Authority Center, QDM Data Elements that can query a local EHR database, and logical and temporal operators.We successfully executed the KNIME implementation of CMS30 using data from the Vanderbilt University and Northwestern University EHR systems.

View Article: PubMed Central - PubMed

Affiliation: Vanderbilt University, Nashville, TN.

ABSTRACT
Electronic clinical quality measures (eCQMs) based on the Quality Data Model (QDM) cannot currently be executed against non-standardized electronic health record (EHR) data. To address this gap, we prototyped an implementation of a QDM-based eCQM using KNIME, an open-source platform comprising a wide array of computational workflow tools that are collectively capable of executing QDM-based logic, while also giving users the flexibility to customize mappings from site-specific EHR data. To prototype this capability, we implemented eCQM CMS30 (titled: Statin Prescribed at Discharge) using KNIME. The implementation contains value set modules with connections to the National Library of Medicine's Value Set Authority Center, QDM Data Elements that can query a local EHR database, and logical and temporal operators. We successfully executed the KNIME implementation of CMS30 using data from the Vanderbilt University and Northwestern University EHR systems.

No MeSH data available.


Implementing the temporal operator (≤ 30 day(s) starts before start of)
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4525225&req=5

f2-2091960: Implementing the temporal operator (≤ 30 day(s) starts before start of)

Mentions: Temporal Operator (≤ 30 day(s) starts before start of): This temporal operator (see Figure 2) takes two input tables (denoted with left and right) each with at least two fields (pid and StartDatetime). First, we create a new column (called LeftAdded) on the left input table with 30 days added to their StartDatetime (Node A on Figure 2). Then we inner join the left and right input tables with pid (Node B), and filter the rows where the StartDatetime (left) is earlier than StartDatetime (right) AND LeftAdded is not earlier than StartDatetime (right) (Node C). Then, we split the joined table back to the left and right output tables (corresponding to input tables) with removing duplicated rows (Node D and E).


A Prototype for Executable and Portable Electronic Clinical Quality Measures Using the KNIME Analytics Platform.

Mo H, Pacheco JA, Rasmussen LV, Speltz P, Pathak J, Denny JC, Thompson WK - AMIA Jt Summits Transl Sci Proc (2015)

Implementing the temporal operator (≤ 30 day(s) starts before start of)
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4525225&req=5

f2-2091960: Implementing the temporal operator (≤ 30 day(s) starts before start of)
Mentions: Temporal Operator (≤ 30 day(s) starts before start of): This temporal operator (see Figure 2) takes two input tables (denoted with left and right) each with at least two fields (pid and StartDatetime). First, we create a new column (called LeftAdded) on the left input table with 30 days added to their StartDatetime (Node A on Figure 2). Then we inner join the left and right input tables with pid (Node B), and filter the rows where the StartDatetime (left) is earlier than StartDatetime (right) AND LeftAdded is not earlier than StartDatetime (right) (Node C). Then, we split the joined table back to the left and right output tables (corresponding to input tables) with removing duplicated rows (Node D and E).

Bottom Line: To prototype this capability, we implemented eCQM CMS30 (titled: Statin Prescribed at Discharge) using KNIME.The implementation contains value set modules with connections to the National Library of Medicine's Value Set Authority Center, QDM Data Elements that can query a local EHR database, and logical and temporal operators.We successfully executed the KNIME implementation of CMS30 using data from the Vanderbilt University and Northwestern University EHR systems.

View Article: PubMed Central - PubMed

Affiliation: Vanderbilt University, Nashville, TN.

ABSTRACT
Electronic clinical quality measures (eCQMs) based on the Quality Data Model (QDM) cannot currently be executed against non-standardized electronic health record (EHR) data. To address this gap, we prototyped an implementation of a QDM-based eCQM using KNIME, an open-source platform comprising a wide array of computational workflow tools that are collectively capable of executing QDM-based logic, while also giving users the flexibility to customize mappings from site-specific EHR data. To prototype this capability, we implemented eCQM CMS30 (titled: Statin Prescribed at Discharge) using KNIME. The implementation contains value set modules with connections to the National Library of Medicine's Value Set Authority Center, QDM Data Elements that can query a local EHR database, and logical and temporal operators. We successfully executed the KNIME implementation of CMS30 using data from the Vanderbilt University and Northwestern University EHR systems.

No MeSH data available.