Reviewing Summarised QC History Observations

• Reviewing Summarised QC History Observations
  • QC History Summary Scope
  • Properties of a QC History Summary
  • Use of QC History Summaries
  • Precision Assessment
    • Configuring QC Precision Assessment of Samples
    • Performing the Precision Assessment
    • Writing Observations of Precision Assessment to QC History
    • Summary View of Precision Assessments for Duplicate and Replicate QC Observations
  • Accuracy Assessment
    • Configuring QC Standard Assessment of Samples
    • Performing the Standard Assessment
    • Writing Observations of Standard Assessment to QC History
    • Summary View of Accuracy Assessments for Blank, Standard and spike QC Observations
  • Process
    • Accessing Historical QC Observations from Laboratory Batch Jobs
    • Returning Duplicate, Replicate or spike QC Samples where a Laboratory Batch Job is Involved
    • Returning Blank or standard QC Samples where a Laboratory Batch Job is Involved
    • Impact of applying a Secondary QC Assessment for a Sample
    • Persisting User QC Status and User Workflow Status into QC History Records

The QC history review allows you to search QC history records to return summaries of the contamination, accuracy and precision readings made for QC types by the laboratory over a period of time.

A QC (quality control) history summary is a statistical summary of all QC history records for a particular analytical and QC type.

A QC (quality control) history record is an historical reading of a duplicate, replicate, blank, standard or spike sample, written due to the analyte being flagged as active for QC assessment and for writing to QC history.

QC History Summary Scope

QC history summaries and records are scoped by analytical and QC type.

Properties of a QC History Summary

A QC history summary can have:

• Multiple QC history records.

Use of QC History Summaries

In addition to performing quality control assessment on QC samples in production and laboratory batch jobs, laboratory managers also need to review long-term QC performance of QC samples to detect any errors or malfunctions, and view the laboratory's performance.

Assessments of QC samples are recorded in QC history and reviewed later to produce reports based on the search grid data, allowing for statistical analysis of those results.

The QC history records are first viewed as a summary for a given analytical and QC type. A summary can be opened to view the QC history records that comprise the summary.

This allows chemists to calculate analytical accuracy and precision based on QC samples included within normal analytical work. QC history review can be used to monitor and refine existing schemes or quantitatively evaluate new methods. Information gained from this analysis can be fed back to more finely turn the quality control checks. Standards and precision limits can be updated after careful statistical analysis is carried out.

QC reports and charts provide indications of improvements required to ensure ongoing accuracy and precision of QC sample analysis results, allowing QC managers to efficiently monitor and tweak various entity and/or instrument settings. Charts can be saved as image files as required.

The types of QC assessments saved to QC history include QC precision assessment and QC standard (accuracy) assessment.

Precision Assessment

Configuring QC Precision Assessment of Samples

When samples undergo sample preparation, a portion of the original sample may be taken and made into a duplicate QC sample. In this case, a sample's Primary Analytical Type is set to Duplicate and the sample's Original Duplicate Sample is set to the original sample.

When samples undergo analytical preparation, a portion of the original sample may be taken and made into a replicate QC sample. In this case, a sample's Primary Analytical Type is set to Replicate and the sample's Original Duplicate Sample is set to the original sample.

When a multiple number of blank, standard or spike QC samples are created for the same primary QC type, the second and subsequent samples can be linked to the first sample to set the sample's Secondary Analytical Type to Duplicate or Replicate, depending upon whether the samples were created during sample preparation or analytical preparation.

If a sample in a production or lab batch job has a Primary Analytical Type or a Secondary Analytical Type of Duplicate or Replicate, when a result is entered for both samples of the pair, and the scheme version analyte QC Active is selected, the system performs a precision assessment for repeatability and sets the sample scheme analyte's Precision Status.

A scheme version analyte holds the limits for precision analysis: one set for the assessment of duplicate QC samples, and one set for the assessment of replicate QC samples.

Performing the Precision Assessment

A test result of sample that has a primary or secondary analytical type of duplicate is compared against the original sample to determine the precision of sample preparation.

A test result of a sample that has a primary or secondary analytical type of replicate is compared against the original sample to determine the precision of analytical preparation or instrument calibration. The outcome of precision assessment is written to the sample scheme analyte's Precision Status.

Writing Observations of Precision Assessment to QC History

If the scheme version analyte Saved to QC History is selected, then any sample scheme analyte that undergoes standard assessment has the outcome of that assessment written to QC history.

Those samples that have a Primary Analytical Type of Blank, Standard or Spike and a Secondary Analytical Type of Duplicate or Replicate, have two records written to QC history; one for the standard assessment and one for the precision assessment.

Summary View of Precision Assessments for Duplicate and Replicate QC Observations

Using criteria based on an Analytical Type of Duplicate or Replicate, or an Analytical Type of Standard, Blank and Spike and a Secondary Analytical Type of Duplicate or Replicate, and a combination of QC status, QC type, standard, section, scheme, job, client, client group, user and/or date, historical QC precision assessment observations are searched to return a simple statistical summary for each unique combination.

Accuracy Assessment

Configuring QC Standard Assessment of Samples

When QC samples are added during sample preparation, analytical preparation or analysis, a portion of blank material may be taken and made into a blank QC sample. In this case, a sample's Primary Analytical Type is set to Blank.

When QC samples are added during sample preparation, analytical preparation or analysis, a portion of standard material may be taken and made into a standard QC sample. In this case, a sample's Primary Analytical Type is set to Standard.

When QC samples are added during sample preparation, analytical preparation or analysis, a portion of standard material may be added to a portion of other sample and made into a spike QC sample. In this case, a sample's Primary Analytical Type is set to Spike and the sample's Original Spike Sample is set to the other sample.

Sometimes, client samples are actually identified as blanks, standards or spikes. In this case, a sample's Secondary Analytical Type is set to Blank, Standard or Spike, accordingly.

If a sample in a production or lab batch job has a Primary Analytical Type or a Secondary Analytical Type of Blank, Standard or Spike, when a result is entered and the scheme version analyte QC Active is selected, the system performs a standard assessment for reproducibility and sets the sample scheme analyte's Standard Status.

A sample's linked standard lot's specification contains the limits for standard analysis, matched by scheme, version, analyte and analysis units.

Performing the Standard Assessment

A test result of a sample that has a primary or secondary analytical type of blank, standard or spike (in this case, with the amount of the original sample deducted) is compared against the standard specification's limits to determine the accuracy of sample, analytical preparation or instrument calibration. The outcome of standard assessment is written to the sample scheme analyte's Standard Status.

Writing Observations of Standard Assessment to QC History

If the scheme version analyte Saved to QC History is selected, then any sample scheme analyte that undergoes standard assessment has the outcome of that assessment written to QC history.

Those samples that have a Primary Analytical Type of Blank, Standard or Spike and a Secondary Analytical Type of Duplicate or Replicate, have two records written to QC history; one for the standard assessment and one for the precision assessment.

Summary View of Accuracy Assessments for Blank, Standard and spike QC Observations

Using criteria based on an Analytical Type of Blank, Standard or Spike, and a combination of QC status, QC type, standard, section, scheme, job, client, client group, user and/or date, historical QC standard assessment observations are searched to return a simple statistical summary for each unique combination.

Process

When historical QC observations are searched, they are displayed as summarised observations.

Where the summary grid search criteria's (firstly) Group by is:

• Analyte Type and QC Type Code—A filter and grouping applies firstly by Primary Analyte Type and secondly by QC Type Code, and the QC Type Code and QC Type Name columns are populated.
• Analyte Type only—A filter and grouping applies by Primary Analyte Type, any filter in QC Type Code is ignored, and the QC Type Code and QC Type Name columns are left empty.

Where the summary grid search criteria's (secondly) Group by is:

• Standard and Lot—A filter and grouping applies firstly by Standard Code and secondly by Standard Lot Code, the Standard Lot Code and Standard Lot Number columns are populated, and the specification code associated with the standard lot displays in the Specification Code column.
• Standard only—A filter and grouping applies by Standard Code, any filter in the Standard Lot Code is ignored, the Standard Lot Code and Standard Lot Number columns are left empty, and specification code associated with the standard's Current Standard Lot displays in the Specification Code column.

Where the summary grid search criteria's (thirdly) Group by is:

• Scheme and Version—A filter and grouping applies firstly by Scheme Code and secondly by Scheme Version, and the Scheme Version column is populated.
• Scheme only—A filter and grouping applies by Scheme Code, any filter in Scheme Version is ignored, and the maximum scheme version from the data that matches the detailed QC history search criteria displays in the Scheme Version column. Note that this is the scheme version used to set chart limits.

QC history reporting can be extended with scripting. Refer to the QC History API.

Content of the Summarised Observation Search Results Grid

Column header	Analytical Type is Standard, Blank or Spike	Analytical Type is Duplicate or Replicate
Analytical Type	AnalyticalType or SecondaryAnalyticalType	AnalyticalType or SecondaryAnalyticalType
QC Type Code	QCTypeId > Code or LinkedQCTypeId > Code, only present if Group By = QC Type Code	QCTypeId > Code or LinkedQCTypeId > Code, only present if Group By = QC Type Code
QC Type Name	QCTypeId > Name or LinkedQCTypeId > Name, only present if Group By = QC Type Code	QCTypeId > Name or LinkedQCTypeId > Name, only present if Group By = QC Type Code
Standard Code	StandardId > Code	SampleName
Standard Name	StandardId > Name	ClientSampleName
Standard Description	StandardId > Description	SampleDescription
Standard Lot Code	StandardLotId > Code	--
Standard Lot Number	StandardLotId > LotNumber	--
Scheme Code	SchemeId > Code	SchemeId > Code
Scheme Name	SchemeId > Name	SchemeId > Name
Scheme Version	SchemeVersionId > VersionNumber	SchemeVersionId > VersionNumber
Scheme Version Name	SchemeVersionId > Name	SchemeVersionId > Name
Analyte Code	SchemeVersionAnalyteId > Code	SchemeVersionAnalyteId > Code
Analyte Name	SchemeVersionAnalyteId > Name	SchemeVersionAnalyteId > Name
Unit Code	UnitId > Code	UnitId > Code
Unit Name	UnitId > Name	UnitId > Name
Count (#Observations)	Count of Records	Count of Records
Target	InternalTargetValue (as in SpecSVA Target)	--
Mean / Mean Abs Diff	Average of Value1	Average of ( ABS(VALUE1-VALUE2) )
Bias / Mean Abs % Diff	Average of ((VALUE2-VALUE1)/VALUE2)*100)	Average of ( 100* ABS(VALUE1-VALUE2)/((VALUE1+VALUE2)/2)
Std Dev / Std Dev Abs % Diff	StdDev of Value1	StdDeviation of ( 100* ABS(VALUE1-VALUE2)/((VALUE1+VALUE2)/2)
RSD	RelativeStdDev of Value1	--
Min / Min Abs % Diff	Min of Value 1	Min of ( 100* ABS(VALUE1-VALUE2)/((VALUE1+VALUE2)/2)
Max / Max Abs % Diff	Max of Value 1	Max of ( 100* ABS(VALUE1-VALUE2)/((VALUE1+VALUE2)/2)

The search criteria can be saved and named for future recall and re-use, or scheduled on a regular basis using a scheduled script (refer to the QC History API).

Reports can be produced based on summarised QC history observations returned to the search grid.

For any row in the QC history summary grid, the detailed, or individual, QC history readings can be viewed to show contamination, accuracy or precision. This allows for failures in any individual reading to be ignored, to remove statistical outliers.

Accessing Historical QC Observations from Laboratory Batch Jobs

QC samples are often analysed in lab batch jobs, however a QC history review may be needed to be done in context of a production job, or for a client who submits production jobs. For this reason, it is possible to trace the original production job for samples analysed in a lab batch job to view historical assessment outcomes.

When searching for QC history summaries, or filtering QC history discrete observations by status or analysed date, QC history records that are related to a specific client, regardless if the QC samples were analysed in a production job or in a lab batch job, are returned from the search. Lab batch jobs can have a Client Code that is different to the production job Client Code, however:

• Blank and standard QC history records are returned if the blank or standard QC sample was created in the lab batch job but the lab batch job contains at least one sample whose associated production sample is linked to the client, and
• Duplicate, replicate and spike QC history records are returned if the duplicate, replicate or spike QC sample was created in the lab batch job from a sample whose associated production sample is linked to the client.

Returning Duplicate, Replicate or spike QC Samples where a Laboratory Batch Job is Involved

Where the Client Code, Client Name, Project Code or Client Group Code search parameter is populated, and there are duplicate, replicate or spike QC samples that match the other search criteria, and the duplicate, replicate or spike QC samples were registered directly in a laboratory batch job (with a laboratory batch job client, or client and project, that may or may not match the search criteria client, or client and project), and the associated production sample for the duplicate, replicate or spike QC sample is linked to the search criteria client, then the matching duplicate, replicate or spike QC sample is included in the search results.

Example: If a QC history summary search is made for an Analytical Type of Duplicate and a Client Name of BHP, then the search results includes:

• Duplicate QC samples from all jobs where Client Code is BHP, and
• Duplicate QC samples from all laboratory batch jobs where Client Code is anything (for example, INTERNAL) and the associated production sample for the duplicate belongs to a production job where the production job Client Code is BHP.

Returning Blank or standard QC Samples where a Laboratory Batch Job is Involved

Where the Client Code, Client Name, Project Code or Client Group Code search parameter is populated, and there are blank or standard QC samples that match the other search criteria, and the blank or standard QC samples were registered directly in a laboratory batch job (with a laboratory batch job client, or client and project, that may or may not match the search criteria client, or client and project), and the laboratory batch job has at least one sample whose associated production sample is linked to the search criteria client, then the matching blank or standard QC sample is included in the search results.

Example: If a QC history summary search is made for a Standard Code of 1601 and a Client Name of BHP, then the search results includes:

• standard QC samples with a Standard Code of 1601 from all jobs where Client Code is BHP, and
• standard QC samples with a Standard Code of 1601 from all laboratory batch jobs where Client Code is anything (for example, INTERNAL) and the laboratory batch job includes samples from a production job where the production job Client Code is BHP.

Impact of applying a Secondary QC Assessment for a Sample

Given the following samples are analysed and undergo QC assessment:

Sample	Primary Analytical Type	Primary QC Type Code	Secondary Analytical Type	Secondary QC Type Code	Duplicate Original Sample
STD1	Standard	STD
STD2	Standard	STD
STD2REP	Standard	STD	Replicate	STD_REP	STD2

Upon QC assessment, STD1, STD2 and STD2REP undergo standard assessment, and STD2REP also undergoes replicate precision assessment. In total, 4 observations are written to QC history.

Sample	Analytical Type	QC Type Code	Secondary Analytical Type	Linked QC Type Code	Parent Sample
STD1	Standard	STD
STD2	Standard	STD	REP	STD_REP
STD2REP	Standard	STD
STD2REP	Replicate	STD_REP	<empty>	<empty>	STD2

For the following searches of QC history summaries filtered by just these samples:

Analytical Type	Secondary Analytical Type	Returns	When opened to show the individual observations
Standard		3 observations	Shows the 3 standard observations
Standard	Replicate	1 observation	Shows the 3 standard observations, that is, the secondary analytical type is ignored at this point
	Replicate	1 observation	Shows the 3 standard observations, that is, the secondary analytical type is ignored at this point
Replicate		1 observation	Shows the 1 replicate observation

Persisting User QC Status and User Workflow Status into QC History Records

When QC history observations are written and updated, the sample scheme analyte's User QC Status and User Workflow Status are included in the record, along with the Workflow Status. As results are entered and user statuses updated, the QC history observation is kept in sync with respect to the these statuses.

QC history summaries can be searched by these statuses, and for a particular QC history summary, the distinct observations can also be searched by these summaries, with the default criteria being inherited from the summary search.