PEPconnect

PEPFAR Quality Control and Method Validation Activity 10: Part 1

In this activity, participants are introduced to the Sigma-metrics QC Selection Tool, available at Westgard.com. Using a case scenario, participants will select appropriate QC rules and design a QC strategy for a workstation.

Note: Any reference to page numbers in the video may not correlate with supporting documents as documents may have been updated.

Is , Analyt d Type number Of sampks ±es+ed ype number Of sampks fes+ed number number sompestes we and lwe number o QC Sigma-Metrics QC Selection numoet- sampes fes+ed Frequency of (AC Frequency o ype and number of fest even and pe ana p detec l) Frequency OF QC even+s 'Analytical system control evalua+e used +0 +0 i. 2 number Of QC sampks fes+ed number 09 used used +0 Sigma-Metrics QC Selection Tool for 3 Levels Contro Stgma.Metrics QC Selection Tool for 3 Levels Contrc 09 3) QC used +0 QC used 0.1 The 0.03 QC QC used 40 number 00 to the to the mean. 3) QC ase.d procedure used +0 ch The trade." that then evaua e limits are less l) Frequency OF +es+ even+s Sigma-Metrics QC Selection Tool for 3 Levels Control Sigma.Metnics QC Selection Tool for 3 Levels Contrcl Sigma-Metrics QC Selection Tool1 for 3 Levels Control Stgma.MetriCS QC Selection Tool for 3 Levels Contrc. to used or 3 Levels Con Frequency +es+ evenfs Frequency of +es+ even+s frequency of QC +es+ evenFs the SEC The that 3) QC used Quality +hen 3) QC used +0 System Analytical evaluaVe used QC used +0 evo On-going Evaluation of Your Quality Control Systems Type number of sompks fesfed to the Detection System The of periodic review QC ased +0 The of periodic revtew for QC used QC used +0 The control number 2) Type number of QC ±es+ed 2) Type and number of GC ±es+ed Control the SEC o Quality Type number sompks fes+ed Desirable Des limits used Sigma-Metrics QC Selection Tool for 2 Levels Control 20fg:2S also I Fcequency of QC sampe even+s evalua+e Activity: Predicting Probabilities Type and number sampks fes+ed accuracy and Frequency OF +esf evenfs q) The of periodic review QC ased 5 of 3 Type and number of QC sampks tes+ed 2 Type and number +es+ed The of periodic will q) The 0.03 have q) The 09 petiodic review for q) The periodic review Error measurer and Quality evalua+e The OF review detection LD The Detection The of pevtodtc revtew The revtew QC used +0 The of pevtodic revtew The of review used +0 Err of periodic review Type number of fesfed The OF periodic review Goals the SE The OF revtew for Quality The OF periodic revtew da to the mean. meast Sigma Scale Quality measur; System ISO 15189 Wider con LD OF periodic revtew for that review I-I) The of periodic review • q) The 09 periodic revtew for OF peviodlC review for OF peviodtc review Quality QC Rule Selection Guidelines Laboratory you will have a 3) QC used Err QC used Quality of periodic Tevtew QC used Messages ID Tine OF review for 2) Type and n: q) The periodic revtew limits are less evaluA+e. numbec sompks fesfed The OF pevtodtc revtew false Sigma-metrics Rule Selection 'sensitive to error evalua+e evalua+e Laboratory QC used +0 The of pevicÅc review The The periodic review Quality q) The of peviod,c review de Quality The of revtew QC ased QC +0 The of Goals QC used 40 q) The OF review q) The of period,c review for q) The of review for Wider con q) The of periodic revtew The of periodic revtew q) The Fretuency Error Sigma Scale Quality QC +0 defec+inj and + you have The of peviodlC review for Laboratory Wider Laborator Goals The OF periodic review of peviodtc review Goals •ator Goats meast measu measurer l) Frequency OF +es+ even+s The of review for detection but Laborato& The of review q) The periodic revtew review q) The OF periodic review rate sensitive to of review for Goals sensitive to e The of pevtodtc revtew for peviodjc revtew to erne Laboratory limits are less QC Rule Attributes The of periodic review The peviodlC review and *rends Frequency OF QC +es+ even+s Fcequency +es+ even+s limits arc less evaluaVe The of penc,d.c revtew Laborator defec+inj and *rends a. System evalua+e Laborator defec and *rends and *rends of peviod,c review 3a Frequency QC sampe +es+ even+s limits are le limits are lim dc sensitive to erro defec+l. Quality Laboratory and +rends When SEC and Sigma-metrics are large, peviod.c review limits The of peviodlC. review and +cends and +cends alts Quality review Quality Tine review I-b pevtcdic e of periodic Tevtew The OF periodic revtew The OF periodic review for The of periodic revtew q) The of peticdiC review and *recd s The of peviodic review q) The of periodic review 46 detectsions Quality Laboratory y of periodic review oals and *rends defec±inj and *rends defec+nj defec+nq and +cends Quality Results q) The Fretaency pertod,c Quality 4cs have fewer false Laboratory x Laborator 46 An Effective QC Design Goals The of periodic review haw fewer false and +rends 3a 'Analyticalj The OF review ID The of periodic revtew Quality q) The of periodic The OF pevicdic review for IDe OF periodic teview for Goals The OF review for Results The periodic q) The periodic Tevtew Error defec+inj and defecHnj and *rends SESP sensitive to detection but Goals Goals def-ec+inj and +rends ch defecfinj and *rends evaluaVe Goals and +rends sensitive Gaussian is the Key to Quality Control review evaluaVe evalua+e Laborator of Goals Results c; Retection The of periodic teview Laboratory and *rends defec+inj and +cends Quality b) Its defec+inj and *rends Quality _de and *rends th Results defec+lnj and *rends 10 false "Goals Sigma Scale b) The achons ±Q6en when resul+s The of review Results and Laborator Goats The review detec+inj and q) ty Type number sampks fes+ed detections and +cends What will you do? What will you need? rejections 2.65 30 3.654 4.65 5 Laboratory q) The of peviod,c review for 40 4.65 50 50 q) The revtew For defecfinj and +rends Error b) The GCA-ions ±Qhen when q) The of periodic revtew q) The Fretaency of revtew q) The of review defec+inj and +rends defecHnj and +rends and *rend s Goals Goals Laboratory b) The achons fahen re os when resul+s S when resul+s Detection »rator defec+lnj and +rends and De cable and *rends and *rends Results Results defecfinj and *rends Laborator Laboratory Error number sampks fes+ed Goals Quality and and +ceods b) The achons ±ahen when and +cends Rejection Laborator Purpose Results The of peviodiC review defec+inj and and +cends q) The periodic revtew What wi 5.6.2 Quality control resul+s q) The OF periodic review for q) The review q) The OF pertcdic review for Rejections Goals 11) The of review lox have Goals Stgmatv'etr•cs QC Selection Toot for 3 Levels Contrcl Quality 10 Desirable The periodic review The Fretuency review b) The achons ±a6en when defec+nj and +rends System Wider con Sigma-Metrtcs QC Selection Tool for 3 Levels Contro Stgma.Metntcs QC Selection Tool for 3 Levels Contro Sigma-Metrics QC Selection Tool for 3 Levels Contrc, Sigma-Metncs QC Selection T 001 3 Leves Contro b) The achons when re: b) The achons +ahen when re 0 3.65 4.65 5.65Pfr 2.65 0 3.65 4.65 5.65Pfr 0 3.65 4.65 Sigma-Metrics QC Selection T 001 for 3 1-evels Contrc. Sigma.Metn•cs QC Selection Tool for 3 Levels Contrc Stgma.Metncs QC Selection Tool for 3 Levels Contrc Sigma-Mv'etntcs QC Selection ToOi fol 3 L-evels Contrc: Stgma.MetriCS QC Selection Tool for 3 Levels Contrc, Stgma.Metncs QC Selection Tool for 3 Levels Contrcl Stgma.MetnjcsS QC Selection Tool1 for 3 Levels Contrc. Sigma.Metnics QC Selection Tool for 3 Levels Contrc, Sigma.Metrcs QC Selection Tool for 3 Levels Contrc Sigma-Metn•cs QC Selection Tool for 3 Levels Contrcl Stgma.Metrucs QC Selection Tool for 3 Levels Contrc, Sigma.Mdetncs CC Selection Tool for 3 Levels Contrc Sigma.Metncs QC Selection Tool for 3 Levels Contrcl Quality S ±Qhen when resul+s Stgma.Metntcs QC Selection Tool for 3 Levels Contrcl Sigma-Metncs QC Selection Tool for 3 Levels Contrc Stgma.Metnics QC Selection Tool for 3 1-evels Contrcl Sigma-Metr•cs QC Selectton Tool 3 Contrc Sigma-Metrtcs QC Selection Tool for 3 L-evels Contrc• Sigma.Metrics QC Selection Tool for 3 Levels Control S'gma.Metncs QC Selection Tool for 3 Levels Contrc Stgma.MetncsS QC Selection Tool for 3 Levels Contrc. b) The achons when fQHen resul+s The 09 pectodvc review b) The achons when resul+s 1 865 2.65s Results 2.65s 3.5s 2.5s 2.05P 2.65 2.65P 2.656 and *rends 5.65 Pfr 2.65s 5.65P Use control materials that assess the measurement The of pevtod.c vevtew Goals The revtew The periodic review The review For Sigma.Metncs QC Selection Tool for 3 Levels Contrc Sigma.Metr•cs QC Selection Tool for 3 Levels Contrc S•gmatvetn•cs QC Selection Tool for 3 Levels Contrc Laborator Goats and +rends Laboratory Stgmayetncs Selection Tool for 3 Contrcl Stgma.Metncs Selection Toot for 3 Contre Sigma-Metncs QC Selection Toot for 3 Leves Contrc• Stgma.Metncs QC Setect•on Tool for 3 Leves Contrcl Sigma.Metncs Selection Toot for 3 Contre SigmaMetncs QC Selection Toot for 3 Contrc. b) The achons +a6en when b) The achons +Q6en when QC Selection Tool for 3 Leves Contrc, have fewer false limits are less rejections Ken when resul+s Sigma.Metrtcs QC Selection Tool for 3 Levels Contro co Laborator Goats Results Goals What Laboratory: limits are less Hen when resul+s en when resul+s Goal ;oals and CA-ions +Q6en when resu Laborator Goals q) The and ale •Vs and +cends +s and *rends reiections defec+ d and *rends Results and *rends when Goals Normal Goals Quality and Goals b) b) The achons when Laboratory Laborator Desirable as b) The achons +ahen when b) The achons +Qhen when resul+s b) The b) The achons when b) The achons +QKen when resul+s b) b) The achons +ahen when resul+s Activity. Predicting Probabilities and +resvis have fewer false ' Laboratory The achons when resul+s False exceed defec ahen when resul+s Sigma.Metncs QC Selection Tool for 3 Levels Contre•i Sigma-Metrics QC Selection Tool for 3 Levels Control Sigma.MetncsS QC Selection Tool for 3 Levels Contrc• Sigma-Metrics QC Selection Tool1 for 3 Levels Control Sigma-Metrjcs QC Selection Tool for 3 Levels Control Sigma.Metnjcs QC Seiectjon Tool for 3 Levels Contral Stgma„'v'etrics QC Select.on Tool for 3 1-eves Stgma.MetriCS QC Selection T 001 for 3 Levels Sigma.Metrcs QC Selection Tool for 3 Levels Contrcl Sigma-MetnjcsS QC Selection Toot for 3 Levets Contrcl Sigma.Metrics QC Selection Tool for 3 L-evels Control Sigma-Metrjcs QC Selection Tool for 3 Levels Contrc,! Sigma-Metrjcs QC Selection Tool for 3 Levels Contrc, Stgma.Metrjcs QC Selection T 001 for 3 Leves S'gma.Metnics QC Setect•on TOO'1 for 3 Leves Contro, Stgma-Metnjcs QC Selection TOO! for 3 Levels Contrci Stgma.MetriCS QC Selection Tool for 3 Levets Stgma.Metrjcs QC Selection Tool for 3 Levels Contrc. Sigma-MetncsS QC Selection Toot for 3 Levels Contrc• Sigma.Metnjcs QC Selection Tool for 3 Levels Contrcl Stgma.MetriCS QC Selection Tool for 3 Leves Control Stgma*detnics QC Selection Tool for 3 Levels Contrc, Goals and +-rends and +cends and Sigma-Metrics QC Selection Tool for 3 Leves Contrc Stgma.Metnics QC Selection Tool for 3 Levels Contro used ed 3) QC used 3) QC ased sensitive to and *rends when resul+s Error The Detection Laborator The achons ±QKen when ±a6en when resu14s S when resul+s Laboratory 3 1-eves have Analytical achons +aKen Laborator Quality defec+inj and +reMs b) The achons when System OS +ahen when resul+s faHen when b) ns when resul+s ActiVJtYT Predicting Probabilities Activity: Predicting Probabilities e achons ±Q6en when achons when resul+s ±ahen when resul+s Results b) The achons +Q6en when ons when resul+s chons +QKen when resul+s Ions when resul+s Results QC ased ResulIts ns ±QKen when resul+s ions when resul+s limits Goals Detection 30 3, 3a 46 SESP 3a 46 5a 3a 0.03 0.0 b) 46 Results 'Analytical sensitive to defec+ln and +cends 0.2 46 SESP s +Qhen when resul+s Sigma Scale work Goals tons when resul+s when S when resul+s The OF revtew 03 Results When SEC and Sigma-metrics are large, QC Rule Selection Guidelines OC Rule Selection Guidelines Results Laborato review Sigma-metrics Rule Selection Works •tor Analytical u)aly+ical and When SEC and Sigma-metrics are large exceed occep+"le Guide Goals defec+inj and +cends Laborator Quality b) The achons as Laboratory False b) The achons +Q6en when resul+s 35 . Worksht Results Rejection Results detec+inj and and Rejection b) defec+nj and *rends 3.0 Desirable Candi Working individually, you will: SESP SAMP b) The achons ±ahen when and and +rends False Laborator Y) The b) The achons ±ahen when 10 QC used Pumoose b) The achons ±QKen when os 'hen when resul+s resul+s en when resul+s Results The achons +ahen when and *rends on l,) Frequency Results detsc%n Rejections b) The achons b) when resul+s Quality as Wider con S fayen when exceed qccep+üle Results ale Results 1 IES limits n when resul+s SSP b) The achons ±Q6en when Results False limits and +reMs S when resul+s b) The achons when resal+s e achons ±QKen when exceed exceed occep+üle rate as Determine the probabilities limits are Hons +ahen when resul+s occep+oCIe when achons +QKen 30 3.0 50 3a 30 40 0.0 +0 35 Frequency c: Frequency c; Frequency - en resul+s exceed Dosin System Frequency Puroose b) The achons ±Q6en when resul+s 0.0 50 30 1.0 exceed Guide c' Qep+ohule b) The achons fal•5en when Workshe when resul+s b) The achions when resul+s Rejection Desirabl Where Ken when resul+s _ _ Z- 13s/20f3 IR 13 /6 Z - 13s/20f3 IR /3 /6 13s/20f3 IR 13 /6 13s/20f3 IR /3 /6 Z - 13s/20f3 IR 13 /6 b) The achons when resul+s b) The when resul+s b) The achons When b) The achons +Q6en when resul+s hen resul+s L l.mA-s occep+oåe control that Assess control that assess control that control controt that assess exceed exceed limi+s that Laborator Results b) Thea +Qhen when Laboratory False F also and *rends OQQep+oåIe Results detection QQQep+061e cap, achons when resul+s achons ±Q6en when hen when resul+s sensitive rejections Goals acceptable b) The J Type 2) Type Z) Type Ken when Dosirable Quality Look for sensitive to 4' Type 2,) Type and Results Rejection exceed occep+oble J) Type exceed exceed 5.6.2.1 General procedure at medical points of interest Where achons +QKen when resu14s 3s 2s 4s Is x 3s 2s 4s 2s 4s Is x Wh occep+ohule 1.0 When Goals we Expect Need exceed the less occurs less erceed the less WHERE WE . WHERE we ARE WHERE, procedure procedure at procedure at medical procedure medical procedure c,' procedure at c' procedure st have Use the information supplied on Worksheet 1 Capabilities Capabiliti use the de we ro WHERE WE ARE 0.9 limits are less exceed accep+üle limi+s when resul+s and percentages of error Results Wider control limits are less sensitive to error the less overlap occurs between the before- QC Use the Wo Are We exceed accep+oåle False Where OC Quality Cape The revtew q) The of periodic review Look for exceed sensitiv 30 of periodic review 3, 09 peviodlC revtew Eas Goals exceed exceed occep%bie li m WHERE WE ARE abnormal fro Hen when resal+s exceed detection WHERE WE A RE WHERE we WHERE WE WC ARC xceed •6 QC WHERE WE EXPECT ro exceed acceptable WHERE WE ro BE. ro BE. WHERE 0.01- WHERE we ro les Accuracy we change and WHERE ro a ro WHERE WE RE exceed occep+üje Rejection exceed 0 m WHERE WE WHERE 'IS change WHERE we Qua eed accep+061e Provide Provide or error Provide or Provide error Prov•de 904. or Ide review rejections have excee exceed PEPFAR QC Workshop WHERE we ro WE ARE • 6 Sigma do • 6 s•gma 0 03 or E we WHERE Identify rule viol Use the ty ERE WE ARE QC we WE ro WE ARE HERE WE WHERE WE ARE WE ro WHERE WE ARE Detection m WHERE ARE Ide r Look for WHERE WE EXPECT ro DE WE ARE Results WHERE WE ARE Handout WHERE WE ro or WHERE WE expect ro QC i' WHERE we ARE WHERE A Re WHERE 1:2s rule as we ro WE WE ro WE ARE WHERE WE expecr ro Laborator Look 0 03 0.07 Accuracy O Be WHERE WE ARE RE we ARE WE ARC WHERE EXPECT ro DE. WHERE ro as WHERE we EXPECT WHERE WE EXPECT ro BE WHERE WE E WE ARE WHERE WE EXPEcr WHERE ro WHERE WE ARE WHERE WE ro WHERE WE ARE -RE WHERE we ro DC WHERE we ARE ERE we i WE ARE WHERE we ro WE WE ro and calculate either the SEC or Sigma value. WHERE WE EXPECT ro , we ARE WHERE WE ARE ARE we ARE we ro WHERE WE ARE m WHERE WE ARE WE ARE ARE WHERE we detection when given either Precision Quality Quality? cr ro WHERE we WE ARE ' WHERE WE ARE r ro WHERE WE ARE ro WE ARE WE WHERE WE ARE E WE ARE clustering ecr ro WE ARE WHERE WE ARE. WHERE WE EXPECr ro data WHeRE, WHERE we WHERE or less take defec+inj and +cends 10 - WHERE WE ro WHERE WE WHERE WE EXPECT ro or less less C and abnormal Quality ; ae WHERE WE ARE 007 oot oor WHERE WE ARE WHERE WE ARE , cy a. or Cm om WHERE RE WHERE we ERE WE ARE RE WE ARE WHERE WE WHERE WHERE WE , 1 /4 /8 13s/2 1 12 /4 /8 1/2 13 IR 1 12 1 /R /4 /8 1/2 /4 WHERE WE EXPECT WHERE ARE WHERE WE EXPECT ro WHERE WE WHERE WE ARE WHERE WE ARE Results Tool and *rends WHERE WE EXPECT ro BE , data measurements and WHERE WE EXPECT ro DE WHERE WE ro BE WHERE and +rends WE ARE • Provide 90% or greater error detection Provide 90% or greater error detection Provide, 90% or greater error detection measurements i WE ARE a significan Ensures quality performance by quickly detecting Detection E WE ARE WHERE WE oats oals Goats Goals 0oats WHERE WHERE we ARE WE 1:2s rule as y' WHERE WE WHERE WE WHERE h. Tool WHERE W, ro WHERE we ARC Assigned on WHERE WE ARE WHERE WE ARE Use the The laboratory shall design quality control rate o es onls Is 3s 2s 4s Is x 2s x we QC • 6 sigma - any QC will do (just don't use 2 SD limits) alts WHERE WE EXPECT ro BE b) The achons + 0.53 0.9 0.4 0.8 o als als don't clustering cr BE WHERE WE ARE clustering Look for WHERE WE EXPECT ro BE -Y cr BE WHERE WE ARE Re QC WHERE WE roE QC the chart z WHERE WE WHERE WHERE AR' WHERE WE Fewest number WE b) The achons ±QKen when —E WE ARE number r_cn: WHERE WE WHERE WE ro BE number • number WHERE WE *hen detection but have fewer false rejections change population and after-change population WHERE WE ro WHERE WE RE WHERE WE 0.2 WHERE WE EXPECT ro ARE WHERE W. WHERE we EÅPEcr ro WHERE WHERE WE EXPECr V WHERE WE EXPEC 02 WHERE WHERE WE ARE WHERE WE , ARE WHERE we ARE WHERe WHERE WE expecr ro BE WHERE WE ro BE WHERE WE ro BE • co Use the WHERE WE EXPECT ro BE' WHERE we , RE WE ARE ARE WHERE WE EXPECT ro WHERE WE roE we ro WHERE ro WHERE WE ro WHERE WE ARE WHERE WE ARE ARE WHERE WE ro WHERE WE EXPECT WHERE WEEXPEc•: RE WHERE WE EXPECT E WE EXPECT ro WHERE we pen( the chart WHERE we EXPECr WHERE we EXPEC' e WHERE WE WHERE WE ARE WHERE WHERE WE EXPECT ro BE WHERE WE EXPECT TO BE we What WHERE WE ARE WHERE WE ARC WHERE WC ARE WHERE WE ARER WE m WHERE WHERE WHERE we ARE WHERE RE WHERE the SEC or Sigma-metric for a WHERE WE roE WHERE WHERE WHERE WE EXPECT ro BE. WHERE WE EXPEcr ro BE Tool rejections WHERE we EXP • WHERE WE the HERE WHERE we ARE (b aped WHERE 02 WHERE RE WHERE 0.03 — WHERE WE candidate (b aped WE EXPECT ro SE ro WHERE WE ARE WHERE WE EX ECT WHERE WE ro BE W 'HERE WE expecr ro BE we ro WHERE WE EXPECT ro QC away from WHERE WHERE WE ARE WHERE WE WHERE we WHERE we ac+ons acho•ns ac+tons ac+nons a scanning sal+s achons wHERe we b) The achons fahen Use Handout 1 for three levels of control WHERE WE EXPECT ro DE WHERE WE ARE Pencils Alerts us to changes in E WE ARE Ken o, 0.04- -0.04 0.97 wr WHERE WHERE we Use the abnormal '1:2s rule as WC on QC device WHERE WE 70 BC WHERE WE ro i WHERE WE ARE alert us run to the mea, WHERE run as WHERE apply move VHERE WE ARE WHERE WE EXPECT ro we ro WHERE WE ARE we we A Re we ARE Candidate Desirable QC Rule as J' l' t WHERE Monitors and evaluates WHERE WHERE we WHERE WE ARE WHERE ARE WHERE Rule Look fo WHERE we ARE WPHERE WHERE WE QUALITY 1:2s rule as the mean ARE WHERE WE ARE we 02 WHERE con WE ARE WHERE we ARE .97 WHERE WE A RE- WHERE WHERE we WHERE we A Re "'HERE WE ARE WHERE RE WHERE we WHERE WE ro DE WHERE WE A RE WHERE WE EXE N R Rulers Y WHERE WE WHERE WE ARE E ARE WHERE WE 0.2 Rule Ped N WHERE WHERE we oro WHERE WE ARE WHERE wi WHERE t WHERE WE WE WHERE WE ARE Pencils Candidate WHERE RE WHERE WE "HERE WE ARE move away from WE WHERE WE ARE QC QC WHERE we ARE we candidate WERE ARE ro WHERE WE ARE WC WHCRe QC WHERE WE ARE With ant WHERE we ARE measurement procedure. QC (D WE ( % Q ped • Have 5% or less false rejection rates Have 5% or less false rejection rates Use the ! WHERE WE WHERE WE ARE, aped N Rule 003 medically significant errors (Ped 2 90%) —E ero abnormal oo 0.03 0.0 0.23 (70 WHERE WHERE Rule runs less QC Rule a scanning WHERE WE ARE WHERE WE les: clustering (b aped N CY Use the abnormal 1/2 /4 1 /4 WHERE WE ARE 02 0.2 device —Erro accuracy and precision —E —E rro CONTROL (QC) QC • 5 sigma - single-rules such as 1:3s or 1:2.5s QUALITY materials to: QC Use the WHERE WE EXPECT 2.5s Look for exceed candidate QC Rule clustering Rule VHERE we • Rule Single procedures that verify the attainment of Error detection will be greater when control data measurements WHERE method performance 3s 2s 4s 3s 2s 4s Is 3s 2s 4s Is x •g 0.7 QC QC QC mater •Is QC mater As QC mater 'AL QC mater •a's QUALITY 1:2s rule as the mean Rulers 0.1 0.5 WHERE WE' away from 0.1 QC used 1:2s rule as 0.0 aped N a scanning QC abnormal 1:2s rul' meet pc Rejection 0.07 0.04 0.7 0.74 7- CY QUALITY apply 0.03(3X) p, Population • Reagents Reagents PATIENT 1:2s rule as 0.03 0.5 (D CONTROL 0 0.3 0.01- 0.03- (D (QC) away from CONTROL Use the 0.0 QUALITY PATIENT Levey-Jennings Graph WHERE WE ARE the mean 2.0 Fewesenumberdconicowrneasurement QUALITY QUAL move Systematic Enor (SE, mulhples of s) cy (D clustering 0.03(3X) a scanning . QUALI Locate the value on the Sigma-metric scale CY CY - (n) CONTROL QUALITY 0.03(38) SAMPLE device Fewest number of control measurements per Consumables • Consumables 'CY CY a scanning CY QC ICY 0 0.3 2.03 0.65P 1.07 0.5 0.9 0.53 0.03- 0.97 0.65 0.4 0.2 0.8 0.04 0.0 can' 200.0 203.0 201.0 0.0 CY ODS CONT' a scanning QC QUALITY CONTROL Messages PATIENT • 4 sigma - multirules CY CONTROL s per a sca CONTF QUALITY PATIENT QUALITY In QUALITY WHERE PATIEN• PATIE' PAT CONTROL CONTE we ARE limits are set closer to the mean. The tradeoff is How to Select Control Rules System Control metric (ASE, mult on PATIENT CONTRC QUALITY away from 0.03 (3X) CONTROL PATIENT regulatory or accrediting bodv's requirements Meet regulator; or accrediting bodv's requirements Meet or accrediting todv•s requirements Meet or accrediting todv's requirements Meet or accred•tlng todv's requirements CY (ox 0.03(39e) apply move 0.03(38) t' Meet regulatory or accrediting requirements Meet or accredittng requirements QUALITY requirements the intended quality of results. CY CONT Messages Abs move Meet or Meet Key Mesayes 'T) CY .3 'T) (ox 0.03 (390 (QC) (QC) CY device co 2.65s 3.5s 2.5s Control CON (ox 0.03(3X) What will you need? What will you needo? What will you need* cot 'T) VT) Meet PATIENT cm 'T) Messayes CONTROI the easier it becomes to select rules that will ICY 20 3.0 apply (QC) PATIENT SAN 00 2 cy 06 essayes (QC CY SAMPLE on SAMPLE QUAL 0.03(3X) ey Messayes CY QUALITY 0.01 SAMP'E es 0.03(3R) Err:: Systematic E Desirable Desirable= X 0.056) SAMP PATIENT (QC) 1 .13 (QC) PATIE dc QUALITY SAMPLI SD CONTROL 'T) 0.0 S; essays Control metri 202.0 CONTROL SAMPLE SAMPL SAMPI QUALITY ey messages the mean (ox 0.03(38) y essayes CY SAM SO essages CONTROL ss s PATIENT ICY PATIENT QUALITY CONTROL SAMN SAMPLE System PATIENT • Place your ruler so that it corresponds to the Place your ruler so that it corresponds to the number of run number of meosurements per run for number of meesurements run number of measurements per run number of measurements run CY CY QUA for number of meesurements per run number of moasurements run number of meuurements run 'CY CONTROL PATIENT SAMPLE PATIENT (QC) rules CONTROL QUA CY essayes messayes ess ayes CY 0.03 (n 'T) CY 0.00 messayes T) QUALITY SAMPLE 0.03(3c) CONT' CM CY 0.93 1.04 1.00 0.00- 0.07- 1.03 1.07 4.07 0.94 0.97 Key Messayes Control metric (ASE. ess aye s (QC) Key Mess Message ' Messars Messayes 201 s f. Messages Messages Key Messayes messages Messares Key Messages Messag€ Messayec Mess! messag€s t {essays Messi3es essa CONT Control PATIE messages QC CY ITRO 0.03 (38) CONTI PATIENT •less ages Abs essayes PATIENT SAMPLE Systematc (SE QUALITY CONTROL CY (QC) (QC) QUALITY s. 0.5 analytical run possible to save on costs associated 0.9 3.5s 0.24 0.8 2.5s CONTRC 0.4 0.0 PATIENT SAMP PAT ssayes oats oals als oalts Goals Key messayes ey essayes ey ssayes 'CY PAT PATIENT CONTRCL ss ages SAMPLE -s CONTROL PATIENT 0.00 7Z*0 SAMPLE CON"' ey CY QC essages 203.0 CONTP co X 0.056) CONTROL • < 4 sigma - multirules with look-back to previous 'T) CONTROL 0.03-) 0.03(38) essayes .72 72 coy om onl os (QC) (1%) co on (QC) that you will also have a higher false rejection SAMPLE Messayes • Worksheet 1: Predicting Ped CY SAMP PATIEN Messages PATIENT SEC value — ensure the ruler is straight. CONTROL CY SAMPLE onl on os co es Is 'CY (QC) CONI SAMPLE SAMPLE CONTROI PATIENT corn cy SAMP'E PA QUALITY o os oals A. SAMPLE SAMPLE O- CY 200.5 203.0 (QC) (QC) Generate few rule violations when there are no SAMPLE Does SAMPLE alert us to a medically significant change The an SA SAMPLT on QUALITY Rejection Detection SAMPLE, CONTROL CON' SAMPLE (QC) SAMP'E 1. SAM SAMPLE Abs on and an S! SS ages (QC) T, and The accuracy SAMPLE Tie •n Tee •n and SS ayes Abs QC) 'T) The •n The In The an an In •n "n SAMPL' Tie The Tee Ti•e 'n on on SAMP 0.02 (TO) ed on The accuracy and 10 cacy and messages -essages (28) The accuracy Q Why _y messages Key messages 0.02(29.) messages fed on . messages Messages messag€s and on SAMPLE Abs Capabilities Ision j.ö3 — on 0.020%) DtecÅSlon ' on an an Abs Messays 0.026) SAMPLE on v Messages ssag€s .32 Key ey Messages ey messages essages tessages Messag€s Key Messages messayes ey essages .72 0.020 accuracy and@€eclsion accuracy and@€eclsion an ccucacy and@vecÅSlon 0.020%) SAMP'E SA .32 72 essayes assisned NC-Cucacy an with cucacy iqd 0.036) ess ayes Key Messages ssayes byes •curacy and an Systematic Enor (SE., multiples of s) Control metric (ASE, multiples of s) Systematic Enor (SE, multiples of s) ssages .s CONTROL Read the probability of rejection from the Y- (QC) essayes sy*em descetbes es MPLE 0.026) v*em descftb es ss a runs, increase the number of controls analyzed PATIENT c sys+em describ es SPFem es sysfem descri b es 0.006) Shi-em descrt b es The accuracy and descra b es descri b es 189.0 can IS Il .y+iesy*em descrt b es SY*em b The accuracy and onl Messages Key Messayes descrt b rate oats have Goals Key messages Shi-em b es Key Messagef descrt b es on VARIATION S, descflb e and an Key message *em descftbes descftbes b e b es 0.02 (7%) Qnaly+ic s significant errors occurring (Pfr 5%) an System Analytical Bic sysfem describes 199.0 198.5 12 es es (QC) descftb es • Handout 1: Sigma-metrics Sysy-em descft b es sys+em es System descrtb es Shi-em es Shi-em descft b es (1%) .65 as SAMPLE b es 1:3s Target 1 TEa Sys+em descrtb es (o descft b e can an describes on 1:3s '*ic Sys+em descflbes 0.03 TEa TEa analy+icsys+em descri b es es Abs less QC effort is needed for methods that easily • less QC effort is needed for methods that easily 101 (1%) analy+ic sysi-em describ es Äsysi-em descftb es (1%) 36 3s 2s 4s 3s 2s 4s Z.5S 2.56 axis where the ruler and the rule intersect. WHERE A WHERe The accuracy and (1%) ecision Abs 0.01 (1%) Gaussianbisfr,bu+ed on (QC) as accncy and .72 Shi-em descf( b es (1%) nfQ tan and an Pteclsion (1%) 0 0.3 2.03- Key Message accuracy and uracy and an Total Error Error (1%) .65 Key Messages WHERE in b es -o.of -0.01 0.01- WHERE WE on on QC materials (o I: 0.026) (1%) .63 2.53 (1%) WHERE we ARE and sys+em descftbes and Tool 3.04 messages Kev Messages 6) and (1%) WHERE WHERE WE ARE WHERE • 3 or less - look for better analytical methods 0.03 03 Key Messages WHERE WE WE A ) descftb es 0.01 (1%) mmcill_ messayes System we WHERE WE ARE z.5s 2.56 2.5s 2.53 we WHERE Easier to detect larger shifts than smaller shifts l: 2.56 Messages c.03( on FR TA Il 2.56 Key hessages 0.0263X) .32 WHERE WE 2.56 (1%) fie specificafi and 35 WHERE we RE we ARE WE A RE ARE fie cision Z.5S 0.01b) SAMPLE 0.00(5 2.5s eclSlon an and an .11 ( 0.02B Identify rule violations when 4. 0.01 Key messages account the quality required for 0.02 (2%) Defines an acceptable change vs. 0.01 0 and on detect Shi-em descftbes 6.01 (1%) HERE WEARE WE ARE WHERE an ecision Record your answers on Worksheet 1. ,c sys+-em descftb es an descftbes concentration Concenyrafion Concentration sy*em describ es es and Messayes Messages that IS the 2.5s on WHERE we 0.03 ( And 0.01 (1%) WHERE 2.56 IS IS fie specihca€ic and 2.5s 0.02B) 2.5s Key messages we ARe meet quality specifications and The accuracy and and and sys+em descflbe Z.5S E we 2.5s V 2.56 2.56 0.01 (D speci an • Pencils RE we mmol/L mmol/l- mmol/L_ Concenyration 0.9 02 0.4 0.8 we ARE WHERE WE E WE ARE on (1%) we 0.026) 30 0.03 IS 0.01 (1%) 0.00 0) 0.01 (1%) '3 +0 b es Cconcentration fie IS Concentration oco IS a significant change occurs (1%) ant • Reagents Resulnts Reagents unacceptable change (TE < TEA) concenyrafion concen±pafion concentration (1%) 35 fie specihca\iens fie specifica speci$ concentration 0.00 (0%) ey WE ARE 0.0 02 0.2 0.03 0.24 0.23 0.27 Concen±trafion 0.01 3 IS 0.0 0.Ø0 (OX IS 35 1:33 aned specificafic l. 2.53 fie concen±rafion 0.5 Concentration -ö.Tö- mmol/L 0.016) 0.00 (Ota) WHERE WE ro — and speciöca+iens and 1:33 +0 +0 move Place a tic mark next to the rules that are fie con Concen ARE specifications WHERE we ARE 0.01 (1%) IS Proportion of data measurements beyond the 0.01- Key Messages IS cision 2.5s IS *he. 0.07 0.016) i. 0.00 o es IS and sy Messages 01 +0 System 0.02 (2%) and IS 0.00@ 0.00 Key Messages is in or precision mmo I 0.01 (It) "Re we +0 — fie speciFica+iers 001 fie speeihcafiens os 001 (1%) mmo\/l_ WHERE WE , mmol/l_ 2.56 Rulers • Rulers sac. in ocuracy or Pfecision mmol/l_ mmol/L_ _ IS •rhe ODS onl mmo\/L b" on os de as 0.01 (It) 3 +0 Analytical WHERE and cy or Concenyra±ion mmo\ mmo mmo\/L- 000 Is IS Tool IL IS bass .15 mmol 0.016) +0 +0 0.00 0) .15 Determines how far the mean can shift before 1:3s 1:33 0.00 (ox 101 (1%) 0.10 (0%) 0.00 (o) 0.00 (ox) +0 and +0 0.026) +0 maKe maKe I •33 l•3s 1:33 mmol/L 0.00 (o%) QC CM concentration appropriate to consider. 0.026ox) and SF&ihc- bens +0 0.01 (1%) IAS 0.00 Concentration fie 0.01 mmo\/L- WHERE is ne+ Concentration WHERE we A WHERE WE each test and the performance fie specihca+iens QC QC 'The 3.5s 0.00 (ox) WHERE IL — is is IS ISZ Consumables is 15 mmol +0 assyed • - is ro+ 0.00 ISZ, the 1:33 0.03@ racy 0.00 (0%) 0.020) IS the is _ IS he I the IS IOS — fie speciöcafiøns 0.00 (0%) Analytical ardy+icnl fie specihca\'ie concentration 0.00 0) meet quality specifications 0) he 1:3s , WHERE we fie specificafiens QC s+ahshcs cm reqec+ concentration .15 0.006) Change thne analy+ical fie is WH WHERE S+ahshcs CM cefleci- 0.00 (0%) QC feqec+ .13 Tie fie specihcafiens 1:3s change maKe control limits increases with larger shifts 1133 0.026) Change to analyiical Change fie analy+ical fie specificc mmol/L QC CM 1:33 Yo fie analy+ical IS is speciRca6ens 0.03 0.00 erroneous results are reported (SEC and - is fie analyiical — is fie mmol/L— mmol/l_ defines fie Analytical specifications sac. fie specifica 0.07- 0.03- 1.03 3.07 0.93 0.63 0.00- 3.04 0.53 3.03 0.84 0.04- WHERE we IS DO limi+s o IS analy+ica\ we QC cm feqec+ +0 +0 maKe +hen *hen VERE WE ro 'ERE we EXPECT TO B and 1:33 +he n ve ro B anqly+ical 000 (0%) Analytical QC cm ne analyiica\ +hen ncen±rafion e anqlyiica\ fie n peck* cai-iens *hen Circle the rule you select limi+s mmol/L WE ARE fie specifica± and -v cm me* beiro 0.00 (0%) +0 maKe 0.01 (1%) (QC) 04 ecihcåions has occutted. Chanee analyiica\ analyiical +0 +0 bens +0 moKe specificat RE 0.9 has on 1:33 inge fie analy+ica\ QC cm 0.00 (ox) — Is +hen 0.04b) on fro fie analyiical Sigma-metric) ardyfe ) *her limi+s mol [L defines fie cm m, maKe applied. *hen limi+s an amly+cal Meet regulatory or accrediting body's requirements Tie Analytical •Ted. Analytical 0.026) .15 152, cal 0.01 (1%) 1.0 4.0 3.03 10 3.0 4.0 2.05 0.9 20 1.0 2.05s 2.65 2.0s specifications 2.05 1.07 30 4.0 1.0 3.5s 0.03 0.97 3.0s 1.00 1.03 0.27 'Analytical 1.04 3.07 WE EXPECT ro change fie analyiica\ ho fie analyiical chanse fro fie analy+ical change fie chanse fie (0%) Chanee th fie analy+ical .15 amly+ioal limits QC tE WE 15 0.02690) ISZ ERE change +0 fie analy+ical Chanse the analy+ical change fro fie analy+ical change fie analy+ical Change Hie analy+ical on 3.0 Analytical QC cm 20 an we ARE IS reflecf me+hod amly+ioat IS Change re analy+ica\ Frequenc+ Analytical analy+ica\ and applied. Prepare to defend your selection 0.006) is has occuvced. QC cm reqec+ 0.0 change *hen change +0 He IS the we Chanse +0o fie analy+ical chanse Yro He analy+ical capabilities of the test method. is QC Siahshcs +0 ardy+ioal cm correc+ly amly+ioal +0o fie analy+ical is ne+ make QC o analy+ical limi+s Select appropriate control rules based on the (QC) )+hen meanifr$cd arnly+icql correctly WHERE WE EXPEC to fie analyiical ed. occuvqed. QC cm ceqec+ me+hed mol IL and applied. meanir+d QC cm QC S+ahshcs cm See+ively ceqec+ applied. cm feqeci- me+hec{ S+ahshcs cm ccuffed. QC S+ahshcs ceqeci- me+hed QC cm ceqec+ S y;tematic Enor (SE, multiple s of s) Systematic Emor (SE, multiples of s) Stic Enor (SC multiples of s) QC S±ahshcs reqec+ me+hed QC reqec+ me+hed Sys+em has WHERE WE EXPECT Systematic Emr (SE., multiples of s) Systematic Emr (SE, multiples of s) STÄematic Emr multiples of s) STÄtematic Emr (SE, multiples of s) STFtematic Emr (SE, multiples of s) Systematic Emr (S E, multiple s of s) SrÄtematic Emr (SE, multiples of s) Systematic Enor (SE, multiples of s) SF+em bas occurted. QC cc Control metric (ASE, multiples of s) Systematic Enr (SE, multiples of s) Control metric (ASE, multiples of sy STÄtematic Emr SE, multiples of s) (A jntrol metric (ASE, multiples or s) Control metric (ASE, multiples or s) etric (ASE, multiples c f s) make has occutted. has occurced. QC cm me+hed Analytical analy+icql noe fro the analy+ical E. ) +hen anse +0 He analy+ical anse fie analy+ical .15 STÄematic Enor (SE, multiples of s) QC reqec+ me++vd Analytical QC cm Sys+em has occuvted. •s+em has occuned. noe analy+ical Sys+em has occuvted Sys+em has occutced. clinicallv and applied. limi+s has occurced. QC feqec+ QC s±ahshcs cm feqec+ me+*vd has occutted. +hen Sys+ern ha applied. SF+em has QC QC fed. ornly+-icnl And curted. Change CM reqeci- 0.01- roly+ical clncally +hen sys+•em has occunfed. sys+em has occuvced. sac sys+•em has occutted. cm has occuned. for number of measurements per run cm t tesql+ WE TO B QC s+ahshcs cm be beir,9 +esfed +0 maKe change the ts occuned. t WHERE WE ARE iical Analytical åe analyiica\ c.limcallv From www.westgard.com/links-to-india-part-i-best-qc-practices-and-westgard-rules.htm. Last accessed on March 5, 2013. he analyiica\ oe +0 the analy+ica\ Analytical feqec+ limi+s SEC or Sigma for that method QC cm m change om has occurred. s+em has occuvted he em has occunced. Sys+-em has occ fie •hange fie analy+ical analyiica\ chanse fie analy+ical o, change +0 he analy+ic +0 fie I make 1.0 10 0.0 30 +0 4.0 3.0 c-linicallv clinicallv analy+ical the analy+ical limßs iCal IS J' 11 hcs cm reflec+ me+hed Image from Westgard, Image from Westgard, James (2007) Assuring the Right Quality Right. page127 Image from Westgard, James (2007) Assuring the Right Quality RigJht. page127 Image from Westgard. James (2007) Assuring the Right Image from Westgatcl, James (2007) Assuring the Right Quality Right. page127 25 minutes limi 'han +0 -fie anal iica Analytical Ted. Sys+em has change hange anqlv+ica\ SigmaMetricsToolHandout.pdf from Westgard.com website. Last accessed March 12, 2012. has

  • control rules
  • QC rules
  • accuracy
  • precision
  • sigma-metrics
  • QC selection tool
  • westgard
  • QC strategy
  • SLIPTA
  • SLIPTA checklist
  • error detection
  • TEA percent
  • probability of error detection
  • PED