Home

The Cohort Paradigm V30 - Clinical Looking Glass from Montefiore

1. BEG DisbetesHgA1c6e3 5 All of IHgbA1c6e3 5 LabTestDate WHEN IN Je 02 to Je 03 WITH Age 21 to 65 AND LT Eventi New Event Definition within 0 to 0 select time units select time direction New Abstract Event Definition New within condition line Note within symbol selected Condition GUI e _ r Edit Selected Condition Eventi NOT Allof New Event Def WITHIN 0 TO 0 Days w After Event DiabetesHgA1cGe9 5 Update Update and Close 42 96 Steps Edit Selected Condition a Eventi 1 CINOT Allof New Def W 2 WITHINJo 3 JTojo Ue ae J DiabetesHigbA1cGeP 5Y es Cy 97 Change All of to Earliest of Edit Selected Condition O Event 1 NOTI Earliest of New Event Def Y WITHIN rop Days After Update Update and Close Event DiabetesHgbAicGe9 5 98 Rename Eventi to RepeatHgbA1c 99 Build a new event that is a Lab test of Hemoglobin Alc You did this before but now do not add the second condition of lab test value _ A e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deffype EVENT amp SelMenuOy Edit Options Name epeat Event Lab Test Vv l Tvpe i Event Definition Options E a Definition u i X Undefined Condition INOT Lab Test Type v HEMOGLOBIN A1c V Update cese 43 100 Update and Close to
2. HEART FAILURE A 428 HEART FAILURE F 428 0 CONGESTIVE HEART FAILURE UNSPECIFIED 428 0 CONGESTIVE HEART FAILURE UNSPECIFIED 428 20 UNSPECIFIED SYSTOLIC HEART FAILURE 428 1 LEFT HEART FAILURE 428 22 CHRONIC SYSTOLIC HEART FAILURE 428 20 UNSPECIFIED SYSTOLIC HEART FAILURE 428 23 ACUTE ON CHRONIC SYSTOLIC HEART FAIL 428 21 ACUTE SYSTOLIC HEART FAILURE z 428 31 ACUTE DIASTOLIC HEART FAILURE 428 22 CHRONIC SYSTOLIC HEART FAILURE 428 40 UNSPEC COMBINED SYSTOLIC amp DIASTOLIC 428 23 ACUTE ON CHRONIC SYSTOLIC HEART FAIL 398 91 RHEUMATIC HEART FAILURE 428 30 UNSPECIFIED DIASTOLIC HEART FAILURE 428 1 LEFT HEART FAILURE 428 31 ACUTE DIASTOLIC HEART FAILURE 428 21 ACUTE SYSTOLIC HEART FAILURE 428 32 CHRONIC DIASTOLIC HEART FAILURE 428 30 UNSPECIFIED DIASTOLIC HEART FAILURE 428 33 ACUTE ON CHROWIC DIASTOLIC HEART FAI 428 32 CHRONIC DIASTOLIC HEART FAILURE 428 40 UNSPEC COMBINED SYSTOLIC amp DIASTOLIC 426 33 ACUTE ON CHRONIC DIASTOLIC HEART FAI 62 398 91 RHEUMATIC HEART FAILURE 428 HEART FAILURE 4 428 0 CONGESTIVE HEART FAILURE UNSPECIFIED 428 1 LEFT HEART FAILURE 428 20 UNSPECIFIED SYSTOLIC HEART FAILURE 428 21 ACUTE SYSTOLIC HEART FAILURE 428 22 CHRONIC SYSTOLIC HEART FAILURE 428 23 ACUTE ON CHRONIC SYSTOLIC HEART FAILI 428 30 UNSPECIFIED DIASTOLIC HEART FAILURE 428 31 ACUTE DIASTOLIC HEART FAILURE 428 32 CHRONIC DIASTOLIC HEART FAILURE 428 33 ACUTE
3. Inpatient Admit Inpatient Discharge Internal MedAdminDate Lab Test Med Order Start IS Med Order Stop Medication Allergies Start Date Medication Allergies Stop Date Microhininayv CollectNate Define an attribute condition 21 21 Hover over the word definition 22 on the definition canvas Note how a box appears around the word definition clueing you into the fact that the area is now primed for a new action e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deffype EVENT amp SelMenuOp za Name HabAte ge 95 i MI Pal Event Definition Options Mi nents 23 Right click on the activated definition button to open a menu This is the definition index line controlling the canvas Name HigbATe ge 95 aly Event Definition Options Name HabAic ge 9 5 Lab Test v Event Definition Options E Definition bie x v Undefined Condition INOT Please Select kul Please Enter the Required Data 22 25 Left click on the downward arrow next to please select Name HobAic ge 93 5 Lab Test ww Uodare Uudutz and Close Event Definition Options EA Definition tl Undefined Condition Please Select Categoric Lab Test Value Grouped Numeric Labs INOT ESCE IT Lab Test Value bh Si Test Clinic B aE Test MD True First Lab Text WHEN INI New 26 Left click on Lab Test Type Please Select
4. MAL NEO LYMPH INGUIN LEG 196 6 MAL NEO LYMPH INTRAPELV a 196 8 MAL NEO LYMPH NODE MULT 196 9 MAL NEO LYMPH NODE NOS 197 0 SECONDARY MALIG NEO LUNG 197 1 SEC MAL NEO MEDIASTINUM 197 2 SECOND MALIG NEO PLEURA 197 3 SEC MALIG NEO RESP NEC 197 4 SEC MALIG NEO SM BOWEL r 119 After clearing the Set Builder the space will look like this 58 E Set Builder Webpage Dialog 2 https clq emerginghealthit com SBApp S hOrioekOfrpatwpsd2kmykvl SetBuilder aspx UserName lstahl amp SetType 14 amp 5etID 84595 Select Type and Action ICD9DxX Diagnosis Explore Pick an TE set Please Choose E C Code Description Grouping hierarchy names _ Set Hierarchy Please Choose a Show dimension values only Show Hierarchies Remove Search for Heart Failure diagnoses Type Heart Failure in the Search field and click 120 Find 59 Select Type and Action ICD9DX Diagnosis View Code List ick an existing set Please Choose Optional earch options Heart Failure L Code Description ve earch Within Set _ Hierarchy Please Choose Show dimension values only 0 Show Hierarchies 121 You may select only the diagnoses codes that you wish to include in your set by holding down the Control key as you click on each one For this set you want all of the codes Add All 7 displayed in the search results so simply click Search result list 398 91 RHEU
5. PINOT Lab TestType V xi Update Update and Close N SET 27 Now you can choose a single lab test by choosing the sign or build a set of lab tests by using the IN SET option We are going to be using a single lab test so we will choose the Un sign 28 Left click on the sign LINOT Lab Test Type ww Please Select Ww cena 49 30 Left click on the next arrow over to see all the available lab tests 23 CROSSMATCH RESULT CRP HI SENSITIVITY CRP HIGH SENSITIVITY CRY CRYO REQUEST CRYOGLOG QUANT ID CRYOGLOB SCREEN CRYOGLOBULIN SCREEN CRYPTOCOCC AG CSF CRYPTOCOCC AG SERUM CRYPTOCOCC ANTIGEN Cryptococcal Ag Titr NOT Lab Test Type v CRYPTOCOCCAL TITER CRYPTOSPORIDIUM E Update and Close _ So CRYPTOSPORIDIUM AG Crystal Fluid Type CRYSTALS i USR outpatientVisitoBMedicaidFirstbi ERYS rAr S CF eR i CSF J anesthesia CSF AdenoV qPCR CSF APPEARANCE T1 CSF APPEARANCE T4 CSF BAND COUNT T1 CSF BAND COUNT T4 J CSF BANDS T1 CSF BANDS T4 CSF BLAST COUNT T1 CSE BLAST COUNT T4 31 Either slide down the window to Hemoglobin Alc or simply type Hemoglobin A1c and the menu will advance to the test Hemoglobin Alc Note the space between words Hemoglobin and A1c CINOT Lab Test Type v HEMOGLOBIN Aic z ee es Ih 850 32 Left click on update and close e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilde
6. i R T e USR New ele 0209 J x y Test MD LJ CensusTract ec B USR or procedure LI city EC Buse orders with text x Lo l v Q Ix Virrue First LI Age at Index Datew anneMearaPopulationBasedReadmission lt gt Output a ee lt gt If the Index Event Line points to the first Condition Line then it is the event in the First condition line that is used to provide attributes In our last section this would be the original Hemaglobin A1c value greater than or equal to 9 5 that made the patient eligible for inclusion in the cohort This is the time zero value of Hemoglobin A1c for each patient 47 If the Index Event Line points to the second Condition Line then it is the event of the Second Condition that is used to provide attributes In our last section this would be the repeat Hemoglobin A1c collected 180 to 365 days after the first In a browse we would see the value of the repeat Hemoglobin Alc some of whom we would hope would be below the 9 5 value that occurred at time zero If the Index Event Line says Earliest of Any not specifying a condition line then the patient s first qualifying event on the canvas will determine the attributes that will be shown Of course in our case the within after condition of the second line guarantees that all the events of the second line are after the events of the first line Therefore the Index Line with Earliest of any would be fucn
7. 3 a 1 A 74 Search options G2 Find Reset earch On Description LI Grouping hierarchv names gt aaa Search Within _ Set _ Hierarchy Please Choose Display Options Show dimension values only Show Hierarchies Search result list Assigned list 428 HEART FAILURE 428 0 CONGESTIVE HEART FAILURE UNSPECIFIED A sackets 428 1 LEFT HEART FAILURE 428 20 UNSPECIFIED SYSTOLIC HEART FAILURE 428 21 ACUTE SYSTOLIC HEART FAILURE Add All gt 428 22 CHRONIC SYSTOLIC HEART FAILURE 428 23 ACUTE ON CHRONIC SYSTOLIC HEART FAILI 428 30 UNSPECIFIED DIASTOLIC HEART FAILURE 428 31 ACUTE DIASTOLIC HEART FAILURE KA SA 428 32 CHRONIC DIASTOLIC HEART FAILURE 428 33 ACUTE ON CHRONIC DIASTOLIC HEART FAII 398 91 RHEUMATIC HEART FAILURE 428 HEART FAILURE 428 0 CONGESTIVE HEART FAILURE UNSPECIFIED 428 1 LEFT HEART FAILURE 428 20 UNSPECIFIED SYSTOLIC HEART FAILURE 428 21 ACUTE SYSTOLIC HEART FAILURE lt Remove 428 22 CHRONIC SYSTOLIC HEART FAILURE 428 23 ACUTE ON CHRONIC SYSTOLIC HEART FAILI 428 30 UNSPECIFIED DIASTOLIC HEART FAILURE 428 31 ACUTE DIASTOLIC HEART FAILURE 428 32 CHRONIC DIASTOLIC HEART FAILURE 1 type in the code 428 2 check box for codes 3 Left click find The cycle you should follow is use word find then code find to build your set 122 Review the Assigned list If it is correct click Save 61 A 398 91 RHEUMATIC HEART FAILURE E 428
8. HgbA1c ge 9 5 LabTestDate WHEN IN Je 02_ Je 03 WITH Age 21_65 Foe repeatHgbAicGE9 Earliest of repeat HgbAic ge 9 LabTestDate within 180 to 365 Days After Event Diabetic EY EY 49 Build Cohort with Good Repeat Test 1 Cohort BadDiabetesWGoodRepeatF2 Earliest Hgbalc gt 9 5 in Je 02 Je 03 Age 21 65 with a subsequent HgbA1c measuring lt 7 within 180 365 days after Focus index date time on the subsequent HgbA1c N 373 BadDiabetesWGoodRepeatF2 Criteria i First hemaglobin A1c in patients seen between Je 02 Je03 with a value Greater than or equal to 9 5 and Age between 21 65 ii a repeat HgbA1c within 180 365 days after the first with a value less than or equal to 7 Result N 373 iii The index event line should point to the second hgbalc so we have a cohort called BadDiabetesWGoodRepeatF2 who have been brought under control by the time of their second study c S INDEX EVENT Earliest of repeatHgbAicLE7 0 And Ee Diabetic Earliest of HgbAic ge 9 5 LabTestDate WHEN IN Je 02_ Je 03 WITH Age 21 651 i Or ey repeatHgbA1cLE7 0 Earliest of repeatHgbAicLe7 LabTestDate within 180 to 365 Days After Event Diabetic 50 Build Two more Cohorts by Changing the Focus of the Index Event Line to the first line BadDiabetesWBadRepeatF1 Bl edit Selected Name Bad Diabetes WRepeathadF1 _C J INCEX EVENT Earliest of DiabetesHgbAicGe9 5 And f
9. ON CHRONIC DIASTOLIC HEART FAIIS428 41 ACUTE COMBINED SYSTOLIC amp DIASTOLIC I 428 40 UNSPEC COMBINED SYSTOLIC amp DIASTOLIC 428 9 HEART FAILURE NOS 123 Name your new set CHF All and click Save Name CHF AII Description 124 Note the confirmation message Click Close Set saved as CHF All Save 63 Use Your New Set in Your Condition Definition Click the set menu and find your new set at the top of the list Select it by clicking on it USR CHF All 22 ACUTE IV Salle a dfletche Acute MI_646c22cc dfletche Acute MI_99e5d725 jweissma CLG_UC4 Diagnosis List2 i 3 A F jebellin mvocardial infarctio_7665cb96 E a ae EE hollin myocardial infarctio_7665cb96_05cbbd07 E e ebellin myocardial infarctio 7665cb96 0b437bbf a l ebellin mvocardial infarctio 7665cb96 2896bde5 In ebellin mvocardial infarctio 7665cb96 744785aa ebellin myocardial infarctio 7665cb96 9dc658b7 ebellin mvocardial infarctio 7665cb96 a73ef96a ebellin mvocardial infarctio 7665cb96 cc204169 a ebellin myocardial infarctio_7665cb96_d4c327c8 fal Edit Options ebellin myocardial infarctio_7665cb96_e4d66117 Name CHF Admissions Event Typ ee ee infarctio_7665cb96_f924d6 c Update and Close Close mmccrosk Polio_ba3abcb1 Pal Event Definition Options Definition Adjustment disorders NOT Inpatient ICD9 Group vs INSET vy Pia Aunzjjo Mi 125 Do not change the settings for Primary or Secondary an
10. WHEN IN march2012 1 Result N 2 patients 2 Now get the same information using a different approach Build a third cohort called CHF March 2012DiedWheninMethod 71 Use the first condition line from the original cohort and then add another condition line using a when in condition with the intent to use the when in to mean during the event of the condition line 1 to which condition line 2 will point When in has two different meanings depending on what you will choose in the GUI Edit Select Event Editor below the event canvas If you choose New Duration Def New Duration Def Select from Copy from gt Select System Duration gt g WHEN IN WITH f LINOT Select System Duration b i WHEN IN Nen Duration Def WITH Update Copy from the event canvas understands the when in to mean when in a defined calendar duration and offers you a window to put in the calendric information Edit Options ee Name QDR1 C391e from O Update Update and Close El to O El Q If you choose Select System Duration from the GUI Edit Select Event Editor the event canvas understands that you are trying to find a durational event on the canvas that you will point to By pointing to this durational event you will be demanding that this condition line event occurs during the duration of the pointed to durational event A du
11. time of the patient As a group these patients constitute the cohort of patients whose care is to be evaluated This cohort can also be represented by two columns of data in a spreadsheet The first column is the patient s medical record number identifier and the second column is the patient s index date time from which elapsed time will be calculated Figure 2 Spreadsheet representation of the cohort Medical Record Number Index Date time 1 10 2010 2 13 2010 3 11 2010 3 11 2010 4 15 2010 6 1 2010 7 8 2010 9 15 2010 9 15 2010 10 6 2010 O CON DM BWYN e 4 O Both illustration 1 and the two columns above equivalently represent a cohort of patients a group of patients with a date time associated with each member In this case the date time is only a date Now we want to illustrate how rapidly these diabetics get their blood sugars under control Achieving control is indicated by a plus sign For the sake of simplicity in this example all patients achieve control 1 1 2010 1 1 2011 1 1 2012 Patient 1 44 i Diabetes Control 3 1 O index date 4 4 start therapy gt efs outcome Pc etree ra achieve lab value y Se ee eee EEN B aces ete ce eect Dooa a a AQ Lsien a ia Figure 2 Diabetics achieving blood sugar control We note that each patient achieves control at a different calendar date and after a different elapsed time from diagnosis Now how might you evaluate medical success for t
12. want to force another event such as a cardiac echo blood pressure to occur during the event of the hospitalization 74 _ c BINDEX EVENT Earliest of Any And fesse pace chfMarch2012 All of chf discharge InpatientDischargeDate WHEN IN march2012 1 Add when in condition pointing to a durational event on the event canvas to force the death event to occur during the hospitalization New Duration Def Ml Edit Select Select from gt pedina Copv from CI NOT Al select System Duration 11719 WHEN IN New Duration Def WITH No Demographics Update When you update and close you get Event Canvas z c BJINDEX EVENT Earliest of Any And ace chfMarch2012 All of chf discharge InpatientDischargeDate WHEN IN march2012 H ke Paice DiedDuringHospitalization All of DeathInHouseOrSocialSecurity SSAorInHouse WHEN IN chfMarch2012 Inpatient Stay Now build The result N 2 The same as the first method 75 Homework Before the second class 1 Rebuild all cohorts of class exercise and make sure you understand how to do it You will absolutely need BadDiabetesWRepeatGoodF2 BadDiabeteWRepeatBadF2 BadDiabetesWRepeatGoodF1 BadDiabeteWRepeatBadF1 2 Read Pages 1 8 of Study Designer Manual 3 In preparation for the second class build the following four cohorts Tutorial 10 Build four pneumonia Coh
13. you will feel comfortable to continue your education on your own To become a CLG expert requires time experience and investment of effort We have four manuals built into CLG which can be sent to you in advance of the class if you send an email to ebellin montefiore org asking for the clg manuals The manuals can be downloaded to your computer or if you have an ipad download to an ipad and save as an ibook avery convenient way to manage the manuals 1 CLG User manual 775 pages selections should be read after each class to solidify what you have learned and enhance it 2 Event Definition Manual Volume 3 Provides a nearly comprehensive listing of all the events and their attributes available for analysis at Montefiore 3 Ad Hoc Reports User Manual 4 CLG Manual Addendum 4 3 pdf 5 Voutube Channel plavlist with 34 recorded Tutorials following the lectures and notes https www youtube com watch v wPmjQmLoKSO amp list PLf7raPnm lLOeAWU2cNf2jIMDR QuUTCum Resources 1 Computers Enabled with CLG a Your mentors you should arrange local access for practice at hours and off hours b Montefiore Library 9 8 45 mon Friday Saturday 9 5 If you want to play videos bring earphones and you will have to adjust the realtek program so that it is not muted Remember to return it to its muted state afterwards c ClS training room above the Montefiore Moses Cafeteria second floor you must make arrangements in advance t
14. 69 Close W You are now back in the event canvase with the demographic defined 70 Now click update and close Po Lo Name Bad Diabetes EH CW INDEX EVENT Earliest of Any And Booms face DiabetesHgbAicGe9 5 L All of HgbAic ge 9 5 LabTestDate WHEN IN Je 02 to Je 03 WITH Age 21 to 65 7A Your gui editor closes and you now have a condition line painted on the event canvas The Index Event line specifies Earliest so it is looking for the first event for the mrn in the condition line it is pointing to However it does not point to a specific condition line It points to Any of the condition lines on the canvas Since there is only one condition line on the canvas it is functionally pointing to only the one line named DiabetesHgbA1cGe 5 That one condition 33 line has All as its criteria so there may be more than one HgbAic lab test for an individual patient Because the Index Event line states Earliest it will choose the Earliest of All the HgbAic results greater than or equal to 9 5 for an individual patient thereby yielding a single 34 event per person a single mrn and date time of that event per person which is a cohort of unique individuals 72 Left click on build There are 3 130 people who meet the criteria on the event canvas The cohort in the management pane changes from yellow striped CiG Event Canvas A Management SATE Collections and C
15. ASMOSIS PNEUMONIA 130 4 TOXOPLASMA PNEUMONITIS 136 3 PNEUMOCYSTOSIS 320 1 PNEUMOCOCCAL MENINGITIS 352 3 PNEUMOGASTRIC NERVE DIS 480 VIRAL PNEUMONIA 480 0 ADENOVIRAL PNEUMONIA 480 1 RESP SYNCYT VIRAL PNEUM 77 480 2 PARINFLUENZA VIRAL PNEUM 480 3 PNEUMONIA DUE TO SARS ASSOCIATED COR 480 8 VIRAL PNEUMONIA NEC 480 9 VIRAL PNEUMONIA NOS 481 PNEUMOCOCCAL PHEUMONIA STREPTOCOCCU 482 OTH BACTERIAL PNEUMONIA 482 0 K PNEUMONIAE PNEUMONIA 482 1 PSEUDOMONAL PNEUMONIA 482 2 H INFLUENZAE PNEUMONIA 482 3 PNEUMONIA DUE TO STREPTOCOCCUS 482 30 PNEUMONIA DUE TO UNSPEC STREPTOCOC 482 31 PNEUMONIA DUE TO STREPTOCOCCUS GRC 482 32 PNEUMONIA DUE TO STREPTOCOCCUS GRC 482 39 PNEUMONIA DUE TO OTHER STREPTOCOCCI 482 4 PNEUMONIA DUE TO STAPHYLOCOCCUS 482 40 PNEUMONIA DUE TO STAPHYLOCOCCUS UN 482 41 METHICILLIN SCPTBLE PNEUMONIA DUE TO 482 49 OTHER STAPHYLOCOCCUS PNEUMONIA 482 8 PNEUMONIA DUE TO OTHER SPECIFIED BACT 482 81 PNEUMONIA DUE TO ANAEROBES 482 82 PNEUMONIA DUE TO ESCHERICHIA COLI E 482 83 PNEUMONIA DUE TO OTHER GRAM NEGATI 482 89 PNEUMONIA DUE TO OTHER SPEC BACTERIA 482 9 BACTERIAL PNEUMONIA NOS 483 PNEUMONIA DUE TO OTHER SPECIFIED ORGAN 483 0 PNEUMONIA DUE TO MYCOPLASMA PNEUMON 483 1 PNEUMONIA DUE TO CHLAMYDIA 483 8 PNEUMONIA DUE TO OTHER SPECIFIED ORG 484 PNEUM IN OTH INFEC DIS 484 1 PNFUM W CYTOMFG INCI DIS 484 3 PNEUMONIA IN WHOOP COUGH 484 5 P
16. CLINICAL LOOKING GLASS The Cohort Paradigm How to Repurpose EMR data A syllabus for a four session Introduction to Clinical Looking Glass A self tutorial Guide Version 30 October 25 2015 Eran Bellin M D Before the First Class Required Reading Before the first class please read 1 Book Eran Bellin M D Riddles in Accountable Healthcare A primer to develop analytic intuition for medical homes and population health available on Amazon as paperback or kindle Read Introduction and Chapters 1 3 4 5 8 11 19 21 Don t worry chapters are short and are a quick read 22 a copy or share with a friend JCO7 1 Appropriate Use of Clinical Looking Glass Montefiore Policy and Procedure JCO 1a Practical Guidance in the use of CLG Montefiore Policy and Procedure Instructions for Registering QI projects Read Chapter 1 of this Syllabus the Cohort Paradigm P 1 15 aa ee am going to assume that you are familiar with the concepts as we will be going quickly to the application itself Print out this material before class Background Objectives of the four session Core Training Clinical Looking Glass is a powerful data retrieval and analytic tool The objective of the course is to enable you to build cohorts extract information and perform time to outcome analyses It is expected that in the course of this work you will become familiar with the manuals and built in videos which you will read along at home so
17. Data 60 Left click on the please select arrow INOT CESES Please Select w e Update I ETTD AS Close Primary Care MD Set iOEXDIread Preferred Language ma selected Cond SES esHgbAicGe9 anInDied Social Security Number Present Zip Code Set 61 Left click on age Update Definition 63 Left click on please select which opens a drop down menu 64 Left click on between nore een VJ eRom TOC Update Definition J Update and Close close 31 66 Type in the ages of 21 to 65 inclusive of 21 by checking the box near 21 This results in the following image PINOT Ase o veen eon fos x update Definition Update and Close close 67 Left click on update and close a e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx2Def vppe DEMOGRAPHICS8SE EA e Name Age 21 to 65 update P close H Even Denton options aE Definition i x y Age BETWEEN included 21 AND excluded 65 You have successfully painted your condition line restricting demographics to age 21 to 65 68 Now left click update and close 32 ram Edit Selected l LI Gave As ll Name Bad Diabetes E NN F CW INDEX EVENT Earliest of Any And Q Eventi All of WHEN IN Edit Selected Condition DiabetesHgbA1cGe9 5 LINOT All of HgbAic ge 9 5 LabTestDate W WHEN IN Je 02to Je 03 WITH Age2ito65 Y
18. MATIC HEART FAILURE 428 HEART FAILURE 428 HEART FAILURE Add 428 0 CONGESTIVE HEART FAILURE UNSPECIFIED 7 428 0 CONGESTIVE HEART FAILURE UNSPECIFIED 428 20 UNSPECIFIED SYSTOLIC HEART FAILURE 428 1 LEFT HEART FAILURE 428 22 CHRONIC SYSTOLIC HEART FAILURE 428 20 UNSPECIFIED SYSTOLIC HEART FAILURE Add All gt 428 23 ACUTE ON CHRONIC SYSTOLIC HEART FAIL 428 21 ACUTE SYSTOLIC HEART FAILURE 428 31 ACUTE DIASTOLIC HEART FAILURE 428 22 CHRONIC SYSTOLIC HEART FAILURE lt Remove 728 40 UNSPEC COMBINED SYSTOLIC B DIASTOLIC 428 23 ACUTE ON CHRONIC SYSTOLIC HEART FAIL 398 91 RHEUMATIC HEART FAILURE 428 30 UNSPECIFIED DIASTOLIC HEART FAILURE 428 1 LEFT HEART FAILURE 428 31 ACUTE DIASTOLIC HEART FAILURE Remove All 428 21 ACUTE SYSTOLIC HEART FAILURE 428 32 CHRONIC DIASTOLIC HEART FAILURE l 428 30 UNSPECIFIED DIASTOLIC HEART FAILURE 428 33 ACUTE ON CHRONIC DIASTOLIC HEART FAI 428 32 CHRONIC DIASTOLIC HEART FAILURE 428 40 UNSPEC COMBINED SYSTOLIC amp DIASTOLIC 428 33 ACUTE ON CHRONIC DIASTOLIC HEART FAI 7 You can search for ICD9 codes using google on the internet and you can then enter the icd9 code into the find box after you click on the code search button 60 CiG Set Builders NU Share Delete Select Type and Action Set Type ICD9Dx Diagnosis Iv View Code List Search ESET Pick an existing set Please Choose VI view l Optional
19. MarrowBiopsy Earliest of BoneMarrowBiopsy ProcedureDate WHEN IN Duration2000_present WITH AgeGE18 2 MonoclonalParaProteinemia Earliest of Any Or Sa Inpatient_MP Earliest of Inpatient_MP InpatientAdmissionDate WHEN IN Duration2000_present WITH AgeGE18 5 5 Rheumatoid Arthritis Earliest of Any Or 6 Saale Inpatient with RA Earliest of Inpatient RA InpatientAdmissionDate WHEN IN Duration2000 present WITH AgeGE18 7 ails Outpatient with RA Earliest of Outpatient_RA ClinicVisitDate WHEN IN Duration2000_present WITH AgeGE18 Lines 1 7 are called condition lines For a patient to qualify as a member of the cohort all conditions 1 7 must be met Once the conditions are met the medical record number representing that patient is included in the cohort As you know a cohort is made of two columns A medical record number and an index date time unique for each patient The index date time is the time from which elapsed time is to be calculated You need to decide which condition line event 1 7 is going to provide the date time to associate with the patient s MRN in the cohort Line 8 is called the Index Event Line The Index Event Line chooses the events that result from application of all the condition lines painted on the Event Canvas below Once condition lines 1 7 are run the resulting set of events are acted on by the i
20. NEUMONIA IN ANTHRAX 484 6 PNEUM IN ASPERGILLOSIS 484 7 PNEUM IN OTH SYS MYCOSES 484 8 PNEUM IN INFECT DIS NEC 485 BRONCOPNEUMONIA ORG NOS 486 PNEUMONIA ORGANISM NOS 487 0 INFLUENZA WITH PNEUMONIA 495 7 VENTILATION PNEUMONIT 482 42 METHICILLIN RSTNT PNEUMONIA DUE TO ST 488 01 Flu dt iden avian w pneu 488 11 Flu dt 2009 H1N1 w pneu 488 81 Flu dt nvl A vrs w pneu 78 ARISE sevens ex Name Pneumonia 2010 Male admits dc alive x HC IN INDEX EVENT Earliest of Any And l EAA admits All of Pneumonia InpatientAdmissionDate WHEN IN 2010 WITH Male Note the numbers of patients detected by using Inpatient Admit differs from the number with Inpatient Discharge The reason for this is the requirement of being in 2010 applies in one case to the inpatient admit date and the other the inpatient discharge date So that in a small number of cases patients admitted in 2010 were not discharged in 2010 and people discharged in 2010 were not admitted in 2010 5 Review Videos 1 Set Builder 79 2 Event Canvas a b C d Build A cohort Within When Subgroup 80 l WHEN IN march2012 81
21. NT Earliest of Any And Ze DisbetesHgAicGe9 5 Earliest of HgbAicGe9 5 LabTestDate WHEN IN Je 02 to Je 03 WITH Age 21 to 651 L AND Lorre Repeat HgbAtc Earliest of RepeatHgbAic LabTestDate within 180 to 365 Days After Event DisbetesHgA1c6e3 5 108 Note only 1821 patients had a repeat HgbA1c within 180 to 365 days 109 The management Pane shows two cohorts because you used save as 45 Event Canvas 110 Both are green because you have the cohort rules as well as the built cohort with mrns and date times 111 Remember post build to exit do not save do not save and exit 112 Congratulations you have now completed a two condition cohort with temporal sophistication of a within command 46 Develop Intuition Index Event Line Tutorial 6 Browse and the Power of the Index Event Line httos www youtube com watch v LboeRILua Now let s develop deep intuition of how the Index Event line works using the browse option in the management pane to focus our attention Remember the purpose of the Index Event Line is to specify the line and which event in that line should provide the date time for the person qualified by the multiple condition lines on the event canvas But what is remarkable is that Clinical Looking Glass remembers which event generated the date time and when you browse a cohort of patients CLG displays the index date event s attributes In the case of our last sec
22. PA Insured Line of Business ___ Event Type Inpatient Admit End Point Event Please Select Close f l Event aj Admit Priority cE pO APR DRG Average LOS APR DRG Weight Admit Nursing Station Discharge Nursing Statio Inpatient Encounter Type MNOT ERRAR i casa Update and Close Please Enter the Required Data http clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Internet Protected Mode Off Please Select an Duration Definition Define a Set 2 You need to define a diagnosis set using Set Builder To access it click Bi Whenever you see a three dotted button the program is making available a window to build whatever is supposed to be entered to the left of the builder button INOT Inpatient ICD9 Group vy IN SET X 7 Primary or Secondary Present On Admission v 57 118 Set Builder opens with the currently selected set loaded Clear this set by clicking the Clear button Select Type and Action l Set Type ICD9DX Diagnosis 2ndary malig vien Set Search options ii Search On C Code Description FI Grouping hierarchv names Search Within Set _ Hierarchy Please Choose Show dimension values only O Show Hierarchies 196 0 MAL NEO LYMPH HEAD NECK A 196 1 MAL NEO LYMPH INTRATHOR E 196 2 MAL NEO LYMPH INTRA ABD a 196 3 MAL NEO LYMPH AXILLA ARM F Add All gt 196 5
23. aa DiabetesHgbAicGe9 5 Earliest of HgbA1cGE9 5 LabTestDate WHEN IN JuneQ toJune03 WITH Age 21 to 651 AND RepestHigbA1cGs3 Earliest of RepestHgbAicGe9 LabTestDate within 180 to 365 Days After Event DisbetesHgbAicGe9 5 BadDiabetesWGoodRepeatF1 Bl Edit Selected Name BadDiabetesWRepeatGoodF1 je Event Canvas E c INDEX EVENT Earliest of DiabetesHgbA 1oGe9 5 And L fEIGp DiabstesHgbA1c629 5 Earliest of HgbA1cGES 5 LabTestDate WHEN IN June02to une03 WITH Age 21 to 65 L AND LeS RepestHgbAlcLe7 Earliest of RepestHgbAicLe7 LabTestDate within 180 to 365 Days After Event DisbetesHgbAicGe9 5 You will need all four cohorts for your next class Challenge question Why is it important which line the index line is focussing on 51 52 Now build a CHF example focused on admission Educational goals 1 Learn how to invoke an inpatient event how to build a diagnostic set 2 Understand why you want an inpatient discharge or inpatient admission You know this from Chapter 5 Riddles in Accountable Healthcare Heads or Tail which end is up Clinical Looking Glass Session Exercise Build a cohort of Congestive Heart Failure admissions using the admit date to represent the hospitalization Tutorial 8 Build a Congestive Heart Failure Cohort using a Diagnostic Set https www youtube com watch v 8WxXMury6w 0 Create a new cohort 53 Event Canvas launches in a new w
24. ach is much better The cohort approach recognizes outcome relative to each patient s own start time relative to the patient s own index date Every analysis begins at each patient s index date the patient s own time zero We now modify our question from do we have control in the year 2010 to do we achieve control within the first year of follow up This is what the cohort approach would graphically look like Enrollment 1 Year 2 Years is Diabetes Control Bice Q index date ma start therapv ml 3 outcome 4 achieve lab value Pa I 5 i0 2 What of new diabetic patients were controlled within 1 year Figure 4 Successfully treated diabetics in one year cohort view Same data but the illustration graphically represents that we are following elapsed time for every patient starting at his own zero time sorted by elapsed time until diabetes control the relevant metric of evaluation Note the time scale is no longer calenderic time but is elapsed time since presentation with outcome evaluated by year s end Note how Five of the patients Patient 3 8 9 1 4 5 10 50 are successes whereas before only 40 were successes Note now patient number 9 is recognized as an obvious SUCCESS This cohort view by respecting each patient s zero time index date time and by evaluating success by elapsed date time is a fairer representation of the care delivered Suggested reading is Riddles in Acco
25. brought under control by 6 months 1 year 1 5 years It compares the success across the years and across clinics by comparing graphs of the cumulative percent brought into a good state by each of these time intervals This analytic pattern is exemplified in Chapter 3 of Eran Bellin Riddles in Accountable Health Care available on Amazon as paperback or Kindle Question Pair 3 Business Volume Centric How much did we spend on drug X Patient Value Centric How rapidly did patients improve on drug X The first question asks how much did we spend on drugs a pure volume of cost question The Patient value question asks from the time we started the drug on an individual patient how long did it take before we saw improvement We are asking a cohort value question where we are following each member of the cohort of patients taking the drug what percent were brought to the clinical target in a year 1 year and 1 5 years To really lock in your motivation for cohorts we will now look at a cartoon evaluation of diabetes care Imagine you are working in a clinic in the year 2010 Over the course of a year 2010 2011 10 people declare themselves to be diabetics 1 1 2010 1 1 2011 Patient 1 1 0 Figure 1 Time of Onset of Diabetes in 10 patients in 2010 Note implicit in Figure one is that each of these patients has an identifier a medical record number and a date of onset of diabetes This date of onset of diabetes is the index date
26. close the GUI editor for the Definition Canvas z e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deffype EVENT amp SelMenuQ Name Repeat dv Type a Cs Cose A Event Definition Options A Definition i fx VI Lab Test Type Equal HEMOGLOBIN Aic 101 Update and Close the definition canvas to return to event canvas s GUI editor and populate the event type 4 Edit Selected Condition Repeat HgbAic MINOT Earliestof RepeatHgbAic LabTestDate WITHIN il TO DJ Davs After Event DiabetesHgAicGe9 5 Update Update and Close 102 Choose within 180 103 Choose To 365 104 Make sure you are pointing to the first condition line as the time anchor for the within condition 105 This results in Edit Selected Condition RepeatHgbAic NOT Earliest of RepeatHgbAic LabTestDate W WITHIN 180 TO 365 Days After Event Diabetes hgbA1CGe9 5 7 Update Update and Close 106 Left Click on Update and Close 44 Edit Selected BUILD IY Save As Q IX Ix Bad Diabetes WRepeat BJINDEX EVENT Earliest of Any And REO DisbetesHgA1c6e3 5 Earliest of IHgbA1c6e3 5 LabTestDate WHEN IN Je 02 to Je 03 WITH Age 21 to 651 L AND Lorre Repeat HgbAic Earliest of RepeatHgbAic LabTestDate within 180 to 365 Days After Event DisbetesHgAlcGe9 5 107 Now build E C INDEX EVE
27. d Present on Admission Update and Close 126 Click to save your condition definition USR CHF All 64 Update and Close 127 Your Definition Canvas should look like this Click to save your Event Definition Edit Option II Name Event Type Inpatient Admit Update Update and Close l Event Definition Options Ix vlinpatient ICD9 Group InSet USR CHF All Position Equal Primary or Secondary 128 Note that your newly defined Event Definition is selected in the Event Condition Editor with a prefix of the event type CHF Admissions Admit w CHF Mar 2012 NOT All of CHF Admissions InpatientAdmissionDate Y WHEN IN New Duration Def WITH No Demographics Update Update and Close Define Time Period Click New Duration Def and select New Duration Def 65 129 Enter a name for your duration Name Mar 2012 from Update Update and Close Close 130 Enter the date range in the from and to fields Click the inclusion check boxe to include March 1 2012 but do not check the inclusion box of 4 1 2012 This will include the last second up to the start of the day 4 1 2012 Name march2012 from EZ 12 00 AM to oja 12 00 AM oO Update and Close Update and Close 131 Note that your duration definition is now selected in your Event Condition Edit Selected Condition Admit w CHF Mar 2012 NOT Algi Define Demographic Condition 1 Click No Demographics a
28. dE General Principle Hover over highlights and shows the area is primed for activation Right Click opens a menu Left click selects Often it opens a GUI editor at the bottom of the screen GUI Graphical User interface 1 Hover mouse over collections and cohorts FE Collections and Cohorts A box appears around the phrase cohort and collection letting you know that the system is now ready for another action Ey Collections and Sel 2 Right click to expose menu FICE Collections and Cohorts l Show Legend 3 Hover over New 15 a e fy Collections and Cohorts ly Folder Save TreeView Layout Event Collection Refresh T Show Legend 4 Hover over Folder He Collections and Cohorts E Order By Cohort Save TreeView Lavout Event Collection Refresh Show Legend 5 Leftclickon Folder a AREER Oer ERX X Name New Folder bbb8 N E E Collections and Cohorts CO New Folder bbb8 gt 6 Type in Name TDiabetes Fa Edit Selected C Save As l A ir _ _ Name TDiabetes x 7 Click on Save Icon Siaa C Save As x 2 1s TDiabetes S x Note the next icons are Save and exit Name ITDiabetes Exit 8 Click on exit Icon 16 CG Event Canvas E E Collections and Cohorts D TDiabetes 72 9 Hover Over folder TDiabetes 10 A box appears around the folder TDiabetes indicat
29. ect an Dura 18 The Event Definition Palette opens in a new window 20 k Name EV Cc6e1 Event Type Please Select V End Point Event Please Select es Co cone cione Ev finition O ns 19 Name the new event HgbA1c ge 9 5 di Te https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deffype EVENT amp SelMenuOptio i dit Option j Name HgbAic ge 9 5 x vent Type Please Select v End Point Event Please Select v Close Mat mem mem Misra ruse r Event Definition Options 20 After the words Event Type left click on the please select box and choose Lab Test as the Event Type in the dropdown menu g Te https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deffype EVENT amp SelMenuOption A Edit Options o Name HgbAtege95 Event Type EEEE End Point Event FiesseSetz O y Ambulatory Surgery Admit A Close Ambulatory Surgery Discharge X Assessment Functional ADL ji Event Definition Options Assessment Functional IADL Cancer Registry First Recurrence Date Cancer Registry Initial Diagnosis Date Cancer Registry Site Spec Factors Init Dx CPT4 Procedure Date Date of Birth Death In House Death Date Death Social Security Death Date Death Social Security or In House Death Date Echo Date ED Discharge ED Seen ED Triage Findings Immunization Date
30. f Any Or 3 EIQ Inpatient MP Earliest of Inpatient_MP InpatientAdmissionDate WHEN IN Duration2000 present WITH AgeGE18 OR 4 M a Outpatient_MP Earliest of Outpatient MP ClinicVisitDate WHEN IN Duration2000 present WITH AgeGE18 Condition Line numbers 2 3 4 are what is called collectively a subgroup A subgroup is like having a parenthesis around a series of conditions We want to find people who carry the diagnosis of paraproteinemia either by Condition line 3 inpatient admission with an icd9 code of paraprotenemia Or Condition line 4 outpatient clinic visit with an ICD9 code of paraproteinemia Condition line 2 is called the subgroup index line runs condition lines 3 and 4 applies a rule to the resulting set of events to choose earliest 2 MonoclonalParaProteinemia Earliest of Any Or a o Inpatient_MP Earliest of Inpatient_MP InpatientAdmissionDate WHEN IN Duration2000_present WITH AgeGE18 OR 4 isle Outpatient_MP Earliest of Outpatient MP ClinicVisitDate WHEN IN Duration2000_present WITH AgeGE18 inpatient OR outpatient event to represent the subgroup on the event canvas You can intuit that condition line 2 controls lines 2 3 because lines 2 3 are indented and line 2 is preceded by a parenthesis warning you that line 2 owns and controls the parenthetical subgroup Further down the event canvas we see condition lines 5 6 7 We have a second subgrou
31. he entire cohort Traditionally Hedis scores perform a calendar window cross sectional analysis answering the question of the 10 diabetics identified in 2010 what percent achieved the goal of sugar control in 201077 Circle those patients who achieve success in 2010 1 1 2010 1 1 2011 1 1 2012 Patient L 8 l Diabetes Control 3 JIS index date 4 E start therapy 5 IMMA outcome IN achieve lab value 7 _ eee eee B A 9 0 10 JEE What of new diabetic patients were controlled in the year 2010 4 10 40 Figure 3 Successfully treated diabetics in 2010 crossectional view We have a 40 success rate in this cross sectional analysis Is it fair to say that all the other patients were badly managed Look at patient 9 Look how fast he was brought under control Why was he not called a success He was not called a success because his success occurred in the year 2011 and the cross sectional analysis looks for success only in calendar year 2010 While his doctor achieved success in short time the success occurred in 2011 and therefore did not count The patient presented to his doctor at the end of the year so there was no time in the same calendar year to demonstrate success What is the consequence of this A sane doctor would never accept a diabetic patient at the end of the year nor would he do blood tests This cross sectional metric creates perverse incentives Now consider how the cohort appro
32. indow and a new blank Cohort is loaded with the Event Condition Editor open I Save As a gt 2 C_J INDEX EVENT Earliest of Any And i fig Eventi All of WHEN IN DJ Eventi ENOT Alof WHEN IN New Duration Def Y WITH Please Select an Event Definition Please Select an Duration Definition Internet Protected Mode Off ev Q10 v 113 Name your new cohort Et Sere BUILD C saveas IBjD xjixi Name Inpatient Admit w CHF in Mar 2012 1 E Event Canvas EH _ H INDEX EVENT Earliest of Any And i BAU Eventi Al of WHEN IN DI 54 Enter condition line name 1 Name the event condition Admit w CHF Mar 2012 NOT All of New Event Def Y WHEN IN New Duration Def WITH No Demographics Update Update and Close Define an Event Definition 1 Click on New Event Def to expand the menu and select New Event Def Admit w CHF Mar 2012 NOT All of n WHEN IM New Dura update 114 The Event Definition Palette opens in a new window F J EVEN DC UUNIIN Fs http clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx DefType EVENT amp SelMenuOption NewEventDefinition amp EntityID ew Name EV C98b5 Event Type Please Select End Point Event Please Select Update Update and Close http clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitio
33. ing that the program is readv to accept instructions Event Canvas lfh Collections and Cohorts HO eas Sn aaa Collections and Cohorts HO TDiabetes Order By gt WAR PnaMav aNd Anai Atr 12 Hover over New The Menu expands to the right 17 b a l Event Canvas E E h Collections and Cohorts TDiabetes LALA Mi Pie ede A Abad Po fink e 13 Left click on Cohort 18 1 Event Canvas launches in a new window and a new blank Cohort is loaded with the Event Condition Editor open CG Event Canvas Management Edit Selected CI save As l Type raw JO Tr A FE Collections and Cohorts a ika A mW L R USR New Cohort db2f EHCE BINDEX EVENT Earliest of Any And CE PUSR New Cohort 4a87 BBG Eventi All of WHEN IN Training Cohorts Class 1_2 5 monoclonal gammopathy _ E BUSR BoneMarrowMonoclonalGammapathyRA L Cc Buse monoclonal gammopathy 6 CancerPredictiveAnalytics L Buse cancer 2010NotdiedPre2015LstContact l EB USR cancer 2010diedPre2015 l Buse cancer 2010 l CE USR New Cohort 5261 bD anneMearaPopulationBasedReadmission L Buse MedicaidDischarge 08 CE Busn medicaiddischarge 12 Edit Selected Condition HOE Buse outpatientvisit12 medicaid EB USR outpatientVisit1 2MedicaidFirstDischarge NOT LEE Bos ssrenvetowaresoichage WHEN IN New Duration De WITH No Demogra
34. lick CINOT LabTestVaue Ss VI gt es IE 25 40 Type in the box 9 5 and then left click on Update and close yielding a Te https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deffype EVENT amp SelMenuQ a e Event Lab Test CY NamelHabAtcge95 J Lab Test v Update and Close J Close A Event Definition Options s XK c Definition x Lab Test Type Equal HEMOGLOBIN A1c me y Lab Test Value GreaterThanOrEqual 9 5 41 Congratulations You have painted your second condition line on the definition canvas 42 Now Left click update and close 43 You are returned to the Event Canvas to the editing GUI with the Event definition populated with the event definition you just built 26 A Zie Eseni BUILD saveas G gt X X hm Name Bad Diabetes Tem F JN INDEX EVENT Earliest of Any And l oftl fi Eventi All of WHEN IN EESE NOT alt 1 WHEN IN WITH Please Select an Duration Definition 44 Now you are ready to define the New Duration Definition Left click on New Duration Definition and a menu opens New Duration Def Select from Copy from LINOT All select System Duration gt WHEN IN Ji WITH No Demographics Y Please Select an Duration Definition 45 Left click on New Duration Def and 27 Te https clg emerginghealthit com CLGNET Modules Eve
35. nBuilder aspx Internet Protected Mode Off 55 115 Name the new event CHF Admissions and select the Inpatient Admit as the Event Type in the dropdown menu l Edit Options Name CHF Admissions Event Type Please Select vs End Point Event Please Select Update Update and Close Close Please Select Ambulatory Surgery Admit Ambulatory Surgery Discharge Event Definition Options CPT4 Procedure Date Date of Birth Date of Death Echo Date ED Discharge 1 Name the Event Inpatient Admit Inpatient Discharge Internal MedAdminDate Lab Test Med Order Start Med Order Stop Microbiology CollectDate Orders With Text Outpatient Visit Define an attribute condition 1 Define the CHF condition right click on a aii and select Add Condition Name CHF Admissions Update Update and Close Event Definition Options ii 116 The Event Definition Editor opens Name CHF Admissions Event Type Inpatient Admit End Point Event Please Select Update und Close Event Definition Options MINOT AEREAS vi gt Update Update and Close 56 117 Select Inpatient ICD9 Group from the drop down menu Coj Discharge Attending DEX EVENT Earliest of Any And Discharge house staff Disposition Facili m Modules EventDefinitionBuilder EventDefinitionBuilder aspx DefType EVENT amp SelMenuOption NewEventDefinition amp EntityID ew I
36. nd select New Demographics Def Admit w CHF Mar 2012 Select from WHEN IN Mar 2012 WITH No Demographics 66 The Demographics editor opens Name your new demographic definition 132 Rightclickon Definition and select Add Contiion Edit Options M Name Age 65 or Greater Update Update and Close Event Definition Options ass sub 133 The Demographic Condition Editor opens Name Age 65 or Greater Uodats Undute and Close Event Definition Options NOT Please Select Update Definition f Update and Close 67 134 Click Please Select and select Age from the dropdown menu A Edit Options eee Name Age 65 or Greater NOT Please Select NOT Ace JA Update Definition E o afaniahe WCF TFYT f dar 136 Type 65 in the value field moe J a Update Definition Update and Close close Update d Close 137 Click 138 Verify that your demographic condition now includes Greater than 65 Click Update and Close 68 oS a Name Age 65 or Greater Update and Close Event Definition Options 45y Age GreaterThan 65 Casas Update and Close 139 Your Event Condition is complete Click Admit w CHF Mar 2012 MINOT Alof CHF Admissions InpatientAdmissionDate WHEN INI Mar 2012 WITH Age 65 or Greater Update and Close Build the Cohort 140 Congratulations You have identified your patie
37. ndex event line which chooses those events and date time to be associated with the qualified cohort member In our case the date time is taken from line 1 bone marrow patients who also satisfied condition lines 2 7 below You can see that the Index Event line is pointing to the earliest bone marrow for patients who have both Paraproteinemia and Rheumatoid Arthritis Condition Line 1 Looks for the Earliest bone marrow from the year 2000 to the present whose bone marrowed patient was 18 or greater at the time of the bone marrow The event is a procedure and the procedure type is Bone Marrow Biopsy If the Index Event line chooses a condition line that has more than one event per person as in the case when the condition line has the word all in its condition instead of earliest or latest then the Index Event line chooses not only the condition line but the event within the condition line It chooses 12 either the Earliest or Latest event so that each patient each Medical Record Number has only one date time associated with it At the end of the build in a cohort build we need each Patient each Medical record number represented only once so if there is a possibility of more than one date time the Index Event line must choose which event for that patient in the cohort Let s now go to the next three lines painted on the event canvas condition lines 2 3 4 2 4 MonoclonalParaProteinemia Earliest o
38. nt group Click o 69 Additional Work in class Tutorial 9 Death during CHF Admission using attribute or when in https www youtube com watch v 6aVHYB 7yHOc 1 Identify which patients died during the admission by building a new cohort called CHF March 2012 died a Hint Look at attributes of the inpatient discharge in the CLG Event manual vol 3 i Which atttribute of inpatient admissoin type would you want to use to select those who died during the admission disposition Make sure you have 70 looked in the manual so you will know how to find the relevant attribute in the future i Which set values would you want to use expired ii Now tell me How many people died What percent died Start by building a cohort of hospital discharges made up of chf primary March 2012 no demographic restriction N 226 Chf primary march 2012 no demographic restriction disposition expired Hint definition canvas looks like e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deflype EVI Edit Options a Name yee patient Dishare Ee int E Ce Ciose Event Definition Options El Definition X Inpatient ICD9 Group InSet USR Heart Failure Position Equal Primary only 0 ai soc Deen Ejf 0 Name ChfMarch2012WhenInDied J Fc J INDEX EVENT Earliest of Any And ice chfMarch2012 All of chf discharge InpatientDischargeDate
39. ntDefinitionBuilder EventDefinitionBuilder aspx Deffype DURATION amp SelMenuOptic lt Eee Name JQDRI C318d from O fa to O ta 5 E f Ciose 46 a duration definition Palette opens A palette is just another type of canvas 47 Type in a name Je 02 to Je 03 48 Put in the starting date next to the from 6 1 02 49 Put in the end date next to the to 6 1 03 50 Left click the box to the right of the word from which will tell the program to include that day in the interval The absence of a check in the box after to will tell the program to 28 include time up to but not including that date If time is not included the program assumes midnight 51 The resulting image is e https cIg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx2DefTvpe DI EEA l Name from 6 1 2001 x 12 00 AM to O 6 1 2003 12 00 AM 5 Event Definition Options 52 Left Click on Update and close This returns you to the event canvas and the GUI editor below E Js INDEX EVENT Earliest of Any And fig Eventi All of WHEN IN DiabetesHgbA1cGe9 5 O NOT All of HgbAic ge 9 5 LabTestDate W WHEN IN l WITH No Demographics 53 The duration is automatically filled with the duration you just built 29 54 We now will go to Demographics and restrict the cohort to patients aged 21 or older 55 Left click on arrow next
40. o have access would suggest that you try to schedule as a group 2 Buddy System Each student should have a CLG buddy who should be the first person you call for help CLG is a community of users and learning how to reach out to others and how to help others in this cooperative intellectual ecosystem is critical for community development 3 Bring Earphones like those you use with iphone or ipod to connect into computer so you can listen to tutorial without anyone else hearing it 4 Critique and recommended improvements to these study guides is appreciated Please send comments to ebellin montefiore org Contents Be TONS ne PI E125 S air E E on une vnnnunaaes aie jd 2 Chapter 1 What Questions do want answered and Why do need cohorts sse eessssenenzesnnnnzennnenza 5 Patient Centricity tie RCW EO is au 5 Pre Training Session b siicicncesncasncescvetecusreusncesiconixGnsvossecwiain Teceaneuseeaiaonsnaresnaeneacantecnsnotersasqnitedyenbusncrsacntucuenaneens 11 Build a Cohort your Mind eeraa ae telvenietiasasnceteuaitonige E 11 In Class example Class 1 Cohort Build ccccccssssccsosssccsssssescoussectcussscncussscneoussectenssscneusssensousressensnss 15 Clinical Looking Glass Session Exercise Build a cohort of Bad diabetics ss essnnerennzennnnneznnnnnzna 15 Derne ai EVEnNECOROIEO f ai grat alko pet gili ipatti ia ti 20 Define a New Event DETINITION vaciccsivesxssavsecconsavesdvssendenenudvenies
41. ohorts S L TDiabetes 73 i def JJUSR Bad Diabetes 35 74 To Green striped indicating that the cohort has both the rules and the medical record numbers and date times of the patients available for additional analytic work CG l Event Canvas eh Collections and Cohorts AO TDiabetes Sb 7 C J USR Bad Diabetes Green 76 Green signifies a built cohort with rules and medical record numbers If you made the mistake of saving the cohort you would have lost all the mrn s and only have the rule set remaining When you leave Event Canvas after a build just exit If you save or if you save and exit then all your MRN patient numbers will vanish and you will only be left with the rules that built the cohort This rules will have to be built again to recreate the cohort 77 Now that we have built a cohort let s see what we can do with it Got to the managemet pane Hover over the cohort and CiG l Event Canvas e i Collections and Cohorts OQ TDiabetes c J USR Bad Diabetes li 4 n a box surrounds the cohort letting you know that CLG considers the cohort active for interrogation 78 Right click on the activated boxed cohort A drop down menu appears 36 79 CG l Event Canvas fE Collections and Cohorts AH TDiabetes c B USR Bad Diabetes Edit 4 Delete Make Copy View Summary Browse Copy to Event Collection O
42. orts https www voutube com watch 7v rL imNiNrri 1 First inpatient Admission with Pneumonia Primary diagnosis in 2010 Males who survived the hospitalization N 762 2 First inpatient Admission with Pneumonia Primary diagnosis in 2010 Females who survived the hospitalization N 932 3 First inpatient Discharge with Pneumonia Primary diagnosis in 2010 Males who survived the hospitalization N 753 4 First inpatient Discharge with Pneumonia Primary diagnosis in 2010 Females who survived the hospitalization N 927 Read Riddles in Accountable Healthcare by Eran Bellin Chapter 5 Heads or Tail Which End is Up Choosing between two sides of a durational event Pneumonia Event 76 a e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deflype EVI Name am inpatient Admit qj End Point Please Select Type CS CES dose Event Event Definition Options El Definition jew x y Inpatient ICD9 Group InSet USR PneumoniaHope Position Equal Primary only Pneumonia Set 003 22 SALMONELLA PNEUMONIA 020 3 PRIMARY PNEUMONIC PLAGUE 020 4 SECONDARY PNEUMON PLAGUE 020 5 PNEUMONIC PLAGUE NOS 038 2 PNEUMOCOCCAL SEPTICEMIA 041 2 PNEUMOCOCCUS INFECT NOS 041 3 Klebsiella pneumoniae 052 1 VARICELLA PNEUMONITIS 055 1 POSTMEASLES PNEUMONIA 073 0 ORNITHOSIS PNEUMONIA 115 05 HISTOPLASM CAPS PNEUMON 115 15 HISTOPLASM DUB PNEUMONIA 115 95 HISTOPL
43. p of condition lines led by subgroup index condition line 5 with indented lines 6 7 below 13 5 g Rheumatoid Arthritis Earliest of Any Or 6 ans Inpatient with RA Earliest of Inpatient RA InpatientAdmissionDate WHEN IN Duration2000_present WITH AgeGE18 OR F aie Outpatient with RA Earliest of Outpatient_RA ClinicVisitDate WHEN IN Duration2000_present WITH AgeGE18 l This second subgroup finds patients with Rheumatoid Arthritis by searching through all inpatient and outpatient events for the ICD9 diagnosis of Rheumatoid Arthritis Condition line 6 inpatient admission with an icd9 code of Rheumatoid Arthritis Or Condition line 7 outpatient clinic visit with an ICD9 code of Rheumatoid Arthritis Subgroup index condition line 5 owns and controls condition lines 6 7 choosing the earliest evidence of Rheumatoid Arthritis as recorded in an inpatient admission or outpatient visit In summary the totality of condition lines find patients with a bone marrow paraproteinemia by icd9 diagnosis and rheumatoid arthritis by diagnosis More sensitive analyses would use laboratory tests to determine paraproteinemia but this example serves its purpose Our first worked example in class will be much simpler 14 In Class example Class 1 Cohort Build Clinical Looking Glass Session Exercise Build a cohort of Bad diabetics Tutorial 4 The Diabetic Cohort https www youtube com watchrv KEI9pqDA
44. phics y ia A l Ec B USR New EventCollection Please Select an Event Definition ie pl gt Please Select an Duration Definition 14 Name your new cohort TM euo Osaveas BExjxe X wW Name Bad Diabetes x 4 c W INDEX EVENT Earliest of Any And fed Eventi All of WHEN IN 15 Go to the bottom of the Event Canvas to find an opened Edit Selected Condition GUI INOT Alof V New Event Def Y WHEN IN WITH Please Select an Event Definition Editor Please Select an Duration Definition 19 Define an Event Condition 16 Name the condition line Edit Selected Condition DiabetesHgbA1cGe9 5 x NOT All of New Event Def W WHEN IN WITH Update Notice the GUI editor Edit select condition has two boxes outlined in red This tells you that the minimal data that must be entered to define a condition line on the event canvas are two a New Event Definition The event which qualifies the individual for inclusion in the condition line b New Duration Definition the time over which the event is to be sought Define a New Event Definition 17 Left Click on New Event Def to expand the menu and select New Event Def DiabetesHgbA1cGe9 5 CNOT Alor V New Event Dee W Select from b b Select from Common Definition Copy from Please Select an Even w Please Sel
45. ptions include 80 81 82 83 84 85 86 87 88 Edit modify the cohort Delete remove the cohort Make copy make a copy of the cohort that you can rename View Summary Documentation of rules you used to make the cohort Browse Ability to create a spreadsheet with the attributes of the cohort as well as demographics Share ability to share the cohort with another member of the clg community Your shared cohort will appear in your friends management pane with YourName Name of cohort instead of Usr Name of cohort Buld allows you to rebuild the cohort Copy to Event Collection Turn cohort into an Event collection For example if you did not just want the first HgbA1c but wanted all you would need to copy the cohort to and Event 37 Collection and when you then editted the event collection you would see All in the IndexEvent Line 38 Assignment 89 See videos in CLG and learn how to use each of these commands 90 If your audio card is not working in your computer then go to the 765 page manual and look for the commands and written description would advise the videos 39 Example 2 Tutorial 5 The Diabetic Cohort with repeat HgbAlc https www youtube com watch v 30a9We ZCeU Let s Find out how many of these Bad diabetics had a repeat HgbA1c within 180 days to 365 days after their first value Note do not care what the repeat HgbA1c val
46. r EventDefinitionBuilder aspx Deflype EVENT amp SelMenuOy lt 7777777777 Name HgbAte ge 9 5 side Tvpe CS dose F Event Definition Options EE Definition i jx VI Lab Test Type Equal HEMOGLOBIN Aic 33 Congratulations You have painted your first condition line on the Definition Canvas Now you must add another condition You have already required that the event be a HgbAi1c you must now require that the test have a value gt greater than or equal to 9 5 24 34 To add another condition hover over the Definition icon ff Psfinition and Right click on the definition icon This opens a menu now vou left click on Add a condition and vou see the following e Definition Canvas Webpage Dialog e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx2Defvpe EVENT8SelIMenuOp mm S Name HgbA1c ge 9 5 della Lab Test vV A Event Definition Options E Definition X y Lab Test Type Equal HEMOGLOBIN A1c L E9V undefined Condition LINOTI Please Select Please Select V ae 35 Left click on please select 36 And then hover over lab test value Please Select w LINOT GEESE En r Categoric Lab Test Value 2 Grouped Numeric Labs Lab Test Type Lab Test Value Test Clinic iMedicaidFirstDisch 38 Left click on Lab Test Value INOT Lab Test Value vii a Sere 39 Now hover over gt and left c
47. rame we are concerned with the experience of the individual patient How many unique people are hospitalized How many times was the same person readmitted While the fee for service business model does not care who makes up the 100 admissions the patient centric model makes a big distinction between Hospital A that admits 100 people one time and fixes them on the first try and hospital B that admits 1 patient 100 times never getting it right The unit of analysis in the patient centric mode is the patient as we are concerned with his longitudinal experience experience across time Question Pair 2 Business Volume Centric How many outpatient visits with the primary diagnosis of Hypertension Patient Value Centric How long does it take to control the individual s blood pressure In the volume question we just count how much service we are delivering in the outpatient for hypertensives It is a pure service volume question not a patient value question In the second question we are looking at the individual patient and asking how much time elapsed before we brought his blood pressure under control We look at the patients individually and then summarize across the whole patient group cohort how long it took for the cohort to be brought under control This summary of the individual cohort member s experience over elapsed time is called a Time to Outcome Analysis It answers the question what percent of the patients are
48. rational event is an event that has a natural duration between start and stop A hospital admission or discharge has a natural duration An ED triage representing the ED visit has a natural duration 72 Clinical Looking Glass offers you each durational event on the event canvas that you might want to point to In this case there is only one other line on the canvas so only one event is made available in the menu to point to chfMarch2012 Inpatient Stay 73 New Duration Def WHEN IN New Duration Def WITH No Demographics Update To build this second condition line a Hover over index event line b Right click choose Add When in c Choose Event of death either in house or social security master death file a Te https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deffype EVENT amp SelMenuOptio1 ea 1771717117 Event Name DeathinHouseOrSocialSecuritv Type Death Social Security or In House Death Date V MT CS ja Event Definition Options Jm Update and Close This will pick up death both reported in the hospital emr and those cases known only to social security For limitations in Social security Death tapes see Riddles in Accountable Healthcare Chapter 29 When the dead are silenced who speaks for the living d New duration b Select system duration from the event canvas i this choice forces the event of death In other studies you might
49. second CIASS sscccccccccsceeesseecccccesseeusseeeccccessauesseeesceeessauausseeceeeessauugaeeeeseesenaaas 76 Chapter 1 What Questions do want answered and Why do need cohorts Tutorial 1 Introduction to CLG self training PPTIntroduction https www youtube com watch v wPmjQmLoKSO amp list PLf7raPnmlLOeAWU2cNf2jIMDRqQu UTCum amp index 1 Tutorial 2 The cohort paradigm PPT 1 https www youtube com watch v nun8rp6d7A Patient Centricity the new goal First a definition Cohort A cohort is a group of patients each one of whom is associated with an index date time This index date time is the time from which elapsed time to outcome is to be calculated For example if you have a group of patients who are diagnosed with hypertension when you ask the outcome question how long did it take to bring their hypertension under control you measure the time from each person s index date time To understand the importance of cohorts we will compare analytic targets in fee for service world We will compare business volume centric targets with related patient value centric goals in the healthcare space Question Pair 1 Business Volume Centric How many admissions were there in 2010 Patient Value Centric How many unique individuals were admitted in 2010 In a fee for service world each time someone crosses the hospital threshold the hospital s cash register rings and money is collected In a patient centric f
50. snunesansacnuiesaunlsarvaduurieheumnsearsdusedsseuntenacsduteiehes 20 Define aia UEC E CONAI N i ceututovannrstuecoa O T 21 PSS US VSIA aaa otuancmsawnis ined ecae vd nccipeconeeunicb E quss st peswnanneteucasnevaurenncseutionronstaeeetesaibieiaaumnesdas 39 EVAN HE 2 os grec occa dete cposuarn so nnncerseoncnet auesdceuseauoccasscu satus cu csaasieecoansane E E 40 pevelopinttrtron Inaex Event LINE ieee eee l 47 Exercise Build Good and Bad Diabetic CONOrts ccccccccccccssssssseeccccccesaeeeseeecceeessaeeuseeecceeessauenseeeeeeeeesaaas 49 Build Cohort with Bad Repeat TES U i a 49 Build Cohort with Good Repeat Test iii tistinka b ke 50 Build Two more Cohorts by Changing the Focus of the Index Event Line to the first line 51 Now build a CHF example focused on admission sssessesessensesreresrrrsssrrresrreessrereserersserreserereseereserereseeress 53 Enter condition lne NAME eeren 55 Define an Event DETINITION sae E 55 Define an attribute condition cccccccssssssseeccccccseeeesseeccccecseaesseeeceecessaeussseeeeceessauuusseeeeceessagaaseeeseeeenaaas 56 PYRG a SC Ci kielet A E 57 Use Your New Set in Your Condition Definition cccccccssssssseecccceceeeeseeeccceessaaeueeecceeeseeaesseeeeeeeesaaas 64 Denne TIME a o rearen E EEA 65 Define Demographic Be p Lo l l o 66 Bulla tie CONT g i a Ea 69 Additional Work IN CIASS essre nn aE NE AEEA r AESA E the AEE ONECARE EE AEE AAE 70 Homework Before the
51. tion one of the index date event s attributes was numeric value so numeric value of the index date event is selectable through the choice of index date made at the index event line CG Event Canvas ARE tog D Output Type List v v File Type Excel 2007 7 Demographics Attributes z TEB Collections and Cohorts a Select Analysis Definition optional Please Select an Analysis C Check All Y Add Advanced TDiabetes i D Bad Diabetes With RepeatHgbA1c L MRN i USR Bad Diabetes With RepeatHgbA1c EO USR Bad Diabetes Description Patient Name P zheng Afib Afl 2011 2013 NO AC 2year after I Primary Care MD i Select Events and Attributes r l ashresth Negative Antibodies CONTROL V1 905f3 aan _ Patient Phone USARWAR PnaMay2014bevCephOnly hey EA Preferred Langua A c J USAaRwaR PnaMay2014tbwVancoOnly V beyl EventType A Race USR Afib Afl 2011 2013 has Ac 2yrbel Kev i x gt eric Z SES USR BoneMarrowBiopsyRheumatoidArthritisMonclonalP LJ rsa ramped Numeric Labs ii USR jan14 discharge _ LI Social Security Nu x T MW CC PUSR New Cohort idi FT Lah Tant Type O State aa USR New Cohort 4a87 M 7i E Bose mover zi Delta Test Value rina EB LI Patient Address EC Psoverhol Taft Campus Visits pus 0 k l Lab Text L Apartment USARWAR PnaMay2014bwCephVanc 5 anesthesi LI CensusBlock ta anesthesia O x y Test Clinic USR New EventCollection LI CensusBlockGrou
52. tionally pointing to the first condition line 48 Exercise Build Good and Bad Diabetic Cohorts Tutorial 7 Build Two Diabetic Cohorts that achieved good and bad Hemoglobin Alc respectively https www youtube com watch v UTHUK75VkXM You will need both cohorts for our next class You will need these cohorts for future work so make sure to build them Build Cohort with Bad Repeat Test Cohort BadDiabetesWRepeatBadF2 Note the naming convention for this cohort camelback is used to occupy the least number of characters while maintaining a descriptive name F2 designates where the index event line is focusing The index event line time is to be focused on the repeat test Focus on line 2 Earliest Hgbalc gt 9 5 in Je 02 Je 03 Age 21 65 with a subsequent HgbA1c measuring 2 9 within 180 365 days after Focus index date time on the subsequent HgbAic N 1010 BadDiabetesWRepeatBadF2 Criteria i First hemaglobin A1c in patients seen between Je 02 Je03 with a value Greater than or equal to 9 5 and Age between 21 65 ii a repeat HgbA1c within 180 365 days after the first with a value greater than or equal to 9 Result N 1010 iii The index event line should point to the repeat HgbA1c so we have a cohort called BadDiabetesWRepeatBadF2 who have not been brought under control by the time of their second study CT 1 E c INDEX EVENT Earliest of repeatHgbA1cGE9 And fa EFi Diabetic Earliest of
53. to Demographics BI esi selected Condition DiabetesHgbA1cGe9 5 CI NOT All of HgbAlc ge 9 5 Copy from WHEN IN e02to Je 03 _ WITH JINSDEmOGRERESTITI No Demographics Select from 56 Left click on New Demographic Def The demographic definition palette opens di e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deflype DEM Name DEMOGRAPHICS1 C16d7 5 es Ce dos Pal Event Definition Options A one Definition 57 As you did in all previous palettes and canvases type a name In this case type the name Age 21 to 65 and then hover over the definition on the palette until it is activated with a box around it Name Age 21 to 65 Event Definition Options m 58 Right click on the boxed definition thereby opening a drop down menu 30 cept Definition MW AME 59 Left click on Add condition a and a condition line is created on the AaAmaaranhinc dAafinitinn nalatta anda Gill aditar Ananc halaa e Definition Canvas Webpage Dialog e https clg emerginghealthit com CLGNET Modules EventDefinitionBuilder EventDefinitionBuilder aspx Deffype DEMOGRAPHICS amp Name Age 21 to 65 Update and Close Clos Pal Event Definition Options Ir BR Definition X Undefined Condition CI NOT Please Select v Update Definition Update and close Close Please Enter the Required
54. ue is just want to know that they had a repeat study Go to the Management Pane hover over the cohort Right click on the cohort to open a menu and left click on the edit command The Event Canvas opens up with the previously built cohort condition lines in place 40 91 A B 1 LI Save As H S a Name Bad Diabetes An Zz a lt i i i i i E c INDEX EVENT Earliest of Any And Fone ale DiabetesHgbAicGe9 5 All of HagbAic ge 9 5 LabTestDate WHEN IN Je 02 to Je 03 WITH Age 21 to 65 92 Check the save as box so when you finish your editing you will save the cohort with a new name and leave the old one intact 93 Rename to Bad Diabetes with RepeatHgbA1c Yielding a edit Selected Name Bad Diabetes WRepeat r l Event Canvas E INDEX EVENT Earliest of Any And L ER DisbetesHgAicGes 5 All of HgbAioGe9 5 LebTestDats WHEN IN Je 02 to Je 03 WITH Age 21 to 651 94 Now hover over the Index Event line and Right click opening up a menu Mr Fm c J INDEX EVENT Earliest of Any And ofsedai DiabetesHgbAicGe9 5 All of HgbA i 2 to Je 03 WITH Age 21 to 65 V 95 Left click on within to paint a new within line on the event canvas and open up a Selected event condition gui editor below 41 P Edit Selected JB Bad Diabetes WRepeat A l Event Canvas r _ _Bjinvex EVENT Earliest of Any And
55. untable Healthcare Chapter 8 Am I mv brother s keeper the longitudinal healthcare paradigm 10 Pre Training Session 1 Build a Cohort in your mind Tutorial 3 Anatomy of a Multiple Event Simple Cohort PPT2 httos www youtube com watch v GenjnskwZ6E First look at the representation of a cohort in Clinical Looking Glass on the event Canvas am purposely choosing an inefficient example to make a number of teaching points You build your rules in a Edit Selected Condition GUI Graphic User Interface rule builder to be shown in class Updating this edited selected condition paints a condition line on the event canvas Suppose you wanted to find all those patients for whom we have a bone marrow and also carried the diagnosis of Monoclonal Paraproteinemia and Rheumatoid Arthritis After using the Edit Selected Condition GUI you eventually get the following conditions painted on the event canvas BI event canvas a c B INDEX EVENT Earliest of BoneMarrowBiopsy And jeee ale BoneMarrowBiopsy Earliest of BoneMarrowBiopsy ProcedureDate WHEN IN Duration2000 present WITH AgeGE18 Figure 5 Event Canvas cohort of patients with Bone Marrow biopsy and diagnosis of Monoclonal paraproteinemia and Rheumatoid Arthritis 11 Now we will review each line numbered for convenience By Event cones afc B INDEX EVENT Earliest of BoneMarrowBiopsy And 8 1 Sai Bone

The Cohort Paradigm V30 - Clinical Looking Glass from Montefiore

Contents

Download Pdf Manuals

Related Search

Related Contents