SNOMED Training Workshop SNOMED CT in primary care

SNOMED Training Workshop SNOMED CT in primary care Denise Downs, Ian Spiers, Helen Beecroft

Welcome to SNOMED Training Workshop Agenda 9. 30 Registration 10. 00 Welcome and Introductions 10. 30 Training others on SNOMED CT 11. 30 Tea/Coffee 11. 50 Training Presentations 12. 30 Data Quality 13. 00 Lunch 13. 45 Data Quality Continued 14. 15 Existing Searches (Tea/Coffee at 3. 00) 15. 15 Supplier updates 16. 00 End of workshop 2

Housekeeping • Toilets • Fire • Refreshments • Introductions to NHSD team 3

Nothing will change Things will greatly improve Don’t need to know anything about SNOMED CT I need to understand SNOMED CT

Year 2000 again ? ? ? Plan for the worst Hope for the best! 5

Objectives Give you time to get more confident and ask questions Understand SNOMED and what is the key information for training practice staff Understand Data Quality and differences between Read and SNOMED Understand SNOMED Searches Update on suppliers 6

Pre-workshop Requirements Webinars/Recorded Presentations • ‘Introduction to SNOMED CT for general practice’ • ‘Exploring Content in SNOMED CT’ Laptop or Smart device for Internet Browsers Example templates and reports 7

Ice Breaker

Ice Breaker • In pairs, find five things you have in common with the other person (5 mins) • This does not include anatomy • In your pairs, what is one thing you want to come away from today with ? 9

Understanding SNOMED CT and training practice staff

Understanding SNOMED CT and training practice staff In pairs: • Workbook Sections 1 -3 (40 mins) In groups of 4: • Design Training for a practice (20 mins) – Create bullet points of what are the key things to cover – Devise a 10 min presentation Tea/Coffee Break 3 of the presentations will be chosen to present 11

Data Quality

Inactive codes SNOMED CT can deal with inaccuracies in a way Read couldn’t: • Concepts can be made inactive • Descriptions can be made inactive • Inactive concepts/descriptions should not be visible for data entry • Inactive concepts can be used in searches 13

In the browser Some Read codes are mapped to inactive SNOMED CT Descriptions Need to choose 14

Inactive concepts 15

Look-up for Read to SNOMED Browsers 16

Some current maps are to inactive codes 17

Inactive code - ambiguous Some Read codes are mapped to inactive SNOMED CT Bill/fee sent inactive in SNOMED CT It is ambiguous, do have Bill sent and Fee sent Need to choose 18

Ambiguous codes Serum bilirubin borderline is inactive in SNOMED CT It is ambiguous, do have Borderline low and Borderline high Need to choose 19

Not interoperable Terms with Clinic A / Clinic B have codes mapped to inactive SNOMED CT concepts These will map but to inactive codes; they are not interoperable, never went through GP 2 GP Different Practices will have different clinics Use specific clinic type administrative terms in future e. g. seen in asthma clinic, referral to asthma clinic 20

NOS, NEC, unclassified … Classification Terms 3519007 |Disorder of synovium (disorder)| 68962001 |Muscle pain (finding)| 26889001 |Myositis (disorder)| 68172002 |Disorder of tendon (disorder)| 10597006 |Disorder of bursa (disorder)| 21

Maps to higher level terms E 2 B. . Depressive disorder NEC 35489007 Depressive disorder Eu 32. [X]Depressive episode X 00 SO Depressive episode 35489007 Depressive disorder Eu 32 y [X]Other depressive episodes 35489007 Depressive disorder Eu 32 y [X]Depression: [oth episode][atypic][single epis masked NOS] 35489007 Depressive disorder Eu 32 z [X]Depression NOS Eu 32 z [X] Depression NOS 35489007 Depressive disorder Eu 32 z [X]Depressive disorder NOS 35489007 Depressive disorder Eu 32 z [X]Depressive episode, unspecified 35489007 Depressive disorder Eu 32 z [X](Depressn: [episode unsp][NOS (& react)][depress dis NOS] 35489007 Depressive disorder Eu 33 z [X]Monopolar depression NOS 35489007 Depressive disorder X 00 SO Depressive disorder 35489007 Depressive disorder X 00 SO Depressive illness 35489007 Depressive disorder X 00 SO Depression 35489007 Depressive disorder X 00 SO Depressed 35489007 Depressive disorder Xa. B 9 J Depression NOS 35489007 Depressive disorder XE 1 Za [X]Other depressive episodes 35489007 Depressive disorder XE 1 Zb [X]Depressive episode, unspecified 35489007 Depressive disorder XE 1 Zb [X] Depression NOS 35489007 Depressive disorder 22

Maps to different hierarchies Template Design 1 Under care of team District nurse Social context 1 Under care of team Practice nurse Social context 1 Under care of team Dietitian Social context 1 Under care of team Hospice Environment / location 1 Under care of team Under care of the elderly physician Clinical finding 1 Under care of team Under care of palliative care physician Clinical finding 1 Under care of team Under care of continence nurse Clinical finding 1 Under care of team Under care of podiatrist Clinical finding This will mean you have to write cherry picked searches in SNOMED CT, ideally all should be Under of care terms so can interoperate (we’ll come back to this) 23

Data Quality • Section 4 Workbook (30 mins) • In groups look at Data Quality Guidance • Will this effect your sample templates/reports? Review those you have brought with you. • What can you do ahead of time? • Lunch (13. 00 -13. 45) • Q&A 24

Data Quality : Questions and Answers • Any questions/observations? 25

Collaborative space - Delen • Area specifically for trainers to share information • If you don’t have access email: snomedprimarycare@nhs. net 26

Searches

Searches • Designing Searches Presentation • Convert one of your searches (Use look up tables and Browser) • Questions and Answers (1 hour) (tea and coffee at 15: 00 ) 28

Concepts and descriptions Synonyms are truly synonymous Searches are done on the concept id FSN is often the term provided in search definitions – Fully specified name – Preferred term – Acceptable synonyms • Preferred term is usually the description seen in the system 29

Language of SNOMED CT specifications Operators in SNOMED CT - Expression Constraint Language (ECL) conceptid <<conceptid ^refset. Id just this concept the descendants of this concept and all its descendants members of refset Then any of the above preceded by MINUS exclude concepts (s) OR include concepts(s) or concept(s) AND include concept(s) AND concepts(s) 30

Expressing a search We can express a search in two ways: Use ECL and the hierarchy structure : Concept and all of its descendants <<233838001 | Acute posterior myocardial infarction (disorder) | Cherry pick the codes (cluster list): List each concept seperately 233838001 | Acute posterior myocardial infarction (disorder) | 70998009 | Acute myocardial infarction of posterobasal wall (disorder) | 15713201000119105 | Acute ST segment elevation myocardial infarction of posterobasal wall (disorder) | Although these produce the same result, that may only be true for a particular release. We’ll see that using the hierarchy is a better way to write the searches when we can. 31

Walkthrough – Example 1 • Hip replacement 32

Search for hip replacement Total Revision Partial 33

Where in the hierarchy do we search: 34

Walkthrough – Example 2 • Contraception – comparison with Read • If we searched on children of Contraception 61… 35

Would get patients on contraception …

Would get patients Not on contraception

Would get patients Not actually person of record term Unknown whether on contraception or not

Have a go yourselves • Either do some you brought with you (either looking for data items for a template or a search) OR • Try the following: – All patients who have had a stroke – All obese patients 39

Moving to SNOMED CT – things to be aware of Searches will continue to work once moved to SNOMED CT UNTIL someone starts recording data that does not have a Read code equiv. Two ways to approach migrating a search from Read v 2/CTV 3 to SNOMED CT 1. Take the previous search definition and write a new SNOMED CT search from scratch • This may include SNOMED CT concepts that do not have a Read v 2/CTV 3 equivalent 2. Take the previous search codes and map these to SNOMED CT • This will need extending and reviewing in future as SNOMED CT concepts without a Read v 2/CTV 3 equivalent are used

Approach 2: Not fully automatic A previous search cannot be converted automatically as contains content not in Read, this needs a human to check Example: C 10. . % Diabetes mellitus TO << 73211009 | Diabetes mellitus (disorder) | May be ok Example: H 33. . % Asthma TO << 195967001 | Asthma (disorder) | For example include Exercise induced Asthma , which may not wish to include 41

Steps for Approach 2 • • Expand current query to all Read codes Expand list to all terms for all codes Map each to SNOMED CT Reduce to the list of unique SNOMED CT concepts • Are there any other Read codes that map to these concepts ? • How do we then extend this to be a SNOMED CT search? 42

Approach 2: example Take the previous search codes and map these to SNOMED CT Following example illustrates the approach and why some results may be different once data entry moves to SNOMED CT How can this be transformed into SNOMED CT? 43

Approach 2: term level and map Expanded to 73 Read codes Which map to 67 SNOMED CT concepts Notice two different Read codes go to the same SNOMED CT concept 44

Considering historical data • In simple terms we have performed the following: Original search • Are there any other Read codes that map to our list of SNOMED CT concepts? • We can use the Delen look-up to find out 45

Back to the example of Diabetes mellitus When checking the forward mapping tables for the 67 concepts in the converted search we find there are other Read codes not in the original specification • • • Original Read v 2 specification: 73 codes Converted SNOMED CT specification: 67 concepts Review codes that also map to these: 138 codes 46

Other mapping features • One Read v 2/CTV 3 code may go to different SNOMED CT concepts as synonyms were not always true synonyms in Read v 2/CTV 3 47

Practical • Working in pairs, take an element of one of your searches and go through Approach 2 to see any differences you get 48

Supplier Update

Supplier Update 50

EMIS Code picker • Available to download via their Support Centre • Not the final version, but close to it 51

Microtest Screens (1/3)

Microtest Screens (2/3)

Microtest Screens (3/3)

Post Coordination

What about: • Left / right • Planned / done • Family history … • Under the care of 60

Post coordination toe mild toe moderate finger severe Lung inflammation infection bacterial chronic acute exertional angina ischaemic cough phlegm viral

Different user interfaces Structured Data Entry File Edit Help FRACTURE SURGERY Reduction Fixation Open Closed Femur Tibia Fibula Ankle More. . . Humerus Radius Ulna Wrist More. . . Left Right Shaft Neck Gt Troch More. . .

Representing context in a health record Context can be expressed in a health record using: 1. Pre-coordinated expressions Family history of diabetes mellitus 2. Context-specific section or field Family History: Diabetes mellitus 3. Separate context field Problem: Diabetes mellitus Context: Family history 4. Post-coordinated expressions 281666001 |Family history of disorder|: 246090004 |Associated finding| = 246090004 |Diabetes mellitus| 63 Ref. A SNOMED International Education Slide

Currently – no post coordination • Replace one Read code with one SNOMED CT – Can get Snomed concepts with more than one ‘thought’ – A concept provided is known as pre coordination • BUT, over time, this will change and post coordination will start to be introduced, primarily for areas not well covered or where there are long list with very similar concept descriptions. . . • Likely areas: – Left /right – Family history – Any suggestions? 64

Next Steps …

Delen : SNOMED CT in Primary Care Trainers workspace This is your space Have you looked ? What would be useful? Task for next 2 weeks – post on forum one thing you find out or a question a practice asks 66

Workshop Roundup • Any questions? 67

Next steps • Data Quality regular webinar (monthly) ? ? – Scope of these ? • QOF webex – walkthrough of QOF in SNOMED CT • Anything else you need? 68

Thank you … • Helpdesk: snomedprimarycare@nhs. net • We will email you: – 69

www. digital. nhs. uk @nhsdigital enquiries@nhsdigital. nhs. uk 0300 303 5678