(Meeting details being handled separately.) I spoke with Shawn on this. We agree that getting some standard code behind each item in the ontology seems important. And where that is not possible we’ve probably modeled observations/value poorly and should change the ontology.
In the case of smoking status, we’ve discovered that the ontology tree is based on a set of SNOMED codes, so each of these can be backed by a SNOMED code. So I think the design is fine here, we should just add the SNOMED codes. Also I’d like to keep a reference to the (not-quite standard) i2b2-ized PCORI codes in a separate column, for people implementing i2b2 on top of the PCORI CDM (yes this does happen) - maybe in an i_pcori_basecode column… or add a pcori_prefix column and realize this can be built from that and pcori_code. Thanks, Jeff Jeffrey Klann, PhD Instructor of Medicine, Harvard Medical School Instructor in Investigation, Massachusetts General Hospital On 5/2/16, 2:06 PM, "Dan Connolly" <dconno...@kumc.edu> wrote: >Jeff, in our gpc-dev call last Tuesday, you took the ball to make sure a >meeting gets set up. What news? > >-- >Dan > >________________________________________ >From: Klann, Jeffrey G. [jeff.kl...@mgh.harvard.edu] >Sent: Monday, April 25, 2016 2:34 PM >To: Dan Connolly; gpc-dev@listserv.kumc.edu; campb...@unmc.edu >Cc: Murphy, Shawn N. >Subject: Re: #455: SCILHS CDM compliant with ONC standard coding? > >Cool. Some initial reactions: > >This intuitively feels very useful, but what¹s the intended use case? In >SCILHS, whatever c_basecodes we provide in the default ontology are mostly >overwritten by local sites with their local terminologies, especially in >demographics and vitals. I don¹t have a full set of the ontologies on >hand, but I don¹t think they generally use LOINC internally. So is the >goal: a) to encourage sites to transform to LOINC? B) to have a set of >LOINCs for the (eventual) day that sites use these codes internally, to >make mapping easier? Or C) to have a reference mapping to LOINC? Or >something else? If c, I think the LOINC codes should go into a separate >column. If B, I worry we need to make sure the list is exhaustive - we¹ve >had a problem with labs where sites assume that since they use LOINC too >they can just do a 1:1 mapping - but we don¹t provide all the possible >LOINC codes, so this fails. > >Also there are some entries that have LOINCs but are not queryable. >birth_date/birth_time are hidden in the SCILHS version because they are >only accessible indirectly. There is a new ³age calculated² with a LOINC >code but it¹s not useful for querying. Not sure what value the LOINCs have >in this case. > >What time is the call tomorrow? We have SCILHS meetings all day but I >might be able to cut out for 15 min depending on the time. > >Thanks for the work! >Jeff > >Jeffrey Klann, PhD >Instructor of Medicine, Harvard Medical School >Instructor in Investigation, Massachusetts General Hospital > > > > > >On 4/25/16, 11:58 AM, "Dan Connolly" <dconno...@kumc.edu> wrote: > >>Everybody please take a look at this update; bonus points for reactions >>in writing before our call tomorrow. >> >>Shawn, Jeff, you are of course welcome to join us (in writing and/or on >>the call). Let me know if there's somebody else you'd like us to invite. >>If we don't manage to sync with you SCHILS folks tomorrow, we'll look for >>other opportunities. >> >>-- >>Dan >> >>________________________________________ >>From: GPC Informatics [d...@madmode.com] >>Sent: Sunday, April 24, 2016 9:33 AM >>To: campb...@unmc.edu; Dan Connolly >>Cc: Nathan Graham >>Subject: Re: [gpc-informatics] #455: SCILHS CDM compliant with ONC >>standard coding? >> >>#455: SCILHS CDM compliant with ONC standard coding? >>--------------------------+---------------------------- >> Reporter: campbell | Owner: campbell >> Type: design-issue | Status: new >> Priority: minor | Milestone: data-domains3 >>Component: data-stds | Resolution: >> Keywords: | Blocked By: >> Blocking: | >>--------------------------+---------------------------- >> >>Comment (by campbell): >> >> I have uploaded an initial revision of SCILHS.DEMOGRAPHIC metadata that >>I >> have started editing to include ONC standard coding. I would hope to >> discuss this on our Tuesday call for UNMC compliance with GPC network >> queries using i2b2 formalisms. >> >> This is in followup of discussion with Shawn Murphy at the Epic spring >> meeting regarding publishing standards-based version of SCILHS that >>would >> allow sites employing national standards to use and avoid mapping >> concept_cd to Harvard's local scheme. >> Jim >> >>-- >>Ticket URL: >><http://informatics.gpcnetwork.org/trac/Project/ticket/455#comment:4> >>gpc-informatics <http://informatics.gpcnetwork.org/> >>Greater Plains Network - Informatics > > > >The information in this e-mail is intended only for the person to whom it >is >addressed. If you believe this e-mail was sent to you in error and the >e-mail >contains patient information, please contact the Partners Compliance >HelpLine at >http://www.partners.org/complianceline . If the e-mail was sent to you in >error >but does not contain patient information, please contact the sender and >properly >dispose of the e-mail. > _______________________________________________ Gpc-dev mailing list Gpc-dev@listserv.kumc.edu http://listserv.kumc.edu/mailman/listinfo/gpc-dev
