>From the site-by-site responses below, it appears people are interpreting the >question differently.
Might help to split this into 3 concepts: (1) how fresh is the tumor registry itself (i.e., what is the lag between diagnosis and registrar abstraction) (2) how often does the tumor registry ETL into i2b2 occur and (3) what is the date that the last ETL was conducted. Betsy -----Original Message----- From: GPC Informatics [mailto:[email protected]] Sent: Monday, September 14, 2015 9:37 AM To: Chrischilles, Elizabeth A; [email protected]; [email protected] Cc: [email protected]; [email protected]; [email protected]; [email protected]; [email protected] Subject: Re: [gpc-informatics] #348: NAACCR Data Profile: How fresh is it? (was: Data Profile over time: How fresh is it (e.g. cancer tumor registry)? how far back does the data go?) #348: NAACCR Data Profile: How fresh is it? ---------------------------+-------------------------------- Reporter: bchrischilles | Owner: bzschoche Type: design-issue | Status: closed Priority: major | Milestone: cohort-char-bc-db Component: data-sharing | Resolution: fixed Keywords: data-quality | Blocked By: Blocking: 297 | ---------------------------+-------------------------------- Changes (by dconnolly): * status: assigned => closed * reporter: dconnolly => bchrischilles * type: enhancement => design-issue * resolution: => fixed * blocking: => 297 Old description: > During the recent MU site visit, a researcher asked how far back our > data goes. Russ explained that it varies from site to site and from > system to system within sites. > > Russ answered on behalf of KUMC by way of a recent HERON release > contents > summary: > > - [https://informatics.kumc.edu/work/blog/heron-calhoun- > update#HERONCalhounContentsSummary HERON Calhoun Contents Summary] > > The code we use to generate the summary above is > - > [https://informatics.kumc.edu/work/browser/heron_load/report_stats.sql > report_stats.sql] which draws from `concepts_stats`, which is built in > - > [https://informatics.kumc.edu/work/browser/heron_load/concept_stats.sq > l > concept_stats.sql] > > Actually, the numbers from that code are only part of the story; we > manually maintain info about when the various systems went live and > when is our most recent snapshot. > > Let's integrate this sort of data profiling into our quarterly data > quality review process. For bonus points, let's actually do breakdowns > by year (or quarter or month). > > But perhaps that's a large goal that we should break down into parts. > A smaller scoped task is something that I owe the breast cancer > research > team: info about how fresh tumor registry data is at each site. New description: The breast cancer research team would like to know how fresh tumor registry data is at each site. -- Comment: Let's scope this ticket to just the NAACCR freshness issue and track the rest as #359. We collected these data 1 Sep (ticket:12#comment:78): 1. KUMC - July - monthly 1. ''CMH - N/A'' 1. UIOWA - every two weeks 1. WISC - July in production; Aug in QA; monthly 1. MCW - July 30; working on an update ... Taylor writes: Our latest refresh was with data extracted on July 30th 2015. We will be delayed on the next one, due to the migration of 3 hospitals to one reporting system. 1. MCRF - July; monthly 1. UMN - July; quarterly 1. UNMC - March; no regular update process yet 1. UTHSCSA - registrar provides data monthly; integration into i2b2 happens quarterly 1. UTSW - diagnoses thru 2014; an update takes about a week 1. MU - working on it (#344) -- Ticket URL: <http://informatics.gpcnetwork.org/trac/Project/ticket/348#comment:3> gpc-informatics <http://informatics.gpcnetwork.org/> Greater Plains Network - Informatics _______________________________________________ Gpc-dev mailing list [email protected] http://listserv.kumc.edu/mailman/listinfo/gpc-dev
