Hi All! Here are the numbers for CMH for part A and part B. Does each site need to fill out part C?
· For "active patients", used the formula: "2 visits in the 3 years where the visits are separated by greater than 30 days". CMH Site Name Site Name Site Name Site Name Site Name Site Name Number of active patients with data in EMR 227,120 Number of patients covered by insurance plan 219,959 Number of patients for which both EMR and insurance claims data are available 219,959 Count of distinct patients with at least one encounter or record in the system 1,444,273 Part B: Insurance Coverage Overview · No distinction in our system whether insurance is "group" coverage or "individual". · Used "active patients" to get percentages · Medicaid insurance includes out of state Medicaid coverage as well CMH Site Name Site Name Site Name Site Name Site Name Site Name Percent group UNKNOWN Percent individual UNKNOWN Percent commercial 43.51% Percent Medicaid 47.45% Percent Medicare 0.08% Percent other public payer 2.46% Percent self-pay or uninsured 6.17% Percent with unknown payer 0% Thanks, Rita Fothergill | Software Architect | Phone: 816.701.4881<tel:816.701.4881> | Children's Mercy Hospitals and Clinics<http://www.childrensmercy.org/> From: [email protected] [mailto:[email protected]] On Behalf Of Russ Waitman Sent: Wednesday, January 07, 2015 4:54 PM To: 'Thomas F Mish'; [email protected] Subject: RE: Due Monday, January 12th: Population characterization (parts A and B) for PCORNet LOI Required Table to Accompany LOI for PCORnet Clinical Data Research Networks (CDRNs) This table (Parts A - C) must be completed for each CDRN applicant. For Parts A and B, provide the requested information by each site participating in the CDRN. Insert additional columns for participating site as needed. Part A: Population Overview Interpretation notes for KUMC discussion: · For "active patients", we applied the same criteria used to characterize enrollment for PCORI milestone 1.1<https://urldefense.proofpoint.com/v2/url?u=https-3A__informatics.gpcnetwork.org_trac_Project_ticket_97&d=AwMFAw&c=Zl2T6vaIOSZ-iGixmidu-Jjpn1CKtCl7U5wJPI4UCTc&r=ffcDtYvvAbHAzh4Sc5NJlakOuXA-_Nff7ehHq9G0034&m=8sgG5_tgMsrJdi4kPtoGLEy9cTUwkga5FjFsLwOGrYA&s=4yAtU1tG4mQdMhBUIDcy8VFsURh2FuYNfN_vf2F26DY&e=>: "2 visits in the 3 years where the visits are separated by greater than 30 days". · For the number of patients covered by insurance plan, we counted patients that had had an insurance payer recorded in the EMR. We don't appear to have payer information for all patients in our system. · KUMC doesn't have claims data as far as I know. · The count of patients with at least one visit encompasses all patients we pull in to our i2b2 system without restricting the types of encounter. KUMC Site Name Site Name Site Name Site Name Site Name Site Name Number of active patients with data in EMR 331,177 Number of patients covered by insurance plan 234,556 Number of patients for which both EMR and insurance claims data are available 0 Count of distinct patients with at least one encounter or record in the system 2,077,184 Part B: Insurance Coverage Overview Interpretation notes for KUMC discussion: · We don't seem to have payer information for everyone in the system. The percentages below were calculated two different ways - both querying Clarity directly. We don't know how to determine if the insurance is "group" coverage or individual. · Over all patients in the system (note that most are unknown) · Using the same method that we used for the original proposal - calculate percentages based on patients who have payer information recorded (parenthesis) KUMC Site Name Site Name Site Name Site Name Site Name Site Name Percent group Unknown Percent individual Unknown Percent commercial 5.38% (46.61%) Percent Medicaid 0.97% (8.41%) Percent Medicare 4.95% (42.88%) Percent other public payer 0.03% (0.28%) Percent self-pay or uninsured 0.01% (0.10%) Percent with unknown payer 88.47 Part C. Data Description 1. Has the site data that will be queried been de-identified or obfuscated in any way? If yes, please describe the de-identification/obfuscation process. Yes. KUMC uses the Safe Harbor definition and removes all 18 HIPAA Identifiers. Dates for every patient are shifted back in time by a random number of days between 0 and 365. Patient IDs, encounter IDs, and Order numbers are replaced by a sequence number. For patients over 90, their age is capped at 90 and their de-identified birth date is moved forward in time to match the capped age. Identifiers such as name, address, phone number, etc. are removed. Zip codes are reduced to a sufficiently non-specific "distance from KUMC" (radius) in 5-mile increments. For IRB approved studies we can link back to fully identified data. See also approaches to securely de-identify and re-identify patient data<https://urldefense.proofpoint.com/v2/url?u=https-3A__informatics.gpcnetwork.org_trac_Project_ticket_73&d=AwMFAw&c=Zl2T6vaIOSZ-iGixmidu-Jjpn1CKtCl7U5wJPI4UCTc&r=ffcDtYvvAbHAzh4Sc5NJlakOuXA-_Nff7ehHq9G0034&m=8sgG5_tgMsrJdi4kPtoGLEy9cTUwkga5FjFsLwOGrYA&s=_GrEsCLFE2PVBsN8v1KScxvFel0NNbIqCxky_x5QxcE&e=>, De-identification Plan<https://urldefense.proofpoint.com/v2/url?u=https-3A__informatics.gpcnetwork.org_trac_Project_wiki_DataSecurity-23De-2DidentificationPlan&d=AwMFAw&c=Zl2T6vaIOSZ-iGixmidu-Jjpn1CKtCl7U5wJPI4UCTc&r=ffcDtYvvAbHAzh4Sc5NJlakOuXA-_Nff7ehHq9G0034&m=8sgG5_tgMsrJdi4kPtoGLEy9cTUwkga5FjFsLwOGrYA&s=n0xNZNMwu_K2NBsXbhPQl1nMT0ndPXbti6IAne8nCQM&e=> and KUMC De-Identification Strategy<https://informatics.kumc.edu/work/wiki/DeIdentificationStrategy>. 2. Does the source data, (i.e., raw data from your legacy data systems) consist of data from multiple sources? If yes, please describe the sources and the combination process. All patients are from University of Kansas Medical Center and affiliated clinics. EPIC EMR, KUH Cancer Registry (linked by MRN), GE IDX clinic billing system (linked by MRN), University HealthSystem Consortium (linked by MRN), Social Security Death Master File (linked by SSN/birth date), KUMC Biospecimen repository (linked on MRN). 3. Does your network impose eligibility restrictions for inclusion in PCORnet, or otherwise have members, populations, or records that are intentionally and routinely excluded? GPC data agreement specifies: "Participants will not include alcohol and drug abuse patient records, or data derived from such records, that are maintained in connection with the performance of any federally assisted alcohol and drug abuse program which are protected from disclosure by 42 C.F.R. Part 2, psychotherapy notes as defined by 45 C.F.R. § 164.501, or where otherwise protected by state or Federal law." 4. Is the patient identification number (i.e., medical record number) in the electronic health record consistent across time, or might a new identifier be generated for the same individual? If so, please explain the circumstances when a new patient identification number would be generated (e.g., change in insurance policy)? Yes for identified medical record number. The de-identified patient identification number may change every month when new data is loaded as part of the ETL process into i2b2 from the various sources. The PCORI CDM tables are built from the de-identified i2b2 tables. 5. What operating system(s) (e.g., Unix, Linux, Windows, Mac) is/are used at your site? Linux, Windows, Mac (LINUX for servers) 6. What database system (e.g., Oracle, Postgres, MySQL, etc.) is used to host and extract the PCORnet data at your site? Oracle (i2b2, Epic EMR, CDM), PostgreSQL (cross-GPC terms-only i2b2), MySQL (REDCap). 7. What SAS version and modules are available at your site? SAS v9.4 though use is mostly limited to the Statistics department. In the medical informatics department, we primarily use R. From: [email protected]<mailto:[email protected]> [mailto:[email protected]] On Behalf Of Thomas F Mish Sent: Wednesday, January 07, 2015 4:54 PM To: [email protected]<mailto:[email protected]> Subject: Re: Due Monday, January 12th: Population characterization (parts A and B) for PCORNet LOI On the GPC Global call (that's going on right now) Russ showed us KUMC's answers to Table C. Is there any chance we could get a copy of KUMC's answers, I'd like to model mine after what Russ showed on the phone call a few minutes ago. -TM On 1/7/2015 1:07 PM, Nadkarni, Prakash wrote: Maybe Debbie's solution is OK - I don't see any language in the template to the effect that the numbers must add up to 100 percent, and multiple insurance is a fact of life in the real world. A footnote to the effect that the numbers are > 100 because of multiple insurance per patient may resolve the issue. From: Nathan Graham [mailto:[email protected]] Sent: Wednesday, January 07, 2015 12:02 PM To: 'Debbie Yoshihara'; [email protected]<mailto:[email protected]>; Nadkarni, Prakash; '[email protected]<mailto:[email protected]>' ([email protected]<mailto:[email protected]>); '[email protected]<mailto:[email protected]>' ([email protected]<mailto:[email protected]>); Alex Bokov ([email protected]<mailto:[email protected]>); Fothergill, Rita, R <[email protected]><mailto:[email protected]> ([email protected]<mailto:[email protected]>); 'Jason Lebsack ([email protected]<mailto:[email protected]>)' ([email protected]<mailto:[email protected]>); Mish Thomas F ([email protected]<mailto:[email protected]>); [email protected]<mailto:[email protected]>; Justin Dale ([email protected]<mailto:[email protected]>) Cc: Steve Fennel Subject: RE: Due Monday, January 12th: Population characterization (parts A and B) for PCORNet LOI Debbie, That's a good question - I somewhat arbitrarily picked one of the groups per patient (Oracle min() function on the ID number associated with the group). Maybe that's not a good approach - I'm open to other ideas. -- Nathan From: Debbie Yoshihara [mailto:[email protected]] Sent: Wednesday, January 07, 2015 10:56 AM To: Nathan Graham; [email protected]<mailto:[email protected]>; '[email protected]<mailto:[email protected]>' ([email protected]<mailto:[email protected]>); '[email protected]<mailto:[email protected]>' ([email protected]<mailto:[email protected]>); '[email protected]<mailto:[email protected]>' ([email protected]<mailto:[email protected]>); Alex Bokov ([email protected]<mailto:[email protected]>); Fothergill, Rita, R <[email protected]<mailto:[email protected]>> ([email protected]<mailto:[email protected]>); 'Jason Lebsack ([email protected]<mailto:[email protected]>)' ([email protected]<mailto:[email protected]>); Mish Thomas F ([email protected]<mailto:[email protected]>); [email protected]<mailto:[email protected]>; Justin Dale ([email protected]<mailto:[email protected]>) Cc: Steve Fennel Subject: Re: Due Monday, January 12th: Population characterization (parts A and B) for PCORNet LOI Hi, When I do my query for Part B: Insurance Coverage Overview, there are patients (25%) which are in multiple groups: for example, Commercial and Medicare FFS. How are other groups marking this? I will get > 100% if I mark these individuals in both groups. The effective date is the same, so I have no choice but to use both. --- Debbie On 1/7/2015 8:58 AM, Nathan Graham wrote: All, As we discussed in the GPC-dev call yesterday<http://listserv.kumc.edu/pipermail/gpc-dev/attachments/20150107/1a231dff/attachment-0001.pdf>, we're working on PCORnet Phase 2<https://urldefense.proofpoint.com/v2/url?u=http-3A__www.pcori.org_announcement_national-2Dpatient-2Dcentered-2Dclinical-2Dresearch-2Dnetwork-2Dclinical-2Ddata-2Dresearch-2Dnetworks&d=AwMFAw&c=Zl2T6vaIOSZ-iGixmidu-Jjpn1CKtCl7U5wJPI4UCTc&r=ffcDtYvvAbHAzh4Sc5NJlakOuXA-_Nff7ehHq9G0034&m=8sgG5_tgMsrJdi4kPtoGLEy9cTUwkga5FjFsLwOGrYA&s=obyLzHGCVPcNuETdXgB2WjkEpaiWPpdLlNwkcyp-qyo&e=> letter of intent (LOI). Much like the quarterly updates we've provided to PCORI, the LOI document template<https://urldefense.proofpoint.com/v2/url?u=http-3A__www.pcori.org_sites_default_files_PCORI-2DPFA-2DCDRN-2DLOI-2DTemplate.doc&d=AwMFAw&c=Zl2T6vaIOSZ-iGixmidu-Jjpn1CKtCl7U5wJPI4UCTc&r=ffcDtYvvAbHAzh4Sc5NJlakOuXA-_Nff7ehHq9G0034&m=8sgG5_tgMsrJdi4kPtoGLEy9cTUwkga5FjFsLwOGrYA&s=7LXJNRp5s6HFVw2AKeS1DFlUqPaIxxRlWFhgbtPjvBo&e=> has a couple of sections with tables devoted to population characteristics: · Part A: Population Overview · Part B: Insurance Coverage Overview Brittany sent out an e-mail to the site PIs on Friday (January 02, 2015) asking that each site fill in these sections (among others) by Tuesday, January 6th. However, as per GPC-dev meeting notes for January 6, 2015<http://listserv.kumc.edu/pipermail/gpc-dev/attachments/20150107/1a231dff/attachment-0001.pdf>, Dr. Waitman has extended the requested deadline to Monday January 12th (for the tables in sections A and B at least). Please let us know right away who from each site will be the primary contact for providing numbers for the tables in parts A and B. For the PCORI quarterly updates, I believe I got population numbers from the following people: · University of Iowa: Prakash Nadkarni · Medical College of Wisconsin: Brad Taylor · Marshfield Clinic: Laurel Verhagen · San Antonio: Alex Bokov · Children's Mercy Hospital: Rita Fothergill · University of Nebraska: Jason Lebsack · University of Wisconsin - Madison: Tom Mish · University of Texas Southwestern: Phillip Reeder · University of Minnesota: Justin Dale There's been some discussion regarding what "Number of active patients with data in the EMR" means. After talking with Dr. Waitman yesterday I plan to use the same method we used to characterize enrollment for PCORI milestone 1.1<https://urldefense.proofpoint.com/v2/url?u=https-3A__informatics.gpcnetwork.org_trac_Project_ticket_97&d=AwMFAw&c=Zl2T6vaIOSZ-iGixmidu-Jjpn1CKtCl7U5wJPI4UCTc&r=ffcDtYvvAbHAzh4Sc5NJlakOuXA-_Nff7ehHq9G0034&m=8sgG5_tgMsrJdi4kPtoGLEy9cTUwkga5FjFsLwOGrYA&s=4yAtU1tG4mQdMhBUIDcy8VFsURh2FuYNfN_vf2F26DY&e=>: "The GPC will use a refined definition of enrollment based upon the criteria we used in our proposal...2 visits in the 3 years where the visits are separated by greater than 30 days" I also plan to use that number for "active patients" as the denominator for part B. Thanks for your help! Regards, Nathan _______________________________________________ Gpc-dev mailing list [email protected]<mailto:[email protected]> http://listserv.kumc.edu/mailman/listinfo/gpc-dev -- Debbie Yoshihara Informatics Specialist SMPH-IT, Biomedical Informatics Services School of Medicine and Public Health, UW-Madison [email protected]<mailto:[email protected]> Tel:(608)263-5643<Tel:%28608%29263-5643> ________________________________ Notice: This UI Health Care e-mail (including attachments) is covered by the Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is confidential and may be legally privileged. If you are not the intended recipient, you are hereby notified that any retention, dissemination, distribution, or copying of this communication is strictly prohibited. Please reply to the sender that you have received the message in error, then delete it. Thank you. ________________________________ _______________________________________________ Gpc-dev mailing list [email protected]<mailto:[email protected]> http://listserv.kumc.edu/mailman/listinfo/gpc-dev ________________________________ Electronic mail from Children's Mercy Hospitals and Clinics. This communication is intended only for the use of the addressee. It may contain information that is privileged or confidential under applicable law. If you are not the intended recipient or the agent of the recipient, you are hereby notified that any dissemination, copy or disclosure of this communication is strictly prohibited. If you have received this communication in error, please immediately forward the message to Children's Mercy Hospital's Information Security Officer via return electronic mail at [email protected] and expunge this communication without making any copies. Thank you for your cooperation.
_______________________________________________ Gpc-dev mailing list [email protected] http://listserv.kumc.edu/mailman/listinfo/gpc-dev
