uzluga.ru
добавить свой файл
Vermont Explor Data Coordinators’ Meeting

March 14, 2005 – Vermont Technical College, Randolph

Meeting Notes


Lauri Scharf reviewed issues affecting data submissions in 2004. The submission process has been going extremely smoothly save for a few issues affecting a hospital here and there.


Ken Kuebler (HIDI) described that UB-04 claim forms can be used beginning in March 2007 and will be mandatory on May 23, 2007, the same date that National Provider ID’s will be mandatory. Originally, the reason for updating to UB-04 was to accommodate ICD-10. However, at this time ICD-10 is very much on hold due to the resistance offered by payers to upgrading their systems from ICD-9. Still, AHIMA has helpful resources for transitioning to ICD-10 (see below).


Discussion followed regarding patient status:


  • “Discharged to jail”: this should be coded 1

  • There is interest in having separate codes for “Left Against Medical Advice” and “Discontinued Care” (basically vanished without discussion). Lauri will ask NUBC about possibility.

  • Is there a code for “Discharged to community care home”? Lauri will ask NUBC.

  • Status code 65 (psychiatric hospital or distinct-part unit) is valid as of January 1, 2005.



Charles Bennett from the Vermont Department of Health talked about the need for accurate and complete race and ethnicity coding. Presently, the hospitalization rates for certain minorities are very low, based on the codes in the hospital discharge data. All other indicators say that these minorities are not healthier than the white population; therefore, the hospital data are incomplete. Hospitals responded that patients are wary of such questions and hospital employees are uncomfortable asking. Ken noted that we may ask the question but we may not demand an answer, under federal law. He also said that we are in agreement about the need for more accurate data but certain realities make their collection difficult. Several hospitals stated that the public needs to be better educated about the need for the data in order to be more comfortable responding to the questions.


Ken then talked about correcting errors in the discharge data. People had several issues:


  • When making corrections in HIDInet (web site), it’s not always possible to mark an error as “OK.” Ken will look into allowing that for more situations.

  • HIDInet also needs to allow editing of revenue data; HIDI will provide a new screen to make these corrections by revenue line item.

  • ZIP codes from foreign countries cause errors even if they are valid and correct. There are no standard edits for foreign codes; hospitals are asked to fill geocodes with X’s. We agreed to add logic on HIDI’s system to ignore the ZIP when CTC (county town code) is 1800 and ZIP is outside US and Canada. If CTC is 1700, ZIP must be a valid Canadian.

  • When an error involves a ZIP or CTC, HIDInet will display the other.

  • Two hospitals have organ donors, which should be a distinct payer code. We will investigate how to implement this.

  • Soon series patients’ claims will be unbundled, meaning we will likely get more records than before. This is fine; we do not collect all diagnostic and therapeutic outpatient visits, so our numbers likely will not go up.

  • Hospitals requested the ability to identify changed/corrected records. HIDI will implement some method to identify records on the screen that have been changed.


Over lunch Lauri demonstrated a new web-based reporting tool. This is primarily intended to serve hospital planners in that it reports market share for specific service areas in a 4-year or 4-quarter trend. Data coordinators expressed interest. As a new venture, we have not yet decided what to charge for it, but a hospital’s subscription will cover all employees.


Dian Kahn and Pat Worcester then explained how the Health Care Administration and the Vermont Dept. of Health designed new hospital service areas for the Monograph Series and Community Needs Assessments and how they serve public health needs rather than hospital planning needs.


Ken reviewed a set of outpatient reports that HIDI produces for Missouri hospitals that we may be interested in obtaining for VT hospitals. The reports cover four categories of patients: ED visits, surgeries, observation, and clinical services (MRI, etc.). They break them down further by age category, gender, county and zip of residence, and hospital. About a dozen reports slice and dice the data in different ways. HIDI also provides the summary data in tab-delimited files.


The services are split on revenue codes. Vermont hospitals submit revenue codes, but in the past we have not obtained them from New Hampshire (the only state from which we get outpatient data). Greg mentioned this may change. Even if not, it would be possible to design a report based on general patient type categories.


Ken then showed us a tool on the HIDI web site for downloading a filtered subset of discharge data. It is not currently available to anyone but it is functional; Ken has not determined a pricing structure. Hospital planners at the meeting did not express interest in it. Lauri has provided some hospitals with spreadsheets of data, filtered and organized, along with a pivot table for easy analysis. They prefer this to downloading raw data. HIDI’s tool would supplement other data services, not replace them. If use of HIDI’s tool is rare, a pay-as-you-go model may be best (rather than annual subscription).


Lauri then provided a quick overview of how Excel pivot tables work. They are an excellent tool for summarizing data in various ways.


Related Resources:

------------------


ICD-10


CMS Overview of ICD-10-PCS

This is a nice summary (but still 237 slides!) of all sections covered by the coding system.

http://www.cms.hhs.gov/providers/pufdownload/icd10pcs.pdf


CMS Data Files for Download

http://www.cms.hhs.gov/paymentsystems/icd9/icd10.asp


AHIMA Position

AHIMA supports conversion to ICD-10 and has various information available.

http://www.ahima.org/icd10/index.cfm


Of particular value is AHIMA's 3-year preparation checklist at bottom of this page:

http://www.ahima.org/icd10/prepare.cfm


UB-04


Proposed UB-04 in Comparison to UB-92

http://www.nubc.org/public/whatsnew/UB04%20DRAFT%2008_31_04.pdf


Patient Status:

This is not on the agenda but is a potentially helpful clarification from NUBC http://www.nubc.org/public/whatsnew/Patient%20Discharge%20Status%20Codes

_02_23_05.pdf


NUBC 3/9/2005 Conference Call Report:


The ANSI ASC X12N Claims Workgroup (TG2 WG2) is in the later stages of developing the 5010 (October 2003 view) versions of the Institutional, Professional, Dental, and Reporting 837 implementation guides. The public comment period for this version of each of these guides ends on March 26th at 2:00pm eastern time. As part of this development process TG2 WG2 is trying to improve the situational notes for the data elements defined in all the implementation guides.


Three such elements are the Admitting Diagnosis, Patient's Reason for Visit, and DRG. They have been working closely with the NUBC to help better define the appropriate situations for reporting each of these elements. The NUBC conference call held yesterday addressed these three issues. It is important to note that the public comment period for each of the 837 guides is still open so there is still opportunity to provide feedback that could reverse any of these decisions. Bottom line if you don't like what is happening you will need to get involved now to affect a change.


Admitting Diagnosis:

The 5010 Version of the institutional and reporting 837 implementation guides have a situational note that this data element is to be reported for inpatient admissions. The front matter in each of these guides will refer to the UB manual to define what constitutes an inpatient admission. The UB manual will contain a matrix for each Type of Bill to define whether that bill type is for inpatient, outpatient, or both types of services. On yesterdays call the NUBC agreed to further limit the reporting of Admitting Diagnosis to the following Type of Bills; 11x (Hospital Inpatient), 12x (Hospital Inpatient Medicare Part B), 21x (Skilled Nursing Inpatient), and 18x (Inpatient Swing Bed).


Patient's Reason for Visit:

The 5010 Version of the institutional and reporting 837 implementation guides have a situational note that this data element is to be reported for unscheduled outpatient visits. Based on the conversation yesterday, there will be a comment posted to change the unscheduled wording to "certain outpatient visits". Again the front matter in the institutional and reporting 837 implementation guides would refer to the UB manual to define what constitutes the certain reportable outpatient visit for this data element. There was a lot of discussion on yesterdays call, but no decision on what circumstances should determine when the Patient's Reason for Visit should be reported. What was agreed is that the decision will be made by the NUBC and the appropriate X12 implementation guides will again refer the UB manual for the specifics of a reportable situation.


DRG:

It was agreed that there should be a situational note that would allow the reporting of the DRG. There had been some discussion about not including a way to report the DRG on the 837, since it is an element that is best derived from existing data elements on the 837. Since not all receivers of the data use the grouper software to calculate DRG's it was agreed that a means to report this data should remain with an appropriate situational note in the institutional and reporting 837 implementation guides. The more interesting issue brought out by this discussion was the situation where a primary payers does not need or what the DRG reported but a secondary payers does need it. In that situation the question was asked whether the primary payer would reject the claim because of the unneeded information from their perspective. Everyone agreed that scenario is problematic and needs more discussion to reach national consensus on what to do with data needed by downstream users of the data.


NOTE:

I believe the direction of having the ANSI ASC X12 implementation guides be more general with the specifics of situational rules being in some external (to X12) source is a good thing. Since the HIPAA requirement to use the claim guides is legislated, changes to those guides is a time consuming process that sometimes creates a hardship on the industry. Having the specifics of situational notes defined by data content groups outside of X12 in my view allows for a more responsive standard.


My thanks to all those that provided input on the question posed last week about the Admitting Diagnosis, the Patient's Reason for Visit, and DRG. It was great help in getting our viewpoint heard during the discussions on the NUBC call yesterday. Please let me know if any of the decisions made on that call are problematic to your data systems.


Bob Davis

=======================================================================


Charles Bennett, Epidemiological Surveillance Chief at VDH, will talk about the following items (he provided these notes). Item C, timeliness of E-codes, would likely be something that Vermont Explor handles.


A. Race & Ethnicity:

There are some reasons to consider going somewhat beyond the minimum

standards:

(1) The State is using (for birth records - soon) the expanded list of Hispanic (nationality) ethnic groups.

(2) There is anecdotal evidence that recent immigrant/refugee populations, as represented by ethnic groups from eastern Europe and various African nations, have special health problems - but that these may not be recognized in a reporting system that groups (or fails to distinguish among) the recent European immigrants in with all other whites, and the resettled populations from Africa in with all blacks.

(3) Given a chance to report more than one race on the 2000 Census,

7,335 persons chose to do so; but the NCHS "assignment" of those persons based on national NHIS data differs markedly from an assignment based on Vermont BRFSS data.

(4) The CDC is pressing hard for all states to demonstrate how well they are caring for minority groups in their population. The answers given will likely affect federal dollars coming into Vermont (or not).

(5) Unless we ask a follow-on question, we cannot well assess how we are doing with health care even for the minority races in Vermont.


Hispanic origin categories as they will appear on the new US standard birth certificate which VT plans to implement later this year:


When asked if mother/father is of Hispanic Origin, the responses are:


No, not Spanish/Hispanic/Latina

Yes, Mexican, Mexican American, Chicana

Yes, Puerto Rican

Yes, Cuban

Yes, other Spanish/Hispanic/Latina

Specify___________________________


B. With regard to E-codes:

The reporting of location (where injury occurred) is currently about 40% complete for unintentional injuries.

The reporting of perpetrator for intentional assaults upon children/youth (less than 18 years old) is, in recent years, running about 33 percent. (The same second field for e-codes might also be a useful prompt for reporting this information as appropriate. Or a third field might be used. The manner of original data collection - e.g., worksheet - may merit review from a quality assurance/uniform approach to data collection perspective.)

C. More timely reporting of injuries:

The CDC is pressing the Departments of Health nationwide to move closer to active surveillance of injuries. More timely reporting (our data sets are currently approx 14 to 26 months after events; ) e.g., quarterly reporting - even of data not fully cleaned - might move us within 4 to 6 or 7 months of events. Hospitals are currently reporting some data to the Department more or less as it occurs - for Bioterrorism surveillance. Studies have shown that some suicides are affected by "social contagion: More timely reporting of suicide attempts may also result in identification of clusters meriting public health, mental health efforts to intervene and stop such contagion.

D. Unique Personal Identifier Number (UPIN):

An ID that does not reveal identity but allows linkage of multiple admissions of the same individual - would be useful for many health surveillance issues - including improved assessment of prevalence of heart disease, diabetes, other chronic conditions, as well as possible re-admissions associated with late sequelae of prior events. Although this is the last on this list of topics, it is of considerable importance to the Vermont Department of Health. Achieving a "break through" in this regard would greatly enhance our efforts to understand conditions of chronic disease and of certain injuries - thereby helping to guide programs in response.