The USRDS and Its Products · The USRDS and Its Products USRDS 1998 Annual Data Report 2 Project Goals The USRDS now has six primary objectives: • Design and implement a consolidated

USRDS 1998 Annual Data Report The USRDS and Its Products

1

Chapter I

The USRDS and Its Products

Key Words:Data files World Wide WebDatabase Standardized mortality ratioData access

his is the tenth Annual Data Report of the UnitedStates Renal Data System (USRDS), which

began operations in 1988. The USRDS is operated bythe National Institute of Diabetes and Digestive andKidney Diseases in conjunction with the Health CareFinancing Administration. This national data systemcollects, analyzes, and distributes information aboutend-stage renal disease (ESRD) in the United States.It includes comprehensive data needed to describe theincidence and prevalence of treated ESRD, modalityof treatment (including both dialysis and kidneytransplantation), causes of death, patient survival,hospitalization, cost and cost effectiveness, andinstitutional providers of ESRD treatment.

What’s New?Regular readers of the ADRs will find this chapter

very familiar but should look for the following newfeatures:

• The full USRDS Annual Data Report is nowavailable on a CD as well as on the World WideWeb (http://www.med.umich.edu/usrds).

• This 1998 ADR generally includes data throughcalendar 1996, although some data for 1996 fromthe patient database must be treated aspreliminary. USRDS Special Study data andother limited data are presented for 1997.

About this ADRThe USRDS ADR is designed to serve a number

of functions:

• To provide an introduction to ESRD, the USRDS,and ESRD data for persons new to the subject.

• To provide a reference for all members of theESRD community.

• To report trends over time and to call attention tochanges indicated by the addition of the mostrecent year of USRDS data.

• To report on findings from USRDS SpecialStudies.

• To report on special topics, which vary from yearto year.

Most analyses presented in the USRDS ADR aredescriptive in nature and involve relatively simplestatistical techniques. Studies which rigorously testhypotheses and which use sophisticated statisticaltechniques are submitted to peer-reviewedpublications.

Because the ADR functions as an introduction anda reference, much of it remains very similar from yearto year. For each chapter, we have tried to identifythe most important facts about the topic and topresent those facts clearly in the graphics and text.Our rule of thumb is that half or more of the graphicsin most chapters will be unchanged from year to year,except for updating to reflect the most recent year ofdata.

Usually, one or two chapters of the ADR willreport new material. In this and the prior two ADRs,the DMMS and Cost Effectiveness chapters haveserved this function.

T

The USRDS and Its Products USRDS 1998 Annual Data Report

2

Project Goals The USRDS now has six primary objectives:

• Design and implement a consolidated renaldisease data system that will provide thebiostatistical, data management and analyticalexpertise necessary to characterize the total renalpatient population, and to describe the distributionof patients by sociodemographic variables acrosstreatment modalities.

• Report on the incidence, prevalence, mortalityrates and trends over time of renal disease byprimary diagnosis, treatment modality and othersociodemographic variables.

• Develop and analyze aggregate data on the effectof various modalities of treatment by disease andpatient group categories. These data will be usedto analyze the prevention and progression of renaldisease with special emphasis on morbidity andmortality.

• Identify problems and opportunities for morefocused special studies of renal research issuescurrently not addressed by the consolidated datasystem.

• Conduct cost effectiveness and other economicstudies pertaining to biomedical andepidemiological aspects of ESRD.

• Support investigator-initiated research by makingdata from the database widely available inconvenient formats to the biomedical andeconomic research community.

Organizational Structure The USRDS is funded and directed by the

National Institute of Diabetes and Digestive andKidney Diseases (NIDDK) of the National Institutesof Health (NIH). The Health Care FinancingAdministration (HCFA) of the United StatesDepartment of Health and Human Services is a majorcontributor to the project, providing expertise andmost of the primary data in the system. The HCFAalso funds the cost-effectiveness and economicstudies performed by the USRDS.

The USRDS is operated by a Coordinating Center(CC) at the University of Michigan in Ann Arbor.Figure I-1 shows the organization of the USRDS.Table I-1 lists USRDS contact persons.

NIDDK's Division of Kidney, Urologic andHematologic Diseases (DKUHD) directs the USRDS

Figure I-1

The organizational structure of the United States Renal Data System.

USRDS 1998I-1

United States Renal Data System (USRDS)

NIDDK

DKUHD

Project OfficerExecutiveCommittee

Data RequestReview Committee Biomedical Scientific

Advisory Committee(BSAC)

CoordinatingCenter

RenalCommunity

Council

Special StudySubcommittees

HCFA

Economic ScientificAdvisory Committee

(ESAC)

Special Studies Review& Implementation

Committee


3

project, ensuring that the scientific and technicalgoals of the USRDS are consistent with the missionand responsibilities of the NIDDK and the NIH. TheNIDDK Project Officer has responsibility formonitoring the Coordinating Center's technicalprogress in meeting the six primary objectives.

A HCFA Project Coordinator monitors cost-effectiveness and economic studies performed by theCoordinating Center, and assists with other issuesrelated to the USRDS project. While the NIDDKmust approve all cost-effectiveness and economictopics, a HCFA Staff Coordinator directly supervisesall such projects.

USRDS Committees NIDDK makes all final decisions regarding the

studies pursued by the USRDS as implemented by theCC. Seven major committees have assisted in thisprocess by providing advisory input to NIDDKand/or the CC:

The Scientific Advisory Committees (SACs)draw on the expertise of researchers and practitionersin the fields of ESRD research, pediatric nephrology,quality of life, transplantation, hemodialysis,peritoneal dialysis, biostatistics, epidemiology and

Table I-1

CONTACT ADDRESS Phone / e-mailNIDDK Project Officer Lawrence Y.C. Agodoa, M.D. Phone (301) 594-7717

United States Renal Data System Fax (301) 480-3510NIDDK [email protected] Building – 6AS-13B45 Center Drive – MSC 6600Bethesda, Maryland 20892-6600

NIDDK Deputy Project Officer Camille A. Jones, M.D., M.P.H. Phone (301) 594-7717United States Renal Data System Fax (301) 480-3510NIDDK [email protected] Building – 6AS-13K45 Center Drive – MSC 6600Bethesda, Maryland 20892-6620

USRDS Coordinating Center (CC) USRDS Phone (734) 998-6611315 W. Huron St., Suite 240 Fax (734) 998-6620Ann Arbor, MI 48103

CC Director Philip J. Held, Ph.D. thru Feb. 8, 1998Robert A. Wolfe, Ph.D. from Feb. 9, 1998 Phone (734) 998-6611

[email protected]

CC Deputy Director Friedrich K. Port, M.D., M.S. [email protected]

CC Data File Contact Randall L. Webb [email protected]

Standardized Mortality/ Robert A. Wolfe, Ph.D. [email protected] Rate Methodology

CC Data Request Contact Dora Smith [email protected] Orders

Internet World Wide Web http://www.med.umich.edu/usrds/

List of USRDS Contacts

USRDS 1998


4

health economics. The Biomedical SAC (B-SAC)has the role of providing epidemiological, clinical andbiostatistical advice to the NIDDK and the CC, whilethe Economic SAC (E-SAC) provides economicadvice to the NIDDK, HCFA, and the CC. Bothrecommend priorities for research by the USRDS andreview and make recommendations to NIDDK onproposed special studies and on USRDS studyresults.

The USRDS Executive Committee (EC) iscomprised of the NIDDK Project Officers, the HCFAProject Coordinator, a staff member from HCFA, theChair of the Biomedical Scientific AdvisoryCommittee, and the Director and Co-Director of theCoordinating Center. The EC advises the NIDDKProject Officers on the overall data management andresearch plan to ensure cooperative participationamong all components of the project and to identifyand address any other major issues related to theproject.

The Renal Community Council (RCC) iscomprised of more than 30 professional, scientific,and advocacy groups with an interest in ESRD andthe USRDS. Until 1996, the USRDS made apresentation to the RCC at the annual meeting of theAmerican Society of Nephrology (ASN). Starting in

1996, the USRDS made an hour-long presentation toa special session of ASN. This presentation isexpected to be an annual event and will take the placeof a report to the RCC.

The Data Request Review Committee (DRRC)advises the Project Officer. It includesrepresentatives from NIDDK, HCFA, and the CC.When they are needed, two additional reviewers areselected from the SAC. The DRRC reviews requeststo the USRDS for release of data files to supplementany research files produced by HCFA or theCoordinating Center for use by the researchcommunity. This helps to ensure that USRDS dataare made available to investigators in the pursuit oflegitimate biomedical and economic research.

The Special Studies Review andImplementation Committee (SSRIC) is chaired byHCFA and is comprised of staff from the NIH,HCFA, the CC, and two representatives of the ESRDNetworks. The SSRIC focuses on the design,implementation and progress of USRDS SpecialStudies.

Figure I-2

Overall structure of the USRDS database with counts of patient and facility records as of January, 1998.

USRDS 1998I-2

Overall Structure of the USRDS Database

PatientCharacteristics

(821,000 Patients)

PatientCharacteristics

(821,000 Patients)

DialysisClaims

(17.7 million)

DialysisClaims

(17.7 million)

Master ESRDProvider Records

(3,738)

Master ESRDProvider Records

(3,738)

Annual FacilitySurveys(32,800)

Annual FacilitySurveys(32,800)

TransplantReports

(159,000)

TransplantReports

(159,000)

TransplantFollowup Reports

(633,000)

TransplantFollowup Reports

(633,000)Updated as of

January, 1998Updated as of

January, 1998

USRDSSpecial Studies

(35,000)

USRDSSpecial Studies

(35,000)Treatment

History(2.6 million)

TreatmentHistory

(2.6 million)

Residence(1.2 million records)

Residence(1.2 million records)

HospitalClaims

(4.9 million)

HospitalClaims

(4.9 million)Oupatient

Claims(16.9 million)

OupatientClaims

(16.9 million)

OtherInsitutional

Claims(7.9 million)

OtherInsitutional

Claims(7.9 million)

Physician/SupplierClaims

(246 million)

Physician/SupplierClaims

(246 million)


5

The USRDS Database The Health Care Financing Administration

(HCFA) provides most of the existing data in theUSRDS database. In addition to all the data from itsESRD Program Management and MedicalInformation System (PMMIS) and the AnnualFacility Survey, HCFA shares data on transplantfollowup and Medicare Parts A and B servicesderived from Medicare claims. These HCFA-supplied data are the core of the USRDS database, assummarized in Figure I-2.

In addition, HCFA helps the USRDS with SpecialStudies. Most of the new, primary data for SpecialStudies are collected through the 18 ESRD Networks,which are funded by HCFA under separate contracts.Data from the Special Studies are fully integratedinto the USRDS database. Data collection began inMarch 1995 for an important new USRDS SpecialStudy, the Dialysis Morbidity and Mortality Study.Data not otherwise contained in the USRDS databasewere collected from a national sample of nearly24,000 patients drawn from all dialysis units.

Since July 1990, selected data on non-Medicarepatients treated by U.S. Department of VeteransAffairs (DVA) facilities have also been incorporatedinto the USRDS database. In July 1994, HCFA andthe Health Resources Services Administration(HRSA) consolidated transplant data into a singlecollection the United Network for Organ Sharing(UNOS) under its contract with HRSA. The expandedtransplant data are shared among HRSA, HCFA, andNIH and thus are available to the USRDS. TheHRSA-collected transplant data include non-Medicare as well as Medicare patients.

Data in the USRDS database collected by HCFA'sESRD Networks, federal insurance carriers and fiscalintermediaries are supplemented by data from theSocial Security System, the U.S. Bureau of theCensus, local and national ESRD provider databases,and international ESRD registries.

The USRDS database is updated every year. Thelast update was in the Winter of 1997-98, using datacollected through August 1997. Because of delays inprocessing data through the Medicare system, theUSRDS has waited 15 months before reportingpatient-specific data for a given time period. Thus,tables in the 1996 ADR for example, generallyreported data through December 1993. Because ofimprovements in the flow of data to the USRDS, this15-month rule is relaxed in the 1997 and 1998 ADRs.

Data generally are reported here through 1996,although the 1996 data should be treated aspreliminary. Data from the HCFA Annual FacilitySurveys are current through 1996. Some USRDSSpecial Studies data are reported through early 1997.

The discussion below of the USRDS StandardAnalysis Files provides a good overview of theUSRDS database. The Researcher’s Guide to theUSRDS Database provides more detaileddocumentation of the database (see Table I-2 ).

USRDS Products Table I-2 shows the various "products" of the

USRDS through which the USRDS disseminates theresults of ESRD research to the renal community andto the general public and provides data to supportESRD research. Except where a cost is mentionedinTable I-2, these products are free. Table I-1 listsspecific contacts for further information the USRDS.

USRDS on the Internet World WideWeb

The entire ADR, including Reference Tables andcolor slides, the Researcher’s Guide, and otherUSRDS publications are available electronically onthe Internet using the World Wide Web. Table I-3summarizes the material available at this site. Theaddress of the site is:

http://www.med.umich.edu/usrds/

The USRDS WWW site is accessed more than1000 times each month, and approximately 600million bytes of data are downloaded from the siteeach month. The site is accessed from all around theU.S. and from many other countries. Figure I-3shows monthly usage of the site.

The full 1997 USRDS Annual Data Report nowavailable on the World Wide Web, will be replacedby this 1998 ADR.

SMR/SHR/STR Methodology The USRDS has developed a methodology for

calculating annual mortality statistics for ESRDpatients. Rates of deaths per 1,000 patient years atrisk are published every year in the Annual DataReport and are grouped according to gender, age,race, primary cause of ESRD, and modality oftreatment. Chapters V, IX, and XIII provide a


6

Table I-2

Annual Data Reports The principal vehicle for dissemination of USRDS data. Available in print and microfiche from the National Technical Information Service (NTIS), United States Department of Commerce, Springfield, Virginia, 22161, (703) 487-4650.. Text portion of the rep

Annual Data Report CD The CD gives you direct access on your PC to the ADR text, Reference Tables, color slides, The Researcher’s Guide, and USRDS 1996 Unit-Specific Reports for Dialysis Patients – A summary.

Annual Data Report Slides 100+ color slides of all graphics in the 1998 ADR with printed Speaker’s Guide. Cost is about $70.

Internet World Wide Web The Reference Tables, text, and color slides of this ADR and other USRDS publications are available electronically on the Internet using World Wide Web at http://www.med.umich.edu/usrds/

“2-Hour” Data Requests The USRDS CC staff responded to more than 700 requests for data and for general information about the USRDS in 1997. Limited resources are available for simple computer tabulations.

SMR/SHR Methodology Chapters V, IX, and XIII describe the USRDS Standardized Mortality Rate (SMR) and Standardized Hospitalization Rate (SHR) methodologies. These methodologies can be used to compare local outcomes with national norms for quality improvement purposes.

SMR/SHR Spreadsheets The USRDS produces a spreadsheet that calculates Standard Mortality Ratios (SMR) for dialysis facilities or other aggregates with up to 500 patients. It allows each dialysis facility to compare their dialysis patients’ mortality to national rates based o

Dialysis Unit- Specific SMR/SHR Reports

The USRDS produces annually dialysis unit-specific mortality, transplantation and hospitalization reports, which have been distributed to over 2,400 dialysis units through the ESRD Networks. These reports are not available except through the individual d

Researcher's Guide to the USRDS Database

This Guide is the basic reference for researchers who use USRDS data files. It provides a detailed description of the USRDS database and of the USRDS Standard Analysis Files.

Standard Analysis Files These data files provide patient-specific data from the USRDS database to support ESRD research, at an affordable price. User must sign a data release agreement with NIH. More information is provided in this chapter and in the Researcher's Guide.

Custom Data Files For research needs not met by the Standard Analysis Files. Researcher pays costs of production and must sign a data release agreement.

Papers, abstracts, and publications

Most USRDS research studies result in published papers or presentations at professional meetings. A list of publications and presentations is in Appendix A.

To request any of these products, contact the USRDS Coordinating Center at (734) 998-6611 or by e-mail at [email protected], or see the contact list in Table I-1. Products are providedwithout charge except as mentioned in the descriptions above.

USRDS ProductsFor ESRD Researchers and for the General Renal Community

USRDS 1998


7

description of this methodology. This methodologyhas now been extended to include standardized ratesfor hospitalization (SHR) and transplantation (STR)as well as mortality.

The availability of published mortality tablesallows dialysis and transplant units and ESRDNetworks to compare their mortality rates with thenational rates published by the USRDS. The USRDSmethodology includes computation of a Standard

Table I-3

Figure I-3

Monthly usage of the USRDS World Wide Web site from June 1996 through February 1998.

U S R D S 1 9 9 8I -3

0

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0

1 2 0 0

1 4 0 0

1 6 0 0

J-96

J-96

A-9

6

S-9

6

O-9

6

N-9

6

J-97

F-97

A-9

7

M-9

7

J-97

J-97

A-9

7

S-9

7

O-9

7

D-9

7

J-98

F-98

0

2 0 0

4 0 0

6 0 0

8 0 0

1 0 0 0

1 2 0 0

U s a g e o f th e U S R D S W o rld W i d e W e bJ u n e 1 9 9 6 - F e b r u a r y 1 9 9 8 *

M o n th

W W W S e s s i o n s

T o ta l B y t e s

A D R B y t e s

* F u ll A D R B e c a m e A v a ila b le S e p tem b e r 1 9 9 6

R e f e r e n c e T a b l e B y t e s

O t h e r B y tes

Est

imat

ed M

onth

ly W

WW

Ses

sion

s

Mill

ions

of B

ytes

Dow

nloa

ded

per

Mon

th

W ith a World Wide Web browser like Netscape, Internet Explorer, or America OnLine,you can view a variety of material from the USRDS and can easily download it for use onyour PC. Contents include the follow ing:

Annual Data Report in W W W browser format, as M icrosoft Word documents,and as M icrosoft PowerPoint slides

ADR Color Slides graphics from the ADR as M icrosoft PowerPointcolor slides can be downloaded for your presentation

ADR Reference Tables text files can be imported into spreadsheet programs

Researcher’s Guide documentation of data files available to researchers andprocedures for accessing the files

E-mail to the USRDS send e-mail to USRDS contact persons

Links to related W W W sites opens the door to a network of sites with material related toESRD

Address and Contents of the USRDS Site on the World Wide Web

http://w w w .med.umich.edu/usrds/

USRDS 1998


8

Mortality Ratio (SMR), which is a comparativemeasure of mortality, adjusted for age, gender, race,primary diagnosis, and treatment modality. SMRs areused to standardize observed mortality in specificpatient subgroups relative to the national death rates.

The USRDS is developing an SMR methodologythat adjusts for other meaningful comorbid conditionssuch as hypertension, a previous history of cardiacdisease, and peripheral vascular disease.

Dialysis Unit-Specific SMR/SHRReports

Early in 1996 and again in January 1997, theUSRDS produced over 2,300 unit-specific reportscontaining information about the dialysis patientstreated in each facility. These reports weredistributed to dialysis facilities through the 18 ESRDNetworks. In January 1997, each facility received an8-page report containing information on:

• patient characteristics

• mortality rates

• hospitalization rates

• causes of hospitalization

• causes of death

• rates of vascular access procedures in inpatientand outpatient settings

• data completeness and validation

The highlight of these reports was a series ofSMRs, SHRs, and STRs for each facility. The SMR,SHR, and STR were calculated for each facility foreach year from 1993-95 and for the combined 3-yearperiod. Each report also allowed facilities tocompare the SMR, SHR, STR, and other informationfor their facility with summaries among all facilitiesin the same state, ESRD Network and the UnitedStates. The information in these reports could thus beinterpreted in the context of local and national norms.

Each facility also received a 23-page documentthat provides further detail about the methodologiesused to prepare the reports (Guide to the 1996USRDS Unit-Specific Reports for Dialysis Patients:Overview, Methodology and Interpretation). Thisguide may be found on the World Wide Web site aswell as the USRDS ADR CD. The methods used tocalculate an SMR, SHR, or STR and some recentanalyses involving unit-specific SMRs and SHRs arealso presented in Chapters V and VIII of this ADR.

Figure I-4

Data requests processed by the USRDS, 1989-1997. These are data requests filled under the “2-hour rule”which limits requests to those requiring up to 2 hours of staff or computer time to fill.

USRDS 1998I-4

Data Requests Processed 1989-1997

88 77 98152

45

251 226

365

743

0

100

200

300

400

500

600

700

800

1989 1990 1991 1992 1993 1994 1995 1996 1997


9

The reports were distributed to the facilitiesthrough the 18 ESRD Networks. The reports are notavailable except through the individual dialysis units.Each of the ESRD Networks was provided with datafiles on CD with all the statistics reported forfacilities in that Network.

For 1998, the Unit-Specific Reports are scheduledto be released in June 1998.

Data Requests The USRDS has a primary objective of making

data available to the renal community. One of theimportant means of making data available is throughtimely response to data requests made by researchers,practitioners, and other members of the renalcommunity.

As shown in Figure I-4 the number of requests hassteadily increased over the years. The year 1993 wasthe year of transition between USRDS contracts, witha temporary interruption of USRDS CoordinatingCenter activities. During 1997 an average of almostthree requests were filled per working day. There hasbeen a seasonal increase in requests around theAmerican Society of Nephrology meeting. Severalrequests were filled regarding specific analysis filesand several have led to scientific publications

(Bloembergen).

In many cases these requests can be answered byproviding data published in the Annual Data Reportor elsewhere. Requests for data not available in theAnnual Data Reports, but that would require 2 hoursor less of computer programmer/analyst time, can beprovided by the Coordinating Center, usually withinone week of the request.

Requests that require more than 2 hours ofcomputer programmer/analyst time will beundertaken only upon written approval by theNIDDK Project Officer. Research needs that cannotbe met by the ADR or by 2-hour data requestsprobably can be met by obtaining the StandardAnalysis Files or custom data files described later inthis chapter.

Because of limited resources for filling datarequests which require a computer run, such requestsmust be made in writing (by e-mail, fax, or classicmail). These requests will be completed as resourcesallow.

USRDS Analysis Files forResearchers The USRDS has been sharing research data withother researchers for many years. Figure I-5 shows

Figure I-5

Researcher requests filled for USRDS patient level data files, 1992-1997. Also shown is the number of 9-track magnetic tapes and CDs supplied. Prior to early 1996, files were supplied on 9-track magnetic tape.

USRDS 1998I-5

Researcher Requests for USRDS Patient Level Data Files,and CD’s Supplied, 1992-1997

3 2

8

22

2

25

3 212 5

79

170

0

10

20

30

40

50

1992 1993 1994 1995 1996 1997 1992 1993 1994 1995 1996 19970

40

80

120

160

200# Requests

Requests

# CDs

CDs


10

the number as well as the size of requests for datafiles for researchers outside of the USRDS that werefilled from 1992 through 1997, including pendingrequests. The figure shows the number of CD-ROMsprovided to researchers by calendar year. Note thatthe data that can be stored on one CD would requireabout 450 high density floppy disks or 4 nine-trackcomputer tapes. There has been a clear increase inboth the number of researchers and in the amount ofinformation released for research during the recentyears.

As experience regarding the most commonrequests for research files grew, the CoordinatingCenter developed a set of Standard Analysis Files(SAFs) designed to meet the needs of a wide varietyof research, including production of all the ReferenceTables in the ADRs. The SAFs were introduced in1994, and at the same time NIDDK began awarding anew group of research grants focusing on researchusing the USRDS data. The result was a sharpincrease in the number of files provided by theUSRDS. The growth continued in 1997. In addition,a complete set of files is sent to NIH and HCFA eachyear. Files also have been provided to the 18 ESRD

Networks containing the data from the USRDSSpecial Studies collected by the Networks.

The Standard Analysis Files make the USRDSdatabase available to researchers in an easy to use andwell documented format. This approach allowed amajor reduction in the production costs and thus acost saving for researchers. These analysis files havepatient-specific information, however patientidentifiers and facility identifiers are encrypted.Further discussion of the SAFs appears later in thischapter.

Prior to 1994, all files were custom files createdbased on the needs of a specific research project.Since the introduction of the SAFs, custom files aregenerally limited to cases in which the researcherprovides a file of patients to the USRDS for matchingwith the USRDS database. Typically, the researcherwants to know which patients in his or her filedeveloped ESRD or had certain outcomes withESRD. The USRDS returns the researcher’s file withspecific key ESRD data items added. In these casesall patient identifiers are removed in order to protectthe confidentiality of the patient data. Mostresearchers requesting custom files have not alsorequested the SAFs.

Table I-4

State Topic State Topic

GA Estimating production and cost models for dialysis services.

VA Cost and util ization patterns for detailed patient and setting profiles, and linking to outcomes results.

MA Case mix models for end-stage renal disease patients.

NY Evaluation of factors related to compliance with hemodialysis & their effects of patient

NC Predictors of DRG day outliers among end stage renal disease

NY Evaluation of risk factors for withdrawal from hemodialysis.

OH Determ ining risk factors for patient mortality following transplantation.

VA Vascular access morbidity by RN staff/patient ratios.

MD Relationship of various factors with risk of hospitalization in ESRD patients.

CA Use of appropriate denominator for expressing dialysis dose.

OK Effect of specific market characteristics on the costs and outcomes of subacute care episodes for ESRD patients.

MA Long-term survival in ESRD patients with ADPKD.

CA Modeling outcomes in renal-replacement therapy.

MA Case Mix Models For ESRD Patients.

MA Effect of P re-ESRD care on dialysis outcomes.

PA Impact of Preemptive Transplantation in Recipients of Kidneys from Live Donors.

Appendix C includes a cummulative l ist of independently supported research which has used USRDS fi les.State is the state of the research institution.

Recent Independently Supported Research Which Used USRDS Standard Analysis and Custom Files

USRDS 1998


11

All requests except custom requests include theCore SAF CD (84 percent). This CD includes basicpatient data, each patient’s treatment history, fulltransplant and transplant followup data, and all datafrom the USRDS special studies. Half of theresearchers using the USRDS SAFs need only thisCD. The hospitalization CD includes data abouthospital inpatient stays except for payment data items.This file is too large to be included on the Core CD.About 40 percent of researchers need this file inaddition to the Core CD. The Medicare paymentSAFs are used by only about 20 percent of theresearchers, but these files account for 79 percent ofthe CDs provided. A full set of Medicare paymentfiles requires 51 CDs, although these files may bepurchased by year. See Table I-7 for details.

CD-ROM technology has been crucial to thegrowth in the use and the usability of the USRDS

SAFs. Half of the researchers have needed only oneCD in order to carry out their research. The fullMedicare payment data would require 170 9-trackmagnetic tapes instead of 59 CDs. Providing thisvolume of data on tape would not be practical.

A wide variety of research topics have beenaddressed by researchers outside the USRDS usingthe USRDS SAFs. Table I-4 shows the researchtopics for the requests for data files which were filledfrom 1994 through 1997 or currently are pending.The Reference Tables in the ADRs also are usedextensively. As an example of the wide use of thesedata, Table I-4 shows titles of abstracts presented atthe 1997 meetings of the American Society ofNephrology which cite the USRDS data in theabstract.

Table I-5

Author AbstractBloembergen1 A description of the delivery of peritoneal dialysis in the U.S.

Factors associated with late referral among new dialysis patients in the U.S.Dancey Cardiac and peripheral vascular morbidity in diabetics: comparison of hemodialysis and peritoneal dialysis.Eknoyan Adequacy of delivered dialysis doses in the HEMO study: an interim report.Fink

Pts from same dialysis unit are not independent observations: estimation of within center correlations in the USRDS.

Held1 A comparison of patients initiating hemodialysis and peritoneal dialysis in the U.S.

Hirth1 Predictors of employment of ESRD patients.

Hulbert-Shearon1 False positives and false negatives in measurement of standardized mortality ratio.Jaar Risk factors for septicemia among diabetic and nondiabetic hemodialysis patients.

Risk of septicemia from reuse of dialyzers in hemodialysis patients with diabetes mellitus.Jaber Prescribed versus delivered dose of intermittent hemodialysis in acute renal failure: a substantial discrepancy.Kutner Physical functioning, dialysis modality, and employment outcomes among incident dialysis patients.

Leggat1 Noncompliance with hemodialysis: predictors and survival analysis.Survival in hemodialysis patients by gender and race: cause-specific mortality and the effect of withdrawal.Withdrawal from dialysis: the relationship with comorbid conditions, albumin, and dose of dialysis.

Matzke A multicenter optimization of anemia management trial: the hospital pharmacy cooperative study.Ornt Analysis of cause of death determination in the end-stage renal disease programScheinman Transplantation for primary hyperoxaluria in the U.S.A.

Wolfe1 Patient survival for wait-listed dialysis versus cadaveric renal transplant patients in the U.S.Differences in access to cadaveric renal transplantation (TX) in the U.S.: [1] rates of wait list (WL) and [2] of TX among patients on WL.

Woods1 Timing of nephrology referral predicts vascular access placement prior to ESRD.

Young1 Predictors of target hematocrit among erythropoietin-treated HD patients.

1 = a USRDS Coordinating Center staff member was first author.Appendix C includes a cummulative list of abstracts presented at American Society of Nephrology meetings citing USRDS data.

Abstracts Presented at the 1997 American Society of Nephrology Meeting Citing USRDS Data

USRDS 1998


12

Table I-6

Patient ESRD Patient 627,000 Patients

Incidence, prevalence, patient survival. Most other files will need to be linked to this file using the encrypted patient ID.

Residence For each patient, one record for each period in a different residence.

Regional analyses.

Treatment History Patient. One record for each spell a patient spends on one modality.

Modality dis tribution and treatment patterns . Treatment modality at a point in time and changes in modality over time.

Transplant Transplant. Can have multiple transplants for one patient. 124,000 Transplants.

Transplant and transplant outcome analyses.

Transplant Waiting List

One record for each patient ever on waiting list. Only data item is date first listed.

Comparison of transplanted patients to dialysis patients who are transplant candidates . Patient selection to waiting list.

Dialysis Mortality and Morbidity (DMMS) (USRDS Special Study)

5,670 patients included in Wave 1.

Comorbid conditions, adequacy of dialysis, dialysis prescription and other treatment parameters , laboratory test values, nutrition, vascular access. See Chapter 4. Data from Wave 1 of this study is now available.

Case Mix Adequacy (USRDS Special Study)

7,096 patients included in the s tudy.

Comorbid conditions, adequacy of dialysis, dialysis prescription and other treatment parameters , laboratory test values.

Case Mix Severity (USRDS Special Study)


Comorbid conditions, adequacy of dialysis, dialysis prescription and other treatment parameters , laboratory test values.

Pediatric Growth and Development (USRDS Special Study)


Growth, development, and other issues relating to pediatric ESRD Patients.

CAPD Peritonitis (USRDS Special Study)


CAPD and peritonitis.

Facility One record for each year during which each facility was in operation.

Merge with the treatment history, transplant, or annual summary SAFs for analyses involving provider characteris tics by encripted ID.

Facility Cost Reports

One record per facility per year. Costs and staffing of dialysis facilities.USRDS 1998


13

Standard Analysis Files The USRDS Standard Analysis Files (SAFs) are

designed to meet the needs of most such research atminimal cost to the researcher. In 1997, a total of170 CDs of data were provided to 25 researchers.

The use of the SAFs is governed by the USRDS“Policy on Data Release for Investigator-InitiatedResearch,” which appears near the end of thischapter. Use of the SAFs requires that the researchinvestigator’s proposal be approved and that theresearcher sign the USRDS “Agreement for Releaseof Data” (included with this chapter), agreeing toobserve the prescribed restrictions.

Most SAFs provide patient-specific data. Allpatient identifiers (name, address, Social Securitynumber, Medicare beneficiary ID, etc.) are removedfrom the files or are encrypted, but the confidentialityof the data is still a serious concern. The “Agreementfor Release of Data” therefore includes restrictions onthe use and disposition of the SAFs. The SAFsinclude an encrypted ID number to allow multipleSAFs to be merged when needed.

Core Standard Analysis File CD-ROM The USRDS SAFs are available on CD-ROM

disks that can be used on virtually any PC that has aCD-ROM reader. The Core CD contains the mostfrequently used SAFs, including those from theUSRDS Special Studies. The Core CD is needed inorder to be able to use the Transplant CD, theHospital CD or any of the CDs based on Medicareclaims data. Table I-5 lists the SAFs on this CoreStandard Analysis CD-ROM. The two central filesare the Patient file and the Treatment History file.

Because of the continued growth in the USRDSdatabase, it has been necessary to put three of thedetailed transplant files on a separate CD.

Patient The Patient file has one record per patient in the

USRDS database and gives the basic demographicand ESRD-related data about the patient.

Residence The Residence SAF provides a longitudinal

record of place of residence for each patient down tothe ZIP code level.

Treatment History The Treatment History file is also referred to as

the Modality Sequence file. For each patient, a newrecord occurs whenever the patient’s treatmentmodality or dialysis provider changes.

Medical Evidence The HCFA Form 2728, Chronic Renal Disease

Medical Evidence Form, is the source of data aboutthe primary disease causing renal failure and the startdate of chronic renal dialysis which appear on thePatients SAF. In April 1995 a new version of theform went into use, including data on comorbidconditions, employment status, lab values at start ofdialysis, and review by the Network Medical ReviewBoards. A SAF was added to make available the fulldata from the new version of the HCFA 2728.

Facility

The HCFA ESRD Annual Facility Survey is thesource of data in the Facility SAF. The survey periodis January 1 through December 31. The Facility SAFcan be linked to the Facility Cost Report files usingthe USRDS provider ID. Because of this linkage, anygeographic variables which could be used to identifyfacilities have been deleted.

Facility Cost Reports

The HCFA hospital and independent facility costreports for the years 1989-1995 are now available asStandard Analysis Files. To ensure confidentiality,all geographic variables have been deleted. The filemay be linked with the Facility SAF by using theUSRDS provider ID, however geographic analyses atless than a regional or ESRD Network level are notpossible.

TransplantWhile the detailed transplant data have been

moved to a separate CD, the Core CD still contains afile with basic data, including graft failure date, for alltransplants. The various transplant files are describedin more detail below.

Transplant Waiting ListThis file contains one record for each patient in

the USRDS database who also can be identified in theUNOS transplant waiting list file. This file is usefulfor identifying dialysis patients who were candidatesfor transplantation. This is useful for studies whichcompare transplanted patients with dialysis patients


14

who were healthy enough to be candidates for atransplant.

This file contains one variable, the date the patientfirst was listed on the waiting list. Because of thecomplexity and variability of the patterns ofmovement of patients on and off of the waiting list,we have not attempted to derive more complexindicators of transplant waiting list experience.

USRDS Special Studies The USRDS has carried out a number of Special

Studies, most of which result in SAFs. All thesestudies have been based on national random samples.Special Study topics are approved by NIDDK, withrecommendations from HCFA, the USRDS ScientificAdvisory Committees, the ESRD Networks, and theRenal Community Council (RCC). For each study,design and sampling plans were developed; sampleswere selected; and data collection forms andinstructions were drafted, tested, and finalized. Thestudies that have resulted in SAFs are describedbelow. The Dialysis Morbidity and Mortality Study(DMMS) is described in Chapter IV. The datacollection forms used for several of the SpecialStudies are in Appendix B.

Case Mix Severity Study The objectives of the USRDS Case Mix Severity

Study were to:

• Estimate the correlation of comorbid conditionsand other potential factors existing at onset ofESRD regarding their association with subsequentmortality rates and hospitalization rates, whileadjusting for age, gender, race, and primarydiagnosis.

• Evaluate possible associations of these factorswith reported causes of death.

• Assess the distribution of comorbid and otherfactors among patients utilizing differenttreatment modalities.

• Compare relative mortality rates by treatmentmodality with adjustment for selected comorbidconditions and other factors.

Data were collected on 5,255 patients incident in1986-87 at 328 dialysis units nationwide.

CAPD and Peritonitis Study The objective of the USRDS CAPD and

Peritonitis Rates Study was to compare peritonitisepisodes in CAPD patients with respect to connection

device technology and other factors. The studypopulation includes all patients newly starting CAPDin the first 6 months of 1989, up to a maximum of 14patients per dialysis unit. All units providing CAPDtraining participated in the study. The samplecontains 3,385 patients from 706 dialysis units.

Pediatric Growth and Development The objectives of the USRDS Pediatric ESRD

Growth and Development Study were to:

• Establish a baseline for assessing pediatric ESRDpatient growth and sexual maturation by modalitychoice.

• Establish a prototype for ongoing collection ofpediatric data.

All patients prevalent in 1990 who were born afterDecember 31, 1970, are included in the study. Thestudy population includes 3,067 at 548 dialysis units.

Case Mix Adequacy Study (CMAS) The objectives of the USRDS Case Mix

Adequacy Study of Dialysis were to:

• Establish the relationship between the dose ofdelivered dialysis therapy and patient mortality.

• Determine the strength of this relationship whenadjusting for comorbid conditions.

• Assess how this relationship changes at differentdoses of dialysis.

• Assess how this relationship is affected by reuseof dialyzers.

• Assess the impact of different dialysis membraneson patient morbidity and mortality.

The study consists of two groups of patients: anincident sample of patients starting hemodialysis forESRD during 1990 and a prevalent sample ofhemodialysis patients with onset of ESRD prior to1990. There are 7,096 patients from 523 dialysisunits in this study. Approximately 3,300 patientshave the pre- and post- BUN values needed tocalculate delivered dose of dialysis. We havematched 94 percent of these cases to the USRDSdatabase, which will allow the data to be used formany extensive analyses. The ESRD Networks havecollected these data in conjunction with their MedicalCase Review data abstraction.


15

Dialysis Morbidity and Mortality Study(DMMS) This study is currently under way and is described

in Appendices A and B. Complete data are nowavailable from Wave 1 of the DMMS study, and areincluded on the USRDS Core SAF CD ROM (TableI-3). Wave 2 will be available early in 1998. Thereare three files from Wave 1:

• DMMSWAV1 patient data from the core,anemia, nutrition, and vascular access forms

• DMMSFACS facility data from the facilityform

• DIALYZER dialyzer characteristics fordialyzers reference on the core patient form

As described in Appendix A the DMMS is anobservational study in which demographic,comorbidity, laboratory, treatment, socioeconomic,and insurance data are collected for a large randomsample of U.S. dialysis patients, using the patients’dialysis records. The plans for this study include 4phases (“waves”) of data collection on 6,000 ESRDpatients in each of Waves 1, 3, and 4 and 4,500patients in Wave 2, for a total sample of 22,500patients over 3 years. Waves 1, 3, and 4 are eachhistorical prospective studies in which data arecollected for patients receiving in-center hemodialysison 12/31/93. In each of these “waves,” data areabstracted from the patient’s medical record withpatient status followed from 12/31/93 through theearliest of date of data abstraction, death, transplant,change in modality, or transfer to another facility.Wave 2, which began in 1996, is a true prospectivestudy of incident hemodialysis and peritoneal dialysispatients for 1996.

Another caveat concerns information in theDialyzer file. These data come from publishedsources available at the time of the study. We believethese data accurately represent the characteristics ofthe dialyzers, but they should be used with caution.

Transplant CD As mentioned earlier, the increased size of the

database has made it necessary to move the detailedtransplant files to a separate CD.

Due to changes in data collection sources over theyears, data pertaining to transplants are nowpresented in five separate SAFs. Because of spaceconsiderations, the first two files listed below areincluded on the Core CD, and the remaining three areincluded on the separate transplant CD.

• TX includes a minimum of details about alltransplants from all sources.

• TXWAIT has one record for each patient inthe USRDS database who also can beidentified on the UNOS kidney transplantwaiting list. The only variable is the date offirst listing.

• TXPMMIS includes transplant detailscollected by HCFA’s PMMIS system for theyears 1976-93.

• TXUNOS includes transplant detailscollected by the United Network for OrganSharing, currently our main source oftransplant data, for the years 1987-1996.

• TXFU includes transplant followup reports,which are completed at discharge, 6 monthsafter the transplant, on each anniversary ofthe transplant, and at graft failure. Theserecords include immunosuppressive therapy,patient status, rejection episodes, and reasonfor graft failure.

In July 1994, HCFA and the Health ResourcesServices Administration (HRSA) consolidatedtransplant data into a single collection by the UnitedNetwork for Organ Sharing (UNOS) under itscontract with HRSA. The expanded transplant dataare shared among HRSA, HCFA, and NIH and thusare available to the USRDS. This has resulted in theaddition of data on a substantial number of non-Medicare transplants starting in 1994. Children aredisproportionately represented in these non-Medicaretransplants.

The HCFA and UNOS transplant data filesoverlap for the years 1987-1993. Additionally, someMedical Evidence Forms indicate transplants whichare not included in either the HCFA or the UNOSfile. The USRDS CC has created three transplantStandard Analysis Files from these three sources.

The main transplant SAF provides a roster of allof the transplants indicated by the three sources. Inorder to resolve the conflicts among the three sources,the following procedure is used. First, all of theUNOS transplants are accepted into the file. Secondall HCFA transplants from before 1987 are accepted.Third, HCFA transplants from 1987-1993 areaccepted if there is not already a transplant for thatpatient within 30 days of that HCFA transplant. It iscommon for the transplant dates to differ by one daybetween these two sources. Finally, Transplantsindicated on the Medical Evidence Forms are


16

accepted if there is not already a transplant for thatpatient within 30 days of the Medical Evidencetransplant.

The main transplant SAF (TX) has a smallnumber of variables describing the recipient and thedonor and indicating the graft failure date computedby the USRDS. The UNOS transplant SAF(TXUNOS) contains the detailed data from theUNOS transplant file, and the HCFA transplantdetails file (TXPMMIS) contains the detailed datafrom the HCFA transplant file. A researcher whoneeds additional variables can merge the file witheither the UNOS or the HCFA details file. The tablesin Section F of the Reference Tables are producedprimarily from the main and UNOS transplant files.

Hospital CDThe hospitalization data from the USRDS

database will not fit on the same CD with the files onthe Core SAF CD, but are provided on a second CD.The hospital inpatient data on this CD are a subset ofthe data in the Institutional Claims file. No paymentor cost variables are included on this CD. This CD isfor researchers who need data on hospital inpatientstays and diagnoses and procedures for those staysbut who do not need payment data.

Medicare Payment DataSAFs containing Medicare payment data are now

available. For institutional claims, data are availablefor 1989 through 1996. For physician/supplierclaims, data are available for 1991 through 1996.The claims data CDs now can be purchased by year.

There are two types of Medicare claims:institutional and physician/supplier. All thephysician/supplier claims are Medicare Part B. Theinstitutional claims consist of all Part A claims(Inpatient, Outpatient, Skilled Nursing Facility, HomeHealth Agency, and Hospice) and some Part Bclaims, notably outpatient dialysis. Physician/supplierclaims account for about 80 percent of the claims butonly 20 percent of the dollars.

The structure and content of the two types ofclaims are different, and so are the files derived fromthem. For institutional claims, there are two types offile: the Institutional Claims (Claims) file and theInstitutional Claims Detail file. The Claims fileindicates the type of claim, the dollar amounts, thetype of dialysis involved (if any), and the dates ofservice. The Claims Detail file contains details like

DRG, diagnoses, and procedures. For many analyses,the Claims Details file would not be needed.

For the physician/supplier claims, there is onetype of file with one record for each claim line-item.The file includes dollar amounts, dates of service,diagnosis and procedure codes, and type and place ofservice.

As indicated in Table I-7, the claims data CDs arenow available by year. Starting in 1999, we will adddata for one year of claims with each annual update.We also will update the preceding four years forinstitutional claims and the preceeding one year forphysician/supplier claims. Thus early in we willproduce files for 1997 and will issue updated files for1994 through 1996 for the institutional claims and for1996 for the physician/supplier claims. We willupdate prior years to make sure that we havecomplete data for past years for patients who havenewly entered the database. We will continue toevaluate how many prior years of claims should beupdated each year.

Not that the claims files have all claims for agiven patient, including claims for dates before thepatient’s first ESRD service date. This allows at leastlimited analysis of medical care in the per-ESRDperiod.

Case Mix Adequacy CDThis CD contains the Case Mix Adequacy Special

Study file and extracts from all the other SAFs for thepatients in this study. All the Medicare paymentsdata for these patients are included. This file is usefulfor analysis using the Case Mix Adequacy data. Italso is useful as test data for developing analyses thatlater will be rerun on the full Medicare paymentsfiles.

DMMS CDThis CD is similar to the Case Mix Adequacy CD.

It contains the files from the Dialysis Mortality andMorbidity Study and extracts from all the other SAFsfor the patients in this study. All the Medicarepayments data for these patients are included. Notethat data from Wave 2 of the DMMS will be availableearly in 1998.

File Media and FormatsThe SAFs are provided on CD-ROM disks as

SAS (Statistical Analysis System) files. The CDs can


17

be used directly by SAS on any 486 or Pentium PCwith a CD-ROM reader.

In order to keep the SAFs affordable, the files areprovided only in SAS format. Researchers whorequire a different format or a medium other than CD-ROM are responsible for arranging for the conversionthemselves and should have little difficulty obtaininghelp in doing so from any university computer center.The USRDS also may be able to convert files toalternative formats or media, but the cost will besubstantially greater.

SAS format was chosen for the USRDS SAFsbecause it is widely used, easily transported, andlargely self-documenting. SAS is a commerciallyavailable data management and statistical analysissoftware system that runs on most computers, frommainframes to PCs. It is almost universally available

on university computer systems. The USRDS SAFstake full advantage of the ability of SAS data sets toincorporate a large amount of documentation into thefile. The current price structure for the USRDS SAFsis listed in Table I-7.

What You Need to Use the SAFsComputer: A 486 or Pentium PC. The USRDS

CC uses Pentium Pro 200s. Smaller runs have beendone on 486/100 PCs. The files can be converted toSAS transport format for use on any computer onwhich SAS runs.

CD-ROM drive: Any PC with a CD-ROM driveshould be able to use the SAF CDs.

Disk Storage: For the files on the basic CD,between 10 megabytes and 600 megabytes,

Current Prices of USRDS Standard Analysis File CDs

Files Description CDs PriceCore CD See Table I-6 for list of contents. This CD is needed in

order to use any of the other CDs.1 $500

Transplant CD Detailed transplant data from UNOS and HCFA, includingtransplant followups.

1 $100

Hospital CD Derived from the Institutional Claims and InstitutionalClaims Details CDs. It contains diagnosis and surgicalprocedure codes for each stay but does not include the costdata that are included in the Institutional Claims CDs.

1 $100

DMMS Claims CDSet

Contains all of the Institutional and Physician/Supplierclaims data for the patients in the USRDS Dialysis Mortalityand Morbidity (DMMS) Special Study. The data from theSpecial Study data collection forms are included on the CoreCD.

2 $200

Case Mix AdequacyClaims set

Contains all of the Institutional and Physician/Supplierclaims data for the patients in the USRDS Case MixAdequacy Special Study. The data from the Special Studydata collection forms are included on the Core CD.

1 $100

Physician/Supplier Institutional Claims Claims

Years ClaimsCDs

DetailsCDs

Price CDs Price

Before 1989 1 1 $2001989 1 1 $2001990 1 1 $2001991 1 2 $300 4 $4001992 1 2 $300 4 $4001993 1 2 $300 4 $4001994 1 2 $300 5 $5001995 1 2 $300 5 $5001996 1 3 $400 6 $600

Table I-7


18

depending on the files being used. The data on eachCD shown in Table I-6 requires from 550 to 650megabytes of disk storage. Keep in mind that youwill need space for temporary work files and for thefiles that you will create.

Software: SAS. If you convert the files to SAStransport format, they can be used by SPSS, or othersoftware that can read a SAS transport data file.

People with software experience: The SAFdocumentation provides some of the basics of loadingthe files into SAS and using them, but you needpeople with SAS experience. The USRDS CC cannotprovide technical assistance with running SAS. Ifyou plan to use other software, then people with otherappropriate software experience will be needed.

CostThe price of the files is intended to cover the

incremental cost of reproducing and shipping the fileand documentation, the administrative cost ofhandling the sales of the files, and the cost oftechnical support to researchers in selecting thecorrect files and in using the files.

The cost of the Core CD is $500. This CD isneeded in order to be able to use any of the otherCDs. Additional CDs cost $100 each. See Table I-7for more specific pricing. Checks must be payable tothe University of Michigan. These prices are subjectto change.

DocumentationThe Researcher's Guide to the USRDS Database

provides most of the documentation of the SAFs. Itincludes a codebook of variables on the files andcopies of the data collection forms used by theSpecial Studies. A chapter on techniques for usingthe SAFs in SAS has been developed.

Acknowledgment for Use ofUSRDS Data

All users of USRDS data should acknowledge thatuse. Publications that use USRDS data should includesuch acknowledgment and the following notice:

The data reported here have been supplied by theUnited States Renal Data System (USRDS). Theinterpretation and reporting of these data are theresponsibility of the author(s) and in no way shouldbe seen as an official policy or interpretation of theU.S. Government.

Policy on Data Release forInvestigator Initiated Research

Since the Standard Analysis Files and tailoreddata files contain confidential, patient-specific data,release of these files requires the approval processdescribed in this section. The investigator maycontact the USRDS Project Officer (PO) at theNational Institute of Diabetes and Digestive andKidney Diseases (NIDDK) to discuss their datarequest before preparing a written proposal. (SeeTable I-1, USRDS contact list). To request researchdata files for analysis from the USRDS:

1. The investigator will provide the USRDSProject Officer (PO) with a detailed description of theproposed investigation. A suggested outline appearsat the end of this chapter. The proposal may be theproject description from an application for a grant orfor other sources of funding. The project summarymust include goals, background data, an in-depthdescription of the study design and analyticmethodology, and resources available for completingthe project. It is necessary for the proposed project tocomply with the Privacy Act of 1974, and the projectsummary should provide enough information toenable assessment of compliance. The guidelines foradherence to the Privacy Act are contained in SectionF of the USRDS “Agreement for Release of Data”which is provided at the end of this chapter.

2. The proposal must indicate which USRDSStandard Analysis Files will be needed, and mustprovide for sufficient funding to cover the cost of thedata files, as determined from the SAF price list. Ifthe USRDS Standard Analysis Files cannot meet therequirements of the proposed research, the proposalmust specify precisely which data elements areneeded, and must budget for a substantially highercost for obtaining the files. The investigator maycontact the USRDS CC with questions about the files(Data File Contact in Table I-1).

3. The project will be reviewed by NIH fortechnical merit and for conformity with the PrivacyAct. The PO will notify the investigator(s) in writingof the approval or disapproval, discussing the reasonfor a disapproval. The PO will send a copy ofapproval letters to the USRDS CC. The process ofreviewing the written data request, generating the datafile, and releasing the data will take approximately 3months.

4. After approval, the investigator will return asigned copy of the USRDS Agreement for Release ofData to the PO. A copy of the Agreement appears at


19

the end of this chapter. The investigator and theUSRDS CC will resolve any technical questions. Theinvestigator will arrange payment with the USRDSCC, and payment must be received before the fileswill be released. Checks must be payable toUniversity of Michigan.

5. When both a copy of the signed “Agreementfor Release of Data” and payment for the files havebeen received by the USRDS CC, the CC will preparethe files and documentation and will send them, whowill release them to the investigator.

6. Any reports or articles resulting from use ofthe USRDS data must be submitted to the PO prior tosubmission for publication for review to assureadherence to the Privacy Act. The PO must respondwithin 30 days. If the report or article is determinednot to adhere to the Privacy Act, it shall not bepublished until compliance with the Act is achieved.Assessment of compliance will not depend on theopinions and conclusions expressed by theinvestigators. On the other hand, approval does notindicate endorsement of the investigator’s opinionsand conclusions by the Government.

7. All publications using the released data mustcontain the standard disclaimer, “The data reportedhere have been supplied by the United States RenalData System (USRDS). The interpretation andreporting of these data are the responsibility of theauthor(s) and in no way should be seen as an officialpolicy of or interpretation by the U.S. Government.”The investigator is requested to send copies of allfinal publications resulting from this research to boththe PO and the USRDS CC.

Caveats1. This policy establishes conditions and

procedures for the release of data from the USRDSand is intended to ensure that data are made availableto investigator(s) in the pursuit of legitimatebiomedical, cost-effectiveness, or other economicresearch.

2. The USRDS will not release data that identifyindividual patients, providers, or facilities. Ifindividually identifiable data are needed, the requestshould be submitted directly to the Health CareFinancing Administration. However, since it mightbe possible to infer the identity of individual patients,providers, or facilities from the data in the StandardAnalysis Files, the data in these files are consideredconfidential. The USRDS Agreement for Release ofData contains a number of both general and specific

restrictions on the use of USRDS data, andinvestigators are expected to abide by theserestrictions.

3. Use of these data to identify and/or contactpatients, facilities, or providers on the files isprohibited by USRDS policy. Identifying orcontacting patients also is prohibited by the PrivacyAct of 1974.

4. The USRDS CC will provide data in any of theusual forms, such as on tape, disk, and/or hard copy.Analysis services by the USRDS CC (other than toreview the proposal and to prepare the data file ifapproved by the PO) will not be provided for thesedata requests under the USRDS contract. However,USRDS CC personnel may participate in analysesfunded by sources other than the USRDS contract.

5. Standard Analysis Files or other data files fromUSRDS Special Studies will become available 1 yearafter the data have been collected, edited, and enteredinto the database.

ReferencesBloembergen WE, Port FK, Mauger EA, Briggs JP,

Leichtman AB: Gender discrepancies in livingrelated renal transplant donors and recipients. JAm Soc Nephrol 1996, 7: 1139-1144.

U.S. Renal Data System, Researcher's Guide to theUSRDS Database, The National Institutes ofHealth, National Institute of Diabetes andDigestive and Kidney Diseases, Bethesda, MD,1996.


20

Research Proposal to the U.S. Renal Data System

Suggested Outline

I Research Topic Title and Date of Submission

II Background

III Study DesignObjectivesHypothesesAnalysis Plan

IV Data RequestedList Standard Analysis Files needed or specify custom data file

V Proposed by

For Principal Investigator and co-authors, supplyNameAffiliationAddressPhone and Fax

Please submit to:

Lawrence Y.C. Agodoa, M.D.United States Renal Data SystemNIDDKNatcher Building – 6AS-13B45 Center Drive – MSC 6600Bethesda, Maryland 20892-6600

Phone (301) 594-7717Fax (301) [email protected]


21

UNITED STATES RENAL DATA SYSTEM (USRDS)

AGREEMENT FOR RELEASE OF DATA

In this agreement, “Recipient” means ___________________________________________________________

_________________________________________________________________________________________

A. The National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), through the United StatesRenal Data System (USRDS) Coordinating Center (CC), will provide the Recipient with tapes, disks, and/or hardcopy containing data extracted from the USRDS research database.

B. The sole purpose of providing the data is the conduct of legitimate and approved biomedical, cost-effectiveness, and/or other economic research by the Recipient.

C. The Recipient shall not use the data to identify individuals on the file.

D. The Recipient shall not combine or link the data provided with any other collection or source of informationthat may contain information specific to individuals on the file, except where written authorization has been obtainedthrough the approval process.

E. The Recipient shall not use the data for purposes that are not related to biomedical research, cost-effectiveness, or other economic research. The purposes for which the data may not be used include, but are notlimited to:

• identification and targeting of under- or over-served health service markets primarily forcommercial benefit;

• obtaining information about providers or facilities for commercial benefit;

• insurance purposes such as redlining areas deemed to offer bad health insurance risks; and

• adverse selection (e.g., identifying patients with high risk diagnoses).

Any use of the data for research not in the original proposal must be approved by the PO.

F. The Recipient shall not publish or otherwise disclose the data in the file to any person or organizationunless the data have been aggregated (that is, combined into groupings of data such that the data are no longerspecific to any individuals within each grouping) and no cells (aggregates of data) contain information on fewer than10 individuals or fewer than 5 providers or facilities. The Recipient shall not publish or otherwise disclose data thatidentify individual providers or facilities, or from which such identities could be inferred. However, the Recipientmay release data to a contractor for purposes of data processing or storage if (1) the Recipient specified in theresearch plan submitted to the USRDS Project Officer (PO) that data would be released to the particular contractor,or the Recipient has obtained written authorization from the PO to release the data to such contractor and (2) thecontractor has signed a data release agreement with the PO.

G. A copy of any aggregation of data intended for publication shall be submitted to the PO for review forcompliance with the confidentiality provisions of this agreement prior to submission for publication and, if notapproved, shall not be published until compliance is achieved. The PO must respond within 30 days.

H. Appropriate administrative, technical, procedural, and physical safeguards shall be established by theRecipient to protect the confidentiality of the data and to prevent unauthorized access to it. The safeguards shallprovide a level of security outlined in OMB Circular No. A-130, Appendix III — Security of Federal Automated


22

Information System, which sets forth guidelines for security plans for automated information systems in Federalagencies.

I. No copies or derivatives shall be made of the data in this file except as necessary for the purpose authorizedin this agreement. The Recipient shall keep an accurate written accounting of all such copies and derivative filesmade, which will furnished upon request to the PO. At the completion of the activities in the research plan, the fileshall be returned to the USRDS CC at the Recipient's expense, and any derivative files and copies shall be destroyed.

J. Authorized representatives of the PO and/or of HCFA will, upon request, be granted access to premiseswhere data in this file are kept for the purpose of inspecting security procedures and arrangements.

Revised June 1994

Signatures:

___________________________________________________________________________(Recipient typed name, title, and organization)

___________________________________________________________________________(Recipient telephone number)

___________________________________________________________________________(Recipient signature) (Date)

___________________________________________________________________________(Contractor typed name, title, and organization)(As appropriate)

___________________________________________________________________________(Contractor telephone number)

___________________________________________________________________________(Contractor signature) (Date)

Lawrence Y. C. Agodoa, M.D., NIDDK, NIH orCamille A. Jones, M.D., NIDDK, NIH(USRDS Project Officer typed name and organization)

___________________________________________________________________________(USRDS Project Officer Signature)(Date)

Documents

The USRDS and Its Products · The USRDS and Its Products USRDS 1998 Annual Data Report 2 Project Goals The USRDS now has six primary objectives: • Design and implement a consolidated