Cv) POSTGRADUATE SCHOOL Postgraduate School I Code Naval Postgraduate School Monterey, 6c ADDRESS (City, ... ICD-9-CM DRGs) for determining reimbursable amounts, this

00Cv)

NNAVAL POSTGRADUATE SCHOOL

Monterey, California

t&

~DTIC

AIELECTE KMA 0 9 198?

D

ARE U. S. NAVAL HOSPITALS OPERATEDEFFICIENTLY: A STUDY USINGDIAGNOSIS RELATED GROUPS

by

Albert Benjamin Long, III

- and

Howard Thomas Osment

December 1986

Thesis Advisor: David R. Whipple, Jr.

Approved for public release; distributon is unlitaited

UNCLASSIFIED ./,. , , " ."

REPORT DOCUMENTATION PAGEI..ePRt" U YTY SSWAT.ON : lb RESTRICTIVE MARK.INGS

2a SIURITY CLASIPiATON AUTHORTY 3 ISi.TRIOUTUON-IAVAILALITY OF RPIIORI,Approved for public release;

2t DECLASSIFICATIONIDOWNGRADING SCHEDULE distribution is ,unlimited

4 PERAORM'NG ORG3ANIZATION REPORT NUMBER(S) 5 MONITORING ORGANIZATION REPORT N MSR(S)

6& NkM OF PERFORMINORGANIZATION "6 OFFICE SYMBOL 7a NAME OF MONITORING ORGANIZATION

aval Postgraduate School I Code Naval Postgraduate School

6c ADDRESS (City, State, and ZIP CodE 7b ADDRESS (City. State, and ZIP Cod*)Monterey, California 93943-5000 Monterey, California 93943-5000

lia NAME OF FQN0IN0/SONSOA!NG 6 OFI7 CESYMBOL 9 PROCUREMENT INSTRUMENT ID0NrIFICATION NUMBERORGANIZATION (11 applcable,

Sc ADORESS (Ci ry, State,an ZIP Cod,) 10 SOURCk OF FUNDING NUMBERS

PROGRAM PROJECT TASK WORKC UNITELEMENT NO NO NO ACCESSION NO

il TITLE (InClUde, Securitj' Cialilficotton)ARE U. S. NAVAL HOSPITALS OPERATED EFFICIENTLY: ASTUDY t ING DIAGNOS 'S RELATED GROUPS,12 PERSONA, AUT OR($)

Long, Albert Benj amin III and Osment, Howard ThomasIa TYPE 09 REPORT 13b TIME COVERED DATE OF REPORT (Year, Month. Day) us PAGE COuNT

Master's Thesis FROM TO .986 December 228'6 S0UPPLEMENTARY NOTATION

7 COSATI CODES 1S SUBJECT TERMS (Corlimue on rtive if V@ctuai and identify by black number)F-ELD I oUP I"SUB-GROUP hospital cost, DRG, diagnosis related groups,

-I L hospital efficiency, naval hospital, NTF, PPS,4 ... . M R. naval treatment facilityB$-S RACTr (Continue on revete of neceuary and #dertify by block number)

In au effort to control rampant hospital-cost inflation, Congress passedthe Tax Equity and Fiscal Responsibility Act of 1982 and the SocialSecurity Amendments of 1983. The result of these two initiatives is theimplementation of a prospective payment system (PPS) that uses diagnosisrelated groups (DRGs) in classifying patients and reimbursing hospitals forMedicare patients. Using the Health Care Financing Administration's (HCFA)methods (i.e., rates, weights and ICD-9-CM DRGs). for determining reimburs--able amounts, this analysis examines the postulation that the typical U.S.naval hospital--if reimbursed for actual inpatient workload--would havereceived more than its incurred expenses. Data for three naval hospitalsover a two-year period (FY83 and FY84) are used. Findings of this analysis,suggest that on the average the typical naval hospital would have been. -

20 0 SRIOUION i AVAILABILITY OF ABSTRACT III ABSTRACT SECURITY CLASSINCATION

M,,:N-%ASSIFIOAJNLIMITED C3 SAME AS MPT 3 ODTIC USERS Unclassified22a NAME OF RESPONSIBLE INDIVIDUAL 22b TELEPHONE (InClude AreaCode) 22c OFFILE SYMBOL

Djvid R. Dioe. Jr. -408-646-2754 _ 54Wn00 FORM 1473,84 MAR 53 APR edion mey be used wimil ehuste .CURITY ,.ASSIICATtON OF Tw.I$l PAGE

All otIhof editiors vt obsol*t4

1.1

UNCLASSIFIEDSECURITY CLAISIVICATION or TIII PAG (WiM Dom Ue1"Imo4o

#19 - ABSTRACT - (Cr-.ntinued)

reimbursed 32 percent ,.-re than actual inpatient expenses hadit been reimbursed under Meuicare.

k N 0102. LF- 014.66012 UNCLASSIFIED

ICCUMITY CLAIIICATIOW OF THIS PAOWfMrhn Die. neIeefeE)

Approved for public release; distribution i; unlimited

Are U. S. Naval Hospitals OperatedEfficiently: A Study Using Diagnosis Related Groups

by

Albert Benjamin Long, IIILieutenant, Medical Service Corps,

United States NavyB.S.B.A., Appalachian State University, 1975

H.H.H.C.A., Saint Louis University, 1977

and

Howard Thomas OsmentLieutenant, Medical Service Corps, United States Navy

B.S., University of New Hampshire, 1982

Submitted in partial fulfillment of therequirements for the degree of

MASTER OF SCIENCE IN MANAGEMENT

from the

NAVAL POSTGRADUATE SCHOOLDecember 1986

Authors:Albert n am Jng

Howard Thomas OsmentApproved by:

David R. WhipleJj esis Advisor

Willis R. Greer, Jr., Second eader

Willis R. Greer, Jr., Chairm4n,Department of Admin strative ISciences

Kneale T. Marshall

Dean of Information a '-4icy Sciences

3

ABSTRACT

In an effort to control rampant hospital-cost

inflation, Congress passed the Tax Equity and Fiscal

Responsibility Act of 1982 and the Social Security

Amendments of 1983. The result of these two initiativeL

is the implementation of a prospective payment system

(PPS) that uses diagnosis related groups (DRGS) in

classifying patients and reimbursing hospitals for

Medicare patients. Using the Health Care Financing

Administration's (HCFA) methods (i.e., rates, weights and

ICD-9-CM DRGs) for determining reimbursable amounts, this

analysis examines the postulation that the typical U.S.

naval hospital--if reimbursed for actual inpatient

workload--would have received more than its Incurred

expenses. Data for three naval hospitals over a two-year

period (FY83 and FY84) are used. Findings of this

analysis suggest that on the average the typical naval

hospital would have been reimbursed 32 percent more than

actual inpatient expenses had it been reimbursed under

Medicare.

4

TABLE OF CONTENTS

I. INTRODUCTION . . . . . . . . . . . . . . . . . 13

A. GENERAL . . . . , . . * . . . . . . . . 13

B. OBJECTIVES OF THE RESEARCH . . . . . . .. 14

C. RESEARCH QUESTIONS . .... ..... . . 15

D. RESEARCH METHODOLOGY ........... 15

E. SCOPE AND LIMITATION ..... ...... 16

F. ORGANIZATION OF THE THESIS . . . . . . . 18

II. BACKGROUND . . . . . . . . . . . . . . . . . . 19

A. COST CONTAINMENT INITIATIVES AND ECONOMICANALYSIS OF HOSPITAL-BASE INFLATION .... 19

B. RETROSPECTIVE VERSUS PROSPECTIVE REIM-BURSEMENT OR EFFECTIVENESS VERSUSEFFICIENCY . . . . . . e . . . . . 25

C. CASE-MIX MEASURES, COSTS, ACCOUNTING, ANDBUDGETING . . . . . . . . . . . . . . . . . 27

D. PRODUCT DEFINITION AND MATRIXORGANIZATION . . . . . . . . . . . . . . . . 32

E. AN OVERVIEW OF DIAGNOSIS-RELATED GROUPS . . 39

F. STATISTICAL METHOD USED FOR FORMING DRGs . . 42

G. STATE PROSPECTIVE PAYMENT SYSTEMS: NEWJERSEY EXPERIENCE . . . . . . . . . . . . . 56

H. OTHER CASE-MIX MEASURES . . . . . . . . . . 63

I. CONTROVERSY OVER DRGs: LITERATURE REVIEW . 69El

13y.............13Y .... ................. .. .Di. t lb itlo'. I

A viiahiilty C odes

A.t :I .ciad rDItA/t--/

0 1. . ..... . .. . . . ...1

III. REIMBURSEMENT AND EXPENSE METHODS . . . . . . . 78

A. INTRODUCTION ............... 78

B. MEDICARE'S PROSPECTIVE PAYMENT SYSTEMUNDER PUBLIC LAW 98-21 .......... 78

1. DRG Payment Determination ...... 79

2. Transition Period ......... 80

3. Calculation of Prospective PaymentRevenue . . . . . . . . . a . * a . . . 83

C. THE DOD'S MEDICAL COST ACCOUNTING

REPORTING SYSTEM . . . . . . . . . . . . . 90

1. Background .. ........... 90

2. Medical Expense and PerformanceReporting System for Fixed MilitaryMedical and Dental TreatmentFacilities ...... ,...... 91

a. Chart of Accounts . . . . . . . . . 91

b. Manpower and Expense Assignment . . 95

c. Reporting Requirements . . . . . . 100

IV. DESCRIPTION OF THE ANALYSIS . . . . . . . . . . 101

A* DATA . . . * . .* .. . . . . . . . . . . 101

B. RESEARCH METHODOLOGY .. . . . . . . . . . 103

C. FINDINGS . . . . . . . . . . . . . . . . . 110

1. Comparison of Naval TreatmentFacilities Expense Levels to MedicareReimbursement Levels . . . . . . . . . 110

2. Comparison of Naval TreatmentFacilities to Veterans AdministrationFacilities .... . . ......... 118

3. Analysis of the Similarity ofDiagnosis Related Groups . . . . . . . 120

6

V. CONCLUSIONS AND RECOMMENDATIONS . . . . . . . . 125

A. CONCLUSIONS . . . . . . . . . . . . . . . . 125

B. RECOMMENDATIONS . . . . . . . . . . . . . 130

APPENDIX A: PROSPECTIVE REIMBURSEMENT UNDER 98-21 . 133

APPENDIX B: DECISION TREES FOR THE ICD-9-CM DRGs . . 148

APPENDIX C: DIAGNOSIS RELATED GROUPS AND SELECTEDRELATIVE WEIGHTS . ... . ....... 180

APPENDIX D: FY83 DRG WORKLOAD AND MEDICAREREIMBURSEMENT LEVELS . . . . . . . . 190

APPENDIX E: FY84 DRG WORKLOAD AND MEDICAREREIMBUR3EMENT LEVELS . . . . . .. . . . 206

APPENDIX F: CONSOLIDATED FY83 AND FY84 MEPREXPENSE AND WORKLOAD DATA . . . . . . . 222

LIST OF REFERENCES ...... . . . . . * . . 223

INITIAL DISTRIBUTION LIST . . . . . . . . . * . . . 226

7

LIST OF TABLES

I. DRG 167: APPENDECTOMY W/O COMPLICATEDPRINCIPAL DX AGE <70 W/O CC, HOSPITAL X . . . . 38

II. MDCs AND ICD-9-CM CLASSIFICATION SCHEMES . . . 47

III. KEY GROUPING VARIABLES FOR ICDA-8 DRGsAND ICD-9-CM DRGs .. . . . . .....0 0a0 51

IV. BODY SYSTEM CATEGORIES FOR DISEASE STAGING . . 66

V. CALCULATION OF PAYMENT RATE FOR DRG 125,URBAN HOSPITAL X, WEST SOUTH CENTRAL REGION:FISCAL YEARS (FYs) 1984-1986 . . . . . . . .. 85

VI. SUMMARY OF FY83/84s AGGREGATE INPATIENTMEPR/MEDICARE DATA FOR CHARLESTON, LONGBEACH, AND PENSACOLA NAVAL HOSPITALS . , . . . 116

VII. FY 1983 INPATIENT MEPR/MEDICARE DATA FORCHARLESTON, LONG BEACH, AND PENSACOLANAVAL HOSPITALS ............... 117

VIII. FY 1984 INPATIENT MEPR/MEDICARE DATA FORCHARLESTON, LONG BEACH, AND PENSACOLA NAVALHOSPITALS . . . . . . . . . . . . . . . . . . . 119

IX. SUMMARY OF FYs 83, 84, AND AGGREGATEINPATIENT MEPfl/VAFs DATA FOR CHARLESTON,LONG BEACH, AND PENSACOLA NAVAL HOSPITALS . . . 121

X. THIRTY MOST FREQUENT DIAGNOSIS RELATEDGROUPS IN THREE NTFs AND IN THE aTATE OFCALIFORNIA HOSPITALS FOR FY 1984 . . . . . . . 123

8

LIST OF FIGURES

2-1 Percent Distribution of Factors Accountingfor the Growth of Expenditures for CommunityHospital Inpatient Care, 1972-1982 . . . . ... 21

2-2 Causes of Hospital Cost Inflation . . . . . . . 24

2-3 Overview of Case Mix Cost Accounting Process . . 30

2-4 Defining the Product of Hospitals . . . . . . . 34

2-5 Clinical Matrix Management * 0 0 . 6 a a a 0 a 36

2-6 Formation of ICDA-8 MDC 52 . 0 . . * 0 0 a 6 . 46

2-7 MDC 7: Diseases and Disorders of thebepatobiliary System and Pancreas . . . . . . . 53

2-8 Criteria to Construct A PatientSeverity Index . . . . .. . . . . . . . . . 68

2-9 Fundamental Characteristics of SevenPatient Classification Schemes .. . . .. . 70

3-1 Steps to Calculate DRG Price Indexes andCase-Mix Indexes . .... . . . . . . . . 81

3-2 Steps to Calculate the Adjusted Standardizedand Hospital-Specific Amounts per Case,FY 1985 * e * . . . . . . . . . . . . . . . . .* * a 84

3-3 MedPiare Inpatient Revenue, FY 1985 . . . .. 88

4-1 Percentage of Funded and Unfunded Workloadfor Charleston, Long Beach, and PensacolaNaval Hospitals for FY 1983 . . . . . . . . . . 111

4-2 Percentage of Funded and Unfunded Workloadfor Charleston, Long Beach, and PensacolaNaval Hospitals for FY 1984 . . . . . . . . . . 112

9

4-3 Percentage of Funded and Unfunded Workloadfor Cha:leston, Long Beach, and PensacolaNaval Hospitals for FYs 83/84 . . . . . . . . . 113

4-4 Aggregate Percentages of Funded and UnfundedWorkload for Charleston, Long Beach, andPensacola Naval Hospitals for FYs 83/84 .... 114

10

ACKNOWLEDGEMENT

This thesis is the result of the assistance of many

individuals who are presently acknowledged. Ms. Velda

Austin and LtCol John Coventry, MSC, USA, Tri-Service DRG

Study Group, U. S. Army Health Care Studies and Clinical

Investigative Activity, Fort Sam Houston, TX, provided the

necessary DRG data by transforming ICD-9-CM discharge data

into meaningful biometric data. CDR Carl Chitwood, MSC,

USN, and LCDR Steve Olson, MSC, USN, Naval Medical

Command, Washington, DC, assisted us in understanding the

DoD's cost accounting reporting system and provided the

expense data for the three NTFs used in this analysis. We

extend our deepest and most heartfelt appreciation to Dr.

David R. Whipple, Jr., Naval Postgraduate School, who,

over these last eighteen months, has been the decisive

factor in making this academic experience first rate. To

the sage professor of our first and, unfortunately, only

graduate course in Managerial Communications, Dr. Roger

Evered, Naval Postgraduate School, who unwittingly yet

nonetheless indirectly shares responsibility for all

grammatical, syntax, and/or diction mistakes contained

herein, we thank you for demanding only the highest

quality products and for encouraging us to strive beyond

what we felt were our maximum capabilities. To our wives,

11

Melissa and Pat, Ben's daughters, Margaret and Laura, and

our Rottweilers, Mack and Nugget, we could not have

completed this thesis without your understanding,

patience, and sacrifices. And to our critics who may say

Qlp.= xiic to us, we say, as did Virgil in the

AniQ, = omnia psum Qmnu to them.

12

I. NTfgD CIQN

A. GENERAL

On 1 October 1983, the Health Care Financing

Administration (HCFA), which is part of the Department of

Human and Health Services, implemented a prospective

prospective payment system (PPS) that uses diagnosis

related groups (DRGs) to reimburse civilian hospitals for

treating inpatients under Medicare. Historically,

hospital-cost inflation has run much higher than general

inflation; yet tentative findings indicate use of DRGs may

be slowing this growth. In the future, a hospital's

financial well-being will be directly tied to its ability

to contain costs. In the civilian health care sector

emphasis appears to be shifting from retrospective,

cost-pass-through methods to one of prospective, fixed

cost based on specific case mixes. This emphasis on PPS

using DRGs is part of the strategy to design better

management/financial control subsystems into the overall

health care delivery system, providing incentives for its

participants (physicians, administratore, trustees, and

staff personnel alike) to provide more efficient care.

Diagnosis related groups are part of a patient

classification system that uses 470 case-mix groupings,

which are largely based on various characteristics that

13

are statistically homogeneous. As a measurement of output

or surrogate of efficiency, the DRG patient classification

system comes considerably closer to assessing the true

nature of a hospital's product than any other proxy used

today. For the first time, DRGs enable measuring the

output of hospitals by grouping various hospital services

into product groups. Moreover, DRGs permit hospitals to

identify DRGs that are profitable (revenues exceed related

expenses), and conversely, categories or case-mix

groupings that are unprofitable (i.e., are more of an

expense center product than a profit center) by employing

the concept of case-mix accounting.

As an extension of this capability, this thesis

investigates and analyzes what three typical naval

hospitals would have received had they been reimbursed

under Medicare's DRG and PPS reimbursement methods, as

contained in Public Law 98-21.

B. OBJECTIVES OF THE RESEARCH

The objectives of this research effort are twofold.

First, the authors want to determine whether a feasible

and meaningful comparison of inpatient care costs can be

made between civilian and naval treatment facilities

(NTFs) using DRG6. Second, if possible, we would like to

develop an algorithm or model that enables comparison

between what the typical NTFs would have received under

14

Medicare's prospective reimbursement scheme and what these

NTFs actually expended for specific inpatient workloads.

Actual workload data for two fiscal years will be used to

make the comparison.

C. RESEARCH QUESTIONS

The research hypothesis is: NTFs' inpatient operating

expenses are less than the reimbursement levels these

naval hospitals would have received under the provisions

of Public Law 98-21. If true, this would imply that NTFs

are afficient when judged by this private sector standard.

Secondary questions are:

1. Will the uniqueness of the U.S. Naval MedicalCommand's NTFs prevent a meaningful comparisonbetween themselves and Medicare's reimbursementscheme?

2. If the Veterans Administration's average adjustedcost per discharge, HCFA cost weights, and DRGsare used for determining reimbursement amountswill NTFs' actual inpatient operating expenses beless than the VA constructed reimbursement level.

3. Are NTFs' thirty most frequent DRGs similar ineach facility and among NTFs from one year to thenext? and are the NTFs' thirty most frequent DRGssimilar to those in California?

4. If NTFs' inpatient care costs are lower thanMedicare's reimbursement amounts, what exactlydoes this suggest?

D. RESEARCH METHODOLOGY

The research methods employed by the authors include

the gathering of information from the most current and

relevant literature, and telephonic and personal

15

interviews. In an effort to gain insight into the Naval

Medical Command's perspective on the role of efficiency

and current methods used to assess efficiency, the authors

personally interviewed the: (1) Surgeon General of the

Navy, (2) Commander, Naval Medical Command, (3) Commander,

Naval Medical Command, National Capital Region, (4)

Director, Research Department, Naval School of Health

Sciences, and (5) various personnel within NAVMEDCOM Codes

13 and 14, who provided the cost accounting reporting

documents. Literature was obtained from the Naval

Postgraduate School Library, Defense Logistics Studies

Information Exchange, Dialog Information Services,

California's Mid-Coast Health Systems Agency, and S

applicable regulations, directives, and instructions that

govarn DOD's cost accounting reporting system. Extensive

telephonic discussions were conducted with the Tri-Service

DRG Study Group at the U.S. Army Health Care Studies and

Clinical Investigation Activity in obtaining biometric

data. Information and data gathered from the above

sources were used to analyze how DRGs were being employed

by the civilian health care sector and how best our

proposed analysis could and should be conducted.

E. SCOPE AND LIMITATIONS

This analysis examines the development,

implementation, and controversy of DRGS, their potential

16

role in controlling hospital costs, and, most importantly,

how HCFA's (or Medicare's) DRGs can be utilized in

assessing the relative efficiency cf NTFs. Accordingly,

this thesis limits the discussion to only those parts of

the DOD cost accounting reporting system that pertain

directly to or in understanding the foundation of the

analysis. The thesis does ngt: address all the nuances

HCFA used in formulating DRG groupings, the esoteric

literature findings that pertain to current DRG research,

or any particulars of the personal interviews. Essen-

tially, only information that is relevant and valid to the

analysis, itself, and to understanding DRGs and the cost

reporting system is provided. The intended audience of

this thesis are those who have a basic familiarity of the

civilian and U.S. Navy's health care system but who are

not necessarily familiar with DOD's cost accounting

systems or provisions of Public Law 98-21.

Since the biometric data for FY85 were replete with

inaccurate and incomplete data, the authors elected to use

only two fiscal years of data for comparative purposes.

Although they used only the most accurate and best

available data for this analysis, the authors were by

necessity limited to a small sample pupulation of three

NTFs. Therefore, the findings are at best preliminary and

should be cautiously interpreted.

17

F. ORGANIZATION OF THE THESIS

The organization of the thesis is designed to present

a logical progression toward a comprehensive understanding

of DRGs, a basic understanding of the DOD cost accounting

reporting system, and specifically, why and how our

analysis is formulated. Chapter II presents a wide range

of information, varying from a conceptual discussion of

the factors behind cost containment and hospital-base

inflation to a discussion of the perceived pros and cons

of DRGs. Chapter III describes the prospective payment

system under Public Law 98-21 and DOD's cost accounting

reporting system as it is used in military treatment

facilities. Chapter IV contains as in-depth discussion of

the data, research methodology, and findings of the

analysis. Chapter V discusses the conclusions drawn from

the findings and proffers recommendations based on these

findings and conclusions.

18

A. COST CONTAINMENT INITIATIVES AND ECONOMIC ANALYSIS OF

HOSPITAL-BASE INFLATION

The health care field over 1-he last several decades

has experienced rising costs, particularly when con-

trasted with other sections and components of the

economy. In 1983, the nation spent $147 billion for

hospital care compared to $39 billion in 1972--an

increase of 277 percent. During this same ten-year

period, per capita costs for hospital care rose from $179

to $604. Today, 11 percent of the GrosL National Product

is comprised of hospital and health care services. Of

this 11 percent more than 4.5 percent is devoted solely

to hospital expenditures and, by 1990, it is estimated

that hospital expenditures alone will be $304 billion.

(Ref. ip. 5]

Hospital administrators, physicians, third party

payers, and numerous regulatory and governmental agencies

have all tried to control these escalating hospital costs

through a wide range of initiatives and cost containment

measures: (1) health planning (i.e., comprehensive health

planning and health systems agencies); (2) professional

standards review organizations (PSROs); (3) health

maintenance organizations (HMOs)j and (4) cost sharing by

19

third party payers [Ref. 2:p. 3]. In large part, these

initiatives have at best met with only limited success.

This remains true today because the design of the health

care delivery system fails to provide the necessary and

holistic incentives in the hospitals' structuze and

process.

But before analyzing the hospitals' incentive design

systems and related management control systems using DRGs

and PPS, let us first examine the factors responsible for

the significant increases in hospital cost. One method

of examination involves disaggregating expenditures into

broad categories [Ref. 3:p.4]. In their book, An.ing

god Internal Control Under PrpAeotive Rayment, Broyles

and Rosko discuss how Freeland and Schendler use disag-

gregating expenditures to establish general patterns of

care. For the period 1971-1981, Freeland and Schendler

identify factors that comprise hospital expenditures

(with the relative importance of each in parentheses)t

general inflation (51.7%), growth in real scrvices per

visit (20.8 %), medical price increases relative to

general. price inflation (11.7%), growth in per capita

visits (8.6%), and aggregate population growth (7.2%).

As depicted in Figure 2-1, Freelana and Schendler

disaggregate hospital inpatient care in a similar manner

for 1972-1982 (Ref. lip. 7].

20

Admtssimu per capita

Hospital Input Pricein Excess of Deflator 051M

Sours.:Mark 0 %r~a~ a0d Ca% Ilh %owndr Hlt Sain fn t% If of ~ oftg noCiica 4kcts *aten vit Maeet 4o *I1 0 vo.5 0Srn 14 W4

%ge5 4 .%0%0 p.01

Peren Ditibto ofFco Acontn fo th 6rwt % #01.IExpenditures~~~~%o fo .omnt Hopia 0 nain Ca0 1970-9#

21 2 %1 %

Essentially there are three factors that affect

aggregate expenditure for hospital care: unit price,

quantity, and quality [Ref. l:p. 6]. If one of these

factors increases, while the others are held constant,

t'- tota) eArenditure will likewise increase.

A.though 64.9 percent of the increase in hospital

inpatient cost is attributable to inflation, ("GNP

Deflator" plus "Hospital Input Price in Excess of

Deflator") only 13 percent of it is hospital-specific

related. The significance of this findincg is that

overall hospital costs (unit prices) are markedly

affected by the general economy.

The second factor affecting hospital expenditure is

quantity. During the period 1972-1982, both population

and per capita admissions increased. The total U.S.

pOpuiation increased by 10.7 percent, which caused a 6.9

percent increase in hospital expenditure. Admissions

also increased for two reasons: (1) patients are more

knowledgeable and demanding, requiring physicians to

practice defensively, and (2) the percentage of elderly

in the total population grew concomitantly with an

increase ira the use of hospital-based care. This latter

fact portends an even greater proportional increase in

the hospital inpatient costs because of the number of

persons 65 years and over is projected to increase 16

22

percent between 1983 and 1990--including a 22 percent

rise in the number of persons over 74 years. (Ref. lp.

8]

The third factor affecting aggregate expenditures for

hospital inpatient care is quality. It is assumed that

higher quality of care only comes with increased costs.

This fact has not been necessarily supported by relevant

literature. Over this ten-year period, the number of

full-time equivalent employees has increased approxi-

mately 22 percent. However, one cannot equate this

increase with a corresponding increase in the quality of

care. What is important is that the continual demand or

insistence for improved quality will probably mean an

even greater intensity level per admission, and, there-

fore, higher hospital costs.

Even though the factors of unit price, quantity, and

quality help to explain what has happened and the ordinal

relationship of these factors to one another, they fail

to explain why.

Another method Jn evaluating the causes of hospital

inflation uses economic analysis based on two predominant

theories to answer whyp these are the "cost-push" model

and the "demand-pull" model. tRef. 3:p. 5]. As depicted

in Figure 2-2, one can see using the "cost-push" model

that this supply-side model has had a shJIting of the

supply curve from Si to S2 as direct result of the causes

23

Panel A: Cst-P'ish Inflation Psnel : emand-Pull Inflation

Fr S2 Price

S1 S2

P2 /

P3

P2

DID2

00 01 uantity O0 02 quantity

Panel C:Causes of cost-push inflation Causes of demand-pull inflation

0 increased prices for factors of 0 increase in general populalionproduction 0 increase in elderly population

0 loreased full-time equivalent employee 6 increase In income(FTE)/patient ratio 9 defensive mediOne

0 expensive neow technology 0 availabilitij of new services* costs of regulatory oomplianoe 0 growth of public and private* lagging labor productivity health insurance

* inflation in the general eccnomyj

* oost-based retrospective

reimbursement

Seure: Adapted from P]InJig nd Internal Centre1 11gici:

f msm9 by R. V. Broyles and M. D. Roske (Reekville:Aspen Publishers) 1965.

Causes of Hospital Cost Inflation

Figure 2-2

24

listed in Panel C. Similarly, using the "demand-pull"

model, one can see the demand curve for hospital

services s.,ifting from D1 to D2. Together these models

help explain why prices for hospital services have

steadily increased over the last ten to twenty years.

The "common thread" to both models/theories lies in the

cost-based hospital reimbursement methods, which have

been done largely on a retrospective, cost pass-through

basis without any price rationing incentives for the

consumer or the provider [Ref. 3:p. 7]. These models and

the listed causes of inflation should help to illustrate

why inflationary costs for hospital-based care might

continue to outpace the general economy in the years

ahead unless some mechanism is put in place to contain

costs. This mechanism may well be DRG-based measurement

under a PPS.

B. RETROSPECTIVE VERSUS PROSPECTIVE REIMBURSEMENT OR

EFFECTIVENESS VERSUS EFFICIENCY

Before discussing the development or evolution of

DRGs and case-mix measures let us briefly examine first

the differences between retrospective reimbursement and

prospective reimbursement.

Both terms--retrospective and prospective reimburse-

ment--are used in conjunction with rate-setting programs

under either a governmental program or a third party

payer program, such as the pre-1983 Medicare and Blue

25

Cross/Blue Shield programs, respectively. Under retro-

spective reimbursement hospitals are reimbursed after the

services are rendered and costs are incurred. Hospitals

are given interim payments throughout the year; at the

end of the payment year, a complete review is conducted

on the costs incurred and services rendered with a final

adjustment made for the differences between approvedcosts/services and payments already made. Inherent with

a retrospective payment system is the design incentive tospend as much as the hospital feels is appropriate. (Ref.l:P. 10] Therefore, the reimbursement system andfinancial management control systems are largely drivenby effectiveness rather than efficiency.

As a rate-setting mechanism, prospective reimburse-

ment essentially preapproves anticipated bervices and

costs, paying in advance a payment based on the expected

case-mix workload. At the end of the year nominal

adjustments are made to ensure hospitals and third party

payers receive an equitable adjustment. Unlike retro-

spective reimbursement, the prospective payment system

gives hospitals incentives to be frugal and cautious in

the pursuit of their programs and objectives. The

underlying design incentive then is to meet an

effectiveness level--be it a "high level of quality"

and/or provision of certain programs--while concomitantly

26

meeting funding constraints imposed by case-mix reim-

bursement. If a hospital provides inpatient care for

less than its costs then the hospital ostensibly stands

to make a profit. With the prospective payment system

the emphasis appears to shift from the effectiveness

model to one driven more by efficiency,

C. CASE-MIX MEASURES, COSTS, ACCOUNTING, AND BUDGETING

Before specifically discussing DRGs, it is important

to understand the concept of how different patient case

mixes can directly affect a hospital's costs. All other

things equal and hospitals have an increase in the

inpatient workload, costs will similarly increase;

however, it is possible for hospitals to have an

increased inpatient workload yet experience lower total

costs. Conversely, hospitals can have a decreased

workload and experience higher total costs. The explana-

tion for this disparity in cost revolves around the

issues of intensity of services and complexity of care

rendered or, in other words, case mix. The issue of

complexity relates to the types of services; whereas,

intensity relates to the number of services per patient

day or hospital stay [Ref. l:p. 21). As Grimaldi and

Micheletti point out, there is no precise consensus on

what comprises complexity. Certainly there are at least

these five relevant factors: ". . . severity of an

27

illness, the prognosis or likely outcome of an illness,

the difficulty of treating the case, the need for timely

intervention, and the amount and composition of resources

used to treat the patient." [Ref. l:p. 641 Again, the

important point is that case complexity and intensity of

service are the two key components of case mix.

Health care researchers, analysts, and hospital

administrators alike acknowledge that historical methods

of measuring output--through such surrogates as

departmental inpatient bed days and number of admissions-

-fail to accurately capture the relationship between

services provided and the costs incurred. That is,

traditional output proxies are poor for purposes of

assessing and monitoring the relationship between

input--manpower, technology, facilities, and

equipment--and output, patient care through hospital

services. Accordingly, traditional managerial accounting

systems are inadequate since they tend to reflect data

and information in the aggregate and on a departmental

level with no accountability for individual patients

being financially managed [Ref. 4:p. 56].

Under the traditional organizational structure of

hospitals, departments are not required to ensure that

individual patients are both efficiently and effectively

managed. Hospital structures that use the case-mix

accounting and budgeting process have an integrated

28

picture of the financial consequences of providing

inpatient care to individual patients in each of the

DRGs, as Figure 2-3 represents. This type of process

enables hospitals to gain an understanding in detail of

the "profile of service requirements" and costs per

patient grouped into product lines. To achieve this,

three inputs are required [Ref. 5ip. 51]

" Patient clinical data must be sufficient todetermine DRG assignment;

. A "bill of particulars" in needed that describesspecific diagnostic and therapeutic services deliv-ered to each patient; and

" Detailed costs per unit of service (laboratory,radiology, dietary, etc.) must be developed basedon whatever are deemed to be appropriate defini-tions of such services.

As will be discussed in the section on product defini-

tion, hospitals that are organized in a matrix-type

manner will possess the capabilities to use the case-mix

accounting and budgeting concept in its fullest applica-

tion.

Although case-mix measures appear to provide a better

method for assessing, monitoring, and evaluating input-

output relationships than these historical methods, there

is lack of consensus on which grouping strategy or

patient classification system using cass mix is optimal

(e. g., DRGs, John Hopkins Severity Score, Systemetrics

Disease Staging, Blue Cross of Western Pennsylvania

Patient Management Categories, etc) [Ref. 3%pp. 15-16].

29

General Patient PatientLedger Cl, nic, Serallesby Cost oat. DeliveredCenter

Allocationof Indirectand OverheadCosts ToDirectPatientService DROCenters Assignment

andPaiientServiceProfile

DetailedCost PerUnit ofService

i DetailedJ

Cosi/Patientin Each DRO I ll ,

Source: Robert 9, Fetter and Jean L. Freeman, "Diagnosis Related Groups: ProductManagement within Hospitals," Aiaiy.. E1 UUgmyn R 1986, vol. I I,no. I, Figure I

Overview of Case Mix Cost Accounting Process

Figure 2-3

30

There is agreement then that patients' clinical require-

ments greatly affect costs but there appears to be a

disagreement on precisely what factors best explain this

variation in cost or length of stay. Even though the

state of development of a optimal case-mix grouping

technique is in flux there is agreement that whatever

case-mix method (patient classification system) is

finally accepted it should contain these properties

(Ref. l:p. 22]1

. be derived from a reliable and readily availablesource,

. be calculated in manner that would precludespurious manipulation to suit one's purpose,

. be accepted by physicians and understood byhospital personnel, and

a be cost beneficial (i.e., the benefits outweigh thecosts).

When hospitals use case-mix measurements and account-

ing techniques, it enables them to produce more actual

and accurate management information. Ca.e-mix approach

to controlling hospital costs ". . . provides a clear,

complete picture of the costs of treating individual

patients grouped into similar case classes based on use

of resources to set norms and standarde for a management

control system" [Ref. 4:p. 57]. This ipproach, as such,

is based on DRGs, which classifies cases into groups and

thus groups into hospital products that use similar

31

amounts of services and resultant resources [Ref. 6:p.

240].

As Collins points out, the key issue with the

hospital management control system using case-mix methods

is to motivate physicians to use resources in an economi-

cal manner [Ref. 4:p. 56]. Using case-mix (DRG) account-

ing measurements, hospital administrators clearly should

be better able to determine which physicians deviate

between their actual costs and the standard costs

associated with the particular DRG. Similarly,

physicians, themselves, will better understand the

ramifications of their medical decisions in an economic

framework. Case-mix methods and measurements allow

hospitals to more precisely identify costs and to gain

insight between these costs (inputs) and services

(outputs) provided. A later section of this chapter will

discuss specific case-mix measurements, the pros and cons

of DRGs, and what incentives exist or do not exist for

physicians to practice medicine more efficiently and

effectively.

D. PRODUCT DEFINITION AND MATRIX ORGANIZATIONS

As previously described, the DRG approach enables

hospitals for the first time to describe their system in

terms of production. Chase and Aquilano define product

as . . . the output from a productive system offered

32

for sale (in the case of a business) or otherwise made

available (in the case of a governmental or philanthropic

organization) to some consumer." [Ref. 7:p. 26] A

hospital provides a wide range of services to its

patients. These include x-rays, surgery, nursing care,

physician care, medications, hotel and social services.

Although these services may seemingly be interpreted as

the final output of hospitals, they are only intermediate

outputs. The final output of hospitals is to treat

individual patientsl therefore, specific sets of these

intermediate outputs constitutes for each patient a

"product" of the hospital (See Figure 2-4).

Fetter and Freeman explain that a hospital is a

multiproduct firm with each product consisting of

multiple goods and services." [Ref. 5ip. 42] This

product line is extensive and is made up of numerous

intermediate outputs, (hours of nursing care, number of

lab tests, meals, medications, etc.) and inputs (capital,

labor, material) (Ref. 6:p. 2311.

Fetter and Freeman view the output of hospitals much

like matrix programs are used in industry, such as in the

development of the U.E. space shuttles or the Apollo

Project. They compare these matrix-type programs to a

hospital's "projects," as "projects" consist of multiple

services (intermediate outputs) based on the types of

patients the hospital treats [Ref. 5sp. 43]. They see

33

Hospitals physician.operat Ions nrm' rs

INTUTr' IT EDIATE nUTPUTS ER.UCTI

Labor Patient D49s Vagtnal di'g iwnoe

Materials Meals smplating dialnses

squipment Laberatory proceduresHanaqeenIlarlioal prooodures Appeo~destomy W/e

mledisa*llons smplistaed prineipal; dienesis, e7O

S1 v/SCC

* 0 Lens prooedures

EFFICIENCY EFFECTIVENESS

foura.: Robert 3. Fetter and Joan L. Frereman, "Diagnosis Related oroups ProduotManaqement within Hospitals," &Uyd tUU 9#m n1 R ylow 1986, vol. Ii,no, I, Figure 1

Defining the Product of Hospitals

Figure 2-4

34

each particulhr product (project) being a function of a

patient's cundition. as eell as his/her treatment regimen.

Briefly, as represented in Figure 2-5, what is

proposed by Fetter and Freeman in the clinical matrix

organization is first that physicians be placed in charge

of specific subsets of DRGs. It is their responsibility

to determine the appropriate mix of resources necessary

in diagnosing and treating each type of patient. Second,

middle managers and administrators are responsible for

the operational results of the intermediate support

centers: lab, x-ray, laundry, etc. Thus, physicians are

responsible for defined groups of patients and adminis-

trators are responsible for clinical support services.

[Ref. 5p. 49]

What the matrix-type organization permits is a means

of measuring performance along whatever product lines are

established. Young and Saltman propose, for example,

that if the average cost for a particular DRG increased

over a set period of time, the reason for the increase

could be explained in one of three ways: increase in use

of resources, increase in resource (input) prices, and/or

decrease in operational efficiency [Ref. 5S:p. 49]. A

variance report is designed and used to detect whether

physicians are using more resources than previously used,

or whether administrators are not as productive. Of

course, the other factor--an increase in input

35

viob 5uLj*.AodO

WOO OUY~~

2 0

6- 0

.. . .. --.- - - - - - - - a - - b - -

0 i1 i I I IILi .S

0

no .

- (N Xo 'pS.

Source: Robert B,/oetter .nd Jeanl L, Freeman, *Diagnosis Ret~ted Oroups: ProductManagement within Hospitals," gj m.Oy. ofIjnaggma Review ,I Wl6, vot, 11,no. I, riur. 4

Clinical Matrix Mlsnagment

Figure 2-5

36

prices--could well be the reason. As with any variance

report, the onus is placed on the person or group

responsible for that profit or activity center to justify

the deviation. For this type of arrangement to be

successful, Fetter and Freeman similarly purport, as do

Young and Saltmtn, that the control system based on the

matrix organization must have a cost accounting system

which clearly distiaguishes between both fixed and

variable co~tP, and controllable and uncontrollable

costs. [Ref. 5:p. 49] H(nce, if the product and matrix

organizational approach is to succeed, case-mix

accounting and budgeting systemi, which provide

information along product lines, are required.

In their discussion, Fetter and Freeman Zlearly

indicate that, even though it is quite possible to have

well-defined case types (with a set of diagnostic and

therapeutic services normally expected), cost vari4.ions

of great significance occur, even for well understood

illnesses where there is great consensus among providers

as to the appropriate treatment process [Ref. 5:p. 43].

Moreover, products are largely identified and broken

down into groups by factors that predict amounts and

types of services required. While the set of services in

Table 1 might well represent the expected values for this

kind of patient (one who is less than 70 years-of-age,

without complications or comorbidities, and with a

37

TABLE I

DR6167: APPENDECTOMY W/O COMPLICATED PRINCIPAL DXAIE 70 WIo CCa , HOSPITAL X

Resource consumption profile Qtg Unit Cost Total Cost

Nursing care dags, level 1 2 $65.70 $131.56

Nursing care dags, level 2 2 09.32 170.64

Dietary, meals (5TD) 12 2.50 30.00

Operating room (minutes) 60 2.48 148.80

Recovery room (hours) 1 30.50 30.50

Anesthesia 1 42.75 42.75

Lab test 198 1 6.95 6.95

Lab test 205 4 11.32 45.20

Lab test 206 3 4.16 12.48

IV therapy 614 2 6.15 12.30

Abdomen x-rag 1 26.55 26.55

Miscellaneous 189.26

Total cost. $870.17

Source: ,fart 1, Fetter, Jdm 0. Thompon, od JM fi. KImberly, fdaLIn

M I th tlI S±MJzggli, (HoNe o, IIl,:ltchrd 0. rwlin, 19).

38

primary diagnosis of appendicitis) some conditions vary

considerably in cost of hospitalization. Cost of hospi-

talization is predominantly a function of length of stay

(LOS); ergo, the longer one stays hospitalized, the more

resources are consumed--though perhaps at a diminishing

rate. Length of stay is almost always a physician-deter-

mined variable, though variation by any one physician is

usually quite small. [Ref. 5:p. 44]

Additionally, variation in the cost of care is not

only affected by physicians' decisions but also by the

efficiency of actual hospital production of intermediate

outputs. As previously mentioned, hospitals must be able

to assess, monitor, control and evaluate their

efficiency, but they must also be able to 9ontl the

level of effectiveness in which these services are

rendered. Otherwise, they will be unable to control the

process and structure and, while great strides may be

made in improving efficiency, these improvements can be

more than offset by efforts to maximize effectiveness.

For this reason, it is paramount to first develop a

conceptual framework which permits analyzing the system

by defining the actual products. Diagnosis related

groups make this identification of products a reality.

E. AN OVERVIEW OF DIAGNOSIS RELATED GROUPS

Although up to this point we have only mentioned DRGs

in rather broad terms and have cursorily defined what

39

.. ---- .L " -" • • m,

they mean, it is appropriate to explain how DRGs evolved

both on the state and federal levels, the statistical

technique used to form DRGs, some other case-mix measures

and, perspectives on the pro and cons of using DRGs as a

patient classification method.

Diagnosis related groups are an outgrowth of what

Fetter and Thompson started to develop at Yale University

in the early 1970s. They realized that to make meaning-

ful comparisons and analyses of hospital management, cost

control, and planning that case-mix information needed to

be included. They further realized that whatever

classification system was developed it needed these four

characteristicF [Ref. 8:p. 5621:

" The number of patient groupings should bemanageable;

. The system should use available medical and demo-graphic data;

• Groupings of medically similar patients should bestatistically stable in terms of the hospitalresources; and

• The statistically similar groups should be similar

medically as well.

They opined that classification based on the above

characteristics would permit DRGs to center on patient

attributes and the treatment process rather than on such

surrogates as bed size, occupancy rate, And service

40

capacity of a hospital or its medical staff's specialties

[Ref. i:p. 22].

Statistical techniques were used by these researchers

at Yale to form medically coherent groups, groups (i.e.,

DRGs) that used similar quantities ad type of resources

and were medically related. Resource consumption was

assumed to vary directly with length of stay and thus LOS

was selected as the dependent variable. In forming the

DRGs, physicians assisted in transforming diagnostic

codes into specific DRG groups. Accordingly, upon

discharge a patient's final diagnosis is used to place

the patient into one of the DRGs.

Diagnosis related groups evolved from the efforts of

Fetter, Freeman, and Thompson as a case-mix grouping

strategy. They based the groupings on diagnostic,

demographic, and therapeutic characteristics of

inpatients using the International Classification of

Disease, 8th revision (ICDA-8) and HICDA-2 diagnostic

codes. The second generation of DRGa, however, uses

ICD-9-CM codes for the basis of its groupings. The first

generation consisted of 383 DRGs and the second has 470.

Both of these groupings are mutually exclusive and

exhaustive. In addition to using different coding

schemes, the biggest diffetence between these two

generations of DRGs is that the newest DRGs are grouped

based upon specific surgical procedures and secondary

41

diagnoses rather than the mere presence or absence of

surg.cal procedures or secondary diagnoses [Ref. 9:p. 2].

A later section of this chapter discusses more specifi-

cally the role of the ICD-9-CM codes in forming case-mix

definitions and their use in differentiating among levels

of hospital resource use and in differentiating clini-

cally among types of patients.

F. STATISTICAL METHOD USED FOR FORMING DRGs

The statistical method used by Fetter, Freeman, and

Thompson in developing these groupings is a variation of

the Automated Interaction Detector (AID) method of

Sonquist and Morgan [Ref. 10]. Marketing researchers at

the University of Michigan Survey Research Center have

often used AID in analyzing complex sample survey data

which is based on income, age, sex, education, etc. [Ref.

ll:pp. 415-4341 (Ref. lip. 231.

As Grimaldi and Micheletti discuss, the AID's role in

forming DRGs is one of statistical testing; however,

unlike marketing applications more information and input

than just statistics is used in forming these terminal

groups [Ref. lip. 231. Specifically, physician input has

been used in formulating groups in order to ensure each

DRG is medically/clinically coherenh and meaningful

conditions ara contained within each. In using the AID

42

package the objective is to identify the "interrelation-

ships of the variables in the database and to determine

which ones are related to some specific measures of

interest, referred to as the dependent variables." [Ref.

4:p. 57] Although the AID package cannot ensure the

groupinigs are clinically related, physicians can. As

Grimaldi and Micheletti relate [Ref. ltp. 23]:

A medically meaningful classification (scheme)stimulates expectations as to the natural history ofthe disease, the appropriate ways to manage the case,the prognosis, the likelihood of complications ofspecific kinds, of the risk of death. Determinationof medical meaningfulness is therefore a subjectiveprocess, best accomplished by consensus of cliniciansfrom the defined population. [Ref. 12tp. 249]

Although the primary disadvantage of forming groups

in this manner is a loss of statistical homogeneity, the

DRG system as a whole stands a much greater chance of

being accepted by those who most effect the use of

resources, namely, the physician. With greater physician

acceptance comes a much greater probability of the health

care delivery system achieving the desired outcome.

The actual computer program that formed the DRGs is

known as AUTOGRP (AUTOmatic GRouPing System, pronounced

autogroup) [Ref. lp. 231. This program groups informa-

tion by minimizing the distance (unexplained variance)

between observations [Ref. 13:pp. 17-31). As previously

mentioned, length of stay is the dependent variable. The

objectlve is to minimize the unexplained sum of squared

43

differences. And, of course, the smaller the unexplained

sum of squared differences the more homogeneous the

group; therefore, the smaller the difference the better

is its ability to predict length of stay and supposedly

resource use. [Ref. lp. 241

According to Grimaldi and Micheletti, AUTOGRP

attempts to minimize the overall sum of squared differ-

ences (TWGSSQ) by partitioning the population into

subgroups based on diagnoses, procedures, sex, age, or

other variables believed to cluster patients homogeneous-

ly, using a series of binary splits to subdivide patients

based on a myriad of partitioning rules. The TWGSSQ is

calculated as follows:

TWGSSQ- EE (Yik - k)2

where ! is the average stay of patients in the ktb

group. Of course, the desired partition is one that

yields as close to a zero group sum of squares as is

possible. Groups themselves are broken down or split

into subgroups based on whatever partitioning rules are

employed. At some point it is necessary to stop forming

subgroups because the statistical contribution is rela-

tively insignificant or the number of subgroups becomes

unmanageable. (Ref. lp. 25]

As briefly discussed above there have been two sets

or generations of DRGs developed. The first set wae

derived frcm the medical records of over 700,000

44

patients. Diagnostic codes were based on ICDA-8. This

first set consisted of 83 Major Diagnostic Categories

(MDCs) and a total of 383 DRGs. Assignment to a category

and subsequently to a subgroup (or DRG) was based on

primary and secondary diagnoses, surgical procedures,

and/or age [Ref. l:p. 25]. Again, the principal diag-

nosis at discharge determined which MDC was assigned.

Figure 2-6 represents the typical grouping of a MDC under

the ICDA-8 DRGs.

The second generation of DRGs is based on data

provided by the Commission on Professional and Hospital

Activities (CPHA) in which a random sample was taken of

over 400,000 medical records from a population of 1.4

million. The results of this sampling procedure are

shown in Table II. With this newest generation of DRGs

there are 23 major diagnostic categories that contain the

ICD-9-CM DRGs, of which there are 470. Again, unlike the

first generation of DRG assignments, the second genera-

tion is based upon specific surgical procedures and

secondary diagnoses rather than the mere presence or

absence of surgical procedures or secondary diagnoses.

What occurs then with the newest DRG assignment is

that correspondence between the MDC and the ICD-9-CM is

not necessarily one-to-one. For example, CPHA indicates

ICD-9-CM diagnostic codes for the circulatory system are

scattered among at least four and perhaps as many as

45

MDC: 52Diseases of Gall Bladder

and Bilt Duct

No YYes

Under 51 Over 50 No

DR22 RG 226 DRU 22.7

, Under 65 ver 64

Souroe: Paul L. Orimaldi and Julie A. Mioheleiti, Erftap ti ,ement The DefinitivGuide to Reimbureement, (Chioo: Pluribus Press), 1995, Exhibit 3-1.

Formation of ICDA-O MDC 52

Figure 2-6

46

5r

TABLE II

IIDCs AND ICD-9-CII CLASSIFICATION SCHEMES

Numbr of NumberSamled of

major Dil Mtio Categerg ICD-9-CH1 Patlers tsSI Diwass and Disorder Dipses of the Nervys 26,392 35

of the Norous Sstem System and Sense Organs2. Diseaw Disrders 9,5" 13

oft thle3, Distae a Disorders 21,456 26

of th Ear, Nose, andThroat

4. Diseases and Dilsorders D se -:Ithe Respiratorq 29,145 26of the Respiratory System Sst4 n

5. Diseases an Disorders arezs of the Circulatory 44,342 43of the Cticulatorj Sstem System

6. Diseases mnDisorders Diseas of th Digestive 25,914 Isof the Digestive Sqstem System

7 Diseases and Disorders 9,086 isof the HepatobillarySystem and Panoreas

S. Diseases and Disorders Diseas of the Nsculo- 51,235 48of tWe M t uoskeletal Sys- skeletal System andter and Comeotlve Connettlve Tissue

9. Dtoss and Disorders Dbems of * hSkin, and 10,3rS 20of the SkIn, S boanoos Subotan TissuTissue, and fbrest

10. Endoorlni, Nutritional, Endoine, Nutritional, and 7,910 1?and Metabollo Diseases Metaboll Disease andand Disorders ImunitV Disorders

II, Diseases and Disorders Disas of the Oenltourlnary 9,64 32of 1he Kidney and Urinary SystemTraot

12. Diseases and Disorders 4,564 19of the Male Reproduoltivesystem

13. Diseases and Disorders 5,579 17of the FrmaleR produatlve sstern

14. Prerny, Chldbirth Compllatlons of Pregnm y, 59,00 15and the PuerW Chflirth, nd thePuerperlur

IS. Newborns and Other Certain C*nMti Originating 47,209 7Neonates with in the Peri &l PeriodConditions originating inthe Perinatal Period

47

TABLE IIcontinued

Nhmbtr of NuwmerSampled of

Major Diagostte Cateries D-I-CI Patients DROS16. DIfns an Disorders Diems. of tt Bloo and 2,291 6

of the Blood and Blood- ood-Forminl OrgnForming Org *nd DisordersknunologM Dsorders

17. Mqlcproilferattve Neoplan 5,552 15Disesan d Lisordersand Poorly Differentiated

10. nfectious and Parastio Ifectlousa md Parasitic 2,374 9Diseases (Systemic or DiseasesULnpdifod Sites)

19. Mental Diseases and Mental Disorders 10,902 9Disorders

20. Su ance Us. am 4,76 6Substanoe ktduoeOrganic Mental Disorders

21. Inj ry, Posoniin,ad Injry andPosnlng 6,245 1774. 9" 145 5

23. factors Influening Classification of Factors I ,79 7Health Status and Influecoing Health StatusOther Contacts with and Contact with HealthHealth Services Service (Suppemnentary

Classification)

GOncee: The New ED-S-CM Diaces Related Orenm Clasiiflealan Scheme User tal (New Haven,tr: Yale Uniyersity Soholl of Organiation and Maaeent, Dqme 1951). Table 32. Upated as perHealth SYStem International Manual.

48

eleven MDCs--1,4 to 9, 11 to 13, and 18. Also, the

number of DRGs within each MDC ranges from 5 in MDC 22 to

48 in MDC 8. Again, the new DRGs use the body system

(e.g., reproductive or nervous) as the primary factor for

determining assignment within the MDC and to a specific

DRG. Greater emphasis is now placed on grouping sub-

groups in some kind of clinical relationship to one

another. Even greater physician and other professional

input was used to develop the ICD-9-CM DRGs. [Ref. l:p.

28]

Unlike the first generation, which used LOS data for

final groupings, the second generation of DRGs reflect

modifications suggested by cost data obtained from

330,000 records for patients discharged in 1979 from a

total of 33 New Jersey hospitals [Ref. lip. 28]. There

are a number of other distinct differences between these

two generations of DRGs. Unique characteristics of the

second generation include [Ref. l:pp. 28-33]:

. initial partition for each MDC except 14,15, 17,20f and 22 depends on the presence of an operatingroom (OR) procedure rather than the principaldiagnosiso

a pationts with an eligible OR procedure are parti-tioned into a group believed to be the mostresource intensive depending on the surgical codereportedI

* the ranking of qualifying secondary diagnoses andprocedures in tetms of resource consumption doesnot affect DRG assignment (i.e., approximately 210DRGs are predicatod on the presence or absence ofcomorbities or comiplications);

49

. the principal diagnosis of any patient who isinitially assigned to MDC 5 and subsoquently has anacute myocardial infarction is classified AMIregardless of the diagnosisl

" different and greater number of variables are usedto form the ICD-9-CM DRGG;

" age is a criterion for grouping patients inapproximately 55 percent of the ICD-9-CM DRGs, with18 and 70 years being the critical ages;

" patients who die are placed into one two DRGs (123or 385).

If one contrasts the two generations of DRGs,

identifying the key grouping variables, as represented in

Table III, it should be rather easy to discern the key

differences.

Even though other partitioning variables (e.g., type

of payer, admission diagnosis, type of admission, number

of complications and comorbities, etc.) were analyzed

using AUTOGRP none of these were employed in forming the

ICD-9-CM DRGs since their contributions were not statis-

tically or medically meaningful (Ref. lipp. 28-33].

The figure that follows is representative of the 23

MDCs and should aid in understanding how assignments are

made to specific DRGs (See Figure 2-7). The entire 23

decision trees and Medicare titles for each DRG are

contained in Appendix B. Figure 2-7, a decision tree,

illustrates the DRG assignment for patients with a

principal diagnosis that places them in MDC 7. The first

partition in the surgical halt of the MDC is predicated

50

TABLE I II

KEY GROUPING VARIABLES FOR ICDA-8 AND ICD-g-CM DRS

ICDA-S DRG. ICD-V-CM DRGob

Principal diagnosis Principal diagnosis

Secondari diagnosis Operating-room procedure

Principal procedure Age of patient at admission

Sscondaer procedure Box of patient

Age of patient Complication or comorbiditg

Clinical services Certain secondarg diagnosis

a Used to form one DNGb Most frequentlg used variables

Source. Paul L. Orimaldi and JuH* A. MicheleW, EciaRton Pme2 3l2Defiitv Guide to L (Chkcsgoll: Pluribus Prom, 1905).

51.

on the type of surgical procedure: pancreas, liver,

shunt, or biliary tract, or exploratory diagnostic

workup, or other OR procedure. When patients have more

than one type of OR procedure they are usually assigned

to the most resource-intensive DRGI intensity of

resources normally decreases as one moves rightward along

the surgical branch of the MDC [Ref. 1:p. 331.

Additional splits are required for surgical patients

before they can be assigned to a DRO. If the pancreas,

liver or shunt OR procedure is considered major then the

patient is assigned to DRG 1911 if not, he is assigned to

DRG 192. If the biliary tract procedure does not require

a total cholecystectomy then the subdivision is made

based on whether the patient is over age 69 or a

complication or comorbidity (labeled 70 CC) is present.

If one or more of these three conditions exist then the

patient is placed in DRG 193; otherwise, the patient is

placed in DRG 194. On the other hand, if a "total

cholecystectomy" is performed then a split is made at the

"common bile duct exploration" looping variable and again

at the "Age 70 CC" looping variables before assignment is

made to one of four terminal DROss 195 through 198.

Patients that have an exploratory diagnostic workup

procedure are subdivided based upon the "malignancy" of

their principal diagnosis. Patients with malignant

conditions go to DRG 199 and those without a malignant

52

Surgioal Partitening

Mjor tal

Of Procedure Cholioysteotomy

AgR 70 CC 11le D)u1t

Exploraiion

DRO 193 IDRU 194 Age 70CC Age 70CC

IIDC 7: Diseases and Disorders of The HepatoblliarySystem and Pancreas

Figure 2-7 (a)

53

MDC 7

OR

p cedure C-alogory

ro >

DRO201

rincioalDis mosis ofMalignancy

DRO0 DDMa

01!99 2060

Figure 2-7 (b)

54

Medical Partltonlng 2

PrincipalDiagnosis

DRG DO ODR 5202 203 12041

Age 70 CC Age 70~ CC

DO O-RG CRO DO205 206 2 20

Sour ce: Th psdC:I.. ~ z Reaed ro W_..ur Manual ( N en,CT: H~ealth Systems Inlernaliona)).

MDC 7: Diseases anld Dlsordors of the HoettaiarySystemn and Pancreas

Figure 2-7 (')

55

principal diagnosis go to DRG 200. The terminal DRG 201

is used for all other diseasas and disorders of the

hepatobiliary system.

The medical partitioning of MDC seven assignments are,

somewhat easier to determine. The principal diagnosis--

malignancy, pancreas, cirrhosis and alcoholic hepatitis,

liver, or biliavy tract--determinee which DRG these

medical conditions are assigned. Only in the later two

principal diagnoses are they further subdivided and then

by the looping variable of "Age 70 CC." Assignment in

the MDC 7 medical partitioning runs rightward from DRG

202 to DRG 208. Accordingly, MDC 7 has eleven surgical

DRGs and seven medical DRGs.

G. STATE PROSPECTIVN PAYMENT SYSTEMS:

THE NEW JERSEY EXPERIENCE

Although a prospective payment system on the federal

level (Medicare) is relatively new, several states have

been using somt. form of PPS for a number of years. The

first state rate-regulating law was enacted in New York

in 1969 and followed in the 1970s by : Connecticut,

Maryland, Massachusetts, Rhode Island, Waihington,

Wisconsin, and New Jersey (Ref. 3:p. 71. Uf these

states, New Jersey has coumanded the greatest amount of

attention because of its 3uccets with not oialy a

prospective payment system but also with one that

incorporates the use of DRG&.

56

Because New Jersey's "experience' is similar in many

respects (i.e., its ostensible objective of cost

containment and use of DRGs in its PPS) to Medicare's,

many health care analyst and governmental officials have

* thoroughly evaluated New Jersey's PPS, drawing valuable

lessons learned so that these lessons could be, appro-

priately applied to other state programs and Medicare

[Ref. 141p. 43]. Additionally, other third party payers

are evidencing a keen interest in DRGs, PPS, and case-mix

accounting, as a promising means of controlling and

containing costs.

Because of the similarities between the Medicare and

the New Jersey DRG/PPS syatems and because of the

involvement (funding) uf the federal Health Care

Financing Administration (HCPA) in the experimental New

Jersey's DPG/PPS, it behooves one to briefly review the

New Jersey "experience" before analyzing the federal

DRG/PPS program. This review will assess the motivation

for the program, its salient characteristics, the

implementation results, organizational pressuren and

incentives, the financial and economic impact and finally

discues the major differences between New .Yersey's

program and the Medicare program.

The premise on which the New Jersey's Department of

Health uses the DRG method of hospital reimbursement is

on the belief that economic incentives can be used to

57

improve hospital efficiency and to contain health care

expenditures. The primary impetus behind the coalition

that passed the 1978 New Jersey legislation was not

strictly cost containment as one might think but rather

two other pressing issuese (1) the escalating and

seemingly uncontrollable growth in bad debts, which were

threatening the inner city hospitals' financial via-

bility, and (2) the increasing differential between Blue

Cross regulated payments to hospi.tals and the uncontrol-

led charges that private insurers faced [Ref. 14:p. 43].

The reimbursement reform intent was to provide greater

financial stability to the New Jersey hospitals and to

all commercial and private third party payers. Inciden-

tally, one of the unique characteristics of the New

Jersey experience is that PPS and the use of DRG case-mix

reimbursement applies to all third party payers--govern-

m6ntal, commercial, and private [Re. 15: p., 548].

As Sapoleky, Greene, and Weiner discuss, New Jersey

state officials selected DRG-type prospective reimburse-

ment methods based upon case-mix [Ref. 14:pp. 43-46].

Beginning in 1980, New Jersey implemented PPS using DRGs

in 26 of the state's 93 general acute care institutions.

Although it was New Jeroey's intent when it passed the

reimbursement reform initiative to pay only one prospec-

tJ.vv rate for each DRG, analyzes indicated that the cost

variation among New Jersey's hospitals was quite great.

58

Even in retrospect it is impossible to ascertain what

specifically accounted for these variations: real cost

differences and/or unlike patients classified together.

However, those ho3pitals who felt they were at a disad-

vantage were most acrimonious. As a result of their

perceived disadvantage and incessant criticism of the DRG

system, hospitals were yranted relief in two forms.

Dirst, throe classes were established for teaching

hospitals because of the demonstrative correlation of

cost and size. Second, the DRG rate itself was

recomputed to reflect a blend of "each hosptial's own

historical costs and its class average." [Ref. 14tp. 44]

Thus, there are three factors upon which New Jersey's DRG

rates are based [Ref. 15tp. 549]:

Cost of Hospital-Based PhysicianServices

DRG Payment Rate for PLUS

Direct Patient Care - Portion of Hospital's Own

Non-Physician Cost

PLUS

Portion of Standard Non-Physi-cian Cost

Effectively what the above modification does to the

reimbursement system is to weaken the incentives for

efficiency. The system designers had thought by allowing

59

hospital-specific costs in the reimbursement formula that

in the "outyears" the revised DRG rates would reflect new

efficiencies and that periodic "rebasing" would

supposedly "ratchet down" costs statewide [Ref. 14:p.44].

As Sapolsky and Wasserman discuss, when periodic

"rebasing" took effect in 1984 the "ratching down" of

costs really meant an adjustment upward to reflect the

increased cost (i.e., the base year of 1979 was replaced

with cost data from 1982). Whatever new efficiencies

were perhaps achieved by using DROo, they were more than

offset by an increase in new services and costs of

technological improvements: the overall result of using

the 1982 cost data "rebasing" is an increase of nearly

six percent above regular inflation [Ref. 14:p. 44].

Moreover, the organizational affect of DRG prospective

reimbursement on New Jersey hospitals has been

significant. [Ref. 15ipp. 553- 554]

0 The quantity and type of information collected inDRG hospitals has expanded, with the development ofsophisticated management information systems/

. Decision making in the DRG hospitals is now muchmore decentralized than in non-DRG hompitalsi

. The importance of the medical records department incomparison with other hospital departments hasincreased dramaticallyl

• The medical staffs' role in managerial decisionmaking has increasedj

a The focus of hospital administrators hasdiscernibly shifted from an input orientation toone of producing or managing outputs.

60

01

While it may be too early to draw comprehensive and

irrefutable conclusions about the New Jersey experience

in economic and financial terms there are several points

worth discussing. One of the foremost of these points

revolves around the question," How much more does it cost

to operate a DRG system"? Certainly extra costs through

additional employees and computer capacity have been

incurred in implementing this system. May and Wasserman

conducted a study, concluding that an extra $7.23 was

added to each patient's bill so that this additional cost

could be recouped. Also, the State of New Jersey in

developing and administering the system incurred a total

cost of $9.35 million of which $4.7 million was funded by

the federal government. [Ref. 15,pp. 553-5541

Perhaps the most important question that needs to be

answered is, "Does the DRG prospective reimbursement

system result in a more efficient system"? or asked in a

different manner, "Is the DRG system more cost beneficial

than other reimbursement systems"? One study concludes

(though tentatively) that each of the 26 institutions

that started DRG rtimbursement in the first year (1980)

received on the average $2.3 million more than they

otherwise would have received under the preexisting

system [Ref. 15:p. 555). Another study indicates the

rate of increase in per capita hospital expenditures

61

during the first four years under the DRG system approxi-

mates the same trend line as that under the previous

reimbursement system [Ref. 14:p. 44]. Again, it is still

too early to draw definitive conclusions about the

efficacy of DRGs, but what can be concluded is that New

Jersey has succeeded in attaining one its explicit goals,

if not exceeding it, namely, improving the financial

solvency of its inner city hospitals (and also all other

hospitals in the state). However, achieving this program

goal may be at the expense of an improved reimbursement

system, which is effective in containing cost [Ref.

15tp. 557].

Although there are many similarities and numerous

comparatively minor differences between the New Jersey's

and HCFA's Medicare reimbursement programs, there are

several distinct differences, differences that make a

one-to-one comparison difficult on some levels and

impossible on other levels. First, as previously identi-

fied, New Jersey purposely and successfully spread the

costs of inpatient bad debt and uncompensated care over

all third-party payers. Second, New Jersey did this by

requiring all third party payers to pay the hospitals'

DRG rates, which effectively precludes cost shifts among

payers [Ref. 15:p. 551]. Medicare does notl The

significance lies in the fact that those hospitals which

treat a greater percentage of the poor lose in the

62

reimbursement game. That is, Medicare does not reimburse

for bad debts. Additionally, since DRGs and the PPS

apply only to Medicare, hospitals can merely shift

shortfalls for DRG patients to other payers. Having this

"flexibility;" of course, weakens the incentive for cost

containment and efficient management.

Indubitably, the insurance companies will not sit

idly by as these costs are shifted to them. They will

most likely establish "preferred provider" arrangements

usino prospectivw DRG payment eates with individual

institutions or lobby state legislators to adopt a

state-administered all-payer program similar perhaps to

the one in New Jersey (Ref. 15,p. 551]. As hospital

rates continue to escalate and as additional coots are

shifted to other payers (e.g., insurance companies and

U. S. businesses) lobbying efforts will likewise in-

crease. It would seem then inevitable that the federal

political bargaining process could precipitate modifio-

cations to the Medicare program such that it too becomes

potentially less effective as a cost containment

mechanism.

H. OTHER CASE-MIX MEASURES

In addition to DRGu, there are other methods that

employ case-mix measures. Briefly, we shall describe

each of the more widely known caac-mix measures, con-

trasting the differences and comparing the similarities.

63

The reason these "other case-mix measures" are

discussed in this separate section is twofold. First,

HCPA is concerned that it may be overpaying hospitals

with a leS complex patient mix, and second, lobbying

efforts by those concerned with the cctiiiufd financial

viability of larger teaching facilities that treat more

complex illnesses are mounting a persuasive drive for

revisions in their payment rates so that they can recover

the full cost of providing services to such patients

[Ref. 3tp. 15]. The case-mix measures to be discussed

aret Disease Stkaing, Patient Severity Index, Patient

Care Uniti, and CHPA List A

The first to be discussed is diseaxe staging. One of

the often heard criticisms of ICD-9-CM DRG method in that

it fails to account for severity of illness, (i.e., the

more severe a patient's condition the more resources he

consumes in treatment). With the present DRG system

patients can be classified into the same DRG but still

consume conciderably different amounts of resources.

Disease staging, however, groups or clusterv by severity

of illness rather than by length of stay or cost.

Disease staging does use the ICD-9-CM ivedical conditions

as do DRas; it does not use AUTOGRP for formation but

rather a priori professlonal judgments of a 23-member

physician panel [R, f. lip. 461. This classification

systen, an well indicated by its name, normally bases

64

group formation on four stages of a specific organ/body

system, stages which range from conditions of no compli-

cations or minimal severity to death. Table IV lists the

15 body-systems that are used in forming disease-stayed

groups.

Several studies compare the homogeneity of groupings

in disease staging with those in DRGs. Grimaldi and

Micheletti conclude that both of theme classification

systems Axplain a large amount of the sum of squared

differences but that the DRG classification scheme

performs better since it is constructed along statistical

guidelines which minimize the unexplained sum of squared

differences. [Ref. lip. 481

The second type of care-mix measure is the patient

severity index (PSI). Developed by researchera at John

Hopkins University, the PSI also incorporates severity of

illness into its grouping process, requiring evaluators

and raters to review the patient's medical record upon

discharge based upon seven variables [Ref. 3:p. 153:

. stage of principal diagnosis,

* concurrent interacting condition.,

. rate of response to therapy or recovery rate,

0 impairment remaining after therapyt.

° complications of the principal diagnosis,

65

TABLE IV

BODY SYSTEH CATEGORIES FOR DISEASE STAGING

Number ofNuj dUstam Categerlan _ffhLL

1 .i. Diseases of the Skin 25" Diseases of the Nervous System and Cerebral Vessels 57

' 3. Diseases of the Eye 13

Diseases of the Ear, Nose, Throat, and SInuses 33

Respiratory Disease 32

Gastrointestinal Diseases 39

r Hepatobillary and Pancreatic Diseases I Ieli Diseases of the Circulatory System 339, Diseases of the Urinary Tract I I

10, Diseases of Male Senitalia II

I I. Diseases of the Female Reproductive System 35

12. Diseases of the Endocrine System 1913. Hemapoistic and RetIculoendothelial Diseases 23

14. MuscuP osketol Diseases and Traumas 64

15. Newborns and Birth Trauma/Disease

408

sore": PvJ L. OrWWkl ml Jw*b A. M1WlW44H B3 f p,-nv 11k gfdftv Oqb

~B , 0MsapmI: Phru Pr.., II). bdxI4 5-.

66,

" patient dependence on hospital staff and other

resources, and

" nonoperating room life support procedures.

Figure 2-8 represents the PSI nmatLix-type approach

used by evaluators when classifying an inpatient's stay.

A prerequisite for this system is to use trained evalua-

tors, who assign composite severity scores based upon the

seven variables identified above and the relative

severity. Surprisingly, homogeneity of resource consump-

tion in terms of charges, length of stay, and total costs

indicates the PSI method of clustering patients is

superior to any other method. [Ref. lip. 561

The third case-mix classification system to be

discussed is the patient care units (PCUs), which are

based on time-and-motion atudies that estimate the cost

of over 600 clinical services [Ref. lip. 581. These are

somewhat similar to the "resource consumption profile"

codos mentioned in the section on product definition and

matrix organization. Blue Cross of Western Pennsylvania

is developing a similar patient management category

system but uses 50 disease-specific physician panels to

cluster and separate diseases/medical problems [Ref. 3:p.

16]. Additionally, admitting diagnosis is factored into

this classification scheme on the assumption that physi-

cians "treat symptoms and suspected conditions," a fact

67

$UaP of "~get sWrab. M~ mm~twralm cma~qm

Ahp - ( bitf pro* 1 M ~a~ dOUq Uwbwi dt Nonspowl

Cu~~saw or 1W Hemu CAN***h~

#wWWlum vrse a m Al-

hpuimmsqw Lw Medat. Hei

NAWW Thrnpofb Nuwowspm hmwgwy(sm.O*3dlapte or kiwulw lif VUWM, I* i~utai

a -

sm : buab. Hims, F1eb D. Shikag *W ad wW A. B*mm, - au'r" Sewrli of flbm 6-pss

Criteria to Cumtut A Patient Sevesrt. Index

Figure 2-0

68

not often borne out by the pritncipal diagnosis [Ref. lip.

58].

The fourth and last of the case-mix measurements to

be discussed is the CHPA List A. Although there at'e only

* 398 diagn~ostic groupo, there are nearly 8,00, subgroupsf

which are divided 'into f ivie age classes' aid cros:3 tabu-

lated by the jOrevehce or -absence of turgery and the

number (single, or multiple) ok diagnoses [Ref. 1:1,. 58].

*Needless to sny, thii method does not appeal' to those-

interested in casfi-mix measurements because it fails to

be -eas. ily inapaggabI4 and to 'measure explicit resource

consumption.

Figure 2-9 provides a 'jcood synopuis of the Plassi-

f.~cotion schemes discussed in this section as well as

these discussed in previous sections. Even though one

inpatient classification scheme may be superiOL in many

respects to another it does not necessarily indicate it

should be chosen in all applications. There is no

urniversal panacea to case-mix measurement. U~ndoubtedly#

case-mix measurements are in an early stage of develop-

ment and must undergo numerous refinements before one is

heralded as the solution to the cost containment problem,

if ever.

1. CONTROVERSY OVER DRGB: LITERATURE REVIEW

As will be discussed in the next chapter on

Medicare's prospective paymoent system, the Tax Equity and

69

USIA IVAN

"=A0 w IP 100 pel.111 . IU4 IVA~ aSww wey

(~*PhVRW go".H CIO"~ 11110 Nb RU CusF~es* es40. ~ desu WO Ph.owe NIIIII. b~ii Ph&

"60 wm dwo MOM

1,00 Ym14 yo Al" 1e Ano

"8 M"mm futein No" MM" 1111114111 Sg Sim hiu iVol hduii A*hei MII 04 ii

*~~110 Mspiw~iI diwiu beis AINi~ "w"i .euSl

A" N I Aiw" AIt&

aw"p wp im #apse heiAg- PON* II Ageie LSs

A"esgsu lm MOMl s ~ Is~Andu ~ ~ ~ ~~1111 1111001"ey aess ~ ih~ Piine

ft101110llie 11M411 Oes m li Iksgi

ft ml" 4-0 -y y-W o-, yssuiy A

Isa* o ome me'

No NoIFI u es ledNolN

AN i ONAMN Ys-dm IN -M Yses YONee~ Y ph Y w

lM16eft 10,1110418 Ifa IMm WINe eieedss u~i~~Do*"es Alegst Iu lidalees eaeileiu oeh

I~ls~n PhPhPh ii h P P

Omimn121 U

AaMwM1"10*es Ph Is Yes Vol Ye IN PhM YesOdb"Olen-AA~d Ak Ph ast ioAI-fwdw" fwot NNOb fr

Smesiloolhaxe I*siMkmaedI 0"WIVS ~ 006* Ys Ye esP

C~~~WmasnIO 011 be .se

Siedi~INlYNemes

Fundamna Pharateistc osf Seen Patent Clasifiation Schme

igure -9 ~g I eie

d~mimei. iaes. asc ehu70i i

Piscal Responsibility Act of 1982 and the Social Security

Amendments of 1983 took effect on 1 October 1983,

requiring a PPS based on illness-specific conditions for

all Medicare patients. The PPS uses diagnosis related

groups (DRGs) as a classif ication method for these

illness-specific conditions. The off icacy of DRGs as a

meohanism for controlliiig costs and improving resource

use is one of controversy. In this section wo will

examine and analyze recent literature and the various

perspectives on this controvt:sy and comment on what

might lie had for DRGa.

Looking first at the positive comments on DRGs# the

discussio4r fotcuses primarily on the hospitals' ability to

maintain financial viability. Spiegfl and Kairalier note

that Rajani views the DRG system as one that provides a

"pro-market discipline," as well as a "pro-competitive

nature." [Ref. 16:p. 83] Fetter and Freeman see a

dacidedoly positive aspect of DRCa in that they enable

hospitalb, whvther they are not-for-profit or for-pLofit,

to organize themselvcs--otructure and process--in the

mannez of selling a product, Similarly, end perhaps one

of the greatest advantages of the DRG system, is it

provides a direct link bjtween finan*ial duta end

clinical infoumati<,n, allowing much better control over

services (products) and costs to provide those services

71

[Ref. 5:pp. 41-541. The goals of efficiency and

effectiveness are better balancedp with greater emphasis

gIven now to marginal analysis and measurement of the

production relationship of inputs to outputs [Ref. 17:pp.

22-27]. The design incentive of the DRG system is for

hospitals to operate more efficiently within the

prospective reimbursement rates rather than merely

passing costn through and onto Medicare as in the old

program, which used retrospective reimbursement.

Proponents of the DRG system contend that its very

design also requires the board, physician, administrator,

and staff to become more closely aligned in selecting and

pursuing common goals (Ref. 18:pp. 677-679]. This goal

consensus supposee more efficient use of resources and a

streamlined pursuit of agreed-upon objectives. Rather

than working at cross purposes there is a greater

incentive to work together. DRGs "appear" to provide

incentives to organizational participants (including

physicians) to move from what was previously an almost

exclusively effectiveness model to one driven, in greater

part by efficiency. The DRG system allows hospitals and

its many participants to focus concomitantly on issues

dealing wihh efficiency and effectiveness by using a

"common product language" and making tradeoffs between

these two models [Ref. 19:pp. 1-37J.

72

There is only a paucity of physicians and

administrators who maintain the DRG system will

ameliorate the quality of care. Kaemmerer, for instance,

believes DRGs enable physicians to better understand

their practice patterns through comparison with other

hospital physicians [Ref. l8:pp. 677-6791. Riddick

thinks DRGs stimulate physicians to evaluate their

"therapeutic customs and rituals," weighing better

measures of effectiveness against resource use [Ref.

20:pp. 17-18). Potentially, then, a hospital nan

sensitize its physicians to evaluate appropriateness of

care not only in terms of absolute quality but also in

the framework of cost effectiveness.

Though a considerable literature supports the DRG

system, there are critics who vehemently hold that the

DRG system is insidious and, in some instances, outright

nefarious. One of the most outspoken of these critics is

J. A. Meyer, of the American Enterprise Institute, who

believes the DRG system is filled with " . . . excep-

tions, appeals, all kinds of loopholes, 467 categories

that will probably turn into 967 categories . . . (and) .

0 . unfairness . . . " (Ref. 16sp. 821. Because of the

severity differences within the DRG cells, he feels the

system as designed encourages hospitals to "skim the

cream": the design incentives motivate hospitals to

accept the most profitable cases and shun those that are

73

not. Meyer also criticizes the DRGs for being an

incomplete cost control device and for failing to address

admissions, preventive care, and physicians in its

efforts to contain costs. That is, as designed the DRG

system fails to address the totality and, while

potentially optimizing subsystems, it is done at the

expense of the system as a whole. Meyer's position seems

analogous to that of Kerr who believes that the system is

not designed to reward behavior it supposedly seeks [Ref.

21:pp. 769-783). To be so, it would include all relevant

health care delivery subsystems, particularly incentives

for physicians.

As a cost-control method for curbing rising Medicare

costs, the DRG system is seen by its critics to be a

control mechanism that curbs primarily the quality of

care and physician treatment patterns and incidentally,

then only potentially, the cost of care. Again, DRGs

seem to "deincentivize" the provision of optimal patient

care by providing incentives for hospitals to seek the

most profitable DRGs and to cut those that are

unprofitable.

Critics maintain that Medicare (DRG) regulations

contain numerous loopholes and gaps which permit manipu-

lation and gaming of patients' diagnoses by hospitals

trying to maximize their DRG reimbursement (Ref. 22:pp.

74

295-3001. Newhauser vividly underscores this point in a

message he makes to physicians:

How to play games with the DRG payment system maybecome a popular parlor pastime, and even though the"feds" have spent quite a bit of time playing thisgame too and thinking about preventive strategies,they will be only partly successful. Prodictably,their lack of success will set the stage for stillanother approach to payment. [Ref. 16:p. 861

Other criticism is levied at the DRG system Ror

discouraging Jag g capital investments, as fixed DRG

rates do not permit this cost to be directly passed on

and borne by the Medicare program tRef. 16ipp. 86-871.

Regarding goal consensus and interactions among the

board, administrators, and physicians, critics believe

the DRG system will create an even greater adversarial

relationship [Ref. 20tpp. 17-18]. Hospital adminis-

trators are largely motivated by the efficiency model

while the physician is motivated by the effectiveness

model. In fact, Bird thinks the physician's individual

incentives are unaffected by the DRG system [Ref. 16:p.

87]. Since the DRG system as designed fails to reward

desired physician behavior, hospital administrators may

well find themselves countinuing in the role of cajoling

medical staff support in an effort to elicit desired

behavior.

Considerable opposition is found in the literature to

the method in which DRGs , re formed. Critics contend

that homogeneity of patients is impossible and that there

75

Is no such thing as the "average" patient (Ref. 16:p. 88]

[Ref. 23:pp. 1195-11991. Moreover, many express concern

that DRGs fail to reflect vaiiations in resources

consumed and in diseaso/illness levels [Ref. 23:pp. 1195-

11991 [Ref. 8ipp. 388-3961. Others contend length of

stay is not an accurate reflection of resources consumed,

or of the costs incurred. Hughes, in a letter to the

Aals.SL Internal Medicine, attacks the DRG system on

the grounds that, "There is a distinct failure to

identifying multiple complications or comorbid condition3

in individual patients." [Ref. 16:p. 89] In other

words, hospitals are reimbursed for only one condition

per patient, regardless how many might be treated.

The critics really lambaste the DRG system for the

perceived affect it may have on lowering medical stan-

dards and in limiting the pursuit of teonnological

advances [Pef. 25sp. 76]. Reimbursement calculations

fail to adequately cover technological progression and

provisions for innovations. In fact, a Presidential

report indicates scientific advances are "likely" to be

stifled [Ref. 26sp. 25-26] [Ref. 16:pp. 87-911.

Depending on where one stand., one can make a

defensible and cogent argument for cr against the DRG

system. For exasnple, on the issue of a more sophisti-

cated and complete dat3 base as a natural extension of

DRG management, the critics say the costs are prohibitive

76

and the data itself of questionable value; whereas, the

DRG advocates indicate the data base ensures accurate

record keeping and insight into hospital activities far

beyond today's capabilities [Ref. 20tpp. 17-18).

The debate over the efficacy of DRGs will continue

for some years. As conclusive evidence becomes avail-

able, and as shortuomings are detected, refined proce-

dures are and should be devised to make the DRG system

both more efficient and effective, such as incentivizing

physicians to demonstrate desired behavior, incorporating

a severity of illness measurement criterion, and modify-

ing Medicare regulations to permit reasonable techno-

logical advances and modest capital expansion. Moreover,

all facets of the health care delivery system must be

incorporated into the analysis in order to obviate

suboptimization of the system for what might be optimal

subsystems.

77

I11. RE! BURSEMEn AIS aXa.H.H_

A. INTRODUCTION

The discussion of Public Law 98-21 and the Medical

Expenoe &nd Performance Reporting System for Fixed

Military Medical arid Dental TreaLment Facilities (MEPR)

presented below is not oli inclusive. Rather, it

attempts to provide a fundamental framework with which

the reader unfamiliar wlt.n PPS and MEPR will be better

prepared to understand the analysis and findings

contained in his thesis.

B. MEDICARE'S PROSPECTIVE PAYMENT SYSTEM (PP8) UNDER

PUBLIC LAW 9a-21.

On April 20, 1983, President Reagan *3ignqd the Social

Security Amendments of 1983 (P.L. 98-21). Title VI of

this law, which applies to all short-torm acute care

hospitals, modifies the traditional retrospective method

1edicare uses to roimburse hospitals. P.L. 98-21

replaces retrospective coat-based reimbursement with a

prospective payment system. The PPS builds on methods

and procedures used to estaolish case-mix indoxis, cost

weights, and target ceilings under the Tax Equity and

Fiscal Responsibility Act (TEFRA) of 982 (P.L. 97-248)

[Ref. 1:p. 99]. UkliKO the cost-per-case limit

established by TEFRA, the provisions of P.L. 98-211

78

1. sever the traditional relationship betweenactual costs and the revenues generated byproviding inpatient care to Medicarebeneficiaries;

2. constitute the basis for establishingprospective prices for each of 470 DRGs;

3. assign financial risk for unfavorable variancesbetween the cost of providing care and thepredetermined DRG price to the hospitall and

4. permit the hospital to retain favorablevariances between the cost of providing thecare and the corresponding predetermined DRGprice. [Ref. 3:p. 81

Starting with cost-reporting periods after September

30, 1983, hospitals are paid prospectively-established

rates for Medicare patients discharged from participating

hospitals. Eventually, with certain exceptions, this new

payment system mandates paying the same DRG rate to all

participating hospitals.

1. DRG Payment Determination

The basis of hospital reimbursement under the PPS

is the disuharge diagnosis of the particular patient.

The payment for each DRG is established on the basis of

three sources of data: the Medicare cost report, the

Medicare dispcharge file, and the MEDPAR file. The

Medicare cost report contains the cost information that

hospitals submit to fiscal intermediaries in order to be

reimbursedi for care provided to Medicare patients. The

Medicare discharge file indicates the number of Medicare

patients admitted to a hospital in a given year. From

29

these two sources, the HCFA determines a na.iU)nal average

cost per discharge. The HE4DPAR file is a 20 percent

sample of Medicare patient bills from short-stay

hospitals. It is used to create the DRG cost weights,

listed in Appendixes C, D, and E, and case-mix measures.

These measures indicate the relative costliness of

providing care for different Medicare patients in

relation to the average cost per patient (Ref. lip. 1151.

For example, if the cost per patient l.EDPAR file

indicates that the care of a patient in DRG 125 is 1.2

more costly than the care of the averagu Medicare

patient, the DRG cost weight is 1.2. If t he national

average cost per Medicare discharge is $1,000, tbe

hospital would be reimbursed for the care oZ a patient in

providing service in DRG 125 at a rate-of $1,000 x 1.2o

or $1,200. The steps used by the PCFA to calculata 1RG

cost weights and case-mix mbasureu are illus3t ted in

Figure 3-1.

2. Transition Period

Congress provided a three-year phase-in period so

hospitals would have an opportunity to adjust to the

prospective system. During this transition period

composite DRG payment rates . re established for each

hospital participating in the Medicare program. Tnic

rate is unique for each hospital during the transition

period, but after FY 86 a standard national p&ynent rete

will be used to reimburse all hospitals. The composite

80

Ii!

iii

861

ILU~

81N

DRG payment rates, as illustrated by Figure 3-2, are made

up of a federal portion and a hospital-specific portion.

The federal portion As made up of regionel and national

average rates, which take into account whether the

hospital is located in an urban or rural area. The

hospital-specific portion is derived from unique hospital

cost characteristics computed on a base year. In most

cases, this base year is 1981. As the health care

industry proceeds through the transition period,

increased emphasis is placed on the federal payment

amounts, with decreasing emphasis on a hospital's base-

year costs.

PERCENTAGES USED TO CALCULATE PROSPECTIVE RATESp 1963-1986

Medicare FederalCost Reporting Hospital-Specific Regional National

Period Portion

1983 75.0 25.0 0.01984 50.0 37.5 12.51985 25.0 37.5 37.51986 0.0 0.0 100.0

For example, in the first year of the transition period

(the fiscal year beginning October 1, 1983 (FY 84), and

ending October 1, 1984) 75 percent of the payment rate

for an individual hospital was based on a hospital's

TEPRA target amount (hospiL l-specific portion), which is

82

adjusted for inflation by the hospital market basket

increase plus a one percent technology factor. The

rbmaning 25 percent is based on a regional DRG rate

(feieral portion). In the second year of the transition

this ratio changes to a 50/50 split between the hospital-

specific portion and the federal portion. The federal

portion is divided between a regional rate and a national

rate on a 37.5/12.5 respective basis. In year three, the

ratio changes to 25/75 with the federal portion being

equally divided between the regional and national rates.

In the final year of transition (beginning October I;

1986) , 100 percent of the payment rate is based on a

national rate rRef. 2sp. 20].

3. Calculation of Prospective Payment Revenue

To determine "who gets what for services

rendered" one must consider the following factcrs:

a. Adjusted Federal Standard Rate,

b. TEFRA Target Amount (hospital-specific),

c. DRG relative cost weights,

d. Regional Wage Indexes, and

e. The ICD-9-CM DRG to which the patient hasbeen assigned.

In general, as illustrated in Figare 3-2 and

Table 7, the federal standardized and hoapital-specific

anounts are combined to calculate an overall average

payment for a hospital. The overall rate is then

83

IAI%A ILu,

00

*l

I! 0

CS 1 4

va ~ .PEI

if CW

84

ZIP

CCLA

w -c

4.A ; No6

_ _ _ 4

- 06

1* I 85

TABLE V

CALCULATION OF PAYMENT RATE FOR DNS 125, URBAN HOSPITAL X.WEST SOUTH CENTRAL REGION: FISCAL YEARS (FVs) 1964- 196

(Roumnied In Dollars)

Tw,0 pn"IM r-Aai ,2

-- m L MMwopttul mmP0 02X4 $1I

stadr mm o r urn-

-x- -- S pS9TwM parent Sc owx

Pb'fewtw m .2.3

boad~$1 AMlL

NatIMAdparim:Standad smu- h lar b-

rolathSm 02,= 0,2M6 0,2M6Yap him X1,J11 .L1U19 X1,11J9

02,40 02,40 UAWStwad" onse hr owr

ns0 -xm wiVeS*Art * -rnU mis ma7

NM.' IOe Isw owls wte nydia Smav FY 134. Target paent aqmib tt. plm arenwtg

Oweo: wampiw -Pe op tar InM.

86

adjusted by the hospital's regional wage index. Next,

the overall average is multiplied by the appropriate DRG

cost weight to obtain the payment for a given DRG.

Once the appropriate DRG discharge rates have

been determined, one can calculate the hospital's total

Medicare inpatient revenue. As demonstrated by Figure

3-3, the total Medicare inpatient revenue is made up of a

DRG inlier portion, an outlier portion, and a allowable

cost portion. To calculate the total Medicare revenue

for a hospital one simply sums the three cost inputs.

Total Inlier Outlier AllowableMedicare - Portion + Portion + CostRevenue Portion

The inlier portion is that part of Medicare

revenue which is included in the composite DRG payment

rates discussed earlier. The total inlier portion is

obtained by determining the appropriate rate per

discharge in the hospital and then summing the results

[Ref. l:p. 108].

Total Inlier - (Federal + Hospital-Specific) x DRG WgtRevenue Portion) Portion)

The outlier portion, which ii reimbursed un a

retrospective basis, is that part of Medicare revenue

which compensates hospitals for patients with unusually

87

rOR

relativecotweihts

wo~his

Amount The TQta1 DRO- Capital re- Totalhospital is paid rlated lated sts Modiars

National and fo ravenue a Diet md- inpatiOentregional particular DORG 4 payMont o41 education revenuestandtardild * #argot rates oosaWiWits * wage index t*Urns -0NonPhldsicicnper Medioare * COLA discharges anesthetistdischarge # DRO price + servies

index *outlier * Indireot MIDpayments education

costsOther costs

Nospital-speciflo

cose-mixadjusted amountper Medicaredischarge

fouroe: Paul L. Orimaldi and Julie A. Micheletti, Prnp. fnve Paument ThAEf4i4 Ou4d to Eewu.mamen, (Chiago, Ill,: Pluibus Press), 1995,[xhibit 5-1I1

Medicare Inpatient Revenun, FY 1g 5

Figure 3-3

88

long (day outliers) or costly (cost outliers) stays for a

particular DRG discharge. Day outliers are patients

whose length of stay exceeds the average (mean) stay for

a DRG discharge by 22 days or 1.94 standard deviations,

whichever results in a smaller number of days [Ref. lp.

1081. Cost outliers are patients whose charges adjusted

to costs exceed the DRG payment rate by the larger of

$13,000 or double the relevant DRO payment rate [Ref.

lip. 1093. To obtain the total outlier revenue, one

simply adds together authorized day and cost otitlier

costs.

Total Outlier - (Day Outlier + Cost Outliers)Revenue

The last source of Medicare revenue results from

allowable costs excluded from a hospital's prospective

payment rates [Ref. 27:p. 20007]. These costs are

reimbursed on a retrospective reimbursable basis and

include the following:

* Capital-related Costs

* Direct and Indirect Medical Education Costs

* Nonphysician Anesthetist Costs

* Bad Debts

* PICA Costs

* Part B Costs

89

A more complete discussion of these special costs, as

well as the cost foundation on which the federal and

hospital-specific rates are based, is presented in

Appendix A.

C. THE DOD'S MEDICAL COST ACCOUNTINC REPORTING SYSTEM

1. Uakron

The evolution of the MEPR as a viable reporting

system has evolved from "any reporting system is better

than none" to a reporting system linked directly to

expenses incurred by military treatment facilities

(MTFs). Prior to the development of the MEPR the

services primarily used two surrogate measures of output

to report performance: (1) occupied bed days and (2)

composite work units. Even though these two output

measures might appear to be viable performance reporting

mechanisms, they did little to aide managers in

determining "how well the job got done." In 1975, as a

result of dramatic health care cost escalations, coupled

with an inadequate management information system in DOD

MTFs, a new reporting system called the Uniform Chart of

Accounts for Military Medical Treatment Facilities (UCA)

was developed by a tri-service health care study group.

The primary objective of UCA wa's to establish a

management information system that standardized cost and

performance reporting through the use of fundamental (#

90

of procedures) , derived (cost per visit), and fiat

(depreciation) measures. Since its implementation in the

fall of 1978, UCA has remained intact with only minor

revisions. Today, however, its name has been changed to

the Medical Expense and Performance Reporting System for

Fixed Military Medical and Dental Treatment Facilities.

2. The Medical Expense and Performance Reortingsystem for Fixed Military Medical and DentalTreatment Fagilitie

The information contain in this section was

primarily extracted from the MEPR Manual, DOD 6010.13M;

therefore, the discussion of the MEPR is best done in the

context of a review of DOD 6010.13M [Ref. 28:p. 5-17].

The MEPR manual is composed of five chapters with each

chapter, other than chapter one, representing integral

elements of the uniform reporting system. The chapters

are titled as follows:

Chapter 1 - General

Chapter 2 - Chart of Accounts

Chapter 3 - Manpower and Expense Assignment

Chapter 4 - Issues System

Chapter 5 - Reporting Requirements

a. Chart of Accounts

"Chart of Accounts" is the heart of the MEPR

manual. Within this section of the manual a hierarchy of

accounts have been constructed wherein all expenses and

91

corresponding workload data are grouped into six

functional categories:

1. Inpatient Care

2. Ambulatory Care

3. Dental Care

4. Ancillary Services-

5. Support Services.

6. Special Programs

Inpatient Care is defined as health care

which provides for the examination, diagnosis, treatment

and proper disposition of inpatients. This functional

category is a summarizing account that accumulates all

inpatient operating expenses. It represents the total

cost of inpatient care delivered in the MTF.

Ambulatory Care provides for the care,

consultation, examination, diagnosis, treatment and

disposition of both inpatients and outpatients treated by

the various ambulatory care clinics. Like the inpatient

care category, it is a summarizing account. It

represents the total cost of ambulatory care.

The Dental Care functional account includes

all the operating expenses incurred in operating and

maintaining a dental center, a dental clinic, or a

prosthetic laboratory.

Ancillary Services are defined as those

services that participate in the care of patients by

92

assisting and augmenting the physicians and dentists in

treating human ailments.

Support services are those services that are

necessary to direct and support the mission of the

medical facility. This account is somewhat like an

overhead account in a manufacturing firm. It summarizes

all operating expenses for support services, including

depreciation.

The last functional category, Special

Programs, represents those activities performed to

support the MTF's military mission rather than direct

patient care.

Functional categories represent the broadest

category for aggregating costs and they appear highest on

the accounting hierarchy. Each of the functional

categories is further divided into summary accounts and

subaccounts. The subaccounts are accumulated into their

corresponding summary account. An example of this

hierarchical arrangement appears:

Level I - Inpatient Care (Functional Category)

Level II - Medical Care (Summary Account)

Level III - Internal Medicine (Subaccount)

There are four elements that are generally

common to each MEPR account regardless of the level of

the hierarchy. The first element is termed "function."

The function contains a description of the type of

93

activity characteristic of the particular account. The

second element is entitled "costs." This element

identifies the expenses that shall be included in the

account. "Performance factor" is the third element of the

account, and it identifies the uniform workload measure

which is to be collected and used for evaluating or

gauging performance. The final element is the

"assignment procedure." This elements establishes the

basis under which the account cost will be reassigned if

applicable. [Ref. 28:p. 811

Having knowledge of the chart of accounts

structure and common generic elements, facilitates one's

understanding of the flow of expenses in MTFs. In brief,

each element of expense generated within the MTF is

assigned to a particular subaccount (work center). The

sum of the expenses in each subaccount represent the

total expenses for each summary account, and the sum of

the expenses in each summary account represents the total

expenses for each functional category. The functional

categories of Inpatient Care, Ambulatory Care, Dental

Care, and Special Programs constitute final operating

expense accounts, which are the final expense

accumulation points in the systems. Ancillary Services

and Support Services accounts are intermediate operating

expense accounts whose expenses are reassigned to one of

94

the final operating expense accounts through the use of a

stepdown allocation process.

b. Manpower and Expense Assignment

The purpose of the "Manpower and Expense

Assignment" is to transform manpower, expense, and

workload data collected by work centers into meaningful

management reports. It has the objective of defining a

basis for distributing the accumulated costs and work-

months to the direct patient care and Special Program

accounts. In other words, through the use of a

sequential stepdown cost allocation process all

subaccount, summary and intermediate expenses are placed

in the final functional account responsible for incurring

the expense or using the manhours.

The stepdown assignment methodology requires

five sequential steps to be taken. They ares

1. manpower and data collection and processing,

2. assignment of expenses and workload recording,

3. pre-stepdown purification of expenses,

4. assignment of expenses to final operating expenseaccounts, and

5. post-stepdown purification of final operatingexpense accounts.

At the manpower data collection and

RrocsQig stage two substeps are performed. First, one

establishes what amount of full-time equivalent (FTE)

work months are to be charged to each account. Next, one

95

determines the appropriate military personnel expense and

the command, management, and administration expense each

account should be charged.

Civilian personnel salary expenses for the

command is calculated on a monthly basis. This expense

consists of the amount of funds obligated due to the

employment of each employee during a month. It includes,

but is not limited to, basic salary, incentive and

hazardous pay, government contributions to benefits,

overtime, and termination pay. The salary expense for

each employee is charged to the appropriate account based

upon the distribution of FTE work months determined in

the preceding paragraph.

Military salary expenses are charged in the

same manner as civilian salary expense. The amount of

expense to be distributed for each military member is

derived from the DOD Annual Composite Standard Rates

Table, which is published by the Office of the Assistant

Secretary of Defense (Comptroller) . The amount to be

charged to each account is derived by multiplying the

standard rate for a member's grade and military

department times the allocated FTE work month. A more

detailed discussion of the distribution of FTE work

months and salary expense are provided in the MEPR

Manual.

96

The a of ex1__z and workload

recQ.di±. has three phases. The first phase consists of

assigning all non-personnel expenses to the intermediate

and final operating expense accounts. These expenses

come from the DOD Operation and Maintenance Appropriation

(OM&N), and they are usually related to program element

eight, "Care in Defense Facilities." However, any

expenses originating from other DOD program elements that

are incurred in direct support of a MTF are also

included. With the exception of indirect expenses, all

non-personnel expenses are accumulated and summarized in

the MTF't job order accounting system. Indirect expenses

are ellocated to indirect cost pools when it is difficult

to identify the work center responsible for the incurring

the exptnse. These cost pools may include both personnel

and non-personnel related expenses.

The second phase of expense assignment deals

with depreciation expenae. As stated in the MEPR Manual,

the costs for modernization and replacement of investment

equipment is funded from Other Procurement Navy

Appropriation (OPN) when costs are more than $5,000 and

directly support a MTF. Depreciation is on a straight

line basis using an eight-year moving average. In

addition, the manual states these costs will be treated

as an indirect expense during the stepdown reassignment

97

process rather than as a di:ect expense at tho time of

acquisition.

The fin a l phase of expense assignment

involves the compilation of the performance data. Such

information is necessary for the assignment of

intermediate operating expense accounts and indirect cost

pools to final operating accounts.

The third step, gre-stepdgWn purification of

exan ,est allocates expenses not previously allocated in

steps one and two. These expenses are allocated to

Support Services and Ancillary Services accounts,

provided there is no overhead included in the expense.

If overhead is included in the expenses, these expenses

are not allocated until one reaches step four in the

assignment process. Upon completion of step three,

performance data for each operating expense account and

expense applicable to the operation of the MTF have been

compiled.

The next step, asJiLment of exggenses. to

Uin.1expense accgunts, involves the reassignment of

expenses from intermediate operating accounts (Support

and Ancillary Services) and indirect cost pools (wards

and clinics) to the final operating accounts. The result

of this process is the identification of direct patient

care expenses by subspeciality work centers and special

programs.

98

The stepdown process begins with the

allocation of expenses that have been assigned to the

intermediate operating expense accounts. These expenses

are allocated to other intermediate operating expense

accounts and final work center subaccounts in which

services were rendered. The prescribed allocation

sequence and awsignment of these expenses is outlined in

the MEPR Manual. In general, however, the intermediate

operating accounts that render the most srvices to other

center (intermediate and final operating expense

account6) are assigned Clrst, and the intermediato

accounts that receive the most services from others are

assigned last.

The assionmgnt of indirect cost pools is the

next phase of the fourth step. Indirect cost pools are

pseudo-final operating expense accounts in that they have

assigned to them the expenses from all Support Services

accounts, except depreciation. These expense are

assigned to the appropriate work center accounts based on

a ratio of workload generated by each receiving account

to the total workload of the indirect cost pool. After

completion of this step, only the subaccounts of the

final operating accounts contain exper,se data.

Step five, post-stepdown purification of

final operatina expense accu&tj, reallocates final

operating expenses based on the performance factor or

99

other unit of service outlined in the MEPR Manual. In

some cases, a MTF can reallocate these expenses based on

some consistently applied local costing practice. Upon

completion of the fifth step, the assignment of expenses

and workload recording, expenses contained in each

account can be aggregated into its appropriate summary

accounts and functional categories.

c. Reporting Requirements

While there are eight reports created from

the MEPR process, the primary vehicle used by activities

to determine "how well they have gotten the job done" is

the DOD Medical Expense and Performance Report. It

provides managers with aggregate expense and workload

data in three general areas: inpatient care, ambulatory

care, and special programs.

100

IV. DESCRIPTION OF THE ANALYSIS

A. DATA

Data used in this thesis were provided by a Tri-

Service DRG Study Group at the U.S. Army Health Care

Studies and Clinical Investigation Activity, Fort Sam

Houston, Texas, and by the Naval Medical Command,

Washington, D.C. The sample population selected for

analysis consists of three naval hospitals: Charleston,

Long Beach, and Pensacola. The operational bed capacity

of these three naval treatment facilities (NTFs) was 223,

166, and 135, with Charleston being the largest naval

hospital and Pensacola being the smallest NTF. These

three hospitals were selected as the sample population

because:

* They have only minimal teaching responsibilities, ifany;

* They aro located in urban areas;

* Their beneficiary population appear similar; and

* The number of inpatient discharges at each NTF isrelatively stable from year to year, yet offersomewhat different relevant ranges of activities.

There are two primary categories of data used in this

thesis: (1) biometric data (inpatient discharges) and (2)

expense data. The biometric data were provided by the

Tri-Service DRG Study Group and the expense data by the

Naval Medical Command. The biometric data contain

101

information for 60,408 inpatient discharges over a two-

year period (FY83 and FY84) for three hospitals. The

primary characteristics of each inpatient discharge were:

* Principal Diagnoses

* Secondary Diagnoses

* Bed Days

* Age

* Discharge Status

* Disposition Date

* Type of Admission

* Sex

* Disposition Code

* Military Treatment Facility

The authors also gathered information pertaining to

inpatient discharges from the DOD's cost accounting

reporting system, called the Medical Expense and

Performance Reporting System for Fixed Military Medical

and Dental Treatment Facilities (MEPR) in order to check

the accuracy of the inpatient biometric data.

The historical financial (expense) (ata were also

drawn from the MEPR. In total, the MEPR produces eight

cost accounting type reports. The primary report used in

this thesis to determine "how well NTFs got the job done"

was the Medical Expense and Performance Report. The

primary category of information drawn from this report was

the amount each NTF expended for inpatient care in its

1.02

facility. Although we had access to four fiscal years

(FY82 through FY85) of expense data and had hoped to

conduct this analysis using three years of data, we only

used cost accounting information from two fiscal years

(FY83 and FY84). The reason for limiting the analysis to

two fiscal years instead of three, revolved around the

problem of attaining accurate and complete FY85 biometric

data. The authors decided it would be better to have two

years of data that were complete and accurate than have a

third year of data that consisted of incomplete and

inaccurate biometric data, which might lead to erroneous

conclusions.

B. RESEARCH METHODOLOGY

Before discussing the specifics of our research

methods, let us first briefly describe our research

hypothesis. Initially, we had hoped that we might be able

to support the hypothesis that NTFs were operated more

efficiently than &jMJjAt civilian facilities.

Unfortunately, two facts were borne out as we progressed

with the analysist First, there are no similar civilian

hospitals, that is, the organization--process and

structure--of these two types of hospitals (civilian and

naval) is markedly different. In fact, so different that

a vis-a-vis categorical comparison between% types of

facilities is ostensibly impossible. Second, military

103

treatment facilities, particularly NTFS, have unique

mission-driven operational and tactical requirements that

effectively preclude a categorical one-to-one, civilian-

to-naval hospital comparison of relative efficiencies.

riso, admittedly (and perhaps somewhat presumably) the

level of effectiveness (quality of care) between these two

types of facilities is treated as if it were similar.

As we progressed with our analysis we realized that,

despite the uniqueness of NTFs and their seemingly

incomparable differences with civilian hospitals, there

are striking similarities: They both use manpower,

facilities, equipment, and supplies in a transformation

process that provides products, which consist of a group

of services, to patients. Accordingly, rather than making

a categorical statement that these types of facilities are

similar and that one is more or less efficient than the

other, we decided to test in a rather direct and

fundamental manner the research hypothesis that NTFs'

inpatient care expenses are less than the funding which

civilian hospitals would have received under Public Law

98-21.

The method used to test the null hypothesis that NTFs'

inpatient care expenses were greater than or equal to the

funding levels a civilian hospital would have received

under P.L. 98-21 consists of four essential steps:

104

1. Determine the number and type of DRG discharges ineach NTF for each fiscal year:

2. Determine the aggregate funding that each NTFwould have received if they were being reimbursedunder the parameters of P.L. 98-21;

3. Determine each NTF's actual inpatient careexpenses for each fiscal year: and

4. Compare these actual expenses to the constructedMedicare reimbursement.

Step one, determining the number and type of DRG

discharges in each NTF, was primarily accomplished by the

Tri-Service DRG Study Group at Fort Sam Houston, TX. The

authors requested that historical biometric data

(inpatient discharge information) for each of the three

naval hospitals be provided for FY83 through FY85. As

req,.iested, the Study Group's senior statistician compiled

the information, assigning inpatient discharges contained

in this data to appropriate DRG categories. After each

inpatient discharge had been assigned to the appropriate

DRG category, we determined the frequency of each DRG

discharge and the total DRG workload. This was

accomplished through the use of two computer software

packagea. We used SPSS-X to determine the frquency of

iAch DRG, and Lotus 1-2-3 to ascertain the total DRG

workloal for each NTF. The results of step one are

presented in Appendices D and E.

In order to accomplish step two, determininc. the

amount of funding NTFs would receive if they were being

105

funded on the basis of the parameters of P.L. 98-21, we

needed three essential pieces of information. First, we

needed to know the frequency and grand total of each DRG

discharge in each NTF for each fiscal year under analysis.

This information was provided in the first step of the

research method. The other two pieces of essential

information needed to test the null hypothesis--HCFA cost

weights and the national federal payment rate--were

obtained from the Federal Register. A detailed discussion

of HCFA cost weights and national federal payment rates is

contained in Chapter III and Appendix A. The cost weights

used in our thesis are presented in Appendices C, D, and

E. The national federal payment rate used was $2837.91

per DRG discharge, as can been seen in our calculations in

Appendices D and E.

The technique used to determine the revenue NTFs would

have received under P.L. 98-21 is the reimbursement method

that Medicare mandates to be used by all health care

providers after fiscal year 1986. This technique uses a

national standard payment rate, also known as average

adjusted cost per discharge, in its calculation of

Medicare reimbursement without regard to hospital-unique

cost characteristics. As explained in the prospective

payment section in Chapter III, P.L. 98-21 establishes a

reimbursement method to be used in the transition years

(FY84, FY85, and FY86), and a method to be used in the

106

years following the transition period. The former method

uses a payment rate that is made up of a hospital-specific

portion, a federal-regional portion, and a federal-

national portion, and was determined to add little to this

research. Therefore, we decided that, without loss of

generality, we could use the reimbursement method

specified for FY87 and subsequent years. To determine the

amount of revenue a hospital would have received using the

procedure outlined in step two, one must perform the

following procedures:

(a) Determine the total number of inpatient dischargesin each DRG category;

(b) Multiply the total frequency of each discharge bythe cost weight for that ORG category;

(c) Multiply the results of procedures (a) and (b) bythe national federal payment ratel and

(d) Sum the results of procedure (c).

TOTAL Number of DRG NationalMEDICARE - DRG X Cost X FederalREIMBURSEMENT4 Discharges Weight Rate

The third step, determining each NTF's actual

inpatient care expenses for the two fiscal years under

study, required the extraction of expense data from the

MEPR. Again, the MEPR is an expense-linked cost

accounting system, which standardizes cost and performance

reporting through the use of fundamental (# of

107

procedures), derived (cost per visit), and fiat

(depreciation) measures.

The primary data drawn from the MEPR was the total

inpatient care expense, clinician salaries, and the number

of dispositions each NTF reported in its quarterly report,

the Medical Expense and Performance Report. As discussed

in Chapter III, inpatient expenses reported in the

quarterly report represent the total cost of inpatient

care delivered in a NTF. This being the case, we removed

clinician (physician) salaries (both military and

civilian) from the aggregate inpatient care expense totals

since physician salaries are not usually included in the

standard Medicare reimbursement rate. By backing out

clinician salaries, we improved the relevance of our

analysis in making comparisons of military and civilian

health care data.

In additioa to the adjustment made for clinician

salaries, we also normalized FY83 inpatient care expenses

to FY84 expense levels. This was done by multiplying the

FY83 inpatient care expenses, less clinician salaries, for

each NTF by nine percent, the growth rate of health care

expenditures fov that year [Ref. 29:p. 30]. The authors

utilized this procedure so that FY84 cost'weights and the

federal reimbursement rates could be applied to FY83

expense data. Again, since there were no cost weights or

federal reimbursement rates established for FY83, we

108

utilized this procedure so that FY84 cost weights and the

federal reimbursement rate for FY84 could be legitimately

applied to FY83 expense data.

The final step of the research method, comparing

actual expenses to the funding NTFs would have received

had they been funded on the basis of the parameters of

P.L. 98-21, was accomplished through the use of the Lotus

spreadsheet software program. The results are illustrated

and discussed in the last section of this chapter.

Once the primary research question had been answered

we extended the use of our research methodology to a

comparison of inpatient care expenses with those in

Veterans Administration facilities (VAFs). We used the

same methodology as discussed earlier with one exception.

We used a standard payment rate of $2775.00, the average

adjusted cost per discharge in VAFs in FY84, instead of

the national Medicare reimbursement rate of $2837.91 [Ref.

30:p. 25]. The method used to determine the VA average

adjusted cost per discharge appeared to be consistent with

the method used by Medicare. The VA average adjusted cost

per discharge included direct, indirect, and education

expenditures and stipends paid to residents [Ref. 30:p.

25]. Also, the VA removed physician salaries from the

average adjusted cost per discharge.

Finally, in order to determine how similar inpatient

diocharges were in the sample population, we compared the

109

thirty most frequent DRG discharges in each NTF. We also

compared the NTFs' and the State of California's thirty

most frequent DRGs, using the same methodology.

C, FINDINGS

The results of the analysis are divided into three

parts: (1) A comparison of inpatient care expense levels

in NTFs to Medicare reimbursement levels civilian

hospitals would receive under the parameters of P.L. 98-21

for the workload performed in the NTFs; (2) A comparison

of NTFs'and VAFs' inpatient care expenses per fiscal year;

and (3) an analysis of the similarity of DRG discharges

between each NTF.

1. Comparison of Naval Treatment Facilities ExpenseLevels to Medicare Reimbursement Levels

First, consider the summary results of our

analysis presented in Tables VI, VII, VIII, and Figures

4-1, 4-2, 4-3, and 4-4 (and supported in detail by

Appendices D, E, and F). In reference to the figures,

"unfunded workload" indicates the difference between what

Medicare would have paid and what the NTFs actually

expended to provide inpatient care. "Funded workload then

is what the NTFs actually expended to provide inpatient

care. Our analysis indicates that annual inpatient

operating expenses in the sample population of NTFs is

notably 1ess than the funding they would have received

under Medicare.

110

Charleston Navel Hospital FT1983

funded vwr*bo

Long Beach Novel Hospital FYI 915314.96M1

Minded Fundd kld

Pensacola Navel Hospital FY 1983

FPWin WrhbadPercentage of Funded and Unfunded Workload for Charleston,Long Beech, amd Pensacola Naval Hospitals for FY 1903

Figure 4-1

113.

Charleston Novel Hospital FY 194

Ikiasd vwkbd

~~~~Long Beach N scl ovel Hospitals For1FI94

2 i.1 0-2

000 ' 6 .

J. 11 A 11 0

Long Bach on Peacl Novel Hospitals for FI 9

Figre4-

LM1~mdud112

Charleston Noval Hospital FVsB3/a4

Long Beach Navel Hospital M~OM/0

UROAWWVffkl*

~j~jg~ w dbW

#1111

PenaclaNovl osita Fs8/8

m..'l

1131~ll

Total Percentage of Funded and Uffunded Workload

Aggregate Percentage of Funded and Urnfunded Workload forCharleston, Long Beech, end Pensacola Navel Hospitals forF~s 63/84

Figure 4-4

114

As depicted in Table VI, if one uses Medicare

reimbursement rates, HCFA cost weights, and DRGs to

determine annual inpatient funding for Charleston, Long

Beach, and Pensacola naval hospitals in FY83 and FY84,

collectively they would have received an additional

$36,145,661. The aggregate MEPR inpatient expense total

for the three NTFs is $77,341,480, which is 31.9 percent

below the Medicare reimbursement level of $113,487,141.

In FY83, the aggregate inpatient expense total

(illustrated in Table VII and Figure 4-1) for Charleston,

Long Beach, and Pensacola naval hospitals is 30.1 percent

($.6,913,745) below Medicare reimbursement levelsj the

FY84 total inpatient expenses (illustrated in Table VIII

and Figure 4-2) is $19,231,916 (33.6 percent) below that

year's Medicare reimbursement level of $57,289,691.

As shown in the previous illustrations, in both

FY83 and FY84, Naval Hospital Charleston has the greatest

difference between MEPR expense and potential Medicare

reimbursement lavels. If one uses the parameters of P.L.

98-21 to fund Charleston, it would receive an additional

$9,870,566 in FY83 and an additional $9,295,626 in FY84.

Charleston is followed by the Naval Hospital Pensacola

with an unfunded workload of 30.8 percent ($4,375,815) in

FY83 and 35.1 percent ($5,151,253) in FY84. The Naval

Hospital Long Beach has the smallest difference between

MEPR expense and Medicare reimbursement levels

115

TABLE VI

SUMMARY OF FYs83/84 AGBREBATE INPATIENT IlEPR/IIEDICARE DATAFOR CHARLESTON, LONG BEACH, AND PENSACOLA NAVAL HOSPITALS

K rjsu L on BU.ffl Pgael AggrgatFYuI3/B4HEPR $20,507,786 $29,444,247 $19,389,447 $77,341,480

FYBB3/04HEDICARE $47,673,970 $36,096,645 $20,916,510 $113,407,141

FYuS3/84UnfundedWorkload ($19,166,192) ($7,452,390) ($9,527,071) ($6, 145,661)

FV8i3/04PercentUnfunded 40.209 20.209 32.95X 31.853

116

TABLE VII

FY1903 INPATIENT MEPR/IMEDICARE DATA FOR CHARLESTON, LONGBEACH, AND PENSACOLA NAVAL HOSPITALS

RE cl.lma LalemuLng peach ApgrusalFY53

MEPR $14,27 1,344 $15,164,893 $9,a47,460 $39,283,705

MEDICARE $24,141,910 $17,832,254 $14,223,286 $56,197,450

FY53

UnfundedWorkload ($9,070,56) ($2,667,36 1) ($4,375,01 ) ($16,913,745)

FY63PercentUnfunded 40.89X 14.96% 30.772 30.1X

117

(illustrated in Tables VII and VIII). Nevertheless, it

would receive additional funding in each fiscal year. In

FY83, Naval Hospital Long Beach would receive an

additional $2,667,361 and, in FY84, an additional

$4,785,037.

Our analysis uncovered an inconsistent factor

relating to inpatient discharge workload. There appears

to be a consistent 2.5 percent difference between the

number of dispositions reported in each NTF's Medical

Expense and Performance Report, and the number of DRG

discharges contained in the biometric data. We were

unable to determine the exact cause for this occurrence.

The Tri-Service DRG Study Group senior statistician

suggested the reason could be because one set of the data

is patient-specific (biometric data), while the other set

of data (MEPR) is NTF-specific (aggregate data). We

elected to use the biometric data for our calculations

since it appears to represent a more accurate one-to-one,

input-output relationship. However, had we used the other

set of data (MEPR dispositions), it would not have

significantly affected the findings.

2. Comparison of Naval Treatment Facilities toVeterans Administration Facilities

The comparison of NTFs' inpatient care expenses to

VAFs' inpatient care expenses revealed findings consistent

with those of the primary research question. If one uses

118

TABLE VIII

FY1954 INPATIENT HEPR/MEDICARE DATA FOR CHARLESTON, LON6BEACH. AND PENSACOLA NAVAL HOSPITALS

iL Chmciugnn Long 1sh PonnnAh Abgrate

ttEPR $14,236,442 $14,279,354 $9,541,979 $38,057,775

FY64MEDICARE $23,532,068 $19,064,391 $14,693,232 $57,289,691

FY54UnfundedWorkload ($9,295,626) ($4,7195037) ($5,151,253) ($190231,916)

FY64PercentUntunded 39.503 25.103 35.06X 33.6X

119

T it .. a

FY84 VAFS' average adjusted cost per discharge, HCFA cost

weights, and DRGs to determine annual inpatient fundin9

for Charleston, Long Beach, and Pensacola naval hospitals

in FY83 and FY84, these hospitals would have received more

money. Specifically, as illustrated by Table IX, they

would have received $15,667,975 in FY83 and $18,287,651 in

FY84. Similar to the previous Medicare results, the

largest difference between MEPR expense and potential VAFs

expense levels was seen in the Naval Hospital Charleston,

followed by Pensacola, and finally, Long Beach.

3. Analysis of the Similarity of DiagnojARelated-Groups

A subsidiary finding was that the thirty most

frequent DRGs in each NTF from FY83 to PY84 varied only 23

percent. In fact, when comparing the biometric data, each

NTF experienced 77 percent of the same thirty most

frequent DRGs in FY84 as it did in FY83. What this tends

to point out is that the case mix is relatively stable

from one year to the next within the same facllity.r

Furthermore, this finding suggests that a

beneficiary population, coupled with a cjIjALat

availability and use of hospital and physician services,

would manifest itself in a rather homogeneous range of

case mixes from one year to the next.

Moreover, although there are 470 DRGs, these

thirty most frequent DRGs in FY84 account for 58.2, 56.6,

120

TABLE IX

SWIDIARY OF FM183, 84 AND AGGREGATE INPATIENT tiEPR/VAFsDATA FOR CH2ARLESTON. LONG BEACH. AND PENSACOLA NAVALHOSPITALS

FYI 98a DATA:fir Cbualoma Lbfug pic eansamim Aggrgal"EPR $14,271,44 $15,64Af "l,47,48 0 9P,=8,MO

VAFs *25.4O,7I" $1,436,99 $3~J.e W,51,440

UnfundedWorkload (s,=,3s 0$2,2?20) MAoboo) (*1M67,9M

PercentUnfunded "M ~ 13.0s 29215 20Mf

EY904 ATA:IIEPR $14,23.44 $142,38 $9A41,979 P8,057,7

VAF* *,010,415 *16,41,777 $14A,=,24 $U"A4,426

UnfundedWorkload (*6,77,975 (*4,2,42) ($5,151 mm5 ($16,87,4)

PercentUnfunded 36.115 MMS1 35.11 32-51

FUIfl3Z8lDAIALIIEPR $28,37,766 $"444 $19,3",447 $77,41 1400

VAFs *46,17,154 $,73 $28,01,22 $111,7,

UnfundedWorkload ($1 6,109,3) ($6634,4U (*9,21 1,775 (P3,926)

PercentUnfunded 36*15 19.4% 32.25 30"5

121

and 55.1 percent of the total workload at Charleston, Long

Beach, and Pensacola naval hospitals, respectively.

Next, in contrasting the thirty most frequent DRG

in these three NTFs to those in California's hospitals,

one finds considerable variation in the inpatient case

mix. Using FY84 data, Charleston, Long Beach, and

Pensacola naval hospitals' case mix differs from that of

California's hospitals by 73, 70, and 70 percent,

respectively, as depicted in Table X. This finding

indicates the typical case mix which comprises the bulk of

the workload differs considerably between the NTFs' and

California's hospitals. If California's hospitals treat

the "average-type" patient, one could surmise that the

NTFs treat rather "atypical" case mixes. That is, the

nature of NTFs services and products are somewhat

different than that found in the civilian sector.

When contrasting the NTFs to one another, however,

the case-mix variation is not as great as it is between

the NTFs and California's hospitals. In fact, in FY84,

of the thirty most frequent DRGs, 70 percent are common

(i.e., only 30 percent are different) between Naval

Hospital Charleston and Naval Hospital Pensacola, 47

percent are similar between Naval Hospital Long Beach and

Naval Hospital Pensacola, and 57 percent are shared

between Naval Hospital Long Beach and Naval Hospital

Charleston.

122

TABLE X

THIRTY MOST FREQUENT DIAGNOSIS RELATED GROUPS IN THREE NTFsAND IN THE STATE OF CALIFORNIA HOSPITALS FOR FY?1904

DROUann a Cham Lng aLh Gallt neDa1 391 391 391 3912 373 373 373 3733 430 436 371 3714 62 243 62 2435 371 254 438 3906 372 371 313 1277 234 162 355 182a 301 232 162 3099 56 234 103 35510 303 56 460 43011 390 153 436 46012 355 249 140 14013 243 383 234 1414 162 428 372 0g15 254 301 90 8016 39 390 143 9617 143 104 301 12210 153 98 167 29419 374 427 262 43020 359 309 198 21521 460 97 122 0222 420 201 270 20923 232 297 40 14024 40 426 243 21025 140 25 39 12126 97 374 222 15427 270 460 361 10620 195 167 467 lie29 154 222 158 130 122 364 337 306

@Sorw: "JtPtkm ad mhu-. 9V M Fr termk: bdwial Het) Db~ap bDaa," Na

123

What these findings primarily suggest is that the

thirty most frequent (and potentially resource consuming)

DRGs vary much more between the NTFs and California's

hospitals than they do among NTPs. Accoxdingly, the

findings support the assertion that inpatient case mix

among NTFS is more similar than it is compared to a

typical civilian hospital's inpatient case mix (i.e.,

naval hospitals treat somewhat different types/categories

of inpatients than civilians, and these types of patients

are common to naval hospitals, in general).

124

V. CONCLUSIONS AND RECOMMENDATIONS

A. CONCLUSIONS

The focus of this research effort was to: (1)

investigate whether a feasible and meaningful comparison

could be made between NTF expenditures in a given fiscal

year for inpatient care and the amount civilian hospitals

would have been reimbursed by Medicare had they

experienced a similar inpatient workload as that of NTFs

using DRGsj and, (2) if possible, develop an actual model

that would facilitate this comparison using real workload

data. Our analysis suggests the following conclusions:

1. Biometric and actual expense data are available,

wh c _allow interested researchers to make relative anf,

meaningful comparisons between NTFs and Medicare'.U

reimbursement provisions. The Medical Expense and

Performance Reporting System for Fixed Military Medical

and Dental Treatment Facilities (MEPR) uses step-down

procedures that capture all relevant inpatient expenses

for each NTF. Because these inpatient expenses can be

readily identified and are isolated from other facility

operations and programs, one can make definitive

statements concerning the aggregate facility inpatient

costs. Additionally, if existing discharge summary data

can be transformed into biometric data, such as the

125

authors were able to obtain for this analysis, one can

determine constructed Medicare reimbursable amounts for

actual NTFs' workload.

2. If comensated for inpatient cgare in a manne

similar to civilian hospitals under Medicare, these three

naval hoopitalF would hag Collectively received 31.9

uercent more than their actual .expensesA__L ...o._.

$6.000.000 each oer year.

As discuised in Chapter IV, the MEPR expense data

contrasted to Medicare reimbursable amounts, which again

are based on the NTFs' workload, indicate that each of the

three naval hospitals over the two-year period would have

received from a low of 20.2 percent to a high of 40.2

percent m than their actual inpatient expenses, or,

expressed in dollar amounts, from a low of $7,452,398 to a

high of $19,166,192 more.

3. When the Vetr-ans Acdministration's average

j.kad__J. st per ciharge. HCFA cot weights, and the

ICD2-9-CM DRgs are utiliztd_.to p

ajjI.or the three NTFs used in this anal i. they

would have received 30.5 -narcent more than their actual

expenses, or approximately $5.65Q.000 each Rer year.

The VA has asserted in recent months that it

provides inpatient care (almost) as efficiently a

civilian hospitals. While our research does not address

126

whether the VA costs per DRG discharge and Medicare

reimbursable amounts are approximately equal, the three

NTFs used in this analysis would have received over

$33,000,000 more than actual associated expenses had the

VA average adjusted cost per discharge been used in

calculating reimbursable amounts.

4. In analyzing the biometric data and comparing the

thirty most frecuent DRGs. the authors found there to be

much greater similarity of case mix among the three NTFs.

themselves. as well as within the same NTV from one year

to the next (PY83 to FY84). than between the NTFs' and the

State of California's thiry most frequent DRGs.

The analysis indicates that the thirty most

frequent DRGs for each NTF varied only 23 percent from

FY83 to FY84. In other words, using Naval Hospital Long

Beach as an example, of the thirty most frequent DRGs in

PY83 exactly 77 percent of these DRGs were among its

thirty most frequent DRGs in FY84. This suggests that the

case mix is relatively stable for a NTF from one year to

the next, as would be expected for NTFs that serve a well

defined beneficiary population having a consistent case

mix. Of course, case mix is largely dependent not only

upon patients' demands but also available medical/surgical

services and physician capabilities. Undoubtedly, if the

available capabilities changed so would the nature of the

services and products; ergo, the case mix would differ.

127

The comparison among NTFs in FY84, however,

revealed somewhat lower levels of similarity than that

found at each NTF (from one year to the next). In fact.,

similarity of the thirty most frequent DRGs between NTFs

ranged from a high of 70 percent between Naval Hospital

Charleston and Naval Hospital Pensacola to a low of 47

percent between Naval Hospital Long Beach and Naval

Hospital Pensacola. Although this finding indicates there

is not as much similarity of case mix between NTFs as

found at a single NTF from one year to the next, the

variability is considerably less than that between the

NTFs' thirty most frequent DRGs and that of the State of

California's, which varied 71 percent. This, in turn,

suggests that the case mix among NTFs is much more similar

than that found between NTFs and the civilian sector.

Accordingly, one can deduce that the beneficiary

population case mix for each NTF, itself is more

homogeneous than that among NTFs. Moreover, the

beneficiary population case mix between the NTFs' and

California's can perhaps best be characterized as being

almost hetergeneous (i.e., case mix varies so much between

the two that it appears as if the preponderance of

inpatient costs are for considerably different types of

cases).

The purpose of identifying the differences among

NTFs, themselves, and, in particular, between themselves

128

and the DRGs for a large representative area such as

California, is to demonstrate case-mix measurements (using

DRG methods) provide a means, perhaps the best means, to

make meaningful comparisons and statements about what it

is hospitals produce. Although this conclusion is not

surprising and intuitively acceptable, its significance

lies in the fact that it enables a valid and relevant

comparison of output among hospitals, even when they

provide vastly different services and products.

5. Our analysis suggaests that either NTs are more

efficient in providing inpatient care than that which is

grovidedji ncivilian hospitals or that HCFA/Medicare

reimbursement rates are too genjrou.., o a combination

We cannot decisively explain the disparity between

what NTFs would have received under Medicare'a

reimbursement provisions and what expenses were actually

incurred. As with any analysis that depends upon non-

experimental raw data, erroneous findings and conclusiona

may be drawn from data that are inaccurate, incomplete, or

wrongly applied. The authors, however, have gone to great

strides to ensure that the data employed for this analysis

are highly accurate, complete, and applied correctly for

testing the research hypothesis; therefore, the results o2

this analysis are preliminary but nonetheless suggest that

129

NTFs dra Jn fact provide efficient inpatient care, perhaps

even more efficiently than the average civilian hospital.

B. RECOMMENDATIONS

The following recommendations are based on this

research effort.

1. Further analysis be conducted to test the rsearb

byl~theia,

Although the findings of this thesis are

consistent and fully support the research hypothesis, the

authors believe a follow-on analysis comprised of a larger

sample population from several years with the most current

coat weights would confirm the findings of this thesis.

2. Consideration should be siren to incororating DRG

mgthods in amaesaing NTa inipatient workload efficiency

and productivity.

Since DRGs enable hospitals to identify and

measure their products more effectively and accurately it

seems only logical that a case-mix approach be employed

for assessing hospital efficiency and productivity.

Rather than utilizing exclusively such surrogate measures

of output as occupied bed days, number of admissions, or

number of operating room procedures, a case-mix patient

classification system would provide a superior means of

identifying what has been produced and ho% efficiently it

has been produced.

130

3. Similar consjidejration should be given to

incorporatin .DeQ thods into the resource allocation and

decision-makingpg&es.

Because DRG case-mix methods enable more precise

product identification, resource allocation methods should

incorporate funding levels that are predicated on what is

actually provided or produced. In developing the resource

allocation method, provision should be made for

identifying controllable and uncontrollable inpatient

costs and for designing incentives into the overall naval

health cave delivery system for effectiveness as well as

efficiency.

4. The Naval Medical Command should consider

development and refinement of the DRG 2atient

classification system so that it can be tailored to meet

It behooves the Naval Medical Command to examine

development of its own cost weights and average adjusted

cost per DRG discharge because of the somewhat atypical

case-mix groupings found in its NTFs. This is especially

true if it plans to apply this mechanism to productivity

And efficiency analysis and to the resource allocation

process. Admittedly, this would be amajor undertaking

but the uses of such information are poLentially profound.

131

5. BtfQre broad and irrevocable commitmentsqrj

decisions are made in the CHAMPUS Rgeform Initiative (CRI)

the Department of Defense should consider that inpatient

care costs angear to be consistently and cgnsiderably less

than Medicare reimbursement amounts in three of its

Uniformed Services Medical Treatment racilities.

Again, based on the preliminary findings, NTFs

appear to be able to provide considerably more inpatient

care at similar funding levels than either civilian or VA

hospitals. Because of this indication of efficiency

within the DoD health care delivery system, it behooves

DoD to maintain as much inpatient care "in-house" as is

consistent with overall operational goals. In particular,

DoD should strive to maintain "in-house" those case-mix

groupings (DRGs) that it can treat less expensively than

the contractor. Ideally, these would be the more complex

and resource-intensive medical and surgical cases.

However, whatever negotiated fixed price contracts for the

CRI are effected, they should reflect the case-mix

groupings that will remain predominantly an "in-house"

responsibility.

132

APPENDIX A

PROSPECTIVE REIMBURSEMENT UNDER P. L. 98-21

INTRODUCTION

This appendix attempts to provide additional detailed

information about "what does" and "does not" make up the

federal and hospital-specific portions of Medicare

reimbursement rates under Public Law 98-21. The intent of

this appendix is not to make the reader thoroughly versed

in Medicare prospective reimbursement. Rather, this

explanation is intended to provide a foundation for those

unfamiliar with Public Law 98-21 in order that an

understanding of the analysis methods contained in the

thesis might be better understood. Most of the

information contain in this appendix has been extracted

from the Federal Register [Ref. 27] and Grimaldi's and

Micheletti's book, Prospective Pavment; TheDjj.ini iy.

GiQjoReimbursement [Ref. 1]

Federal Portion of the Prospective Payment Rate

The federal portion of the prospective payment rate is

based on the average cost per Medicare discharge. These

standardized costs, also referred to as adjusted payment

amounts, are developed for each DRG in the PPS. They are

133

derived from the following cost-input factors:

* Base-Year Costs

* Inflation Adjustments

* Cost Standardization

* Case-Mix Adjustments

* Indirect Medical Education Costs Wage Adjustments

* Cost-of-Living Allowance Adjustments

* Budget Neutrality Parameters

* Patient and Cost Outlier Adjustments

* Medicare Part B Costs

* FICA Tax Adjustments

* Nonphysician Anesthetist Service

As discussed in Chapter III, the payment rate for each

DRG discharge is established on the basis of three sources

of data: the Medicare cost report, the Medicare discharge

file, and the MEDPAR file. During the transition period,

the federal portion of the adjusted payment amounts are

based on regional and national average payment rates. The

national rate is comprised on a single rate for urban

areas and a single rate for rural areas. The regional

rate is made up of 18 regional rates, one rate for each

urban and each rural area in each of the nation's nine

census divisions.

As the health care industry proceeds through the

three-year transition period, Medicare prospective rates

will increasingly depend on national rates and less on

134

regional rates. Eventually, the national rates for urban

and rural areas will be the only adjusted payment rates

used by Medicare to reimburse hospitals.

A. COST-INPUT FACTORS

The discussion of cost-input factors in the subsequent

paragraphs attempts to better illustrate the cost

foundation on which the federal rate is based.

1. Base-Year Costs: As Figure 3-2 illustrates the

calculation of the standardized-payment amount begins with

the establishment of allowable inpatient operating

Medicare costs in the base year. Reported baseyear costs

are taken from calendar year 1981 Medicare cost reports.

These costs are subsequently modified as a result of the

inclusion of authorized adjustments and exclusions under

P.L 98-21. These costs include:

(a) capital-related items,

(b) approved direct medical educations programs,

(c) nonphysician anesthetist service,

(d) nursing differential, and

(e) routine costs in excess of Section 223 limits.

The net result is divided by the number of Medicare

discharges during the cost reporting period to obtain the

adjusted allowable cost per Medicare discharge.

135

COST PER (Allowable Base-Year (TotalMEDICARE = Medicare Costs) + Adiustment)DISCHARGE (Medicare Discharges)

2. Inflation Update: The updating (inflation) factor

attempts to transform base-year costs into current-year

dollar terms. Adjusted base year costs are updated for

inflation expected to occur between the base and rate

years. First, the costs are updated to the fiscal year

ending September 30, 1983, so that cost reports covering

different periods can be expressed in comparable dollars.

Second, updated costs are increased, through September

1985, by the target percentage; the projected inflation in

the hospital market basket plus an allowance to improve

the intensity/quality of care is included in the standard

payment rate. One should note, however, there is no

retroactive adjustment made if there is variance between

actual and projected inflation.

3. Cost Standardization: Standardization of costs is

done to minimize the effects of certain factors on costs

so a comparison of hospital performance can be made on the

basis of product line (i.e., DRGs). Inflation adjusted

cost per Medicare discharge are standardizes fo..:

(a) differences in case mix among hospitals,

(b) indirect medical education costs,

(c) interhospital differences in wage levels, and

136

(d) cost-of-living differences for Alaska and Hawaii

hospitals.

4. Case-Mix Adjustment: A 'se-mix index is used to

adjust for interhospital differences in the types of

inpatients treated. The index is derived from 1981 cost

and billing data. Case-mix complexity is said to vary

positively with the size of the index number. An index

greater (less) than one indicates that the case mix is

more (less) than average.

Case-mix indexes and DRG cost weights are based on

information obtained from the Medicare Provider Analysis

and Review (MEDPAR) cost reports. The MEDPAR file stores

20 percent of the bills that hospitals submit for payment

for inpatient services rendered to Medicare beneficiaries.

These sample bills contain the patient's age, length of

stay, diagnosis, and surgical procedure. This clinical

information is used by HCFA personnel to place Medicare

discharges into the appropriate DRG using the the 1CD-9-CM

coding methodology.

After patients are assigned to the appropriate

DRGs, the cost of their care is estimated. This involves

transforming hospital charges into costs of services

rendered. The information is obtained from the cost

report a hospital completes and submits to Medicare for

annual reimbursement determination. Cost reports contain

the routine and special care per diem costs and

137

departmental cost-to-charge ratios needed to convert

charges into the costs of services received by Medicare

beneficiaries. Ratios and average costs derived from 1981

reports were used to calculate Medicare case-mix indexes

and DRG price indexes. Table V illustrates the steps

involved in calculating DRG cost weights and case-mix

indexes.

The cost of treating a Medicare patient assigned

to a DRG is calculated as follows:

(a) The cost of routine care is found by multiplyingthe number of days the patients spent in a regularroom by the hospital routine cost per day;

(b) The cost of special care is found by multiplyingthe days spent in a special caro unit by thehospital's special care cost per day; and

(c) The cost of ancillary care is found by multiplyingthe charge of the service by the applicable cost-to-charge ratio.

5. Indirect Medical Education Costs: An adjustment

is made for the tests, procedures, and other indirect

costs generated by the medical education programs. The

ratio of full-time equivalent (PTE) interns and residents

in approved programs to beds and the effects of teaching

activity on operating costs are used to standardize

indirect medical education costs. HCFA estimated the

effect teaching activity on operating costs to be 11.59

percent.

The adjustment for indirect medical education

costs is made by dividing the case-mix standardized cost

138

per Medicare discharge by a hospital-specific education

multiple (EM), calculated as follows:

EM - ((DAW/FTEs)/O.I) x .1159] + 1.0

6. Vdage Adjustment: The amount determined by the

adjustment for indirect medical education cost is divided

into labor and non-labor components, respectively. The

labor-related portion is then standardized for wage

differences among various hospitals.

7. Cost-of-Living Adjustment (Cola): For Alaska and

Hawaii only, an adjustment is made for nonlabor costs due

to the relatively higher costs of living in these two

states. Similar to the labor component, nonlabor costs

are divided by the applicable adjustment factor.

8. Bludet Neutrality, P.L. 98-21 mandates that in

fiscal years 1984 and 1985 the prospective payment system

be "budget neutral." Specifically, Medicare is mandated

not to spend any more or less than it would have under the

1982 TEFRA. If budget neutrality is violated, the federal

share of the amount involved is spread proportionately

among the DRGs.

9. QOuierft Additional payments expected to be made

for outlier patients are subtracted from the standardized

amounts developed thus far. Outliers are patients with

unusually long (day outliers) or costly (cost outliers)

139

stays for a particular DRG. Day outliers are patients

whose length of stay exceeds the average (mean) stay for a

DRG by 22 days or 1.94 standard deviations, whichever

results in a smaller number of days. Cost outliers are

patients whose charges adjusted to costs exceed the DRG

payment rate by the larger of $13,000 or double the

relevant DRG payment rate.

10. Part B Costs. The standards are then adjusted

upwprds for services previously billed under Part 8 but

now included in the DRG payment rates. This is

accomplished by multiplying the standards by 1.0013.

11. FICA axes; Similar to Part 8 costs, an upward

adjustment is made for the FICA taxes previously not paid

by certain hospitals. The multiplier for 1985 waa 1,0018.

12. Non)hysician Anesthatists: The costs of these

services are recognized by reducing the national

standardized amount by a specific percentage. In 1985,

Medicare adjusted the nationel standardized amount by 0.32

percent and the regional standardized amounts by 0.42

percent.

B. FEDERAL RATE CALCULATION

The payment applicable to a particular DRG can be

obtained by multiplying the overall standard rate, by the

cost weight (or DRG price index) associated with the DRG,

listed in Appendixes C, D, and E. The cost weights shown

140

in Appendixes D and E apply to all participating hospitals

for each DRG. For example, if the national average

adjusted payment was $2,000 for each Medicare discharge

and the cost weight for a specific DRG discharge was 1.50,

the amount of revenue the hospital would receive would be

$3,000.

Hospital-Specific Portion of the Prospective Payment Rate

The hospital-specific portion of the prospective

payment rate is based on a hospital's historical cost

experience. For the first cost reporting period under the

PPS, a hospital-specific rate is calculated for each

hospital, derived generally from three cost-input factors:

(1) base-year costs, (2) case-mix index, and (3) updating

factor.

HOSPITALSPECIFIC ( CBase-Year Costs) x UpdatinM FactorRATE (1981 Case-Mix Index)

1. Base Year Costs: Base-year costs for the hospital-

specific rate are derived in almost the same manner as

base-year costs for the federal rate. One additional

adjustment, however, is required in the determination of

the hospital-specific base-year costs. An adjustment is

made for higher costs resulting from changes in accounting

141

principles initiated in the base-year and other actions

designed to raise base-year costs.

Base-year costs for most hospitals are derived

from cost data for the next to last year (or longer)

preceding the first cost reporting period subject to the

new PPS. In other words, the hospital-specific portion of

the payment rate is estimated from the twelve-month

Medicare cost period ending on or after September 30,

1982, and before September 30, 1983. Thus, if a

hospital's reporting period began October 1, 1983, its

base-year would be October 1, 1981 to September 30, 1982.

With certain exceptions, once base-year costs have been

established, they are generally applied throughout the

entire three year transition period.

2. Case-Mix Adjustmantt This adjustment is made so

that case-mix changes occurring between the base and rate

years can be fully recognized in calculating aggregate

Medicare prospective payments. The adjustment cost per

Medicare discharge is divided by a hospital's case-mix

index.

3. Uqatlg" actort The updating factor attempts to

transform base-year costs into current-year dollar terms.

This being the case, case-mix adjustments are increased by

a target percentage, which equals projected inflation plus

an allowance to improve the intensity or quality of care

in the institution. If budget neutrality is violated,

142

however, the target rate is adjusted so that Medicare

spends no more nor lear under the prospective payment

system than it would have been spent under TEFRA.

Cost Exclusions and Adjustments

Certain historical allowable costo are excluded

from the calculation of a hospital's prospective payment

rates, Other historical costs are adjusted to make the

base year inpatient costs comparable to operating costs

covered by Medicare's prospective payment system. In

general, the exclusions and adjustments fall under the

following headings:

* capital-related costs

* direct medical education costs

* nonphysician anesthetists

* nursing differential

* malpractice insurance costs FICA adjustment

* Section 223 adjustment

* Part B costs

1. Capital-Related Cottso These costs are

excluded from the prospective payment rates in FYs 84, 85,

and 86,and they are reimbursed on a retrosactively-

determined reasonable cost basis. These costs include net

depreciation, leases and rentals, improvements, certain

interest and insurance expense., and taxes. In the case

of investor-owned hospitals, these costs include a return-

143

on-equity capital. (Under Medicare's prospective plan, the

return will be calculated by multiplying allowed equity

capital by the rate of interest the federal treasury pays

on loans from the Hospital Insurance Trust Fund).

Capital-related costs do not include repair or

maintenance cost, interest expenup incurred to borrow

working capital, taxes paid on land or depreciable assets

not used for patient care, insurance that does not apply

to depreciable assets not used for patient care or the

payment of capital-related cost if business is

interrupted, and the costs of minor equipment that are

expensed rather than capitalized. Additionally, one

should note that hospitals are not permitted to change

their capitalization and expensing-of-assets policies

during the transition period.

2. Direct Medical Education Costali These costs

are mlso excluded from the prospective rates and are

rLLIbursed on a retrospective, reasonable coat basis.

Apprnved educational activities consist of formally

organized or planned programs of study typically aimed at

enhancing the quality of care in the institution. These

activities may include nursing schools, radiologic

technologist schools, arid the medical education of other

paraprofessionals. They do not include patient education,

general awareness programs for the community, and on-the-

job training.

144

3. Nonphysician Anesthetists: The costs of

services rendered by certified nurse anesthetists (CRNAs)

and anesthesiologist assistants (AAs) are excluded from

th._9payment rates, and like _apital-related and direct

mejical education costs. are reimbursed on a

Ltospecti3.ve-goatia.. This exclusion is designed to

eliminate the incentive that hospitals have to substitute

higher-costing anesthesiologists for nonphysician

anesthetists when cost of CRNA or AA services are in the

rates. Since anesthesiologists can bill under Part B of

Medicare, hospitals could enhance their financial position

by having physicians administer anesthosia while CRNA or

AA costs are left in the payment rates. The exclusion

eliminates the potential "double payment."

4. Nursing Differential; TEFRA abolished the

nursing salary cost differential for general inpatient

routine services for cost reporting periods on or after

October 1, 1982. Thus, these ots ar removed from the

base year in order to establish the prospective payment

amounts.

5. FICA Adjustment: Some hospitals did not pay

social security taxes during the base period, but they

were required to pay them beginning January 1, 1984. T

recognime this legally mandated increase in compensation

cost. an aopropriate amount is added to the regorted base--

145

6. Section 223 Adjustment: Since the mid-1970s

Medicare has imposed a limit (Section 223) on reimbursable

per diem costs for general impatient routine care. Costs

in excess of the limit are excluded from the calculation

of the standardized payment amounts.

7. Part B Costs: Prospective payment rates are

intended to cover all costs associated with covered

inpatient care furnished to Part A beneficiaries, except

physician services. Prior to P.L. 98-21 many nonphysician

services furnished to inpatlents were billed under Part B

rather than Part A. Por the most part, the new law

prohibits this practice for services rendered after

September 30, 1903. In other words# the payment rates

represent full payment for all covered nonphysician

inpatient services. These services must be supplied

either directly by the hospital or another entity under

arrangement made by the hospital. In order to compensate

for costs formerly billed under Part B, reported base-year

costs are adjusted upward by a specific target percentage.

HEALTH CARE PROVIDER EXCLUSIONS

Under P.L. 98-21 certain types of providers

are not subject to the prospective payment system but will

continue to be paid on a reasonable cost basis. Some of

the types of excluded providers are:

146

* children's hospitals,

* long term hospitals with an ALOS greater than 25days,

* sole community hospitals (SCH)

* psychiatric and rehabilitationhospitals, and

* hospitals operating under alternative state paymentsystems.

147

RPPENDIH U

DECISION TREES FOR THE ICO-9-CM DRGs

DEFINITION OF SYMBOLS USED IN DECISION TREES

Symbols Definitions

<Decision operation

Looping uariable

0

Hierarchy of operatingroom procedures

Connector

KTerminal

Souros: The Reised ICD-9-CM Diagnosis Retlated Orps: Grouper User Mwanul (14w Haven,CT: Helth Sylem Interm1oal). Adted fom Paul L. Orinwidl a1 Juie A. Miohletti,Propte Payjmet The Definitive Gi to Reftkbsement, (Chbap; Pluribus Press), 198., A-I

148

J*4101 I

PM-

149

la

PI~q toul

3.

150

-wgig

mmmIit

onIi O

151

R~kWIM14"411

7q~ - *

1 j

IF IWA

iAA I~ II

153

im

ii

154

1o41

155

KID,

UIa WNlW "a%

WIN

MA

157

IV

1 145M

cw~u[ V4tlv

s8eN alo

r.-.Lchl filmil-

1.59

i aI

ft [wFlI

~- ii HMu rtaq

600

IFI

tol I

161

gul .II

Il"1 l- UIPAIZh 4

162

NUWAM

Awl 8410 ti

PON IdIU3,42jo

0 pm ftm i p m $no

IPA

-- ol Aqjo

L-7w .. M

163

10340iM~INIJ

Pk1101010110 flW I

PO qIl 10

AI CI K lf~.i l~ II18II~jpta'IAI

- ~ ~ ~ ~ . M. ssrol

164

SAONII.I&*

165

166E

IONpi11l 7m

1I

-- - '- ----- ---wool-

16

MOAI

169

170

171

"Naw

"IUaNWAlW

172

ism"

117

ii

II *11I lit

I~1 ~3EIll Ii / A~IaIeJie~j (IN1'1iii

AIIaggiIaMg ~.uq I ~

K

H'

I

*S. a1!

K

174

If'i AWWPIIit

go

I17-

ON Im~m

0"Imal

iijill I E

176

sog~ minIV % I

loo &$ I.I1

1 ii"I17

tI-410

au mllj

I 1Ud

178

*17

APPENDIX C

DIAGNOSIS RELATED GROUPSAND SELECTED RELATIVE WEIGHTS

O A HCFA1903 NCFA 1983

RELATIVE 1903 CUT01 HBC TYPE TITLE WINT ALO$ OFF

001 001 a AIt TORT ACE )17 EXCEPT FOiR TlAUMA ,3548 19.4 39002 001 S CIANIOTON' FOR TRAUMA All )17 3.2829 15. 3600 001 S CRANIOTONY Am (IN 2.9419 12.7 33004 001 $ SPINAL PRO0CIURE 2.2452 16.0 36005 001 1 ETRACRNIAL VASCULA PROCEDURES 1o670 9.8 30006 001 S CAIPAL TIINEL RELEASE .393 2.6 8007 001 1 PERIPt i CRAMIAL NERVE 4 OTHER RENY 3YST PROC AGE )9 *O C. C, 1.0279 5.3 25008 001 S PERIPH * CRNIAL NERVE 4 OTHER NERY SYST PROC AGE (70 W/o C. C. .7239 4.1 23009 001 N SPINAL DSO3EDRAS * INJURIES 1.3950 9.1 29010 001 N NERVOUS SYSTEM NEOPLASMS AGE )69 AND/OA .C., 1.3007 0.6 300I1 001 M NERVOUS SYSTEM NEOPLASMS AE (70 W/O 0. C. 1.2540 8.3 29012 001 M DEGERATIVE NERVOUS SYSTEN DISORDERS 1.1136 9.4 29013 001 A MULTIPLE SCLEROSIS + CEOGUELLAR ATAXIA 1.0130 0.9 2014 001 H 3PECIFIC CEREINOVASCULAR DIONDERS EXCEPT TIA 1.3127 9.9 3001 001 N TRANSIENt ISCHEIC ATTAC93 .6673 5.6 24016 001 N NONSPECIFIC CII[ROVAE CULAR DISORDERS WIT C. C. .1592 7.4 27017 001 M NONSPECIFIC CEEIROVASCOLAR IISORDERS /o 0. C. .0392 7,2 270I 001 M CIANIAL + PERIPHERAL NERVE DISORDIRS AlE 69 AND/OR C. C. .715 6.6 27019 001 N CRANIAL PERIPHEIAL NERVE DISOIDERS AGE (70 Io C. C. .6975 1.7 26M25 001 M NERVOUS SYSTEM INFECTION EXCEPT VIRAL MENINiBITIS 1.3141 7.6 28021 001 M VIRAL MENINGITIS .6301 4.5 is022 001 M HYPERTENSIVE ENCEPHA0.OPATHY .7649 6.4 26023 001 N NONTAUATIC STUPOR * COsA 1.1568 5.9 26024 001 N SEIIURE + HEADACHE AGE )69 AND/OR C. C. .7279 5.6 26025 001 N SE URi + EADACHE AGE 18-69 N/o C, C. .392 4.9 25026 001 N SEIZURi * EADACIE AGE 0-17 .4349 3.3 13027 001 N TRAUNATIC STUPOR * CONA, COAA)! NR 1.1368 4.1 24

8 001 N TIAUNATIC STUPOR + CON, CORA (INNl Ai[ )69 AND/01 C, C. 1.0701 5.9 26M 001 N TRAUMATIC STUPOR o COMA (1 HR AGE 18-69 /O C. C. .175 3.8 24030 001 N TRAUKATIC STUPORf COA (1 HR AGE 0-17 .3576 2.0 08031 001 A CONCUSSION ARE )63 AND/JO c. C, .6051 4.6 25032 001 R CONCUSSION AGE 18 - 69 IB C. C. .4519 3.3 19033 001 N CONCUSSON AGE 0-17 .248 1.6 0S034 001 H OTHER DISORDERS OF NERVOUS SYSTEM A8E 69 AND/ON C. C. .9927 7.1 27035 001 N OTHER DISORDERS OF NERVOUS S STEM ARE (70 W0O C. C. .0460 b.2 26036 002 5 RETINAL PROCEDURES .7093 5.0 13037 002 5 ONRITAL PROCEDURES .5630 3.4 11038 002 S PRIMARY IIIs PROCEDURES .4325 3.0 9039 002 3 LENS PROCEDURES .s010 2.6 6040 002 $ EXTOAOCULAR PROCEDURES EXCEPT 0117 ABE )17 .3977 2.4 7041 002 5 EXTRACCULAI PROCEDURES EXCEPT OjllI ABE 0-17 .3695 1.6 4042 002 5 INTAOCULAR PIOCEDURES EXCEPT RETINA, 31I3 4 LENS .0906 3.0 12043 002 A NtYPIREA .312 4.2 12044 002 N ACUTE MAJOR EYE INFICTIOS .6290 6.5 22045 002 N NEUROLOGICAL EYE 010ORIES .5641 4.3 Is046 002 N OTHIER IORRER11 OF TIE EYF ABE )17 WITN C.C .5964 4.1 23047 002 N OTHER DI$ORDEIS OF THE EYE AGE )17 k/O C.C .0064 3 12048 002 N OTHER DISORDERS OF THE EYE ABE 0-17 .4060 2.9 13049 003 3 MAJOR HEAD * NED[ PIOCEDUIE3 2.5270 13.i 34050 003 S SIALOADENECT(WIT 7160 4,6 14

180

DIAN0NI5 RELATED 6P"UPS AND L[CT)[ REL.TIVE NEISHT$HNF0 NCVA1903 HFA 190'

RELATIVE 1983 CU0

DRO HOC TYPE, ,TIll HEIGH ALOS OFF

051 003 ILL VARY 2LA PROCEDUIES XCEPT SIALOAD94ECTOHY ,6702 i,i 15052 003 S CLEFT L1V4 PA'W.1T REPAIR 0.64k, La 11053 003 S MINUS hoTOlp PlOCEDURES AlE )17 0.2095 3', 1a04 003 3 C'OuS 4 MASTuIO PROCEDURES AGE 0-17 0.6961 3W 11W.. 003 S 1SCELLANEOUS EAR, HOSE # tHROAT PROCEDURES 0.4153 2.5 7056 003 3 ROMNOPLAOTY 0,414 2.8 8057 003 8 Y + A PVC EXCEPT TONSILLECTOHY +/OR ADENOID[CTOHY ABE )1? 0.5251 2,7 ;050 003 1 1 # A PROC EXCEPT TONSILLECTOMY i/OR ADENCIDECTOMY AGE 0-17 CAP1 1.5 3059 003 5 IONS1LLEC ORY AR5,OR ADENOIDECT01T ONLY Ail )17 0.314% 2 4060 003 S IOXILLECTOY W0OP ADENOIDECTOMY ONLY ACC 0-17 0.2643 1. 3061 003 5 MYIINCOTOHY ARE )17 0.4273 2. 9062 003 S MYRIROTOMY ARE 0-17 0.3121 1.3 3063 03 3 OTHE EAR.4 OSE # THROAT OR. PROCEDURES 1,109 5.0 11604 003 N EAR, NOSE'j:THROAT MALIRIANC 1.0892 5.7 26041 003 N DOSEOOILPRIUM 0.4957 4.6 17w6 003 N EPISYAXIS 0.4114 3.7 i061 003 r ME[ILOTTIO A: 4,3 17069 003 N OTITI MEDIA 4 URI ARC )o9 ATO/OR C. C. 0.4219 6 22049 003 N OTITI Milli 4 URI Ate 11-65 V/0 C, 0. 0.5417 4.0 19070 003 N OTMS ME14 0410 AlE O.17 A.3497 3.1 10071 003 N LARYG CORACHEITIS 0.3 0 1.9 9072 003 N KAFAL TRAUMA 4 DEFOMM 0.467 3,8 16073 003 N OTHEP EAR, NOSE + THROAi DIAGNOSIS AlE )17 0.5217 3,5 17074 003 M OTHER EAR, NOSE 4 THROAT DIANOSES AGE 0-17 0.3463 0.1 9075 004 5 MAJOR CHEST PROCEDURES 26044 14,4 34076 004 S O.R. PROC ON THE RESP SYSTEM EXCEPT MJOR CHEST Ol1N 0. 0. 1.0714 10, 31077 004 I O.1. PROC ON THE REP SYSTER EXCPT HAJOR CHFST N/C C. C. 1.6170 M5 30070 004 1 PULMONARY EiOLISM 1.4005 10,4 30.079 004 M RESPIRAIORY I ECTIONS INFLAMMATIORS AiE 49 ANDOR C. C. 1.7982 11.2 31060 004 M RESPIRATORY INFECTIONS ' INFLAMMATIONS AGE 108-69 /O C. C. 1.7445 Alp 31061 004 H RESPIRATORY INFECTZONS 4 INFLAMMATIONS ARE 0-17 0.0743 6.1 26082 004 P REStIRATORY NEOPLASMS 1.14 7.4 27003 004 P MAJOR CHEST TRAUMA AgE )69 AND/OR C.C. 0.9809 Il1 28084 004 H MAJOR CHEST TRAUMA ACE (70 R/O C, C, 0.7738 5,3 22UOS 004 H PLEURAL EFFUSION AGE )69 AND/OR C. C, 1.1461 0,4 28086 004 A PLEURAL EFFUSION ANE (70 W/O C, C. 1.1217 7.6 26007 004 H PULNONARY EDEMA + RESPIRATORY FAILURE 1.129 1.7 20080 004 N CHRONIC ORiTRUCTIVE PULMONARY DISEASE 1.0412 7.5 28089 004 N SIMPLE PNEUMONIA + PLEURISY AGE )69 AND/OR C, C. 1.1029 0.5 29090 004 N SIMPLE PNEUMONIA * PLEURISY ARE 10-69 0/0 C. C. 0.9049 7.6 20091004 H SINPL P U.I IA + PLEURISY ARE 0-17 0.5)31 4.6 14092 004 H INTERSTITIAL LUNG DISEASE AGE )69 AND/OR C. C. 1.037 7.6 26093 004 N INTERSTITIAL LUNG DISEASE ACE 170 N/0 C. C. 09724 6.9 27094 004 M PHENIOTHORAX ABE )69 AND/OR C. C. 1.4374 4.2 29095 004 N PHERROTHORAX AGE (70 V/0 C. C. 1.1232 7.7 20096 004 IRONCHITIS 4 ASTHMA ABE )69 AND/OR C. C. 0.7996 6.9 24097 0414 N IRCHITIS + ASTHMA AGE 18-69 A/0 C, C, 0,7256 6.2 21098 004 N iRONCHITIS + ASTHMA AGE 0-17 0.4275 3.7 11099 W4 M RESPIRATORY S1NS 4 SYMPTOMS AGE 69 AND/OP C. C. 0.035 5.5 26100 004 N RESPIRATORY SIMNS 4 SYHPTOMS ABE (70 0/0 C. C. 0.773 5.1 24

181

010A10313 AILATED GROUPS AND SELECTED RELATIVE WEIGHTSHCFA HCFA1903 NPFA 1983

RELATIVE 1913 CUT

DAi NOC TYPE TITLE HEIGHT ALOS OFF

101 004 9 DHER RESPIRATORY DIAGNOSES %GE 169 AND/OR C. C. 0.9035 6.0 27102 004 H T0E RESPRATbY DIAGNOSIS Ali (70 0.9014 6.1 26103 005 5 HEART TPAN0LAHT 0 0 0104 005 8 CARDIAC VALVE PROCEDURE WITH PUMP 4 WITH CARDIAC CATH 6.807 20.9 41105 005 S CARDIAC VALVE PROCEDURE WITH PUMP 1 +/0 CARDIAC OATH 5.2300 14.2 361o6 005 S ODRONARY,IYPASS NITH CARDIAC OATH 5,2624 20.4 40IC7 0O15 0 CRONARY YPASS W/0 CARDIAC CATH 3.991 13.b 34208 005 S CAR0IOTHOR PROC, EXCEPT VALVE f CORONARY IYPASS, WITH PUMP 4.054 13.3 33109 005 0 CARDIOTHORACIC PROCEDURC! Wo/ PUMP 3.6963 12 1 32110 005 8 MAIOR RECOOST7tIVE VASCULAR PROCEDURES AGE )69 AND/OR C, C. 2,9320 14.3 34111 00. S MAJOR RECONSTRUOTIVE VASCULAR PROCEDUREY AGE (0 W/O C. C. 2,5051 13.2 33112 005 3 VASOLLAR PROCEPUREREXCEPT MAJOR RECOHSTRUCTION 2.35 11.2 31113 005 t AMPUTAYION FOR 0IR SYSTEM DISORDEkS EXCEPT UPPER LI01 + TOE 2.68 21.6 42114 005 S UPPER LIN! + TOE AMPUTATION FOR CIRC SYSTEM DISORDERS 2,1047 16,6 371135 003 PERMO INT CARDIAC PACEMAKER IMPLANT WITH 4MI OR 0HF 3,915 IS. 36114 00 PERMANENT CARDIAC PACEMAKER IMPLANT W/0 AMI OR (HF 2.0645 0.3 29117 005 1 CARDIAC PACEMAKER REPLACE + NEVIS CXC PULSEGEN REPI ONLY 1.021 6.4 24111 005 S CARDIAC PACEMAKER PULSE GENERATOR REPLACEMENT ONLY 1.709 4.2 14110 005 5 VEil LIGATION # STRIPPING 1,061 7.2 27120 005 8 OlER O.R. PROCEDURES ON THE CIRCULATORY SYSTEM 2,5204 15 35121 005 N CIRCULATORY DISORDERS WITH AMI + 0.V, COOP. DISCO. ALIVE 1.0641 11.9 32122 005 h CIRCILATORY DISORIERS WITH AMI V/0 0.. CNP. DISCH, ALIVE 1.36 1 9, 30125 005 CIRCULATOY DIORDERS VIM ART, EXPIRED 1.136 3.1 23124 005 CIRCULATORY I18S0111S Ex1 Al, PITH CARD OATH + COMPLEX DIAO 2.22 8.4 28125 005 9 CIRCULATORY DISORERS (XO All, 8ITH CARD CATA 1/O COMPLEX DIAG 1.6455 5 25126 005 M ACUTE 4 SUIACUTE ERDOCARDITIS 2.6643 18.4 38127 00 M HEART FAILURE i SMOCK 1.90 7.6 28120 005 A DEEP VEIN THROMOPHJ.EIITIS 0.0439 0.6 28129 000 CARDIAC ARREST 1.5506 4.6 25130 005 A PERIPHERAL VASCULAR DISORDERS AGE 69 AND/OR C. C. 0,945 7,1 27131 005 A PERIPHERAL VASCULAR DISORDERS ARE (70 N/0 C. C. 0.9491 .4 26132 005 A ATHIEROSCLEROS AGE )69 AND/OR C. C. 0.9182 .7 27133 005 A ATNEROSCLEROSIS ARE (10 WD C. C. 0.8599 5.2 25134 005 M NYPERTENSION 0.7049 6.1 26135 805 0 CARDIAC CONGENITAL 4 VALVULAR DISORDERS ARE )69 AND/OR C. C. 0,9922 6.1 26136 005 M CARDIAC CONGENITAL 4 VALVULAR DISORDERS AGE 18-69 IO C. C. 0.9674 4.9 25137 005 M C01AC CONGENITAL 4 VALVULAR DISORDERS AGE R-17 0,4581 3.3 20138 005 N CARDIAC ARRHYTHMIA 4 CONDUCTION DISORDERS ABE )69 AND/OR C. 0. 0,9297 5.7 26139 005 M CARDIAC ARRHYTHNIA 4 CONDUCTION DISORDERS ARE (70 H/ C, C. 0.8:03 4.6 23140 00 M ANIIA PECTORIS 0.7541 5.5 21141 005 M SYNCOPE f COLLAPSE AGE "49 AND/OR C. 0. 0,441 5 21142 005 M SYNCOPE + COLLAPSE AGE (70 I/0 C. C. 0.$"6 4.3 Ie143 005 M CHEST PAIN 0.6814 4,4 19144 005 M OTHER CIRCUILATOIY DIAGNOSES WITH C. C. 1.167 7 27145 005 M OTHER CIRCULATORY DIAGNOSES U/0 C. C, 1.002 4.4 26146 006 8 RECTAL RESECTION AGE )69 AND/OR C. C. 2.7062 19.1 30147 006 3 RECTAL RESECTION AGE (70 /O C. C. 2.3087 17.9 38146 004 5 MAJOR SMALL 4 LARGE lOREL PROCEDURES AGE )69 AND/OR C. C, 2.5493 17 37149 006 3 MAJOR SMALL 4 LARGE BOWEL PROCEDURES AGE (70 6/0 -. C. 2.2154 15,2 35150 006 5 PERITONEAL ADHESIOLYSIS AGE )69 AND/OR C. C. 2.3746 11.3 36

182

DIAGNOS5 RELATED GROUPS AND SELECTED RELATIVE WEIGHTSNIFA HCFA1983 NHFA 1983

RELATIVE 1903 CUTDAG HOC TYPE TITLE HEIGHT ALOS OFF

111 006 5 PERITONEAL AONEIOLYSIS ARE (70 91/ C C. 2.0274 13.4 33152 006 S NINO0 SHALL 4 LARGE IOEL PROCEDURES AN 3069 AND/OR C. C. 1.4051 10.6 31133 006 5 HINON SALL + LARGE ION[L PROCEDURES AGE 170 NO D. C. 1.2599 9.3 29154 006 5 3TOHACH, ESOPHAGEAL + DUODENAL PROCEDURES AE 39 AND/OR C. C, 2.6901 14.8 35155 006 S STVAC0, ESOPHAGEAL + DUODENAL PROCEDURES AGE 10-69 U/0 C. C. 2.3334 13 33156 006 3 STOMACH, ESOPHAGEAL + DUODENAL PROCEDURES AGE 0-17 0.847 6 20157 006 5 ANAL PROCEDURES AGE m69 AND/OR C. C. 017905 6 25158 006 ANAL PROCEDURES AGE 170 V/D C. C. 0,6400 5.2 19159 006 5 HERNIA PROCEURES EXCEPT IUINAL f FEMORAL ABE 069 AND/OR C. C. 0.9297 7.1 23160 006 5 HERNIA PROCEDURES EXCEPT INGUINAL + FEHORAL AGE 18-69 0/0 C. C, 0.7676 6 181 006 8 INGUINAL * FEHORAL HERNIA PROCEDURES AGE 69 AND/OR C, C, 0.7060 .7 16162 006 8 INGUINAL FEHORAL HERNIA PROCEDURES ABE 10-69 9/0 C. C. 0.5654 4.0 12163 006 5 HERNIA PROCEDUR3 AGE 0-17 0.4350 2.1 6164 006 $ APPENDECTOrY WITH COMPLICATED PRINr. DIAl AI1)69 AND/OR C. 0. 1.032 11.9 32165 006 5 APPENDECTOMY WITH COMPLICATED PRINC. DIAE AGE (70 4/0 C. C. 1.6154 11.3 29166 006 5 APPENDECTOMY 0N/ COHPLICATED PRINO, 5110 AGE )69 AND/OR C. C. 1.4320 9.4 2?167 006 3 APPENDICIORY N/O CORPLICATED PRINC. 01A0 AGE (70 4/0 C, C. 1.0018 7.4 22168 006 S PROCEDURES ON THE MOUTH AGE 65 AND/OR C.C, 0,8631 4.3 24160 006 5 PROCEDURES ON THE ROUTH AGE (70 N/O C. 0. 0,8992 4.2 24170 006 1 OTHER DIGESTIVE SYSTEM PROCEDURES AGE 39 AND/OR C. C. 2.6602 14.6 35171 006 5 OTHER DIGESTIVE SYSTEM PROCEDURES AGE (70 U/0 C. 0. 2,3976 13.3 33172 006 N DIGESTIVE MALIGNANCY AEi 69 AND/OR C. C. 1.2269 0.2 28173 006 N DIGESTIVE MALIGNANCY AGE (70 /0 C. C. 1.0517 6.7 27174 006 H I[ RAIIA ARE 309 AND/OR C. 0. 0.9201 W.) 27175 006 N I.I. HEMORRHAGE AGE (70 R/O C. C. 0.8236 5.8 24176 006 R COMPLICATED PEPTIC ULCER 1.2430 0.1 20177 006 M UNCOMPLIOATED PEPTIC ULCER )69 AND/OR C. C. 0.7422 6.6 24171 006 N UNCOPLICATED PEPTIC ULCER (70 8/0 C. C. 0.6141 0.5 20179 006 M INFLAMMATORY IOEL DISEASE 1.0133 0 20100 O A 6.R .OISTRUCTION AE 09 01/O C. C. 0.8197 6.2 26101 006 N 0.1. OlSTRUCTI00AE (70 0/0 C. C. 0.7045 5.9 26182 006 0 ES0PHAGITIS, ASTAOENT, 4 RISC. DIGEST. D1 ARE 69 +/OR C. C. 0,61R5 5.4 22103 006 0 ESOPHAIITIS, ASTROENT. 4 RISC, DIGEST, DS ANE 10-69 9/0 C. C. 0.652 4.6 19184 006 R 13OPHAOITIS, RASTROENTERITIS + MISC. DIREST. DISORDERS AGE 0-17 0.3022 3.3 11105 006 H DENTAL 4 ORAL DIS, EXC EOTRACTION + RESTORATION, AGE )17 0.66e] 4.2 24186 006 M DENTAL + ORAL 01S, EXC EXTRACTIONS + RESTORATIONS, AGE 0-17 0.4155 2.9 11107 006 0 DENTAL EXTRACTIONS 4 RESTORATIONS 0.3" 2.7 0100 006 N OTHER DIGESTIVE OYSTER DAINOSES AGE )69 AND/OR C. C. 0.7444 5.1 25109 006 N OTHER DIGEISTIVE SYSTEM BIANSRES AGE 18-69 0/O C, C. 0.6576 4.5 23190 006 N OTHER DI1ESTIVE OYSTER DIAGNOSES AGE 0-17 0.3379 2.1 i191 001 S MAJOR PANCREAS, LIVID * SHUNT PROCEDURES 4.1791 20.8 41192 007 5 NINON PANCREAS, LIVER + SHUNT PROCEDURES 3,9197 20.1 40193 007 5 IILIARY TRACT POC ETC TOT CHOLEDYSTECTONY ARE 369 f/OR C. C. 2.4013 17.3 37194 007 II ILIARY TRACT PROC EXC OT CHOLEDYSTECTOMY ACE (70 W/O C. C. 1.9881 13.9 34195 007 5 TOTAL CNOLECYSTECTORY WITH .C.E. AGE )69 AND/OR C. C, 2.169 16 36196 007 8 TOTAL CROLECYSTECTOY N1H C.D.E. AGE 170 W/O C. C, 1.0594 15.1 36197 007 5 TOTAL CNOLECY3TECTOHY W/O 0.0.1. AE 69 AND/OR C. C. 1.4860 11.5 29198 007 3 TOTAL CNOLECYSTECTORY 0/0 C.D.[. ABE (70 W/0 C. C. 1.2752 15.1 24199 007 5 HEPATONILIARY DIAGNOSTIC PROCEDURE FOR NALIGNANCY 2.4574 17.9 30200 007 5 HEPATOIILIARY DIAGNOSTIC PROCEDURE FOR RON-MALIHANCY 2.591C 15.1 35

183

DIAGNOSIS RELATED GROUPS AND SELECTED RELATIVE WEIGHTSHCFA MCFA1903 HCFA 1903

RELATIVE 983 CUTDIG "PC TYPE TITLE MIGHT ALOS OFF

201 007 5 OTHER HEPATOIILIARY OR PANCREAS 0.R. PROCEDURES 2.7291 16.9 37202 07 N CIRRHOSIS + ALCOHOLIC HEPATITIS 1.1965 9.3 29203 007 N NALIGNANCY Of NEPATONILIANY SYSTEM OR PANCREAS 1.0937 8 20204 07 N DISORDEI OF PANCREAS EXCEPT NALI RANY 0.9682 7.S 20205 007 N DISORDERS OF LIVER FXC 0ALS, CIRO, ALC HEPA AlE )49 AND/OR C. C. 1.0822 7.9 28206 007 N DISORNERS OF LIVER SXC NALIE, DIRN, ALC HIPA AE (70 M/ .C. 0.9247 6.8 27207 007 N DISORDERS OF THE IILIARY TRACT AGE 349 AND/OR C. C. 0,9492 6.6 2720R 007 N DISORDERS OF THE ISLIANY TRACT AGE (70 W/O C. C. 0.7315 5,5 24209 00 8 MAJOR JOINT PHOCEDURES 2.2912 17.1 37210 000 S RIP 4 FElOUR PROCEDURES EXCEPT NAJOR JOINT AGE )69 AND/OR C. C. 2.0033 17,3 38211 008 HIP 4 FEfRdE PROCEDURES EXCEPT NAJOR JOINT AGE 10-69 W/O C, C. 1.953 15.9 36212 003 $ HIP 4 PIRUR PROCEDURES EXCEPT MAJOR JOINT A6E 0-17 1.7132 11.1 31213 008 1 ANPUTATIONS FOR HUSCULOSKELETA SYSTEM + CONN. TISSUE DISORDERS 2.1313 14.3 :4214 00 8 RACK 4 NECK PROCEDURES AlE )9 AND/OR C, C. 1.0427 11.6 6215 00 5 lACE 4 NECK PROCEDURES AGE (70 N/0 C. C. 1.492 13 33216 008 3 1IOPIE[ OF NUSCULOS[LETAL SYSTER + CONNECTIVE TISSUE 1,5596 11.3 31217 0018 lND PIJIID 0 SEN IRFT EXCHAND, F0 NUSCULOSIELETAL 4 CONN. TISI. 01 2.2824 13.1 33210 008 LONER EITER HUNER PROC EXC HIP, FOOT, FENUR ARE 349 +/OR C. 0. 1,42S 10.9 341219 000 3 LONER EXTREN + HUHEp PROC Lin HIP, FOOT, FENDR AGE 10-69 0/0 C, 0. 1.079 8.3 27220 00 S LOVER [XTRM + HUMER PROD EXC HIP, FOOT, FENUR AlE 0-17 0.9339 5.3 25221 008 I KNEE PROCEDURES ARE 349 AND/OR C. C. 1,2727 8.3 28222 OD 8 KNEE PROCEDURES AGE (70 W/O 0. C. 0."97 1.4 26•223 O S UPPER EXTREMITY PROC EX NUNERUS 4 HARD All )69 AND/OR 0. C. 1.0723 6.9 27224 DON I UPPER EXTREMITY PROD EX HUMERUS + HAND AGE (70 W/O C. C. 0.1952 5.6 24225 003 I FOOT PROCEDURES 0.6476 4.8 15226 003 S SOFT TISSUE PROCEDURES AE )69 AND/OR C. C. 0.7934 5.1 25227 W08 5 SOFT TISSUE PROCEDURES ABE (70 0/0 C. C. 0.6337 4.2 18228 00H I 8 ANLION (HAND) PROCEDURES 0.3626 2.2 7229 00 8 HAND PROCEDURES EXCEPT IANILION 0.5998 3,4 14230 00 O LOCAL EXCISION 4 RENOVAL OF INT FIX DEVICES OF HIP + FEHUR i.3594 3.9 29231.0 3 LOCAL EXCISION + RENOVAL OF INT FIX DEVICES EXCEPT HIP + FEMUR 0.95319 5.3 25232 0on S ARSHEOSCOPY 0.6063 3.6 15233 O0OS OTHER HUSCULOSKELET YS + COHN T1SS O.0. PROC ARE )69 +/OR C. C. 1.7737 13.1 33234 ON 8 OTHER NUSCULOSIELET SY 4 COHN TSS O.,, PROC AGE (70 N/O C. 0. 1.2454 8.2 28235 008 H FRACTURES OF FENUR 1.7584 13.6 34236 0 N FRACTURES OF HIP 4 PELVIS 1.3855 11.9 32237 000 N SPRAINS, STRAINS, 4 DISLOCATIONS OF HIP, PELVIS + THIGH 0.7929 6,4 26238 001 N OSTEOIYLITIS 1,511 12.3 32239 008 N PATHOLOGICAL FRACTURES 4 MUNULOSIELETAL 4 CONN, TISS. NALINANCY 1.0979 9.2 29240 000 N CONNECTIVE TISSUE DISORDERS AlE )69 AND/OR C. C, 0.9709 8.6 29241 008 H CONNECTIVE TISSUE DISORDERS AiE (70 W/0 C. C. 0.9040 3 20242 ON N SEPTIC ARTHRITIS 1.88 11.2 31243 003 N MEDICAL IACK PROILE48 0.731 7.5 20244 009 H JONI D18ASEO 4 SEPTIC ANTHROATHY ARE 69 AND/OR C. C. 0,7792 7.3 23245 000 SONE DISEASES * SEPTIC ARTHROPATHY AUE (70 N/O C. C. 0.7177 6.3 26246 003 N NON-SPECIFIC ANTHROPATHIES 0.7147 6.8 27247 008 M 51IM4 4 SYNPTONS OF NUOCULOSKELETA., OYSTER + COHN TISSUE 0.6559 5.N 26240 003 N T|NDONITIS, MYDSITIS + IURSITIS 0.6136 5.4 24249 008 0 AFTERCARE, MUSCULOSKELETAL SYSTEN I CONNECTIVE TISSUE 1.0203 7.6 20250 000 9 Fg, SPRNS, STIRS * DISL OF FOREARM, HAND, FOOT AE 349 #/OR C. C. 0.7420 6 26

184

DIAGNOSIS RELATED GROUPS AND SELECTED RELATIVE WEIGHTSHCFA HCFA1903 HCFA 1983

RELTIVE 1903 CUToil NOD TYPE TITLE HEIGHT ALOS OFF

251 000 M FX, SPINS, STANS # DSL OF FOREARM, HAND, FOOT AGE 11-69 W/O C. C. 0.5464 1.2 24252 009 k FX, SPINS. STRNS f DIS OF FOREARM, HAND, FOPT AGE 0-17 0.3533 1.0 7253 009 FX, SPANS, STAH + D1SL OF UPAR , LOULE EX FOOT AE 69 4/OR C. C. 0.74iw 6.6 27214 000 N FX, SPANS, STRHAR DIlL OF UPARN, LOEI EX FOOT AGE 10-69 W/0 C. C, 0.62" 5.3 25215 O N FX, SPRN$, 3TRN VINL OF UPARM, LOWLEG El FOOT AGE 0-17 014687 2.9 15256 000 N OTHER DIAGNOSIS OF MUSCULOSRILETAL SYSTEM 4 CONNECTIVE TISSUE 0,8)06 6.5 27USi 009 3 TOTAL NASTECTOMY FOR MALIGNANCY AGE 469 AND/OR C, 0. 1.1091 P.3 232SO 009 S TOTAL MASTECTONY FOR MALIGNANCY AGE (70 H/O C. C. 1,0729 U.9 21259 009 S SUiTOTAL MASTECTOMY FOR NALISNANCY A11 069 AND/OR C. C. !,0141 7.4 27260 009 S SUOTOTAL NASTECTOMY FOR NALIGNAHCY ACE (70 0.9325 6.4 262d1 009 5 IREAST PROC FOR NHO URAL E XCEPT BIOPSY * LOC EXC 0.7329 4.0 19262 00' $ BREAST BIOPSY i LOCAL EXVI9IOM FO0 NON-MALiNANCY 0,617 3 10263 009 5 SKIN GRAFTS FOR SKIN ULCER OR CELLULITIS AGE )69 AMD/OR C. C. 2,4737 21.3 41244 009 S SKIN GRAFTS FOR SIfN ULCER OR CELLULITIS AE (70 4/0 C. C. 2.2031 18.2 39265 009 1 SIN GRAFTS EXCEPT FOR SI1 ULCER OF CELLULITIS WITH C. C. 1.4959 e.6 29266 009 5 SKIN IRAFTS EXCEPT FOR SKIN ULCER O CELLULITIS 4/O 0. C. 0.9495 5,9 2626) 009 5 PEAIANAL + PILONICAL PROCEDURES 0.6113 5 19268 009 S SKIN, SURCUIANUI TISSUE 4 IRASf PLASTIC PROCEDURIE 0.1384 3 is269 o 5 THER SlIN, SUSCUT TISS + IREAST O.0, PROC All (69 +/D C. C. 0,9947 5.? 28270 009 5 OTH SKIN, SUICUT TISS 4 BREAST 0,0. PROC AlE (70 W/O C. C. 016]23 4.5 25271 009 M SKIN ULCERS 1.302 12.1 32272 009 M MAJOA 011H DISORDERS ARE )69 AND/OR C, C. U.1062 7,0 20273 009 N MAJOR 011 DISOIDERS AGE (70 0/O C. '. 0.8286 7.3 27274 009 0 MALIHANT BREAST DISORDERS AGE 369 AND/OR C. C. 1.010 7.5 29275 009 N MALIGNANT IREAST OIORDERS AGE (70 9/O C. C, 0.9014 6.4 26276 009 M ION- ALIGNANT IIAST DISORDERS 0.6066 4.2 22277 009 0 CELLULITIS AGE )69 AND/OR 6, C. 0.0063 0.3 20270 009M TELLILITI$ AGE 10-69 9/0 C. C. 0.0096 7.2 27279 009 N ELLULITIS AGE 0-17 0.4789 4.2 13210 009 N TRAUMA TO IE 3I14, SUUT TISS+ IREAST AGE )9 i/OR0 C. C. 0.6201 5.4 25281 009 0 TRAUMA TO THE SKIN, SUICUT TISS + IREAST ABE 10-19 /O 0. C. 0.5377 4.2 23212 009 H TRAUA TO THE SKIN, SUICUT 115 4 BREAST AGE 0-17 0.346 2.2 9283 009 N MINOR SK1N DISORDERS APE )69 AND/OR C. V. 0.6394 5.3 25284 009 N MINOR SKIN DIJORDERN AgE (79 A/D C, C. 0.5971 4.4 24203 010 1 AMPUTATIONS FOB ENDOCRINE, NUTRITIONAL M ETABOL7C DISORDERS 2.0650 24 44216 010 S ADRENAL 4 PITUITARY PROCEDUNES 2.3932 16.1 36207 010 1 SKIN GRAFTS + ROUND DEIIDE FOR ENDOC HUTRIT 4 NETAI OSORDERS 2.0143 22.0 43208 010 5 C.R. PROCEDUkES FOR OBESITY 1.5695 10 24209 010 5 PARATHYROID PROCEDURE$ 1.3736 1.3 20290 Ol S THYROID PROCEDURES 0.9549 6 17291 010 3 THYROiLOSSAL PROCEDURES 3,4909 2.9 0292 010 1 OTHER ENDOCRINE, NUTRIT 4 NETAI OR. PROC AIE )69 4 DRE. C,. 2.0307 10,0 31293 010 1 OTHI ENDOCRIHE, NUTRIT 4 METAI O,, PROC ABE (70 0/N C. C. 1.4951 N 20294 010 H DIAiETES AGE E 36 l 0007 7.7 20295 OIC N DIABETES ARE 0-35 0,7417 5.6 26296 DI N NUTRITIANAL + MISC, lETA10.IC DISORDERS AGE )69 AND/OR C. C. 0,979 7.3 27297 010 M NUTRITIONAL + MISC, METABOLIC DISORDERS AGE 18-60 W/O C. C. 0.7923 6 26298 010 M 9 UTRITIOdAL 4 MISC, METAILIC DISORDECS AGE 0-17 0.530 5.4 25299 010 N INBORN ERRORS OF METABOLISM 0.9407 6.0 27300 010 N ENDOCRINE DISORDERS AGE )69 AND/ON C. C. 0.973) 7.0 28

18 5

DIAGNOSIS RELATED GROUPS AND SELECTED RELATIVE HEIGHTSNCFA HCFA1993 (HCA 1993

RELATIVE 19083 CUTPRO MOD TYPE TITLE MEIGHT ALOS OFF

301 015 0 ENDOCRINE DISORDERS A W (70 /C C. 0.814. 6.A 26302 011 S KIDNEY TRANSPLANT 4.22 24.1 44303 011 S KIDNEY. URETER 4 MAJOR $LADDER PROCEDURE FOR NEOPLASM 2.530 16. 36304 01 3 KIDNEY, URETER 4 0AJ GLOAPROC FOR NON-GALIG AlE )9 4/OR C. . 1.7952 12.6 31305 011 5 KIDNEY, URETER 4 KAU ILON PROD FOR NON-RALIO (70 /D C. C. 1.7043 11.9 3306 ol 0 PMOSTATICTOY AlE )69 AND/ON C. C. 1.13" 8.3 29307 011 3 PROOTATECTOY ABE (70 U/U C. C. 0.9513 7.2 26300 011 8 MINOR ILADR PROCEDURES AGE 69 AND/OR C. C. 1.0441 7.1 27309 Oi 5 MINOR ILADODiR rROCEDURES AGE (70 9/0 C. C. 0,92 5.7 26310 Ol S TRANSURETHRAL PROCEDURES AGE )69 AND/OR C. C. 0.7071 4.9 20311 Oil 0 TRAN3URETHRAL PROCEDURES AGE (70 U0 C. C. 0.71 4.1 11:12 01 5 URETHRAL PROCEDURES, AGE )69 AND/OR C. C. D.7424 5.2 22313 011 5 URETHRAL PROCEDURES, AGE 16-6P W/O C. C. 0.6997 5.1 21314 01 5 URETHRAL PROCEDURES, AGE 0-17 0.4360 2.3 11310 011 OTheR KIDNEY + URINARY TRACt O.R. PROCEDURES 2.4884 9.8 303;6 01l M RENAL FAILURE 1.3314 6 ,7 27317 01 M ADMIT FOR RENAL DIALYSIS 0.2305 1.2 331601 M KIDNEY 4 URINARY TRACT NEOPLASMS AGE 69 AD/OR C. C. 0.9142 5.5 26;19 Oil N KIDNEY 4 URINARY TRACT NEOPLASMS AGE 70 /O C. C. 0,7942 4.2 24320 Ol KIDINEY 4 URINARY TRACT INFECTIONS AGk )69 AND/OR C, . 0.R123 7 27321 011 0 KIDhNE 4 URINARY TRACT INFECTIONS AGE 16-69 0I C. C. 0.6803 5.6 23322 01l KIDNE + URINARY TRACTINFECIIONS ABE 0-17 0.453 3.7 13323 Oil B URINARY I NES Alf )69 AND/OR C. C. 0.7131 4.9 25324 Oi I URINARY DIONE AGE (70 W/O C. C. 0.5472 3.9 19325 011 N MIDNEY 4 URINARY TRACT SIGNS + SYmPTOOS AGE)69 AND/OR C. t. 0.7247 5,4 25326 011 N KIDNEY + URINARY TRACT SIGNS 4 SYMPTOHM AGE 1069 RIO C. C. 0.3075 4.3 21327 O1l N KIDNEY + URINARY TRACT SIGN 4 SYMPTMfl AAF 0-17 0.3r.7 3,1 14329 011 l UETHAL SIRICTURE AlE )69ND/OR C. C. 0.65% 4.0 22329 01 N URETHRAL STRICTURE ARE 10-A5 N/O C. C. 0.542 3.9 17330 011 N URETHRAL STRICTURE AGE 0-17 0,211 1.6 5331 011 N OTHER KINTEY 0 URINARY TRACT DIAGNOSES AGE )69 AND/OR C. C. 0.9919 6.3 26332 Oil K OTHER KIDNEY 4 URINARY TRACT DIAGNOSES AGE 18-69 9/0 C. C. 0,7763 5 25333 011 M OTHER RIDNEY * URINARY TRACT DIAGNOSES AGE 0-17 O.Si46 3.2 18334 012 5 MAJOR BALE PELVIC PROCEDURES WITH C. C. 1.5412 12.7 30335 012 5 MJOR HALE PELVIC PROCEDURES N/ C. C. 1.359 11.0 29336 012 5 TRANSURETHRAL PAOSTATECTORY AGE )69 AND/OR C.C. 1.0079 0.4 22337 012 5 TRANSURETARAL PROSTATECTOMY AlE (70 A/0 C. C. 0.8401 7.2 17338 012 8 TESTES PROCEDURES, FOR MALIGNANCY O.9094 6.3 26339 012 5 TESTES PROCEDURES, NON-MALIGNANT AGE W17 0.6091 4.5 15340 012 5 TESTES PROCEDURES, NOW-MALIGNANT AGE 0-17 C.431 2.4 7141 012 5 PENIS PROCEDURES 0.9903 d 23342 012 3 CIRCAMISIPN AGE )17 0.4220 2.0 10343 012 S CIRCUMCISION Al 0-17 0.3820 1.7 4344 012 S OTHER MALE REPRODUCTIVE SYSTEM O.R. PROCEDURES FOR MALIGNANCY 1.1204 7.4 27345 012 0 OTHER MALE REPRODUCTIVE SYSTEM O.A. PROC EXCEPT FOR MALlS 0.N334 5.6 26346 012 N NALIGNANOCY HALE REPRODUCTIVE SYSTEM, ARE )69 AND/OR C. C. 0.9395 6.9 27347 012 N MALIGNANCY, MALE REPROVUCTIVE SYSTEM, AGE (70 U/O C. C. 0.0304 5.7 26340 012 M ENZIN PROSTATIC HYPERTROPHY AGE )69 AND/OR C. C. U.0964 6.2 26349 012 M BEHIGN PROSTATIC HYPEATROPH AUE (70 U/O C. C. 0.4999 4.9 22350 012 M INFLAMMATION OF THE MALE REPRODUCTIVE SYSTEM 0.6096 5.2 20

186

DIAGNOSIS RELATED GROUtS AN0 SELTED RELATIVE WEIGHTSHCFA HCFA1953 HCFA 1983

RELATIVE 1903 CUTus moc TYPE TITLE MKINT LOS OFF

351 012 N STERILIZATION, ALE 0.2655 I.q 3332 012 N OTHRi PALE 1EPRODUCTIVE SYSTEM DIAGNOSES 0.6385 4,4 20353 013 3 PELVIC EVISCERATIOR, RADICAL HYSTR|CTOMY + VULVEITOY 1.9116 12.4 32354 013 5 SOS-RADICAL HYSTEBGCTOflY AGE )9 AND/OR . C. 1.1108 9.6 20355 013 S SOS-RADICAL HYSTERECTOMY AGE (70 WO C. C. 1.0156 0. 17356 0l S FEALE REPRODUCTIVE SYST1 RA1ONSTUCTIVE PAOCOEURES 0.046 $.1 11357 013 U UTERUS AENXA PROCEDURS, FOR MALIGNANCY 1.9180 13.9 34358 013 3 UTERUS 4 ONW5A PROC FOR 0 -MAL1dANCY E19CPT TU AL INTErEUPt 1.009 8 218359 013 S TUIAL INTERRUPTION FOR 0-ALIGNAACY, 0.4279 A.3 7360 013 5 VAGINA, CoRViC 4 V~tqVA PROCEDURES 1,3905 4,2 19361 013 1 LAPAROSCUPY + ENDOSCOPY (FEMALW) EXCEPT TUBAL INTERRUPTTON 0.4864 2.6 10362 013 3 LAPAROSCOPIC TUAlI INjERRUPTION 0.3126 1.4 3363 013 5 P + C. CONIlhTIOH + RAOD-4PLT. FOR MALIINANCY 0.5016 4.3 18364 013 1 D4C, CONfZATION EXCEPT FOR MALIGNANCY .. 4020 2.6 9365 013 8 OTHER FEMALE RSPRODUCTIV( ITN OR. PROCEDURES 1.7965 12.7 3136 013 R OALIGNANCI, FEMALE REPRODUCTIVE SYSTEM AGE 69 AND/OS C. C, 0.9444 5.2 25367 013 N MALISNANCY, EALE REPRUOUC TI IYSTEM AGE (70 W/O C. C. 0.5706 3.5 2431 013 N INFECTIONS, FEMALE RCPRODUCTO/E 3YS(M 0.7944 6:7 2)369 013 4 N HPTRUAL + OTHER FINALE REPP3000TVE SYSTIM DISORWARS 0.6959 5,1 25370 014 0 CESAR1AK SECTION 01TS C. C, 0,9912 7.6 15371 014 5 CSAREAk SPTIOS 0/0 C. C, 0,7531 6.1 10372 014 N VAGINAL DELIVERY S3.H OOOPLICATIHR DIAGNOSES 0,5534 3.8 9373 014 A VAGINAL DELIARYT ;/0 CDOPLICATIAG 0IA3IlO3E 51401 3.2374 O4 0 AAIINAL DELIVERY 11H ITEAILIZATION AND/OR +0 0.54+ '.6 7375 014 S VAGINAL DELIVERY HITH O.R. PROD EXOEPT STEkIL MD/O 0Ot 0.60"9 4.4 15376 014 r POSTPARTUM DIAGOSES /0 (,.A, POO4OURE 0.4158 2.9 10377 014 3 POUTPAATL K DIAf4OSES IT1H 0.A, PROCEDURE 0,476i 2.2 0378 014 h ECTOPIC F[IBdAaCY 0.1094 1.5 11379 014 0 THREATEKIH ABORTION 0.3169 2.2 9300 ol K AIBrTION V/C 14C 0.2705 1.5 4311 014 ABORTION 99tH 0T * 0,3602 1,4 4302 014 N FALL LASIO 0.1842 1.2 2383 0!4 N OTHER AOTEPARTUM IAINOSES SITH MEDICAL CORPLICATIONS 0.431? 3,4 14304 014 H OTHER ANTEPARTUM DIFOSES U/0 MEDICAL CnflPLICATIONS 0.324. 2,2 9385 NIS HEONAIKS, OIEO OR TRANSFERRED 0.6883 1,0 14386 015 EXTREM- INNATURITT NEONATE 3.6061 17.9 303J7 C11 PRERAFURITY WITH MAJO PROBLEMS 1.8459 13.3 3368 015 PENATURITY 0/0 MAOR rOOILEMS 1.1693 8.6 29309 018 FULL TERN IJIONATI RIT MAJOR PROBLES 0.5482 4.7 16390 015 SEONATES 011S OTHER SIGNIFICANT PROBLEMS 0.3523 3,4 9391 0)5 NORMAL NE BORNS 0.2211 3.1 7392 016 S SPLENECTOMY ABE )17 2.7744 16.4 36393 016 3 SPLENECTOMY AGE 0-17 1.5366 9.1 29394 016 5 OTHER OR. PROCEDURES OF THE BLOOD + BLOOD FORMING ORGANS 1.1146 6.1 26395 016 N SIP BLOOD CELL DISORDERS AGE )17 0.7039 6.1 26396 016 N RED BLOOD CELL DISORDERS AGE 0-17 0.6295 4.1 19397 016 R LOAGULATION DISORDERS 0.986J 6.7 27390 016 N RETICULOENDOTHELIAL 4 iSRUNITY DISORDERS AGE )69 AND/OR C. C. 0.89 6.1 26399 016 N RETICULOENDOTHELIAL + INMUNITY DISORDERS AGE (70 0/0 C. C. 0.0459 3.6 26400 017 S LYMPPtOMA ON LEUKEMIA SITH MAJOR 0.0, POCEPURE 2.272 16.9 37

187

DIAONOSII RELATED GROUPS AND SELEUT~b RELATIVE WEIGHTSHCVA HCFA1983 HCFA 1903

RELATIVE 1983 OdT014 MRC TYPE TI TLE REIGHT ALO OFF

401 017 5 LYNPHMA OR LEUXEHI WITH MINOR O,. PROC AGE )69 AHD/OR 0. C. 1.2409 89 '4402 01, S LYMPHOfI %OR LEUKEMIA WITH HINOR O.R. PROCEDURE ABE.(O U/0 C. 0. 1.1316 7.1 27403 017 N LYnINO% R EU E IA AGE 69 AND/O C. C. 1.1715 7.1 27404 017 DS LY'PNA 0R LEDCEMIA AGI 10-61 W/O C. C. 1. 1737 4.1 26405 017 M LYAPNHOA Of LEUKEMIA AG 0-17 . '.0511 4,9 25406 017 S NYELOPROLIF DISO D OR POORLv 1]FF NEOPLASM V MAJ OR. PRO0 4 C, C. 2,2621 -I 3540O 81 9 MfELOPROLIF D1SOAD OR POORLY D1FF NEOPk V HAI O.R. PA W/O C. 0. 2.1;46 13,3 33408 017 $ HYELOPROLIF DNORD OA POORLY DIOF TI[OPL WITH MINOR O.0. PROC , 39 7.1 274m Ov1 N RADIOTHERAPY 0. %U 5.7 26410 017 M CREMOTNERAPY 0.3527 2,4 12All 017 N HISTORY OF MALIGHANCY W/O ENDOSCOPY 0,72n 4,7 .5412 017 f HISTORY OF RALIHNAO, WITH ENDO.OPY 0,34 2 1413 017 M OTHER HYELOPROLIF 015080 OR POORLY D1FF NEOPL OX ZE/O0 41/R C, C. J,O)75 7.3 27414 017 H' 04k HYLOPROLIF DISORD OR POORLY D0FF htCPL OX ARR1(0 A/OR. C: L03W .4 26415 019 C 0.1. PROCEDURE FOR INFECTIONS * PARASITIC DZSEASES 3.bU27 15.1 3S416 010 H SEPTECEMIA AGE )17 .1.5504 9.2 P417 018 0 SEPTECERIA All 0-1; 0f1 I2 5.2 20418 010 0 POSTDPERATIVE 4 POST-TRAUNATIC INFECTIONS 0. 1 0.4 2" I '419 01 f FEVER OF UHONOW ORUIDN All )6$ RHD/OR C. C. OSAS 6.1 27420 01e f FEVEP OF UNKNONH ORIGIN AGE 18-69 N/0 0,1 . O.O. 6.2 0.421 010 M VIRAL ILLNESS AlE )17 0 6,4Y 0.4 21422 010 M VIRAL ILLNESS 4 M3OER OF UNKHOWN 081019 AlE 0-17 0,456 3.2 .1423 010 H OTKR INFECTIOUS + PARASITIC DISEASES DIAGNOSIS 1.2107 8.0 29424 019 $ O.0. PROCEDURES PITH PRINCIPAL DIAGNOSIS OF HINIMAL ILLNESS 2.193H 14.1 34425 019 M ACUTE ADJUST REACT 4 DISTURIARCOE OF PSYCHOSOCIAL DYSFUNCTION 0,4012 5.0 26426 019 f DEPRESSIVE NEUROSES 0,9491 9.4 29427 019 R NEUROSES EXCEPT DEPRESSIVE 0.7010 6.9 27428 019 ft DISORDERS OF PERSONALITY 4 IMPULSE CONTROL 0.9741 1.3 28429 019 M ORGANIC DISTURIANCES + RENTAL RETARDATION 0.9523 k.1 29430 019 M PSYCHOSES 1.0934 10.6 31431 019 N CHILDHOOD RENTAL DISORDERS 2.2511 3.4 35432 019 M OTHER DIAGNOSED OF MENTAL DISORDERS 1 0525 7.2 27433 020 SUISTANCE USE + SUDST INDUCED ORGANIC MENTAL DISORDERS, LEFT ANA 0.4457 2.5 17434 020 D8UG DEPENDENCE 1.0404 9.1 29435 020 DRUG USE EXCEPT DEPEUDEHCE 1.031 H 2v436 020 ALCOHOL DEPENDENCE 0.0651 8. 28437 020 ALCOHOL USE EXCEPT DEPENSENCE 0.6103 3.5 24438 020 ALCOHOL + SURSTANCE INDUCED ORGANIC MENTAL SYNDROME 0.842 .9 27439 021 8 $lIN GRAFTS FOR INJURIES 1.9219 5.9 29440 021 5 WOUND DEIRIGERENTS FOR INJURIES 1.4007 7.2 27441 021 5 HND PROCEDURES FOR INJURIES 0.719 3 16442 021 5 OTHER O.D. PROCEDURES FOR INJURIES AGE 19 AND/OR . C. 1.9026 9.1 29443 021 5 OTHER O.0, PROCEDURES FOR INJURIES AGE (70 W/O C. C. 1.5211 6.6 27444 021 A MULTIPLE TRAUMA AGE 369 AND/OR C, C. 0.083 6.7 27445 021 f MULTIPLE TRAUMA AGE 18-65 W/O C. C. 0.753 5.2 25446 021 f MULTIPLE TRAUMA AGE D-17 0.446 2.4 10447 021 M ALLERGIC REACTIONS AGE )17 0.4785 3.7 19448 021 M ALLERGIC REACTIONS AIE 0-17 0.35n5 2.9 9449 021 f TOXIC EFFECTS OF DRUGS AGE )69 AND/OR C. C. 0.7331 5.6 26410 021 f TOXIC EFFECTS OF DRUGS AGE 30-69 8/0 C. C, 0.3957 3.9 23

188

DIAGHOSIS RELATED GROUPS AND SELECTED RELATIVE HEIGHTSNCFA HCFA

1983 HCFA 183RELATIVE 1983 CUT

PRO HDC TYPE TITLE EIGHT ALOS OFF

4I 021 H TOIIC EFFECTS O DRUGS AGE A-17 0.2)12 2.1 1452 021 N COMPLICATIONS Of TRIATMENT AWE 69 AND/OW C. C. 0.8492 5.5 26453 021 " CONFLICAT IONS OF TREATMENT AGE (70 W/0 C. C. 0.902 5.1 25434 021 N OTO 1IIIJ'RSu, ?UISOH!HGA * TOXIC EFFUZAO AlE )69 AND/DR C. 0. 0.1224 5.3 21451 021 N OHER IRJURIZD, POISONINGS TOXIC EFF DAG AGE 70 HiO C. C. 0.6185 3.5 224U 020 1URNS, TRANSFERRED TU ANOTHER ACUTE CAR! FACILITY 2.D902 11.6 32457 022 EXTENSIVE lURkS 6.W631 12.6 33j)8 02 S WON2XTUSJIV BURNS WITH SKIN GRAFTS 2.8572 18.3 3849 022 S NON-EXTONSIVE BURNS WITH WOUND DEIRIDEMENI + OTH R 0.0, PROC 2.7568 12.7 33460 022 0 NOO*EXTEINSIVE BURDNS /0 U.N. PROCEDURE I.4225 $ 29461 023 5 O.0. PROC WITH DIAGNOSES OF OTHER CONTACT WITH HEALTH SERVICES 1.6307 0 20462 023 H REHABILITATIOH 1.0268 13.5 34463 023 N SIG S 4 SYMPTOMS WITH C. C, 0.7702 6.3 2664 023 M 55645 + SYMPTOMS 9/0 C. C. 0o7 ? 6 26465 032 N AFTERCARE RITH HISTORY OF RALIGNAOCY AS SECONDARY OX 0.2071 1.S 4446 023 . AFTERCARE 8J/ HISTORY OF MALIGNANCY AS SECOADARY DX 0.6377 3.7 24467 023 N OTHER FACTORS INFLUENCING HEALTH STATUS 0.97A9 o.1 2t463 A23 M UNRELATED OR PROCEDURE 2.137 11.2 3146P D 0 PRYA OX INVALID AS DISCHARGE DIAGNOSID 0 0470 0 0 UNGROUPA8LE 0 I I

189

APPENDIX D

FISCAL YEAR 1983 DIAGNOSIS RELATED GROUPS WORKLOAD ANDk

MEDICARE RIBREETLVL

CA -q

0 06

w1' 4046 Nt

N 14w 14

4 mom

* * * * I *"w. - I j 4 1

14 W W W* N) - aN"%.0 S

1190

1 61

0~ N a.0N' N-0m0tm m0 0u4'0wtl. % 4 4A N 044 , 0a

C.4 % 04 0% ti- t e N %MU' km9V r4 * 1 WE U *MN P

N U.- * LM -Q. 0 t4 .s oft 4* *.***4.0 .N. ON 000 M V, 00000 0009 00i w

10 t4 4*M41 N0t 41 * 4b

IN 4W f*'.- 0 r t+4S4M 0.*

'a wNe W440* 4L r5%r I

CA. 4*a*4

4W" ''I 4f &P * 4.0 0'C. . 40 0,0*0r.~I

'U 0

Ma W, 4 *C P vMM

02a0 m NO 0, 10 N0; PN4: . 0 0% afIM0 4

WU 4* Sa 44.4 4* Mo 14 0 iii U41- WO, 4 40p 0 (4 60 r4 a% N ; N~o in4 Aa 00 * 4044 N 4.4.* 1^ 4.. *. 1*4 1" 1

U Jr. 4

ag 0

so Nm 01L0Co NW on

- Urn U

a. a

144

Z 61

.j N -. 041 U.4. a .. N 01,I-0WM N O 10 MI4 0C NOJ4 MO *40.4 (4 4 ;A= W o; rr 1;w- 1% 41D t40 4l.N 0'Pm*.VD40 M%

25 m 00 -0 N- aO L0 M 0 P- - M0 00--' 40a

11 Z iiW 4 40o40, 401U- M 4 RN *%UM %V # 40 N NN0 0%

.aa

2 m0 4, 4,% N 4nW 00 11 01 0 ^; -, 41,MM; 0,P

allb

I.IT 401." 44U 4 4 f N O . 000-1 4+ OMI"..- 40. 0

11~~~4 w 0% 6J %09Ne 11 140 40.1 Z

w0 41 4, 401

L12

0 ZC"Oolo Cli lbnT4CN 'CA mN - i

t-4W mrO%~ 000%mM1D9-N?0" vlir 0 MAW'M@

0n 40 cc 0, 4, 4, a,& I n4nNNg

im - t

a at

m. P4 "Ir O - " 1 -4 I 1 l O

44 '0 It 0% a1010 0 0~ 4- w SNM0 M C7

.. UD x

C61

u~)%g-P c -O - 4 OC go

- ~ ~r r-4O ~ OmO@'UNN C4 N N NU mmo wr1 a --40 00 0 .OI~k~,4Cr~4N N NelN l

V; 001 & 4^ NO 0 U40' WC M M 0NW W.' d .,w ^ too 4 *" OA

InI

I :c0 411

:c 404

4J M4~~~~~~~~4 mJ. m S S i C C00 0 0 0 00 , 0 0 0 0

coU ~ ~ .CJ0C 499 4~~9999 44444459 -

.5 0

a ad

'Ic

00 P Nil- N~.04'0 C4 N N 0 0O 0 O 0. 0 W O .. 9

o ~ a w0* a, 0,-I mOa.- 0%'N0 f9% 0 z~ 0 110-~ U 0MS@".9Log

LdN V~ .

'i6T

40MN04U0Q W04OO%%U4.#-I 4 -r %IO wl&4I~~ .00Ia M 0%

000 OM- -44 ~ 0ON 90 OC O - 0 - NNNNV4M M

N. .4i. C"Ca

I .3 . . .. . . .. ..0 -4 -.*0 M - mNMM N 4'a~C N M

w. w 1p49 -419 4 0 *409- 4 9 94 4* 494.4 mU m to4949 4 400

400wb N N * V a r- 0 *o 0% .

cN 4004V 0N 0 t4t.NU0M -M

490 4w 494 I-- 4 4 "W,00114*1

60.

4ma

m NANU * " qV V) w ~ ~ EE0 11, L43 ' N 4 9 6 0.n 6 - N ~

* ~6M 49 4949 94944944949 49.4949494 494494* in 49 49 P%4 49 M t mN0MN %0MM q NN N

:1

bi 14 0 C4 -0tg N0 N l F6LW 0M40N ~-N 0 Nu- 0 t"..m

AK II.

fo

U %m u . Nm Q 0 % N a,0 ; 4 m-e % 0 C 010 1

961

-~~~~l Ci 0.14' * .f1 '

-* li..0 00 000 0 p "M % F -. O N -0 - 0 Vo

L. wU- cm 4014,U ~ 4 4 O' 4' 41-0-4 *. 6..l.- *V - o

~~~~~~~~~~~~L NM 444 0-l~ M ~ 4 4'4v*-0 9" *. ~W 1.-@ N14 00D0I1(M T4@N '# 0%0-00

4w L d

v w~ 44404a .

= 401

C-4.W

1A

W11 NM (1 %0 C

la au4 W*0 114 .'% ell I.*~ N .14.. .9

bea 2 5

IJI

I. U

961

wo 0 w 40o- r NMN

*P N, NN0.0 co. N D mw %D 0 m o P .

~ 120- 0N NZ WM 00N4* 4 4 04 000M- " 41.a N

@1~~ wL& %m N 0 m lAN 0

41 Oc NN ,P9O 4,4 4 N4 4, .'0 LM -0 0 - S 8; N9 - 04,Nt 4:. 41442 4 40P 404-00 , 40, 00-44 41b.

u 446.4

w 9 0-0.N1 0. 4.41 0. N 01 4 .

to 4, 4* .4

Lm N

1.lhN a) 4 4 L

lu

m ITd 611,.. ~ - 4,,4 N 2 0 Nm.101 .w P9 N 4 tAN..*4K m41 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ r CaM 44N,,, 4444 , , 4 ,1~ ,4

*u uE

£61

- D u r--w0 .'e * * No~o wc'*omi or~u r1 eap-rtmnoJ'99 m0a .. N . ... iN... .-. OO O O- O .--.- . - U

v I O Ow~- r *-omhm~00 C~ o N tiO Nf P%IwZ

al ~ ' N4 0 -el ft0 'ta 0CIDMt tr--0 M'N nM46

*0 c* -

on % ~ l r 0t IM 00 0f '0 -- nt- M-' l f Wr- -ll w

Go m- 1- P 6% M N P--MSu

US

- IN In Qt M" - LI, VN N.

0 CA0 i 4CI " ,09uU

?4 C4NN?4V4r N0

86T

ccw- C ;C 0 O0 00 %'N Ii0 0 0 c

.4 E~0 0 0 0O'O 0 0--~0 ~ 44 0 400

1.1~~W MO O OLI M~ 4 0 0 D 0 0 0 C N ' 0 OM o.

4 Ln N 40. 44 ; M '4*4"%0 P%4*m '4,4* P0: t "M -VZ;NN

0 3C0 4

.00 ell

to w0 -* 4 * 1*41

cc 0 MP -, N.

U 4j 400 0 P(4 O'MdN4 00 N 0L 400. 4

-W UU N4 N~ ' 4* P t v40IN m m4~ N '4, *I' ***'4iU4 tn '4 6- 4o ( N

C-4.

0 0m

C43 N 4 NNNmN W

661

U .0 1i .**Q .. ..... . . .WW - 0-40000000 0NN~0O0C

Wi 1a. 04 "4 ~00C1.0 .4 vkP.0000. -OCo ON..0Uv's 00 m .kM N 444 Wi %,0 f M W-* 400- %-4 "N P. W4"CC00 N 0

W. gmJ~ 40 00 4.4.4 4010 A 4 .Ob44.44 44

-

mo 4=1 N a4.

0 WM 3c

CA A

W 4 % -0 0 0 a 0 M 9-, N m 0 W MN W P'w W' 0~.i 4~I 0

I,~ ~ L,% 44 P.4 Me~~* v M. M~ "~9 M." Nr W" E * 444444b4#... 4V L. 4 *4...44.

Z0 4 41 4*4to Z

41 O4

0. am'0a.MN

'LI

u . zc"N44g0p.4 MCI 0;"m.4 ~ M. ;N9P mq l om

ooz

ILI-

~ ~ N

a*I4-p zU NO -4V 4*44

Q 0 1 e NN lo - P*t..O-N-N WIN IV) obd N N K444444N44E4 J444N to NW 4 4 446. w~ 29 *+44 44 4 410 61, 441 P

I V =

41 X.6 WI-

21 0J .0 f-6000 4 %1

'1 .4 41.5 oA D00N0Mf f " C0rdM0M0

11 VA3 ON% 0 - ?N " r K V 0N 0 0P % ;

:C 43 0L WP 1% , 4- 10

* 0Z Op V9,4N~ 4* 4 0,4k %

ad4

M N -4 C

V - - N

In-

w O NI~~.N. @ N Ne r N e, Nr- l04 e... m r" en r"mr- -

4j ~ M

TOZ

Ofr- ~ ~ ~ 1 C4r LINaeN ve4c LMrv r0cN .*ar.w D

-~ ~ ~~ .mr . .c ~ t'toa~).C~ .-. .0 .4f .N

Cc W 00 0

4)O "k C0 C *); 401, (I LP NN N . r- '104404C P.Ut ..... .. .. . . ~

4 dW NN..4MfWMLfl...0m ?%.# -1N 0 r ".'o hflN r- %= V V -1/ *.00 6 04NN-40 -Wj f- -N 4 0 A #- mVm Ot....N

. 4a0 01, to)0 4 4 1 * .f01,S 14*

4 .

Q 40W enO N 'tP.NN N In~f m4 .0 4n 4,,q UZ * .* .V W N 0 N4*90N0 0*00*' 0 N Cr *N0N 4)

w *w 6"W N " #

LM Ch n

a*a

A~f Mom- NC*IM.N qI.. NNNP -M m r CN 0UIn c:0 -4 0 0 0* * 04 * 4

C-

C 4

W CJij .4 LM N - NO - MN._ A0 N - - .

a. a a

SO

oM' N It- 1 Ce l n11enmf 0w

;:: m S , 1mr s c

Vh 'il Cc~I N*. 1 S -f 'Iu i.)N il- I? NI

0-q oNN 04 .4M 004 " - MO0 4 hoN . M N M

. .q- . . . . . . . . . . . . . MW cc C10 00 .0C0 0 00 C0 0 0 M -- 400 0;

OJ F13- 00N4P # CMoC)P f N1~ 0% 400N 00 00 ChM0 W m N l0 WU M N 0O 00o 0 0 WUC%0t~ 1 LM 61 O 0Cr 41, % N AM 0@ 0

41 .4 k3 %D M M 114 - m v m o4 4 r , -OM4 -S o,0 ChN N uIO* LIX . . a . . a . . N

.4 r-4 o NN 4 0 W 0 0 .40 N 09 NO 401l C0 UMI% 1S ~ -- MI 0M.*4M4.N4LP W0fN M 4 40.M 9 o UN 4 4-4

A.. "M 4.01-. 4.4*4Ma. -CC 09 01 44, 001 4. C44* 41.401, 4

N OD a c- 0* oV %M .c%0foMa- - %f

41.01.

caO

m 07 e-4 o MIN-0 0 N M . 0 0oM4 m0 M 0

v1 fl14 N 64. oM O.40 -W 10%MNCh410. " -M -fN r .4f NNNO 9.M44W Wb I IN ** 4PO6, 4.4 N 40b--~- 441104401- 40 40.41. 4* P4 4. I

ZN * .40b. A* .

N 0

*N "MN DQ N

u CA M N0N004m"M 4 N o N0N PI N

.41 u=4

U- oN NO 4Ohi -4 N L . l

611 0

(n NJ1 m4 It 6m N O '%'. -N en NEI N ,M N

M1 0. 0 00oo0 0 r -f

~f ~ - 000 ' r'0 0 O1 0 Ch tSIS1 0 0 % 4~ -. It 0, 0, F-S C; (v co

N4 4.O4 M)0' O1 0 O D.. .4 ;:4 N~ N P:, N O 6M4NN 0

0 4 -.(N~ 0 0 0 O . N . 6 O M , * % 6 0 - 0 . 0- 0 0 0 0 N . 0 0 0 u.0 O 0 d

w . m4" 4 @3ULMM MN4 0 VN. i 0 0 N 0t%0 0 -t ~0000 2

*4n CJC.

In 0 Ch". 4M4a4.N-*4. 0 Z .4... 4.4. .NN.4..6. w~ ad 4.4 4 .40 . 4.0 "44. w.

L'a4

@3 3c

0 0 .N 0.00. N%,o 0004.- cc- maE cc mo.40 .0% N o U, N cc4..

'46.3 4F o , o C c0 4.4 M %J . 4 400L .. d04 n iW 4 0 'NNO .W,'40N 4"- 4 IS * ON. N N a cc 1 .

Q P . OD N NO 4.M m.. %a4 4. %;..:4.4W CalU .0 4.4. 1 4. N 4 ".4 4.4.0Do a. 40* %. a w 04

V 22

l C13

U 6CU

O '46. -W f~'0 1%00 ? 0 0 M 0 t, 0 01d9'CNt a M

0 w m~- ?. % O% -40 0 '00 M a, 6n 0%%NN0 .m ,

0 U N 0; e: C;- I. - -; s lsv4. 4.16; 44-.40--IV 1.4 N" 4444. 44 a -&.44 401 . W...... 4

u 0

w c P. M en -T 4 N 'o

O IC

hi'4

U3

-

C1- r. N N N 0 m l0 W.4 0 ,~0 0 q W0 mUin 0%

* 400 7 l P 44. 0 *4CP.N 0 M *W .00 .1 *W * *

~ bg- MM 0' O040 0N0 000 .4 ~ '0~ V

*~~~4 40*.0. a h

09I L 496U.- l

a~ all N Ea4 %0NmM11 441 f ~ 0040 Nw4

V 44

V, AIta t to Sail 0N N It lo000 '0 "00..tlt'C .o

:I M.) ON a a .W %V Ch0 W l w P ooo .- 'h kNO OCONN N -Nil. C 0* iI. Nw oNtV 4 093 40I 4* W N 401-u N c * 40 -4 CC 4* 4OW49- 4 4W i 4 44 4 44 .4 ft *409,01

01 top 4w

U g 4.N

U hi!.. NC N-4cc0 N I*.40s-MVNNN 0-400'u N'MID a

M uI b 'C W Lm 0~...N m4N 6 1.44 10 w 0 m C % P .* b1 -NNNNN N N *&q in40 4 *

mm :NN O -N o O N r oz r

01 t.r4

Oh 00b 0 0 Gr 0P'COO N 4A - 4D 0 o0000 n 000 6

k3 O'DA.

U - "N m44@NN P Mc hf4O4I * N N 0 m* 0 4M 4I46401% 4o* ~4: N N I

ba 49 See 44, 40 4*3-

U ~ ~ ~ ~ ~ 49 012- N -. *4 C I

-. NI 44 0 44' CC C P 0h-. 0~ Ph -4

04 1I

v 04 W 44g N0 IA I-- D -* -t 0 0 Q00 C4 1U) a W 0*4 1* %a 4-41 w k0 m I

U hIM * * CC 40406

13 Im

1 .3 MI%1 C 4 i l 0

I. " 1.N M4M 4: 4N

40

N ~. 0 00 m -1 - 04 t. 0% M 00 o I

10 a, 41 to 0% .- iM1 N.d o 0o 0m WI

u N N0 'ON P44 M. 0.4CO

0m *~ C4M I44

O .Z .41. UCII6 M D a % D a% 0

I.--

APPENDIX E

FISCAL YEAR 1984 DIAGNOSIS RELATED GROUPS WORKLOAD AND

MEDICARE REIMBURSEMENT LEVELS

0% 6-aI 4-0 . y,6 .ww . ',

M

0 ( a

kalEN 0

EqqU0-e4

1 -4 t o

10 m

toa

4 6101 IA~ 101 .C-S -0aC~~ M SOb %ho WWW

Eq206

L OZ

00 Id Coc a 0 . 0 C CO 0 * * ft 0t ft ft ft; * 0, * .

- 03 000 0JP Omct 0.000 00 0 00000006__ cooF

1.4 1^ 40 44*p 4

I.l nft @t 40t ft000 4' @4- *U IV S -0c0 4 ** * pv N

In4 ~ i @4 040f ~l 4 ~ 44M4G qI* .w = a .t f . . a a a t

Ciig N@9 *e'8N N 00 4 VM M4 '.N 4 0

in AI * " * "0** 4* 44 ifI' 4 EO 44b4 4F4 0 44

* 4=

N h"40"N 018 MW Nb". 4M ~o O

12 5

4..4 4 4W Nf40 4*~~4'9 S ~ M *1, V 4 - 41W-

E U .4

c i -

a w

-. ~~w fo MS ON * * 41 4n4

0 94M N 4 *~ 1.

0 Ui

U.tm*0

a CAN P 0 f m 10 0 M e 4 .4 l 0o c '. 4 i

-0 rzb aM00 01 . ~r O *-N" 0P 001" f IMMMM( e4- e4 4k CL lU MI 1 1 .

80 Z

M41W NOM 3 - d-~rm~ N o w t 0if 49o- or-N-Dr4M* a

OD W

%Den0~ N' 0f F4r A ChNM ' 9; Mha'-

Ln 4 4* 4 4. 4 4. 0 440 i. N 6M-4.. 140 49- 4w 4. 4. 40*

ba

.4 0'

.0 0 r4% 00N ?4 .m 0 N aVi1 g.NN N Na0.0O~fi'~ N 4n% %0 oo a c %CO

low4,4, cc

, x

WE 1

0 4 bV 00041 o. MP0% 44'I UM

N***O * 1,4 4* 4'C M6 401- 46 40 401-.w0 I 40P440

am N UNL4 CID

* UMe en

Ia ago N C- h aA a

co

60Z

I-W

CI- MMNN -~ff N14M.M 94 N NCN ~c0m -4mcuamw 0.00%00 I'M'0.40 * *m f N OO -cc a 4

-W * * * * 0 -*0,0- 4041 * 4 400 4 * 40 *.

q 1 1.a. a.

4.8~W W, 0 @ 9 M N .4 . I 4 4 * 4& lW 0

a~~~~o -Waa.-

4

.4 f1 0,"t P

In NM N 0N .* M.

14 ~ 40 4M N -44 *40 N If4 4*N

N NM *N M*- In NNNfU . a*

It I

P44

-4 %D Lm0L 1I

W4U S:

olz

Um 6M IM w m c N61%LO 1 0*2 %0 *O.- K ; W~~M O94 M- 4 40 40 40 4* 40 * 0 4*4

th.

~ 5 M.. 41i N-S-~~.4.0VM co 0 ( N0 M M M -t4* NOiO Ch N( D f N . D M

.4 6.3 4*4 0 N o*44* M 0 M 4* 4 0 M t40 NIM * IN *4 0 11 2v " * P4* 40 w4 i*k 0000 -44, 4 0 401 w

S 4

10 b3- I0g0a4 0Nq g aI .P

P4 183 4l,0 to 04 e w- *t T *6 b% F

00M Uu4*4*N0 SO 4 M*ItN4A6C4S6C. 6

1i41 M14 Cr4S 4 Nr N 4N(ifS-0M I1 00

6 am C4 N 4NNM * .0 - 4 * 4* M 0 )

93 4

u 4 4 " M -44Pv 120

* NINts ;= 4 M400 o N

4-

TT Z

CA a- .N s, % ID0%%P.Let-* 00-t t(l1..I0%0 OON r %DCOOD M r4 40 c* *; ccc; * * ** ce;~

Nfmhi 0:Ni 040M 100%% -4 M W. nC000 4* "400. -1 1*0N N 0 W O0%P wilV 0 & ~~0b- Mo.

?A.4.

4- 040N0 to 44

404

di %M% 04%o0 D

-WM'~m~0% 40 406 Sop .q 4

-0 M M MW N 4.1 N M % P. N 1%.4 i* inL 4..N*4%a

I InUM

.0 4h ISuN9n cOc ~ * ~ l~'W o c m c m m m . . . c . c . . .

~ ~ 00S0%~9S~u C S~NNV~0%02 2260 4 1 4 .N O~V 0 *.

II-z

IA 4 .. . . .0 . . . .fN~ %b* 00m 0 40q ed N0.i:4401 . oMo N N r- o40 4*"40 0 0m 0 6 4o" t~ Fat

0 ".- N -*4 4*M o. 4 **0*

ic Ek

w4 aIm

Id. 40.4i P0 ; t00 0"0 NP 04 D %v %r 0 4 f Cm% -N 4 0

*~~~~~~ .S a..aaa . . . . a . . a~~.J1 -4 .0~h en %a.I.4 p N in N 0a0.4 00of* 40- ~ * N I .. 06-4' N 40 a NU0 r-400 4f

bb *44* 401,*e CD*t *

. a (MgoC c .a 14

A. al

hi ~ ~ ~ ~ ~ ~ ~ a bi .A"O%00PU1O'C4O 4o*j tib 04 Is4@uS- 4CSN4atqi'C

a i 04N**-***~w

M MM M NL QP D9% t6 DPcu "

wo M O O nO% " O4- ,40 mN44&e- 0

01- 404 *0~@9 -0aP caa0Nt 40041 %000 ,

.4 ... 0 * ~@Q Sal 64 W N 400 01P W 0 -4 ~ W

4. -b

0 0 "- N4.Cj 0* 4. .4 "* 0 4 .a S 0R fMM00,4W,0 4 4~~~~~~G 0a N 0 mE0N@ I9 -~P @ 0 rAa ~ ~ m 4r MW 0 *N 5, 4* Vt, - a -0 a

99 QA W M .e.4 WA~*W .N S 4N -*se 4

W ES .. .. , 4

tol 4=411-Im

A N oww M O N N

Wl4- 0 C N 0 00 0 S N 40 i m4 l , i V PRd Mw *..O.h90.H4t4

45N M ~*~0~W e~ 5@O SMNSMo-4 U

Uw Ha . i!MS *4 Nm94lN400 ~N N" N N N as

ME

VT Z

0 6 kK OUt- e-4 44 @IM ~ 14 N- tN% -0NN0C

91w. Ic CC O 4c~o C O LIN0q N IO 0 00000wo ft

di m. UvI 4,V pN% -

41,

* a O0 W a 4*- 0 0 *- 4 0 04v1 0g

*l VP "o

N IN

.4 AU a 1%&At

ON to -Q.iM S~3t4~IP N * I; t .4- 0 9t4 A 1* 401 M 1-MM- 1 * 00 0 V 4 40 M- A4

53 4

.Cl hi 0 U 11

a i.1001 4300 at

9%~ W q.045N 0 bN4 - .02O * M %40b41 49404 0 410 4 N *-4 % 401-. 4* 4W W1

4005.40 NbN..U

ta NN

o S

M. M NQ MUtN Nb N 00CC'N'041Q..4N45a Nef*1 OD . C1# 1J14W 4 . . 4 N VI iisi 0 ' 0 sa0WC4 NNE NNNNNNN P

w -c

4;04 .0 .. .r4 .N.. .k.e. .m f% %. N N enMaN c* Mcc W ' '4 ~ ~ ~ 4"P444 Wi, V# 0.0 0 0 0 0 N N.t.t ri.f00 000 OW

"m 440

I I

% M "%a 40 0 ch a N 0 0 0 m" 000 4cr 0 "cfNMF- 0 N *

-. 0- 411-4 -. NN 9lN N Mw:0 0. Vo a. aM... MR N0%- 0 0tog * 4"4 " 0*** 400 401 * 4 00 W

h.

F- Ma " M1%O 4- MI4* *P* 04- 146 4 t b o

a gU-0 aa. Rd f45 '4VFtEtP i

f44

91 all

a ".40 P Cc 0 0 0 41% . enI 0% 04 M .P-f0 PNa ~ ~ ~ ~ ~ ~ ~ ~ ~ m wi a s a . . ............ f

****.e**...* * ***9

.fti0 P-4N4 W ad

Lo b.IDN0cy 00N." "k 9 P

91 Z

IW

01"- 100 m m e me me eo eNooe'414 t.NN

C;n . ~ '-uM n 1 ln- ~ in

4* 0ADM O mN.'aa 4 009 0 0 0 0 0 0 0 0 m o1-moMP.**Nm folSU

I% Mm O* e% N NNeMOO r-4011.-* ~ N " 4* 1*2""4 ** #* 4*"NmN N"40

Q1 wU I * 44 44 44 440 40* U.a Xo :2

i* me 0 ;4 4

3a I

N ae 0 a0.C 04-. M% CP% 0 14 W, m

*~4 go ..

#*: "4* 24*0U .mfS. *M "N 10 N N'1% Mf 4r W,4 " +14 ~ ~ ~ P 40d 44 U.4 ... e4. 44* -4 444 odm. 4V 4 *

'V MU44 * 4 444* 4

1I'-

4.4.W 0- NU M W I CMN-

u 4

a d~ N No4N0 um-P.-W %% M.44I mC I~1

co o * ** 4o el me P.O'1. U N4N~ DN e C4 mmlmfl 4 me in

LIZ

4 11 -.88 It'C O' % I

adw 10-. took 0000V. 4(p 000 01.0 .N010 -4 400 I'.

a,0 I Q4%-N d- 0 -N 0 M0~ ~ N 4* 41*40; "M PM0 OU1i'O W* 1 0t "i Col "' Wq ~ N

V* 440- 44- 4010 'V 4*4 =0 m.

1

= 4101

V .

-t A* W

to kqNr e W n r 04,4 . N. Vj' IN ,

P-

g -4N m2 4 na1.w0 00 i Aat 00

01 r iMM1 nmI r4. 4.0%Pv VW nV nU n1 -

P.C .MI V "MIl( "MM I l l . me

No

WtA 4 m s- I Di 4

= 400 40 N .40W',

*4 t 4 *** . 4*

W3 I-. LvM.MD0,%M V nmc% o cvrlr - ml

&h 04C. 4 CD sIV :2N i 4 * N 0av -r..M1 0 m 1 0 a- 0 .0~. 0 7%

* CA Il 4 4 . 40 4-4. 4 4. 01,40

~ 40

hil-0 Cap- MCO4

*W 40 4-4 4 l 4b-4uN 46~b 46 r 1,4 66.0P* = 4W * 61 4.4..

11 0 . .

* i

04 0v MI 0I'

m*0 'EM CN0~N ~ .N 0 0 N ieoa 0 [email protected] 0% 0 g

a

co adI 0140p- go

61 Z

uU NNM 40NO - FI~VN*@MU.4 %

U ~ 0

01.4 Mo." N oDMM0 V A n cc NNI o~ "="-M MM am N

am 0 . 0 11 Vl N * 4 I" .04N OI'm" .4 M

SE Im

dJad

4.4~ ~~~~ 0 .440 O 4 .4 . 0 0 a0 a M6 0 M 0 M r. 0% M IIp~ ~ ~ f. 1 40141d0 N4@w 00 444 *6 Iwo o in0 0 w .0

o ~ ~ "m M MO N W0 .0 ell-44N4 0.4 -'44 N4444N. N "4

0U Nd 4..44 w.4-4

o 0 Z

f%** WON WN 4MOINM 9w~

* urn Av llZc tal~t"~ttrc ~ iGl

V z 40

Sow M04 16 1-

nk S

-I~ ~ Q5 41% Q.C N -0. M IV N C~ m9i9 V N@ .4* o0 1 .4 4-

o g UzOoI* 0C.~ -. h

I-N

* 4. *Q** 0 6 .0 * .0 0 6 6 .

go N.

'm %-4 N m CIN 0'. 0.-'u0 0.4 -,0 0.4- 0-VN 0 M *i4 441 .4f0 % 14 -04 0 PC m 4'.o W % (U 0 44.0 P-4 0 P 4 40 0

-4441-4 N N N 44, 4W M 40 4 N% %40 1. 0 9M *1- w s

W *4 q * 4 0 - 401 M ow .

a4 r~

w. 4 c119m I 0 n 4w f % co0 m4D 10 .0 4 m 0 . I% & 00 4 a

" 0N 4W I -- 4- 0 0"ND%0 6 r-IO ' 0 4NE 4 1-0 40 M 0

"I :(9~~ No 40.4 4fl*IN 4''D r!- a. O.a. .wl .'41 INU =Na. -mN4 M* "MaoNN ~~. .,

X I .- a- P4

a1 U P %N1 - P 40h 4 %M---0 00o -4 9

d1C UN :. C; N0 oz ' 0 *a4 4 P0* % .. IWOA .n - 4m m -1 .4 %a a- im VaIa-*, 0.4e . a aook i-ibl K 4* 0 N 404+404N- M 1* 0%19 10 40,0 401-

Na 0 -m P- d

t, - %D?. . & 1u

M- %Mo4 4w94 W4i.40 -4 '0qrplt 04 o- NaN 4~. 0r04g. (.1u NV N N0.~ N4 a M

A 0 o c 10W M 4 O0 4 N

1. W % f~~' 00W 00 ~ 'N 4~ 0%0 P. L

e ad

N N

00@ 40N0 - iWW ,01M01-@0 0 N

4J ~ I 0 'M M 44pg%.~ 4 400.

4 %af 64*.4+M *N 4N 401 km V ?. Nm IPP 0 %

0-4 N 0401O40 u,0I-I

U -0 MI 0%4 '.n 1% 4'. M C C 0-i

as0 a96.

CC -4 a Sf " 0Mw

la~i

ow Ba *

.4 0V %NN mN0*, (iN 0 q% I

*4 O . V M M 1 4 0%'4 M 0' f ".0 IM%0VIoco wl1& i.nMt. kmIO , ) n a 'D% 0 %NINDa1P

U OY2 N*4~N N.-WN * (n

APPENDIX F

CONSOLIDATED FY83 AND FY84 MEPR EXPENSE AND WORKLOAD DATA

MEPR INPATIENT EXPENSES AND DISPOSITIONS

INPATIENT CARE CHARLESTON LONG BEACH PENSACOLA

FY84 EXPENSES $14,929,978 $15,041,916 $10,093,468

FY84 DISPOSITIONS 12,856 9,436 7,237

FY84 ALOS 4.4 4.7 4.3

FY84 PHYSICIANSALARIES $693,536 1762,262 $551,489

FY84 EXPENSES (less)PHYSICIAN SALARIES $14,236,442 $14,279,354 $9,541,979

FY83 EXPENSES(ACT) $13,784,155 $14,822,302 $9,704,460

FY83 EXPENSES(ADJ) $15,024,729 $16,156,309 $10,577,861

FY83 DISPOSITIONS 13,024 9,341 6,998

FY83 ALOS 4.3 4.4 5.0

FY83 PHYSICIAN(ACT)SALARIES $691,179 $909,556 $670,086

FY83 PHYSICIAN(ADJ)SALARIES 1753,385 $991,416 $730,394

FY83 EXPENSES(ADJ)(less) PHYSICIANSALARIES $14,271,343 $15,164,893 $9,847,468

222

LIST OF REFERENCES

1. Grimaldi, P. L. and Micheletti, J. A., ProsflgtivePament The Definitive Guide to Reimbursement,Pluribus Press, 1985.

2. Smith, H. L. and Fottler, M. D., ProsgectiyeR Management for Operational Effectiveness,Aspen Systems, 1985.

3. Broyles, R. W. and Rosko, M. D., Planning andInternal Control under Prospective Payment, AspenSystems, 1985.

4. Collins, J. A., "Case-Mix Accounting Can HelpHospitals Control Costs," Manaaement Accounting,November 1985.

5. Fetter, R. B. and Freeman, J. L., "Diagnosis RelatedGroups: Product Line Management within Hospitals,"Academy of Manacement Review, V. 11, 1986.

6. Fetter, R. B., Thompson, J. D., and Kimberly, J. R.#Cases in Health Policy and Management, Richard D.Irwin, 1985.

7. Chase, R. B. and Aquilano, N. J., Production andOperations Management, 2d ed., Irwin, 1977.

8. McMahon, L. F. and Smits, H. L., "Can MedicareProspective Payment Survive the ICD-9-CM DiseaseClassification System," Annals of Internal Medicline,V. 104, April 1986.

9. Naval School of Health Sciences Research Report 2-85,An Attempt og Refine DRGs for Navy Medical DepartmentQ"e by Includina Military Uniaue Variables and AnEstimate of Disease v, by T. L. Kay and K. A.Rieder, February 1985.

10. Sonquist, J. A. and Morgan, J. N., "The Detection ofInteraction Effects," Ann Arbor, Institute forSocial Research, University of Michigan, 1964.

223

11. Moran, J. N. and Sonquist. J. A., "Problems in theAnalysis of Survey Data and a Proposal," Jgurnl ofthe American Statistical Association, June 1963.

12. Woods, W. R., Ament, R. P., and Kobrinski, E. J., "AFoundation for Hospital Case Mix Measurement,"Inguily, V. 18, Fall 1981.

13. Mills, R., and others, "AUTOGRP: An InteractiveComputer System for the Analysis of Health CareData," lUh , V. 14, July 1976.

14. Sapolsky, H. H., Greene, R., and Weiner, S. L., "DRGsin Theory and in Practice," Business and Health, June1986.

15. May, J. J. and Wasserman, J., "Selected Results froman Evaluation of the New Jersey Diagnosis-RelatedGroup System," Health Services Research, V. 19,December 1984.

16. Spiegel, A. D. and Kavaler, F., "The Debate overDiagnosis Related Groups," Journal of CommunityHlt h, V. 10, Summer 1985.

17. Conner, R. A., "Linking Payment Information Systemsfor the 1990s," computers in Health Care, V. 7,February 1986.

18. Kaemmerer, C., "A New Reimbursement System: DRGs AreComing," connecticut Medicine, V. 47, November 1983.

19. U. S. Department of Health and Human Services,Institute for Health Planning Monographs, The Use ofDRGs in Health klani , HRP-0906218, by C. Oviatt,1985.

20. Riddick, F., "The Doctor and the DRG," TiLatenist, V. 24, June 1983.

21. Kerr, S., "On The Folly of Rewarding A, While Hopingfor B," Academy of Management Journal, May 1975.

22. Wennberg, J. E., and others, "Will Payment Based onDiagnosis-Related Groups Control Hospital Costs?"New England Journal of MedcjLn.e, V. 311, 2 November1984.

224

23. Horn, S. D., and others, "Severity of Illness withinDRGs: Impact on Prospective Payment," AericanJgurnal of Public Health, V. 75, October 1985.

24. McMahon, L. F. and Newbold, R., "Variation inResources Use Within Diagnosis-related Groups,"Meigal Care, V. 24, May 1986.

25. Alper, P. Roe "Moderating Hospital Costs," Medicalorl eiwsI , V. 25, 28 May 1984.

26. McIllrath, So, "Revised Grace Panel Report: No DRGsfor Physician," American Medical News, V. 27, 27January 1984.

27. Office of the Federal Register, Federal Register,Washington, D. C., 1986.

28. U. S. Department of Defense, Medical Expense andPerformance Reporting System for fixed Military.Medical and Dental Treatment Facilities, Washington,D. C., 1986.

29. Reinhardt, Uwe E., "Battle Over Medical Costs Isn'tOver," The Wall Street Journal, October 22, 1986.

30. "Costs of VA Acute Care Comparable to PrivateSector," Modern Healthcare, 1 August 1986.

225

ZNITIAL DISTRIBUTION LIST

No. Copies

1. Defense Technical information Center 2Cameron StationAlexandria, Virginia 2.3G4. .6145

2. Library, Code 0142 2Naval Postgraduate SchoolMonterey, California 93943-5002

3. Professor David R. Whipple, Jr. 15Code 54WpDepartment of Administrative SciencesNaval Postgraduat SchoolMonterey, California 93943-5000

4. Assistant Secretary of Defense 1(Uealth Affairs)Room 3D200, PentagonWashington, D.C. 20301-2000

5. Surgeon GeneralOffice of Chief of Naval Operations(OP-093)Navy DepartmentWashington, D.C. 20350-2000

6. CommanderNaval Medical CommandNavy DepartmentWashington, D.C. 20350-5120

7. CommanderNaval Medical CommandMEDCOM-05Director, Medical Service CorpsNavy DepartmentWashington, D.C. 20372-5120

8. Chief, Health Care Studies DivisionDepartment of the ArmyU.S. Army Health Care Studies andClinical Investigation ActivityPort Sam Houston, Texas 78234-6060

226

9. LT H. Thomas Osment, MSC, USN 2Code 14Naval Medical CommandNavy DepartmentWashington, D.C. 20372-5120

10. Commanding Officer 2U.S. Naval Hospital Rota, SpainATTN: LCDR Albert B. Long, III, MSC, USNFiscal/Materials Management ServiceFPO NY 09540-2500

11. Department Chairman, Code 54Department of Administrative SciencesNaval Postgraduate SchoolMonterey, California 93943-5000

227