вход по аккаунту


Diagnosis-related group regulations. Implications for the practicing rheumatologist

код для вставкиСкачать
Implications for the Practicing Rheumatologist
Data from our university hospital on the 2 most
common rheumatologic diseases that require hospitalization indicate that total charges for patients admitted
with rheumatoid arthritis or systemic lupus erythematosus break down as follows: room, board, and nursing,
55%; testslprocedures, 30%; drugs, 5%; physical and
occupational therapy, 5%; and miscellaneous, 5 %
Thus, shortening the length of stay would be the most
effective mechanism to reduce total charges. We also
found marked heterogeneity in the single diagnosisrelated group containing rheumatoid arthritis and systemic lupus erythematosus inpatients under 70 years old
who had no comorbid conditions. Planners should be
sensitive to this as prospective reimbursement is extended beyond Medicare patients.
During the current fiscal year, the federal government will, for each Medicare admission, reimburse
hospitals at a prespecified rate based upon the diagnosis-related group (DRG) into which the patient’s admitting diagnosis is classified. Psychiatric, rehabilitaFrom the Regenstrief Institute for Health Care and the
Department of Medicine, Indiana University School of Medicine,
Supported in part by a Multipurpose Arthritis Center grant
(AM 20582) from the National Institutes of Health.
Morris Weinberger, PhD: Associate Professor of Medicine,
Indiana University School of Medicine; Marilyn K. Potts, MSW:
Director, Community Component, Indiana University Multipurpose
Arthritis Center; Kenneth D. Brandt, MD: Professor of Medicine
and Head, Rheumatology Division, Indiana University School of
Medicine and Director, Indiana University Multipurpose Arthritis
Address reprint requests to Morris Weinberger, PhD, Regenstrief Institute for Health Care, 1001 West Tenth Streethth
Floor, Indianapolis, IN 46202.
Submitted for publication April 25, 1984; accepted in revised form August 27, 1984.
Arthritis and Rheumatism, Vol. 28, No. 2 (February 1985)
tion, and childrens’ hospitals, as well as hospitals in
states with alternative diagnosis-based prospective
reimbursement systems (i.e., Massachusetts, New
York, New Jersey, and Maryland), are exempted. This
system is intended to encourage the delivery of costeffective care. While only Medicare inpatients are
currently affected nationally by DRGs, there is every
indication that third-party payers will adopt similar
pricing systems. For example, Blue CrosdBlue Shield
in Arizona, Oklahoma, Kansas, and Cleveland, Ohio
are already using modified DRG systems, and Medicaid of Pennsylvania began such a system on July 1,
1984.* Additionally, similar reimbursement programs
for physicians’ fees for Medicare inpatients are currently being considered ( I ) .
Papers on the development of DRGs (2) and
criticism of this system (3-8) appear elsewhere. Our
concern is the potential effect of DRGs on practicing
rheumatologists. To address this, we have analyzed
data on all rheumatoid arthritis (RA) or systemic lupus
erythematosus (SLE) patient admissions to the Medicine Services of University Hospital (UH) at Indianapolis during a recent 1-year period.
RA and SLE, the most common rheumatic
diseases requiring hospitalization to the UH Medicine
Services, provide excellent models for rheumatolo-
* Personal communications: Mr. Joseph Villegas, Director
of Professional and Provider Programs, Blue Cross/Blue Shield of
Arizona; Mr. Richard Howard, Vice-president for Public Affairs,
Blue CrosdBlue Shield of Oklahoma; Mr. Robert Pearcy, Senior
Director of Professional and Institutional Affairs, Blue Cross/Blue
Shield of Kansas; Mr. Edward Shamrock, Director of Audit and
Reimbursement, Blue Cross/Blue Shield, Cleveland, OH; Mr. Richard Lyman, Program Analyst, Bureau of Eligibility, Reimbursement
and Coverage, Health Care Financing Administration, Baltimore
gists to study DRGs. These diseases are prevalent: a
population survey indicated that 1% of adults have
definite RA (9). A retrospective analysis of members
of the San Francisco Kaiser Permanente Health Plan,
covering 1965-1973, revealed a prevalence rate of
SLE of 1/1,960 members; a similar SLE prevalence
rate was reported in Minnesota (9). In 1978, average
diirect medical charges for RA patients living near San
Francisco were $2,319 (10). Hospital charges for these
patients accounted for 66% of the medical costs,
despite the fact that only 26% were hospitalized.
Management of patients with RA and SLE
often emphasizes multidisciplinary care. Referral of
the RA patient to an occupational therapist (OT) for
instruction in energy conservation and joint protection
techniques or for provision of adaptive equipment for
activities of daily living, and to a physical therapist
(PT) for instruction in muscle strengthening and range
of motion exercises, is common (11,12). Studies indicate that in the treatment of RA, such care may result
in more favorable health outcomes than care by the
physician alone (13-16). Under retrospective pricing
systems, there had been few financial disincentives for
such referrals, since inpatient PT and OT services
were generally reimbursed in full. Under DEGs, however, ancillary services, such as PT and OT, are
purchased as would be any other item. Therefore, this
would encourage such measures as reducing staff and
limiting time spent with each patient. Nurses and
social workers may be open to similar scrutiny, leading to staff reductions and overburdening of hospital
personnel (17). Such changes may affect the nature of
interactions between patients and allied health professionals; for example, inpatient education may be sacrificed for a more expedient discharge.
Patients hospitalized with RA and, to a lesser
extent, SLE may have a variety of special equipment
needs (e.g., aids for self-care and personal hygiene,
walkers, canes, wheelchairs). Such equipment may
enable patients to perform specific activities and/or
reduce stress on affected joints, possibly reducing the
likelihood that deformities will develop. Under the
DRG reimbursement system, provision of such equipment on an inpatient basis may be decreased.
Finally, clearly defined markers of dischargereadiness for RA and SLE patients are not available.
Thus, in non-acute situations, these patients may be
discharged before home care instruction is completed
or drug therapy is adjusted adequately. Without intensive inpatient instruction, outpatient visits and readmissions may increase.
We gathered data on all nonsurgical inpatients admitted to UH for RA or SLE between July 1, 1982 and June 30,
1983, who, if classified today, would be assigned to DRG 240
(Connective Tissue Disorder, Age Greater Than 69, and/or
Comorbid Condition Present) or DRG 241 (Connective Tissue Disorder, Age Less Than 70, and No Comorbid Condition Present). Thus, it should be emphasized that our analysis is not restricted to Medicare patients.
The 400-bed UH is the only academic referral hospital in Indiana. Over 95% of patients who would be classified
in DRG 240 or 241 are admitted to the Rheumatology
Service, where a staff rheumatologist is responsible for their
care and directs a team composed of a Rheumatology
Fellow, house officers, and medical students.
UH’s computer provided us with each patient’s age,
length of stay (LOS), hospital diagnosis (RA or SLE), DRG
category (240 or 241), and total charges incurred during the
hospitalization. We inflated these charges by approximately
6% annually to estimate 1984 charges. We classified the
charges into subcategories, which were then combined into 4
Table 1. Age, length of stay, and 1984 estimated charges for University Hospital inpatients with
rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE), by diagnosis-related group
DRG 240
DRG 241
(Age Greater Than 69 and/or
(Age Less Than 70 and No
Comorbid Condition Present)
Comorbid Condition Present)
(n = 26)
58.50 (13.71)
(n = 12)
33.58 (12.60)
(n = 48)
52.92 (10.41)
(n = 10)
37.20 (15.87)
Age, years
Length of stay,
range, days
2-3 1
Length of stay,
9.65 (6.33)
8.58 (5.52)
7.85 (3.52)
9.70 (6.1 I )
mean, days
Estimated 1984 charges,
3,646 (2,098)
3,726 (2,277)
3,038 (1,185)
4,124 (2,371)
* Numbers in parentheses = standard deviation; charge estimates are based on 1982-1983 RA and
SLE admissions.
Table 2. A comparison of Diagnosis-Related Group (DRG)-based
lengths of stay (LOS) and charges and mean LOS and charges found
in this study
DRG-estimated LOS, days
University Hospital LOS, days
1984 DRG base rate for
University Hospitalt
University Hospital’s DRG
Mean University Hospital
DRG 240
DRG 241
* Based on rheumatoid arthritis and systemic lupus erythematosus
inpatients at University Hospital between July 1, 1982 and June 30,
t These are transition rates for 1984. Since 75% of these rates are
based on our experience, they apply only to our hospital. The base
rate does not include adjustments, such as direct medical education,
capital expenditures, and case-mix, while the University Hospital
DRG rate includes these adjustments.
broader categories: (a) facility charges (e.g., room, board,
nursing); (b) tests/procedures (e.g., radiology, electroencephalograms, clinical pathology, laboratory tests, cardiac
studies, and pulmonary function studies); (c) drugs (pharmacy and blood bapk); and (d) other services (e.g., occupational therapy, physical therapy, and medical supplies) (18).
T-tests were performed to examine differences between groups for average LOS and total charges; however,
we were limited when analyzing differencesbetween individual charge categories (e.g., facilities, tests/procedures). The
computer could only provide the total amount spent on a
charge item, for example, the total charges for radiographs
among all RA patients in DRG 240. This allowed us to
compute the mean charge per patient, but not the standard
deviation, a statistic necessary to perform t-tests, Therefore,
we estimated the SD using the coefficient of variation (19), a
statistic which assumes that the ratio between the mean and
the SD for the total cost is approximately the same as the
ratio between the 2 for each category. For example, if the
mean charge per hospitalization were actually $4,000,with
an SD of $400, the SD would be 10% of the mean. If the
average facility charge were $2,000,we would estimate the
SD to be ($2,OQOx 0.10)= $200. Four separate coefficients
of variation were calculated, depending upon the DRG (240
or 241) and the disease (RA or SLE).
Ninety-six patients were analyzed (Table 1).
The average age of patients in DRG 240 was 47.2
years, while the mean age of patients in DRG 241 was
50.2 years. This indicates that patients in DRG 240
were classified by comorbid conditions rather than by
age. Only 15 patients (15.6%) were 65 years or older,
and all had RA. For both DRGs, RA patients were
older (P< 0.001) than SLE patients, although LOS did
not differ. The mean LOSS (9.3 days for DRG 240, 8.2
days for DRG 241) were not significantly different
from those established under current federal guidelines (Table 2). The charges for RA patients within
Table 3. Categorization of mean and percent of total estimated 1984 charges for Diagnosis-Related
Grow 240 lAee Greater Than 69 a n d o r Comorbid Condition Present). bv disease*
(n = 26)
(n = 12)
All patients
(n = 38)
Cardiac studies
Other studies
Clinical pathology laboratory
Anatomic pathology laboratory
Blood bank
Other services
Occupational therapy
Physical therapy
Materials management
Total (dollars)
* RA
2,005 (54.99%)
95 (2.61%)
321 (8.62%)
2,lI5 (56.76%)
1,938 (52.81%)
166 (4.52%)
298 (8.17%)
97 (2.66%)
69 ( 1 .89%)
444 (12.18%)
12 (0.33%)
920 (25.23%)
314 (8.43%)
45 (1.21%)
92 (2.47%)
760 (20.40%)
39 (1.05%)
303 (8.26%)
81 (2.21%)
76 (2.07%)
544 (14.82%)
20 (0.54%)
169 (4.64%)
32 (0.88%)
201 (5.51%)
169 (4.54%)
80 (2.15%)
249 (6.69%)
169 (4.60%)
47 ( I .28%)
216 (5.88%)
132 (3.62%)
168 (4.61%)
91 (2.50%)
34 (0.93%)
425 (11.66%)
0 (0.00%)
29 (0.78%)
68 (1.82%)
15 (0.40%)
112 (3.00%)
90 (2.45%)
124 (3.38%)
84 (2.29%)
28 (0.76%)
326 (8.88%)
rheumatoid arthritis; SLE = systemic lupus erythematosus.
Table 4. Categorization of average and percent of total estimated 1984 charges for Diagnosis-Related
Group 241 (Age Less Than 70 and/or N o Comorbid Condition Present). by disease*
Fac i I i t y
Cardiac studies
Other studies
Clinical pathology laboratory
Anatomic pathology laboratory
Blood bank
Other services
Occupational therapy
Physical therapy
Materials management
Total (dollars)
* RA
rheumatoid arthritis; SLE
= 48)
= 10)
All patients
(n = 58)
1,664 (54.77%)
0 (0.00%)
1,664 (54.77%)
2,055 (49 83%)
0 (0 00%)
2.055 (49 83%)
1.731 (53.67%)
0 (0.00%)
1,731 (53.67%)
348 (11.45%)
57 ( 1 .88%)
42 ( I ,3896)
402 (13.23%)
13 (0.43%)
862 (28.37%)
569 ( I 3 8OYr)
152 (3 68%)
155 (3 76%)
885 (21 46%)
38 (0.92%)
1,799 (43 62%)
386 ( I 1.977~)
73 (2.26%)
62 ( 1.92%)
485 (15.04%)
17 (0.53%)
I .023 (31.72%)
76 (2.50%)
12 (0.40%)
88 (2.90%)
124 (3 01%)
9 ( 0 22%)
133 (3 23%)
84 (2.60%)
12 (0.37%)
96 (2.98%)
168 (5.53%)
160 (5.27%)
84 (2.76%)
12 (0.40%)
424 (13.96%)
32 (0 78%)
38 (0.92%)
48 ( I 16%)
19 (0 46%)
117 (3.32%)
145 (4.50%)
139 (4.31%)
78 (2.42%)
13 (0.40%)
375 ( I 1.63%)
3,038 (100.00%)
4,124 (100.00%)
3.225 (100.00%)
systemic lupus erythematosus.
DRG 240 were not significantly different from those
with SLE ($3,646 and $3,726, respectively). However,
wiithin DRG 241, the average charge for SLE patients
($4,124) was higher ( P < 0.001) than that for RA
patients ($3,038).
Table 3 presents the distribution of DRG 240
charges among the 4 previously described categories.
The mean estimated 1984 charges for RA and SLE
patients who would have been classified in DRG 240
(Connective Tissue Disorder, Age Greater Than 69,
and/or Comorbid Condition Present) were $3,646 and
$3,726, respectively. Room and board accounted for
over half of these amounts (Table 3); laboratory tests
and procedures made up approximately 30% of the
total. Less than 15% of the total charges was accounted for by the combination of the drugs and other
services categories (including PT and OT). The only
significant difference between RA and SLE patients in
DRG 240 was in the other services category ( t = 4.33,
df = 36, P < 0.001), and this was due to the increased
use of PT and OT by patients with RA.
Table 4 presents parallel data for DRG 241
(Connective Tissue Disorder, Age Less Than 70, and
No Comorbid Condition). Here, however, differences
in charges were pronounced. The total estimated 1984
charges for SLE patients (mean $4,124) were signifi-
cantly higher than for RA patients (mean $3,038; t =
2.17, df = 56, P < 0.05). As with DRG 240, facility
charges accounted for over half of the total, and there
were no differences between RA and SLE patients in
this area ( t = 1.48). However, compared with RA
patients, SLE patients used more tests/procedures (f
= 5.22, P < 0.001) and drugs ( t = 2.95, P < 0.01), but
fewer other services ( t = 4.33, P < 0.001). Additionally, SLE patients had more charges in each subcategory of tests/procedures, as well as in pharmacy
(Table 4). RA patients used significantly more OT, PT,
and materials management ( P < 0.01).
Presently, DRG payment rates reflect a combination of local and national charge estimates. During
the first year of DRGs, 75% of the rate will be
determined by local charges; however, this will be
reduced to 50% and 25%, respectively, during the
second and third years. In the fourth year, there will
be a single national rate paid to all hospitals. Local
charges depend on the diagnosis-specific makeup of a
hospital’s caseload. These rates are then adjusted for
medical education, capital expenditures, and bad
debts, although such subsidies may be disallowed in
the future. Therefore, the 1984 rates for UH are
weighted heavily (75%) by our local experiences, and
include adjustments specific to UH. Thus, these rates
are applicable for our institution only. For RA and
SLE, the government estimates the 1984 base rate for
all UH medicine admissions to be $4,487 and $4,181,
respectively. However, due to adjustments described
above, our UH will receive $5,694 for each admission
in DRG 240, and $5,360 for each admission in DRG
Our results lead to several conclusions. Notably, most admissions for DRG 240 are based on the
presence of comorbid conditions rather than the patient’s age. Despite this, the only difference in charges
is that RA patients used more other services than SLE
patients, reflecting the extent of OT and PT that they
received (Table 3). This seems to be appropriate.
In contrast, there are large disparities in DRG
241 (Table 4). Since Medicare inpatients are usually at
least 65 years old and will therefore be classified
predominantly in DRG 240 because of the age criterion, this does not present an immediate problem.
However, as prospective reimbursement is extended
to non-Medicare patients, planners cannot assume that
DRGs which include younger patients with less comorbidity (e.g., DRG 241) will be similar to DRGs
designed for older and/or sicker individuals (e.g., DRG
240) with respect to resource allocation. Indeed, our
data indicate that, on average, SLE patients accounted
for 35.7% more charges than RA patients within DRG
One interesting finding is that DRGs may not
present an immediate impact on rheumatologists. Only
15% of inpatients in our study were 65 years or older,
the group currently most likely to be affected by
DRGs. If UH is representative, hospitals will initially
have marginal gains from most inpatients with RA and
SLE (Table 2). However, as categories are refined,
this gain will be temporary. Also, if allowances for
medical education and capital expenditures are eliminated, the picture will be altered dramatically. The
rate will also change as national rates are established
in 4 years.
Another question concerns the most likely targets for reducing hospital costs. For both DRG 240 and
24 1 , room, board, and nursing account for over half of
the expenses, followed by tests/procedures (approximately 30%), drugs (approximately 5%), and other
services (10%) (Tables 3 and 4). There are several
ways to reduce the room/board/nursing category. One
is to lay off nursing staff; however, since this will
increase the remaining nurses’ workload, this strategy
needs to be evaluated in terms of the quality of patient
care and the retention of competent nurses.
The strategy with the greatest potential yield is
reducing LOS. This will directly affect the facility
charge, the largest component of the bill, and will have
an indirect impact on the other categories. However,
planners must consider the implications upon the
quality of patient care and subsequent health outcomes resulting from reduced hospital stays. For
example, is it cost-effective to keep a person in the
hospital for an additional day to improve hidher
medical status if this reduces the risk for rehospitalization? Under retrospective reimbursement, reducing
rehospitalization is the most effective way to reduce
costs (18); however, under DRGs, hospitals may encourage short-stay, uncomplicated rehospitalizations
to remain financially solvent (3).
Physicians may be asked to reduce the number
and types of tests and procedures they order. While
some laboratory tests and procedures are certainly
unnecessary and should be reduced, such a strategy
will make only a small dent in total inpatient charges
Hospital pharmacies may alter the medicines
they dispense. For example, although aspirin is a
widely used, inexpensive treatment for RA, in many
individuals it produces dyspepsia, hepatotoxicity, or
tinnitus. In such instances, other nonsteroidal antiinflammatory drugs (NSAIDS) may be indicated. Some
NSAIDs have the further advantage of requiring fewer
doses per day than aspirin, which may improve patient
compliance. Since NSAIDs are considerably more
expensive than generic aspirin, hospital pharmacies
may be encouraged to limit their use under DRGs.
Notably, at least 18 states have already deleted, or
have considered deleting, NSAIDs from their Medicaid formularies (personal communication: Mr. John
McCray, Vice-president for Government Affairs, Arthritis Foundation).
The effect of DRGs on occupational and physical therapists may include increased caseloads and/or
staff reductions. If this occurs, therapists may be
unable to provide adequate instruction about adaptive
equipment and assistive devices, since contact time
with each patient would be shortened. OTs and PTs
would need to initiate therapy early in the hospital
ccurse and use the outpatient setting for followup. It is
important to note that in our analysis, combined OT
and PT services for RA and SLE patients accounted
for less than 10% of the total charge. Therefore, for
inpatients with rheumatic disease, OT and PT services
may not be the most appropriate target for cost
savings. Further, this may lead to patient education
being increasingly conducted in outpatient settings. If
so, patients who fail to keep their appointments may
increase the likelihood of future rehospitalization. Finallly, cost-shifting to the outpatient service is only a
ternporary solution. Future prospective reimbursement systems are likely to encompass all health care
How will DRGs affect academic rheumatologists? If university-affiliated hospitals, such as UH,
absorb the sickest patients, as may be evidenced by
the makeup of DRG 240, national rates will be inappropriate for these institutions. Unless major adjustments
are made, there will surely be impending danger.
It must be emphasized that our data should be
interpreted cautiously. DRG rates for the first year are
dominated by local experiences. Further, DRG 240 at
U H consists largely of RA or SLE patients who have
cornorbid conditions, rather than persons who meet
only the age criterion. Therefore, we cannot evaluate
whether the current DRG classification system is
appropriate for Medicare patients. While the calculations presented are specific to our UH, we strongly
urge other hospitals to perform similar analyses and
arrive at their own conclusions.
The authors wish to acknowledge the assistance of
Dr. Stuart Kleit, Kenneth Beardsley, and Kendra Meetz in
the preparation of this manuscript.
1. Public Law 98-21: Prospective payment for Medicare
inpatient hospital services. Amendment passed: March
14, 1983, Federal Register 39752 (September 1, 1983)
2. Vladeck B: Medical hospital payment by diagnosisrelated groups. Ann Intern Med 100576-591, 1984
3. Iglehart JK: Health policy report: New Jersey’s experiment with DRG-based hospital reimbursement. N Engl J
Med 307: 1655-1660, 1982
4. Keith SN: Prospective payment for hospital costs using
diagnosis-related groups: will cost inflation be reduced?
J Natl Med Assoc 75:609-623, 1983
5. Public Comment of the Committee on Rheumatologic
Practice, American Rheumatism Association, to the
Health Care Financing Administration. Atlanta, GA,
American Rheumatism Association, October 6, 1983
6. Lockshin MD: DRG legislation and rheumatic disease.
Submitted for publication
7. Kreitzer SL, Loebner ES, Poveti GC: Severity of illness: the DRGs’ missing link. QRB 8:21-34, 1982
8. Bergen SS. Roth AC: Prospective payment and the
university hospital. N Engl J Med 310:316-318, 1984
9. Rodnan GP, Schumacher HR, editors: Primer on the
Rheumatic Diseases, Atlanta, GA, Arthritis Foundation,
1983, pp 33, 49
10. Meenan RF, Yelin EH, Henke CJ, Curtis DL, Epstein
WV: The costs of rheumatoid arthritis: a patient-oriented study of chronic disease costs. Arthritis Rheum
21:827-833, 1978
1 1 . Katz WA: General approach to the management of
rheumatic diseases, Rheumatic Diseases: Diagnosis and
Management. Philadelphia, JB Lippincott Co., 1979, pp
12. Melvin JL: Rheumatic Disease: Occupational Therapy
and Rehabilitation. Philadelphia, FA Davis Co., 1982
13. Feinberg J , Brandt K: Team management of rheumatoid
arthritis patients in functional class I or 11: results of a 2year study. Am J Occup Ther 38:613-620, 1984
14. Katz S, Vignos PJ, Moskowitz RW, Thompson HM,
Svec KH: Comprehensive outpatient care in rheumatoid
arthritis. JAMA 206: 1249-1254, 1968
15. Duff IF, Carpenter JO, Neukom JE: Comprehensive
management of patients with rheumatoid arthritis. Arthritis Rheum 17:635-645, 1974
16. Shope JT, Banwell BA, Jette AM, Kulik CL, Edwards
NL: Functional status outcome after treatment of rheumatoid arthritis. Clin Rheum Pract 1:243-248, 1983
17. Grimaldi PL, Micheletti JA: Recent Developments in
Diagnosis Related Groups: A Practitioner’s Guide. Chicago, Pluribus Press, Inc., 1983, pp 208-218
18. Smith DM, Robert SD, Gross TL: Health care costs of
medical patients at an urban care center. J Community
Health 7:21-32, 1981
19. Colton T: Statistics in Medicine. Boston, Little, Brown
& Co., 1974, p 43
Без категории
Размер файла
558 Кб
implications, rheumatologist, practicing, group, regulation, related, diagnosis
Пожаловаться на содержимое документа