close

Вход

Забыли?

вход по аккаунту

?

Why health care costs more in the USComparing health care expenditures between systemic lupus erythematosus patients in Stanford and Montreal.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 39, No. 6. June 1996, pp 979-987
Q 1996, American College of Rheumalology
979
WHY HEALTH CARE COSTS MORE IN THE US
Comparing Health Care Expenditures Between Systemic Lupus Erythematosus Patients
in Stanford and Montreal
GAIL GIRONIMI, ANN E. CLARKE, VIVIAN H. HAMILTON, DEBORAH S. DANOFF,
DANIEL A. BLOCH, JAMES F. FRIES, and JOHN M. ESDAILE
Objective. Recent studies to identify the causes of
higher health care expenditure in the US versus Canada
have relied on population-based measures of health care
utilization and have restricted their analysis to one
sector, such as physician or hospital expenditures. We
present a detailed comparative analysis of the direct
costs (health services utilized) of treating systemic lupus
erythematosus (SLE) patients in Stanford, CA and
Montreal, Quebec.
Methods. Using the self-report Stanford Health
Assessment Questionnaire, we assessed 6-month direct
costs incurred by 174 American and 164 Canadian SLE
patients. We explored 3 potential reasons for the differential expenditure. These were 1) higher prices for
health care inputs, 2) more severe disease in the patient
case mix, and 3) greater resource utilization.
Results. The direct health care costs for the
American SLE patients exceeded those for the Canadian
patients by almost 2-fold ($10,530 versus $5,271, expressed in 1991 US dollars). The higher direct costs
were explained by the higher price of health services in
the US and the more severe disease mix. In fact, for all
health resource categories studied, Canadians utilized
Supported by grants from the Arthritis Society of Canada
(85009), NIH grant AM-21393 to the Arthritis, Rheumatism, and
Aging Medical Information System, Stanford University Medical
Scholars Program, and the Fonds de la Recherche en Sante du
Quebec. Drs. Hamilton and Clarke arc Montreal General Hospital
Research Institute Scholars.
Gail Gironinii, BS. Daniel A. Bloch, PhD, James F. Fries,
MD: Stanford University. Stanford, CA; Ann E. Clarke, MD. MS,
Vivian H. Hamilton. PhD, Deborah S. Danoff, MD, John M. Esdaile.
MD. MPH: McGill University, Montreal. Quebec, Canada.
Address reprint requests to Ann E. Clarke, MD, Montreal
General Hospital, 1650 Cedar Avenue, Montreal. Quebec H3G 1A4.
Canada.
Submitted for publication June 12, 1995; accepted in revised
form December 12. IY95.
at least as many services as their American counterparts. Canadians had longer hospital stays, made more
emergency room visits, and used more medications.
Conclusion. Despite significantly greater per capita health care expenditure in the US, our data show
that Canadian SLE patients actually receive more medical services.
Many recent studies have attempted to identify
the causes underlying the higher per capita health care
expenditure in the US versus Canada (1-3). Most of
these studies have relied on population-based measures
of health care utilization, such as the annual number of
hospitalizations per capita (2,3). These studies can be
misleading, because the majority of the population in the
US and Canada is healthy. Thus, aggregate measures of
health care utilization may mask important differences
in care for those who are chronically ill in each countiy.
Very few studies have compared treatment of individuals with discrete medical conditions who are likely to
require and benefit from regular health care. In addition, most comparative analyses have been limited to
only one sector, such as physician (1) or hospital expenditures (2,3). These single-sector studies cannot control
for substitution across resource categories; for example,
more outpatient care instead of inpatient care in one
country versus another.
We present a detailed comparative analysis of the
direct costs of treating a sample of systemic lupus
erythematosus (SLE) patients in the US (Stanford, CA)
and Canada (Montreal, Quebec). The data identify
whether cost differences arise from differences in prices,
patient populations, or resource utilization. A transborder comparison such as ours, involving a single disease,
will establish the database and methodology for future
longitudinal studies that could examine the relationship
GIRONIMI ET AL
980
between health care expenditure and health outcome.
Population-based studies are able to address this question only in a very limited way. They examine general
outcome measures such as infant mortality or life expectancy, and therefore a r e unable t o capture much of the
impact of current medical care, which primarily mfluences
quality of life.
Chronic illnesses a r e particularly appropriate for
a transnational cost-comparative study because they are
responsible for significant expenditure, which is certain
to rise as the population ages (4,5). Among all chronic
conditions, rheumatic disorders are a major consumer of
health care resources and cause substantial work disability (6-10). Given the considerable impact of rheumatic
diseases and the number of studies that have evaluated
costs of rheumatic diseases in general, and rheumatoid
arthritis in particular (&lo), the number of studies that
evaluate costs of SLE is remarkably limited (11-15).
SLE can serve as a model for studying the
management of many severe chronic illnesses. It is of
particular interest for 4 reasons, as follows: 1) It is a
prototypic autoimmune condition that primarily affects
young women, a population underrepresented in most
studies of disease-induced disability; 2) It is a multisystem disease with complicated and potentially toxic therapies. Most patients are treated primarily by specialists,
and management strategies a r e relatively homogeneous.
Therefore, unlike conditions that have controversial
management strategies o r a r e treated by physicians with
varying levels of expertise, differences in the management of SLE patients are more likely to be d u e to
differences in health care delivery systems; 3) It is of
interest to study a condition in which management is
primarily noninvasive. F o r conditions in which management is more invasive, greater availability of technology
in the US is likely to dictate more intensive use of
resources; and 4) D a t a are available about the determinants of direct costs in SLE (14-16). Therefore, appropriate adjustments for confounders may be made when
assessing the association between the health care delivery system and health care costs.
PATIENTS AND METHODS
The cost analysis of the Montreal SLE patients has
been described previously (14-17). We describe herein an
economic analysis of the Stanford SLE patients using a similar
methodology, and then compare health care costs between the
SLE patients in Stanford and those in Montreal. This analysis
explores 3 potential reasons for the differential expenditure, as
follows: 1) higher prices for health services, 2 ) more severe
disease in the patient case mix, and 3) greater resource
utilization. This is done by 1) comparing prices for health care
inputs, 2) adjusting for disease severity, and 3) examining
utilization in several health resource categories.
Study population. In January 1991, 162 patients in the
ARAMIS (Arthritis, Rheumatism, and Aging Medical Information System) Stanford lupus databank who met the American College of Rheumatology (formerly, the American Rheumatism Association) revised criteria for the classification of
SLE (18) were invited to respond to a self-report questionnaire
on health services utilization over the preceding 6 months (i.e.,
July-December 1990). One hundred forty-one patients replied. Between January and July 1991, an additional 44 patients were recruited, and 33 of these patients participated in
the self-report survey regarding their health services utilization
and employment over the interval of January-June 1991. The
responses from either of these 2 questionnaires were used in
the determination of 1990 costs. There were no differences
between the January and July 1991 questionnaires.
One hundred forty-six of the 174 ongoing patients (i.e.,
141 from January 1991 and 33 from July 1991) responded to a
similar survey administered in January 1992 (to assess 1991
costs). Of the 28 who did not respond, 4 had died. Therefore,
146 patients completed 2 questionnaires: either a January 1991
or July 1991 questionnaire and the January 1992 questionnaire.
In Montreal, 164 (of 169) patients completed surveys
similar to those administered to the Stanford patients, that
were mailed in January 1990 (covering the interval of JulyDecember 1989), and 155 (of 158) patients completed surveys
in January 1991 (covering the interval of July-December
1990). The first questionnaires completed by 2 patients in 1990
were excluded because <6 months had elapsed between
completion of their 1990 and 1991 questionnaires. Participants
who responded to a single questionnaire did not differ significantly from those who responded to both.
Survey components. The present study in the US
utilized the self-report Stanford SLE Health Assessment Questionnaire (HAQ) (19) to assess direct costs, disability (scored
from 0 [no difficulty] to 3 [unable to do]), and global well-being
(0 [very well] to 100 [very poor]). Direct costs referred to all
the resources consumed in delivering care to the patient and
included hospitalizations, surgeries, emergency room visits,
nursing home care, diagnostic procedures, visits to physicians
and other health care workers, medications, assistive devices,
and nontraditional therapies (20-22). The validity of patient
self-report data on health services utilization has been previously reported (14,23).
The Canadian study also used the HAQ to assess direct
costs (modified for the Canadian health care system by removing references to health insurance plan) and global well-being.
Disability was measured using the Arthritis Impact Measurement Scales (AIMS) (24) (scored from 0 to lo). This work was
not originally designed as a comparative study, and therefore
different health status instruments were used in each country.
However, the correlation coefficient between the HAQ disability index and the AIMS physical subscale has previously been
shown to be 0.91 (25), enabling us to transform from one scale
to the other.
SLE HEALTH CARE COSTS IN STANFORD AND MONTREAL
Table 1. Assignment of costs to health services by country
Stanford*
Montreal
Physician visits
Physician charges
Radiology/diagnostics
Center charges
Medications
Hospital stays
Pharmacy charges
Hospital charges
Physician charges
Outpatient surgery
Center charges
Physician charges
Emergency room
visits
Hospital charges
Physician charges
Negotiated fee
schedule
Downcosting
procedure
Pharmacy charges
Severity-weighted
per diem costs
Physician fees
Equated to
inpatient day
Physician fees
Downcosting
procedure
Physician fees
Health service
* All charges for the Stanford center, except pharmacy charges, were
adjusted for the appropriate cost:charge ratio.
Assignment of costs. In both countries, patients recorded their utilization of health services, and each unit of
service was then assigned a dollar value. Due to differences in
the funding of the health care systems, methods for assigning
costs to the units of health services utilized differed somewhat
(Table 1). For the Stanford center, charges for medical services
reflected the appropriate cost:charge ratio, i.e., the adjustment
to measure more accurately the proportion of charges billed by
a health care professional or health institution that were
actually received. Inpatient charges at the Stanford center were
adjusted to reflect costs using the institution's 1991 Medicare
inpatient cost:charge ratio. Remaining charge data were adjusted to reflect costs using the 1991 Medicare/Private Fee
Index ratio, reported by the US Physician Payment Review
Commission (26).
Total direct costs were then determined by multiplying
the unit price of a service by the number of units of each
service for each patient, and summing the multiplicands. This
sum represented the total direct costs for a 6-month period.
For both centers, mean 6-month costs were calculated by
averaging the costs for the two 6-month study intervals. Annual
direct costs were then calculated by doubling this sum. All costs
have been expressed in 1991 US dollars using purchasing
power parity for Canada versus the US to convert across
currencies (27). Purchasing power parity represents the
amount in Canadian dollars required to purchase the same
quantity of goods and services as a US dollar can purchase in
the US.
In the US, the unit prices for the physician visits were
obtained by surveying 5 randomly selected San Francisco Bay
area physicians in each medical specialty to determine the
average amount billed per visit by specialty. Costs for physical
therapists, social workers, and visiting home nurses were
determined in a similar manner. In Canada, physicians are
reimbursed by the provincial government according to a government fee schedule.
Costs for outpatient diagnostic procedures for the US
patients were determined in a similar manner to the physician
costs, by surveying 5 Bay area diagnostic facilities on their
981
charges for defined procedures, and then calculating a mean
charge for each procedure. In Montreal, costs for diagnostic
procedures as well as physical therapists and social workers
were determined by a downcosting procedure. i.e., determining
the proportion of the hospital global budget consumed by
these departments and the number of services they rendered
(14,17,28).
In both countries, pharmacy charges were used for
prescription and nonprescription medications. Actual retail
charges were also used for medical assistive devices and
nontraditional therapies.
Institutional charges and professional fees represented
the costs for hospitalizations, outpatient surgeries, and emergency room visits for the Stanford patients. In Montreal,
hospital and nursing home costs were calculated by using the
total operating expense per patient day, stratified for each
province by type and size of institution (29). To adjust for the
varying inputs used to deliver the hospital service, a service
intensity weight, determined by the Canadian Hospital Medical Records Institute (30).was applied. Since only very limited
data exist on the costs of outpatient surgery in Canada, it was
equated to R portion of an inpatient day, as suggested by the
Ontario Ministry of Health (31). Emergency room visit costs
were determined by dividing the portion of the hospital budget
apportioned to the emergency room (including that from
overhead departments) by the annual number of visits to the
emergency room. Professional reimbursements were also included in the cost calculations for hospitalizations, outpatient
surgeries, and emergency room visits.
Statistical methods. Stepwise multivariate linear regression was used to generate an explanatory model for the
variation in each component of health services utilization. The
predictor variables included in the models were those that had
been previously shown to influence costs in the Canadian
patients (14). The study center was a '-class categorical variable. In each regression model, second-year values for health
services utilization were regressed on baseline predictors.
RESULTS
Patient characteristics. T h e A m e r i c a n patients
w e r e similar t o t h e C a n a d i a n patients in age, education
level, a n d ma1e:female ratio (Table 2). In Stanford, t h e r e
w e r e m o r e t h a n twice as m a n y nonwhite patients (34%)
as t h e r e w e r e in M o n t r e a l (16%). A g r e a t e r proportion
of t h e Stanford patients w e r e married, a n d a g r e a t e r
proportion w e r e unemployed. T h e Stanford patients h a d
longer m e a n disease duration, higher s e r u m creatinine
values, a n d w o r s e HAQ disability and global well-being
scores.
Direct costs. The m e a n a n n u a l direct cost f o r t h e
Stanford patients was $10,530, m u c h higher t h a n t h e
M o n t r e a l m e a n a n n u a l direct cost of $5,271 (Table 2).
At baseline, 9 of t h e Stanford patients (5%) h a d severe
renal impairment requiring dialysis, versus 3 of t h e
Montreal patients (2%). The mean annual direct cost f o r
GIRONIMI ET AL
Table 2. Characteristics of Stanford and Montreal patient populations at study entry*
Characteristic
Demographic features
Mean (SEM) age, years
Mean (SEM) education level, years
White, %
Female, %
Employed full time, %
Married, %
Clinical features
Mean (SEM) disease duration, years
Mean (SEM) serum creatinine, mg/dl
Health status
Mean (SEM) HAQ physical disability
score (0-3)
Mean (SEM) HAQ global well-being
score (0-100)
Costs, in 1991 US dollars,
mean (SEM) direct costst
* HAQ
=
Stanford
(n = 174)
Montreal
(n = 164)
43.9 (1.0)
14.0 (0.2)
65.5
90.2
29.0
69.0
45.0 (1.2)
12.8 (0.3)
83.5
88.4
37.87
55.4
15.4 (0.7)
1.6 (0.2)
13.5 (0.6)
1.0 (0.1)
0.7 (0.1)
0.4 (0.0)
32.6 (1.8)
25.0 (2.0)
10,530 (995)
5,271 (691)
Stanford Health Assessment Questionnaire.
? Differs from that previously reported (14) because it refers to
patients employed >20 hours/week.
$ Annualized average for the two 6-month survey intervals.
the 9 Stanford dialysis patients was $39,216, exceeding
the average of the entire Stanford patient group by more
than 3.5-fold. The 3 Montreal dialysis patients also had
higher direct costs, averaging $25,328.
Because the Stanford patients were more ethni-
cally diverse, they were stratified based on ethnicity.
Black patients, 10% of the population, had worse mean
disability and global well-being scores and higher mean
direct costs ($12,798), compared with nonblack patients.
Twelve Stanford patients (8%) had direct costs
below $1,000, versus 48 of the Montreal patients (31%).
One Stanford patient (0.7%) and 1 Montreal patient
(0.6%) had direct costs exceeding $100,000.
In Stanford, outpatient diagnostic/therapeutic
procedures and hospitalizations combined accounted for
60% of direct costs (Table 3). Similarly, diagnostic
procedures and hospitalizations made the greatest contribution (64%) to the direct costs incurred by the
Montreal patients (Table 3). However, the percentage of
costs attributed to diagnostic procedures was much
higher in Stanford (35%) than in Montreal (18%).
Hospitalizations represented 46% of Montreal direct
costs. Costs due to ambulatory visits to health professionals, emergency rooms, and Outpatient surgery comprised 13-15% of direct costs in both locations.
The direct health care expenditures of the American SLE patients exceeded those of the Canadian
patients by 2-fold. This differential may have arisen
because of differences in prices, case mix, and/or resource utilization. For example, the price of a physician
visit may be more expensive in the US than in Canada,
or the US patients may be sicker on average, so that
their higher costs are attributable to greater utilization
Table 3. Annual medical costs for Stanford and Montreal patients with systemic lupus erythematosus, in 1991 US dollars*
Montreal
(with Canadian
prices in US $)
(n = 164)
Stanford
(with US prices)
(n = 174)
Health service
Mean (SEM)
Ambulatory care
Diagnostic procedureshherapeutics
Dialysis
Medications
MD/other professionals
Transportation?
Emergency room visits
Outpatient surgery
Babysittert
Miscellaneous (medical devices,
nontraditional therapies)
Hospital care
Rehabilitation facility care
Direct costs
* NA
=
3,688 (252)
1,511 (574)
1,036 (46)
1,209 (84)
NA
44 (16)
234 (65)
NA
168 (33)
2,641 (517)
0
10,530 (995)
% of direct
35
14
10
11
Mean (SEM)
Stanford
(with Canadian
prices in US $)
(n = 174)
7'; of direct
Mean (SEM)
2
939 (61)
241 (119)
453 (34)
662 (69)
93 (12)
68 (8)
108 (41)
16 (8)
27 (6)
18
4
9
12
2
1
2
0.3
0.5
988 (48)
1,036 (393)
420 (20)
708 (48)
NA
10 (3)
102 (25)
NA
89 (16)
25
0
100
2,416 (599)
247 (179)
5,271 (691)
46
5
100
1,094 (218)
0
4,448 (540)
0.4
2
% of direct
22
23
9
16
0.2
2
2
25
0
100
not applicable.
t Not included in the Stanford
calculations for direct costs.
Health Assessment Questionnaire completed by the Stanford patients, and therefore not included in the Stanford
SLE HEALTH CARE COSTS IN STANFORD AND MONTREAL
of all health care inputs, or patient care may be more
heavily weighted toward higher-cost resources in the US
than in Canada.
To answer this question, we first examined the
price differential between the 2 countries (Table 4).
American prices exceeded Canadian for all categories of
health services by at least 1.4 times. This price differential was largely due to the different mechanisms of health
care funding between the nations. We observed the
largest differential with diagnostic procedures and hospitalizations. Other studies (2) have reported much
smaller US:Canadian price ratios for hospitalizations;
this may in part be because they did not incorporate
physician costs.
Table 5. Mean 6-month utilization of health care services per patient
Health service
Stanford
(11 = 174)
Montreal
(n = 164)
Diagnostic procedures/therapeutics*
Physician visitst
Hospitalizations
Days in hospital
Emergency room visits
Outpatient surgeries
Patients in nursing home
6.6
6.5
0.21
1.1
0.06
0.06
0
6.6
5.9
0.20
2.5
0.26
0.06
0.02
* Includes radiographs, nuclear medicine scans, ultrasound tests, magnetic resonance imaging, computed tomography scans, electrocardiograms. echocardiograms. pulmonary function tests, colonoscopies,
gastroscopies, blood tests, and urinalyses.
t Includes rheumatologists/immunologists, internists, family physicians. nephrologists, dermatologists, ophthalmologists, surgeons,
emergency room physicians, and miscellaneous specialists.
Table 4. US:Canada price ratio across health resource categories
us
Health service
Health care worker
visits''
Rheumatologist
Internist
Family physician
Dermatologist
Ophthalmologist
Surgeon
Physical therapist
Diagnostic procedures?
Chest radiograph
Ventila tion/perfusion
lung scan
Joint scan
Echocardiogram
Abdominal
ultrasound
Computed tomography scan
(brain)
Electrocardiogram
Endoscopy (upper)
Colonoscopy
Complete blood cell
count and
differential
Antinuclear antibody
test
Heniodialysis session
Medication (per tablet)
Aspirin
Hydroxychloroquine
Prednisone
Azathioprine
Hospitalizations
(1991 US $)
Canada
(1991 US $)
109
US:Canada
price ratio
81
60
41
91
83
39
56
56
42
25
29
26
20
2.0
1.4
1.4
1.6
3.1
3.2
2.0
96
501
25
I23
3.8
4. I
66
350
264
90
56
69
0.73
6.2
3.8
673
121
5.6
79
360
458
27
8
65
242
6
9.9
5.5
1.9
4.5
47
26
1.8
I92
132
1.4
0.03
1.31
0.21
1.33
0.04
0.42
0.05
0.84
-
-
0.75
3.1
4.2
1.6
2.4$
* All physician fees refer to initial consultation.
t All procedures include professional fees where applicable.
1Represents the ratio of the hospital component (includes professional fees) of the annual medical costs for Stanford patients calculated using US prices and Canadian prices, i.e.. $2,641:$1,091.
To determine if the price differential entirely
explained the expenditure differential, we recalculated
the American costs by assigning Canadian prices, converted to US dollars, to the units of health services
utilized by the American patients (Table 3). American
direct costs decreased from $10,530 to $4,448. These
costs were now less than the Canadian direct costs, (i.e.,
$5,271), demonstrating that the expenditure differential
was largely explained by the unit price differential.
The American patients had more severe disease
than the Canadian patients, as evidenced by a higher
mean serum creatinine level, greater percentage of
dialysis-dependent patients and black patients, poorer
global well-being, poorer physical functioning, and lower
percentage of employment. Adjusting for disease severity would further lessen the costs for the American
patients (which decreased to $4,448 when calculated
using Canadian prices), and would therefore further
widen this differential expenditure between American
and Canadian patients.
The excess expenditure among the Canadian
patients can be explained by either 1) greater utilization
of most or all medical services by Canadians, or 2)
greater utilization of only the more expensive services by
Canadians. To determine the most appropriate explanation, we profiled the utilization in the different health
service categories.
Utilization of health services. In comparing the
patterns of health services utilization, the greatest difference was in the number of hospital days (Table 5). In
both centers, there were about 0.2 admissions per person
over 6 months. However, for those hospitalized, the
mean length of stay was much greater in Montreal, i.e.,
GIRONIMI ET AL,
984
12.3 days, versus 5.2 days at Stanford. The Montreal
patients made more emergency room visits, i.e., 0.26 per
person in a 6-month period versus 0.06 per person in a
6-month period for the Stanford patients.
Outpatient diagnostic procedures averaged 6.6
per person in a 6-month period in both centers. Ambulatory visits to physicians averaged 6.5 per person in a
6-month period in Stanford and 5.9 in Montreal. In both
centers, patients visited rheumatologists/immunologists
most frequently. Stanford patients visited nephrologists
about 3 times and internists 4 times more frequently
than the Montreal patients, and dermatologists less
frequently (data not shown).
Canadians made more visits to the emergency
room and had longer lengths of stay in acute-care
hospitals. Given that the American patients likely had
more severe disease, adjusting for disease severity would
further widen this differential. Yet, American and Canadian patients reported a similar number of diagnostic
procedures and physician visits, and medication expenditure. To compare the utilization of these services,
controlling for disease severity, we estimated multivariate regressions using number of diagnostic procedures,
number of physician visits, and medication expenditure
as dependent variables, and included age, sex, marital
status, ethnicity, education level, employment status
(14,17), SLE disease duration, level of physical functioning, global well-being, and serum creatinine level as
potential explanatory variables. In the first 2 cases, the
coefficient on the Canadian dummy variable was not
significantly different from zero; in the third, it was
significantly greater than zero (P = 0.01). Thus, Canadian patients with disease severity comparable with that
of their American counterparts used a similar number of
diagnostic procedures and made the same number of
physician visits, but used more medications.
DISCUSSION
We have performed the first study comparing
health care costs incurred by patients with a chronic
rheumatic illness in the US and Canada. American SLE
patients incur direct costs that are 2-fold greater than
those incurred by their Canadian counterparts. This
estimate is likely conservative given that the US cost:
charge adjustment that we used is less than that reported
elsewhere (1). The higher direct costs are explained by
the higher price of health services in the US and the
more severe disease in the patient mix. In fact, for all
health resource categories studied, Canadians utilized as
many services, and in some cases more, than their
American counterparts. Canadians had longer hospital
stays, made more emergency room visits, and used more
medications. Critics of Canada’s nationally funded
health care system claim that many cost-saving measures
have compromised quality by leading to excessive delay
in providing health care services, and have restricted
access to necessary technologies. Despite significantly
greater expenditure in the US, this study suggests that
Canadians actually receive more medical services than
their American counterparts. Although Canadians may
experience restricted access to highly priced technological innovations, they appear to have at least similar
access to routine medical services.
We studied a single tertiary care facility in each
country, which potentially limits the generalizability of
our study (32). However, we believe that many patients
with SLE are cared for in specialized facilities, and that
treatment may not differ significantly in comparable
facilities across each country.
It is possible that our results were biased by 2
factors. The greater paperwork burden imposed on the
American patients because of private health care insurers may increase patient recall of health services utilization. Furthermore, the majority of Stanford patients
were privately insured, potentially increasing resource
utilization. However, both of these factors-greater
recall and private insurance-would serve to narrow the
differential utilization between American and Canadian
patients. Therefore, the excess utilization we observed
among the Canadian patients may be even greater.
Adjusting for differences in case mix further
strengthens our conclusions. The adjustment may be
incomplete because psychosocial factors, previously
shown by our group to be important determinants of
direct costs (14,16), were not measured in the Stanford
patients. However, our previous studies have shown that
these are highly correlated with physical functioning and
global well-being, factors which were taken into account.
Furthermore, our findings which demonstrate
less health care expenditure and greater utilization of
health services among Canadian SLE patients are consistent with results of studies documenting hospital and
physician utilization among the American and Canadian
general populations, which largely comprise well people
(1-3). In our study, we observed that mean per capita
expenditure on acute-care hospitals and emergency
room visits in the US exceeded that in Canada ($2,685
versus $2,484). This excess expenditure in the US is
explained by higher prices and not by greater utilization
SLE HEALTH CARE COSTS IN STANFORD AND MONTREAL
of medical services. Although annual Canadian expenditure on acute-care hospitals was less, Canadian SLE
patients had admission rates similar to those of their
American counterparts over the two 6-month study
intervals (0.2 per person), longer hospital stays (12.3
days versus 5.2 days), and more emergency room visits
(0.26 visits per person over a 6-month interval versus
0.06). In a population-based study, Redelmeier and
Fuchs (2) also observed no difference between the
countries in admission rates for persons of similar age to
our patients, and observed that Canadians have longer
hospital stays (11.2 days versus 7.2 days) and more
emergency room visits (0.68 per person per year versus
0.32).
Longer lengths of stay in Canada have, in part,
been attributed to cost containment measures (33).
Canadian hospitals, operating within the constraints of a
prospective global budget, save money by delaying hospital discharge in lieu of admitting new patients. The
intensity of resource utilization tends to diminish with
length of stay. Redelmeier and Fuchs observed that
during a hospital stay, Americans may use more highly
technical services such as magnetic resonance imaging,
but fewer routine services such as biochemical assays.
Although we did not itemize inpatient hospital services,
we did observe that Canadians used at least as many
outpatient diagnostic procedures as their American
counterparts. Further, we observed no difference between the Canadian and American patients in the
number of outpatient visits to health professionals (visits
to both hospital clinics and external offices). Redelmeier
and Fuchs (2) also observed no difference in outpatient
visits to hospital facilities.
The average per capita expenditure differential in
health professional services that we observed ($1,209
versus $662, including only non-urgent outpatient care)
is also due to higher prices in the US (US:Canada ratio
for health professional services is 1.71). Fuchs and Hahn
reported a higher US:Canada cost ratio for physician
services of 2.39 (1985 data) (1). They deflated US
physician fees by 20% versus the 35% we used, reflecting
the 1991 Medicareprivate Fee Index ratio (26). Although American patients may substitute the more
highly paid internists for family physicians as primary
providers, the price differential is small and is only
minimally responsible for the greater US expenditure. In
contrast to our findings, Fuchs and Hahn (1) observed
greater physician utilization among Canadians, while
others (34) have reported less utilization among Quebec
residents (4.6 visits for Quebec residents versus 7.1 visits
985
for all Canadians versus 5.4 visits annually for Americans). It is speculated that this increased utilization
among the general Canadian population may be due
to universal first-dollar coverage increasing patientinitiated demand, and lower physician reimbursement
rates increasing physician-initiated demand (35). These
factors may play a lesser role in a chronically ill population, which explains the similar frequency of physician
visits that we observed in our 2 SLE populations (approximately 6 outpatient visits per 6-month interval).
In conditions in which management is more
invasive, patients in the US are managed more aggressively. Such management may (36) or may not (37-40)
influence patient outcome. Since the management of
SLE is primarily noninvasive, the well-documented
greater availability of medical technology in the US
(41-48) did not significantly influence the health services utilization by the SLE patients.
The present short-term transnational comparative study is one of the first to examine expenditure and
utilization across several health resource categories for a
specific chronic disease involving patients likely to require and benefit from regular medical care. One must,
however, interpret the results with a certain amount of
caution, in that they reflect care provided at single
centers in Canada and the US. To the extent that
comparison is possible, the results for SLE mirror the
results from population-based studies contrasting health
care in the US and Canada (1-3). The higher direct costs
resulted primarily from the higher price of health services in the US. However, fewer resources were employed to care for the Stanford patients, particularly
when greater disease severity is taken into account. This
2-center study of SLE may reflect differences inherent in
the structures of the US and Canadian health care
systems. The Canadian system is less expensive, but the
US system may be more efficient in health care delivery.
However, patient outcome data. imperative in judging
efficiency, are currently lacking. Longitudinal multicenter studies examining clinically relevant long-term
outcomes are required to answer the critical questionAre the costs commensurate with the benefits?
ACKNOWLEDGMENTS
We thank Dr. Diane Lacaille for assisting in data
collection:Dr. Steven Grover and his research associates of the
Montreal General Hospital Center for the Analysis of CostEffective Health Care for providing costing data, Tina Panari-
GIRONIMI ET AL
tis, Sonia Mata, Mary Chong, Karen Braida, and Dianne
Moi-feld for technical assistance, Barbara Cont for expert
secretarial assistance, and Dr. Deborah Lubeck for helpful
discussions.
REFERENCES
1. Fuchs VR, Hahn JS: How does Canada do it? A comparison of
expenditure for physicians’ services in the United States and
Canada. N Engl J Med 323:884-890, 1990
2. Redelmeier DA, Fuchs VR: Hospital expenditures in the United
States and Canada. N Engl J Med 328:772-778, 1993
3. Newhouse JP, Anderson G, Roos LL: Hospital spending in the
United States and Canada: a comparison. Health Aff 7:6-16, 1988
4. Lubeck DP, Yelin EH: A question of value: measuring the impact
of chronic disease. Milbank Q 66:444-464, 1988
5. Waldo D, Lazenky H: Demographic characteristics and health
care use and expenditures by the aged in the US. 1977-1984.
Health Care Financing Rev 6:l-29, 1984
6. Felts W, Yelin EH: The economic impact of the rheumatic
diseases in the United States. J Rheumatol 162367-884, 1989
7. Yelin EH, Katz PP: Labor force participation among persons with
musculoskeletal conditions, 1970-1987: national estimates derived
from a series of cross-sections. Arthritis Rheum 34:1361-1370,
1991
8. Yelin EH: Work disability and rheumatoid arthritis. In, Rheumatoid Arthritis: Pathogenesis, Assessment, Outcome, and Treatment. Edited by F Wolfe, T Pincus. New York, Marcel Dekker,
1994
9. Badley EM: The economic burden of musculoskeletal disorders in
Canada is similar to that for cancer, and may be higher. J
Rheumatol 22:204-206, 1995
10. Lubeck D: The economic impact of rheumatoid arthritis. In,
Rheumatoid Arthritis: Pathogenesis, Assessment, Outcome, and
Treatment. Edited by F Wolfe, T Pincus. New York, Marcel
Dekker: 1994
11. Finn D, Allen P. Hoffstein P, Goldman D. Petri M: Economic costs
of systemic lupus erythematosus: a five year prospective study
(abstract). Arthritis Rheum 36 (suppl 9):S192. 1993
12. Petri M, Genovese M: Incidence of and risk factors for hospitalizations in systemic lupus erythematosus. J Rheumatol 19:15591565. 1992
13. McInnes PM, Schuttinga J, Sanslone WR, Stark SP, Klippel JH:
The economic impact of treatment of severe lupus nephritis with
prednisone and intravenous cyclophosphamide. Arthritis Rheum
37:1000-1006, 1994
14. Clarke AE. Esdaile JM, Bloch DA, Lacaille D, Danoff DS, Fries
JF: A Canadian study of the total medical costs for patients with
systemic lupus erythematosus and the predictors of costs. Arthritis
Rheum 36:1548-1559, 1993
15. Lacaille D, Clarke AE, Bloch DA, Danoff D, Esdaile JM: The
impact of disease activity, treatment, and disease severity on
short-term costs of systemic lupus erythematosus. J Rheumatol
211448-453, 1994
16. Clarke AE. Bloch DA, Danoff DS, Esdaile JM: Decreasing costs
and improving outcomes in SLE: using regression trees to develop
health policy. J Rheumatol 21:2246-2253, 1994
17. Clarke AE. Esdaile JM: Systemic lupus erythematosus: disease
activity, severity, treatment, and costs (letter). J Rheumatol 22:
189-190. 1995
18. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield
NF, Schaller JG, Tala1 N, Winchester RJ: The 1982 revised criteria
for the classification of systemic lupus erythematosus. Arthritis
Rheum 25:1271-1277, 1982
19. Fries JF, Spitz P, Kraines RG, Holman HR: Measurement of
patient outcome in arthritis. Arthritis Rheum 23:137-145, 1980
20. Hodgson TA, Meiners MR: Cost-of-illness methodology: a guide
to current practices and procedures. Milbank Q 60:429-462. 1982
21. Scitovsky AA: Estimating the direct costs of illness. Milbank Q
601463-491, 1982
22. Luce BR, Elixhauser A: Estimating costs in the economic evaluation of medical technologies. Int J Techno1 Assess Health Care
657-75, 1990
23. Lubeck DP, Spitz PW, Fries JF, Wolfe F, Mitchell DM, Roth SH:
A multicenter study of annual health services utilization and costs
in rheumatoid arthritis. Arthritis Rheum 29:488-493, 1986
24. Meenan RF, Gertman PM, Mason JH: Measuring health status in
arthritis: the Arthritis Impact Measurement Scales. Arthritis
Rheum 23:146-152, 1890
25. Brown JH, Kazis LE, Spitz PW, Gertman P, Fries JF, Meenan RF:
The dimensions of health outcomes: a cross-validated examination
of health status measurement. Am J Public Health 74:159-161,
1984
26. Physician Payment Review Commission: Annual Report to Congress. Washington, DC, 1995
27. Statistics Directorate: Main Economic Indicators. November 1993.
Paris, OECD, 1993
28. Drummond MF, Stoddart GL. Torrance GW: Methods for the
Economic Evalu’ation of Health Care Programmes. Oxford, Oxford University Press, 1987
29. Statistics Canada: Total Operating Expenses per Patient-Day,
Reporting Hospitals. by Type and Size of Hospital: Canada and
Provinces, 1987-88. Vol. 1, no. 1. Ottawa, Hospital Statistics
Preliminary Annual Report, 1989
30. Carmichael D: Case weighting methods: a review by the HMRI
Database Committee. Ontario, Don Mills. 1988
31. Detsky AS, O’Rourke K, Naylor CD, Stacey SR, Kitchens JM:
Containing Ontario’s hospital costs under universal insurance in
the 1980’s: what was the record? Can Med Assoc J 142:556-572,
1990
32. Henke, CJ, Epstein WV: Practice variation in rheumatologists’
encounters with their patients who have rheumatoid arthritis. Med
Care 29:799-812, 1991
33. Evans RG. Lornas J, Barer ML, Labelle RJ, Fooks C, Stoddart
GL, Anderson GM, Feeny D, Gafni A, Torrance GW, Tholl WG:
Controlling health expenditures: the Canadian reality. N Engl J
Med 320571-577, 1989
34. Sandier S: Health service utilization and physician income trends.
Health Care Finance Rev 11:33-48, 1989
35. Ginzberg E: Men, Money, and Medicine. New York, Columbia
University Press, 1969
36. Mark DB, Naylor CD, Hlatky MA, Califf RM, Topol EJ, Granger
CB, Knight JD, Nelson CL, Lee KL,Clapp-Channing NE, Sutherland W, Pilote L, Armstrong PW: Use of medical resources and
quality of life after acute myocardial infarction in Canada and the
United States. N Engl J Med 331:1130-1135. 1994
37. Rouleau JL, MoyO LA, Pfeffer MA, Arnold JMD, Bernstein V,
Cuddy TE, Dagenais GR, Geltman EM, Goldman S, Gordon D,
Hamm P, Klein M, Lamas GA, McCans J, McEwan P, Menapace
FJ, Parker JO, Sestier F, Sussex B: Braunwald E: A comparison of
management patterns after acute myocardial infarction in Canada
and the United States. N Engl J Med 328:779-784, 1993
38. Pilote L, Racine N, Hlatky MA: Differences in the treatment of
myocardial infarction in the United States and Canada: a comparison of two university hospitals. Arch Intern Med 154:1090-1096,
1994
39. Roos LL, Fisher ES, Sharp SM, Newhouse JP. Anderson G,
Bubolz TA: Postsurgical mortality in Manitoba and New England.
JAMA 26312453-2458, 1990
SLE HEALTH CARE COSTS IN STANFORD AND MONTREAL
40. Hamilton VH. Hamilton B. Mayo N: What are the costs of
queuing for hip fracture surgery in Quebec? Working Paper,
Department of Economics, McGill University, 1994
41. Coyte PC, Wright JG, Hawker GA, Bombardier C, Dittus RS, Paul
JE. Freund DA, Ho E: Waiting times for knee-replacement
surgery in the United States and Ontario. N Engl J Med 331:10681071, 1994
42. Rublee DA: Medical technology in Canada, Germany, and the
United States. Health Aff8:178-181. 1989
43. Inglehart JK: Canada’s health care system faces its problems. N
Engl J Med 322562-568, 1990
44. Krasny J. Ferrier IR: A closer look at health care in Canada.
Health Aff 10:152-158, 1991
987
45. Katz SJ, Mizgala HF, Welch HG: British Columbia sends patients
to Seattle for coronary artery surgery: bypassing the queue in
Canada. JAMA 266:1108-1111, 1991
46. Higginson LAJ. Cairns JA, Keon WJ, Smith ER: Rates of cardiac
catheterization, coronary angioplasty and open-heart surgery in
adults in Canada. Can Med Assoc J 1463921-925. 1992
47. Morin JE, Symes JF, Guerraty AJ, Poirier NL, Sampalis J:
Coronary arteiy bypass profile in Canada and t h e United States.
Can J Cardiol 6:319-322. 1990
48. Carrier M, Pineault R, Tremhlay N, Pelletier C: Outcome of
rationing access to open-heart surgery: effect of the wait for
elective surgery on patient outcome. Can Med Assoc J 149:11171122, 1993
Документ
Категория
Без категории
Просмотров
0
Размер файла
856 Кб
Теги
cost, care, lupus, health, patients, systemic, stanford, erythematosus, montreal, expenditures, uscomparing
1/--страниц
Пожаловаться на содержимое документа