Embed Size (px)
Citation preview
. . . . . Hospitals . . . . .
© 2006 National Rural Health Association 331 Fall 2006
Rural hospitals not only play a vital role in providing health care in their communities but also serve as an important economic force. 1 Hospitals in rural areas are central to coordinating health care delivery in the
community, 1-3 considered necessary to recruit and retain physicians, and critically important for attracting new employers and residents considering relocation to
rural areas. 1 Because of the chronic lack of physicians in rural areas, 1 rural hospitals take on an even more important role in providing health care 3 for the 49 million (17%) US residents who live in rural areas. 4 As they are an integral component of the health care system, it is important to understand the number and types of patients rural hospitals serve and the care they provide.
In addition to offering inpatient acute care, rural hospitals provide many other services, including swing beds, skilled nursing facility and home health care, and ambulatory surgery. 5-7 While the role of outpatient services provided by rural hospitals is very important, inpatient acute care is still a major and costly component of the care rural hospitals provide. It is this inpatient care that is the focus of this article.
Much research on rural hospitals ’ inpatient care to date has focused solely on Medicare patients who are primarily 65 years and older. 8-10 Rural hospitals are very dependent upon Medicare payment because a large share of the rural population is elderly; 5,9,11 the rural elderly have higher hospitalization rates and are more likely to choose rural over urban hospitals for inpatient care. 8 Medicare data (largely fi nancial) have been published. 9 The few studies that include all-payer inpatient data, for patients of all ages, are based on a limited number of states. Research focusing on the nature and share of hospitalizations for patients
Inpatient Care in Rural Hospitals at the Beginning of the 21st Century Margaret Jean Hall , PhD; 1 Maria F . Owings , PhD; 1 and Judith A . Shinogle , PhD 2,3
ABSTRACT : Context: National data documenting the
role that rural hospitals play in providing inpatient care
to patients both younger than 65 and 65 years and older
has previously been unavailable. Purpose: To present
descriptive nationally representative data on the numbers
and types of inpatients, and the care they received, in
rural hospitals. Methods: This study includes inpatient
data from the 2001 National Hospital Discharge Survey,
a nationally representative survey of short-stay,
nonfederal hospitals in the United States. Inpatients in
rural hospitals were compared to those in urban hospitals
in terms of demographic and clinical characteristics and
patterns of utilization. Among the variables examined
were age, number and type of diagnoses, avoidable
hospitalizations, comorbidity, procedures received, source
of payment, average length of stay, and discharge
disposition. Findings: Seventeen percent (5.7 million) of
hospitalizations were in rural hospitals in 2001 and a
similar percent of the US population lived in rural areas.
Rural hospitals provided 23 million days of inpatient care
and 4.7 million inpatient procedures. Despite the
emphasis placed on Medicare ’ s role in supporting rural
hospitals, half of rural hospital inpatients were younger
than 65 years. Rural hospital inpatients had shorter
average stays and received fewer procedures on average.
Seven percent of rural hospital inpatients were transferred
to other short-stay hospitals. Conclusions: National data
on the broad scope of patients served and inpatient
services provided by rural hospitals illustrate one
important role these hospitals play in serving rural
communities .
1 Hospital Care Statistics Branch, Division of Health Care Statistics,
National Center for Health Statistics, Centers for Disease Control
and Prevention, Hyattsville, Md.
2 National Center for Health Statistics/Academy Health Fellowship
Program , Division of Health Care Statistics, National Center for Health
Statistics, Centers for Disease Control and Prevention, Hyattsville, Md.
3 RTI International, Washington, DC.
For further information, contact: Margaret Jean Hall, PhD,
Hospital Care Statistics Branch, Division of Health Care Statistics,
National Center for Health Statistics, Centers for Disease Control
and Prevention, 3311 Toledo Road, Hyattsville, MD 20782;
e-mail [email protected] .
. . . . . Hospitals . . . . .
The Journal of Rural Health 332 Vol. 22, No. 4
younger than 65 years, not on Medicare, is conspicuously lacking. This article fi lls that void by providing nationally representative all-payer data on the number and type of inpatients, both younger than 65 and 65 years and older, in rural hospitals. The purpose of this descriptive study is to document existing utilization differences between these patients and those in urban hospitals. By examining the inpatient population in each setting, it is possible to describe the role these hospitals play in meeting the health care needs in their communities.
Methods Data Sources. This research used data from the
2001 National Hospital Discharge Survey (NHDS), a nationally representative survey conducted by the National Center for Health Statistics, Centers for Disease Control and Prevention, annually since 1965. Data were collected from a sample of inpatient records obtained from a national probability sample of nonfederal, short-stay hospitals. In 2001, data were gathered on approximately 330,000 discharges from about 450 hospitals. A full description of the survey design and data collection procedures, as well as the estimation methodology, has been published elsewhere. 12
The major variable of interest in this study was location of the hospital: rural versus urban. There are several different methodologies currently used to defi ne rurality. The defi nition used in this research was based on the June 2003 Offi ce of Management and Budget classifi cation of counties according to metropolitan status. 13 This new classifi cation is based on results of the 2000 Census. Metropolitan counties are classifi ed as urban and all others (micropolitan and noncore counties) are rural. Hospital ZIP codes were used to identify county Federal Information Processing Standards (FIPS) codes, which provided the link to the Offi ce of Management and Budget metro/nonmetro classifi cation. FIPS codes are standard numeric codes used for identifying US geographical areas (ie, states and counties).
Data analyzed from NHDS included age, gender, and race of patients; administrative information, such as days of care, principal expected source of payment, and disposition upon discharge; and clinical patient characteristics such as medical diagnoses (up to 7) and surgical and nonsurgical procedures (up to 4) performed during the hospitalization. Data on newborn infants were excluded from the analyses.
Average length of stay refers to days spent as an inpatient and was calculated by dividing the total number of inpatient days by the number of inpatients
discharged. The principal expected payment sources included Medicare, Medicaid, private insurance, health maintenance organization/preferred provider organization (HMO/PPO), other government, self-pay, and other/not stated. There were not enough rural records with secondary payment sources to produce reliable estimates of the dual eligibles (ie, patients with both Medicare and Medicaid coverage). Disposition of inpatients upon discharge included (1) routine discharges (ie, discharged home or to usual place of residence), (2) transfers to another short-term hospital (eg, to another acute care hospital), (3) transfers to a long-term care institution (eg, a nursing home or swing bed), and (4) in-hospital deaths. Inpatients with all the above discharge dispositions were included in this study. Transfers were not excluded because it is important to know whether additional inpatient care was necessary following hospitalization. This adds to an understanding of the hospital ’ s role within the context of the overall health care delivery system. Data were not available on whether patients were admitted from a rural or urban emergency department, so this was not included in our analysis.
Diagnosis and procedure data in the NHDS were coded according to the International Classifi cation of Diseases, 9th Revision, Clinical Modifi cation (ICD-9-CM). Using the fi rst-listed diagnosis, an indicator for “ avoidable ” hospitalization was created. The following were indicative of hospitalizations considered avoidable: pneumonia, diabetes with ketoacidosis or coma, asthma, congestive heart failure, cellulitis, pyelonephritis, hypokalemia, malignant hypertension, gangrene, ruptured appendix, perforated/bleeding ulcer, and immunizable conditions. These conditions are also referred to as “ ambulatory care sensitive ” because early intervention in the outpatient setting may prevent complications or avert more serious disease 14-17 that requires hospitalization.
The Charlson index, as adapted by Deyo, 18 was used to measure the burden of comorbidities among inpatients in urban and rural hospitals. This widely used method of quantifying comorbidities was chosen for this study because it is one of the few measures that rely on administrative data like those collected in the NHDS. Among the components included in the index are chronic illnesses such as diabetes, heart disease, chronic obstructive pulmonary disease, and cancer. Scores range from 0, for the absence of any comorbid diseases, to a maximum of 16, with higher values indicative of greater patient comorbidity burden and more intensive needs.
The NHDS was designed to measure inpatient utilization in US acute care hospitals. Because some patients are admitted and discharged multiple times
. . . . . Hospitals . . . . .
Hall, Owings and Shinogle 333 Fall 2006
during the year, it is possible for them to be sampled more than once. Survey procedures do not track these readmissions, so it is not possible to identify these patients. Nevertheless, for the purpose of this research, all of these contribute to hospital utilization, and so the terms inpatient and discharge and hospitalization are used interchangeably.
To obtain contextual data on urban and rural counties, including information related to the potential patient and labor pool for sampled hospitals, we linked the NHDS data with county-level data from the 2003 Area Resource File (ARF). 19 Using data for 2001, the rate per 10,000 population of active physicians (not including doctors of osteopathy) and the rate per 10,000 population of short-term, general hospital beds (set up and staffed) were calculated and included as general measures of the supply of health services. As indicators of socioeconomic status, we examined the percentage of residents with at least a college education and the median household income of the county. Merging was accomplished using the county FIPS code for each hospital. Counties rather than ZIP code areas were used for contextual information because extensive health services information was only available at the county level through the ARF.
Data Analysis. Sampled data were weighted to produce national statistics using multistage estimation procedures. Differences between estimates for rural and urban inpatients were tested using the 2-sided t test with a critical value of 1.96 (.05 level of signifi cance). Because of the stated descriptive purpose of this study, unadjusted tests were performed. However, we did stratify by 2 age groups (younger than 65 and 65 and older) in order to examine age differences in patterns of utilization. All statistical tests were performed with standard errors of estimates derived from SUDAAN (RTI International, Research Triangle Park, NC), which takes into account the complex design of the NHDS. 20
Results Rural hospitals deliver a considerable amount of
inpatient care. Of the 32.7 million hospitalizations in 2001, 17% or 5.7 million were in rural hospitals. Seventeen percent of the population live in rural areas. 4 Fourteen percent (23 million) of the 159 million inpatient days of care and 11% (4.7 million) of the 41 million inpatient procedures performed were in rural hospitals ( Figure 1). Examining utilization by age indicated that rural hospitals provided a greater share of the inpatient care for the elderly than for those
younger than 65 years. Among the elderly, 22% of hospitalizations and 18% of days of care were in rural hospitals. For patients younger than 65 years, these fi gures were 14% and 11%, respectively.
The demographic profi le of discharges from rural hospitals was similar to that of urban hospitals with respect to gender, but different by race and age ( see Table). In both settings, about 60% of the discharges were female. Black or African American persons represented a higher percentage of discharges from urban hospitals compared to those from rural hospitals (13% vs 8%). Rural hospital inpatients were older on average than urban inpatients (57 vs 51 years old). About half of rural inpatients were 65 years or older compared to about a third of urban inpatients ( Figure 2). Rural hospitals had almost double the percentage of discharges for inpatients 85 years and older (13% vs 7%).
Expected principal source of payment for rural hospital inpatients was primarily from public payers ( see Table ). Two thirds of the inpatients in rural hospitals, compared to half in urban hospitals, expected Medicare or Medicaid as their principal payment. HMO/PPO payment was relatively more common for urban (18%) than for rural (6%) inpatients. Twenty-fi ve percent of urban hospital patients younger than 65 years had HMO/PPO listed as the principal expected source of payment, which was double the 12% of rural hospital patients expecting this payment source.
As would be expected, disposition upon discharge was quite different for older and younger patients.
Figure 1. Utilization of inpatient services in short-stay hospitals, by location of the hospital: United States, 2001.
Rural hospitals Urban hospitals
33 million DISCHARGES
159 million DAYS OF CARE
17%
83%41 million PROCEDURES
11%
89%
14%
86%
Source: National Hospital Discharge Survey.
. . . . . Hospitals . . . . .
The Journal of Rural Health 334 Vol. 22, No. 4
Cha
ract
eri
stic
s o
f In
pa
tie
nts
in R
ura
l an
d U
rba
n H
osp
ita
ls b
y A
ge
Gro
up
, Un
ite
d S
tate
s 20
01 *
A
ll P
atie
nts
Pat
ien
ts Y
ou
ng
er T
han
65
yP
atie
nts
65
y an
d O
lder
Ru
ral H
osp
ital
s (N
= 5
,687
,000
)U
rban
Ho
spit
als
(N =
26,
966,
000)
Ru
ral H
osp
ital
s (n
= 2
,882
,000
)U
rban
Ho
spit
als
(n =
17,
245,
000)
Ru
ral H
osp
ital
s (n
= 2
,804
,000
)U
rban
Ho
spit
als
(n =
9,7
21,0
00)
%SE
†%
SE†
%SE
†%
SE†
%SE
†%
SE†
All
100
0.0
100
0.0
100
0.0
100
0.0
100
0.0
100
0.0
Sex
M
ale
391.
139
0.4
381.
538
0.6
411.
541
0.4
Fe
mal
e61
1.1
610.
462
1.5
620.
659
1.5
590.
4A
ge
gro
up
(y)
Yo
un
ger
th
an 1
56
0.6
81.
111
1.2
131.
6‡
‡‡
‡
15-4
425
1.4
330.
750
1.6
521.
0‡
‡‡
‡
45-6
420
0.7
231.
439
1.3
351.
0‡
‡‡
‡
65-7
416
0.6
140.
3‡
‡‡
‡32
1.0
390.
5
75-8
421
0.9
150.
4‡
‡‡
‡42
1.2
420.
3
85 y
an
d o
lder
130.
87
0.3
‡‡
‡‡
261.
020
0.4
Rac
e
Wh
ite
658.
862
2.1
609.
359
2.0
708.
267
2.5
B
lack
82.
113
0.8
102.
815
0.9
51.
58
0.6
O
ther
rac
es§
§4
0.6
§§
50.
6§
§3
0.6
R
ace
no
t st
ated
§§
212.
2§
§21
2.0
§§
222.
5Ex
pec
ted
so
urc
e o
f p
aym
ent
M
edic
are
531.
836
0.9
121.
19
0.3
951.
084
1.0
M
edic
aid
130.
915
0.9
251.
222
1.3
10.
12
0.3
P
riva
te in
sura
nce
211.
821
1.1
393.
330
1.5
30.
66
0.7
H
MO
/PP
O6
1.6
181.
012
2.7
251.
4§
§6
0.5
O
ther
go
vern
men
t2
0.5
20.
23
0.8
30.
2§
§1
0.2
Se
lf-p
ay4
0.4
50.
27
0.7
70.
3§
§1
0.1
O
ther
/no
t st
ated
§§
40.
6§
§5
0.8
§§
10.
3D
isch
arg
e st
atu
s
Ho
me
721.
779
0.7
870.
888
0.6
562.
662
0.8
Tr
ansf
er t
o a
sh
ort
-ter
m h
osp
ital
70.
74
0.3
50.
63
0.3
91.
06
0.4
Tr
ansf
er t
o a
lon
g-t
erm
car
e in
stit
uti
on
111.
07
0.4
20.
32
0.1
201.
817
0.7
D
ead
30.
42
0.1
§§
10.
14
0.4
50.
1
Oth
er/n
ot
stat
ed7
1.4
80.
64
0.7
60.
510
2.3
110.
9O
ther
mea
sure
s
Ave
rag
e ag
e57
0.9
510.
836
0.6
360.
779
0.2
770.
1
Ave
rag
e n
um
ber
of
dia
gn
ose
s4.
60.
14.
50.
13.
80.
23.
90.
15.
40.
15.
70.
0
Ave
rag
e n
um
ber
of
pro
ced
ure
s0.
80.
11.
30.
00.
90.
11.
40.
00.
80.
11.
30.
0
Ave
rag
e le
ng
th o
f st
ay4.
00.
15.
10.
13.
20.
14.
50.
14.
70.
26.
10.
1
Ch
arls
on
co
mo
rbid
ity
ind
ex m
ean
1.0
0.0
0.9
0.0
0.6
0.0
0.6
0.0
1.4
0.0
1.4
0.0
P
erce
nt
avo
idab
le h
osp
ital
izat
ion
s15
0.8
110.
312
0.8
80.
319
0.9
150.
4C
ou
nty
-lev
el c
on
text
ual
mea
sure
s ||
A
ctiv
e p
hys
icia
ns
per
10,
000
po
pu
lati
on
¶
19.8
4.9
33.0
1.2
‡‡
‡‡
‡‡
‡‡
Sh
ort
-ter
m h
osp
ital
bed
s p
er 1
0,00
0 p
op
ula
tio
n48
.15.
334
.11.
5‡
‡‡
‡‡
‡‡
‡
Med
ian
ho
use
ho
ld in
com
e$3
4,00
0$7
06$4
4,00
0$5
70‡
‡‡
‡‡
‡‡
‡
Per
cen
t o
f p
op
ula
tio
n w
ith
co
lleg
e ed
uca
tio
n15
.21.
726
.90.
6‡
‡‡
‡‡
‡‡
‡
* So
urc
e: T
he
Nat
ion
al H
osp
ital
Dis
char
ge
Surv
ey. E
stim
ates
exc
lud
e n
ewb
orn
infa
nts
. † S
E, s
tan
dar
d e
rro
r. ‡ C
ateg
ory
no
t ap
plic
able
. § F
igu
re d
oes
no
t m
eet
stan
dar
d o
f re
liab
ility
or
pre
cisi
on
. || D
ata
fro
m t
he
2003
AR
F. ¶
Do
cto
rs o
f o
steo
pat
hy
are
no
t in
clu
ded
.
. . . . . Hospitals . . . . .
Hall, Owings and Shinogle 335 Fall 2006
Nearly 90% of both rural and urban patients younger than 65 years were discharged home (87% and 88%, respectively). This share was much smaller for older patients, with rural elderly less likely than urban elderly to be discharged home (56% vs 62%). Transfers to other short-term hospitals were relatively more common among rural hospital inpatients compared to those in urban hospitals. This was true for both the elderly (9% vs 6%) and the younger patients (5% vs 3%).
Avoidable hospitalizations were relatively more frequent in rural than in urban hospitals (15% vs 11%), both for patients younger than 65 years and the elderly (see Table ). In rural hospitals, 19% of elderly inpatients and 12% of nonelderly patients had “ ambulatory care sensitive conditions. ” The comparable numbers for urban hospitals were 15% for elderly and 8% for nonelderly.
Several indirect measures were used to compare the severity of illness of rural and urban inpatients, and these produced inconsistent results. The Charlson/Deyo comorbidity index was similar for rural and urban hospital inpatients in both age groups. The average number of diagnoses for inpatients younger than 65 years in rural and urban hospitals also did not differ signifi cantly. But there were some indications that the urban hospital inpatients were more severely ill than the rural hospital inpatients. Namely, urban inpatients 65 years or older had a signifi cantly higher average number of diagnoses compared to elderly rural inpatients, and urban inpatients in both the nonelderly and the elderly age groups had longer average lengths of stay than rural hospital inpatients (5.1 compared to 4.0 days for all ages).
First-listed or principal diagnoses for rural and urban hospital inpatients, classified by ICD-9-CM
codes, indicated that patients with a wide variety of medical conditions were treated in both settings. However, for inpatients of all ages, there were higher percentages of respiratory diseases in rural hospitals and higher percentages of neoplasms in urban hospitals. For inpatients younger than 65 years, there were relatively more digestive and genitourinary diseases in rural hospitals (data not shown).
The general types of procedures performed on inpatients in rural and urban hospitals also showed some systematic differences. At least 70% of the inpatient procedures performed in both settings were cardiovascular, digestive, obstetrical, and miscellaneous diagnostic and therapeutic procedures (according to the ICD-9-CM). For both settings , the largest share was 30% for miscellaneous diagnostic and therapeutic procedures, which includes Computerized Axial Tomography (CAT), scan, diagnostic ultrasound, chemotherapy infusion, respiratory therapy, and arteriography and angiocardiography. A larger share of inpatient procedures in urban hospitals were cardiovascular (16% vs 9%), but a larger share were digestive in rural hospitals (16% vs 13%). A similar share (16%) of inpatient procedures was obstetrical in both settings (data not shown).
Apart from differences in the type of inpatient procedures performed, overall procedure utilization was quite different in rural and urban hospitals. On average, urban hospital inpatients received 60% more procedures per patient than rural hospital inpatients (1.3 vs 0.8, respectively, see Table ). In fact, over half (54%) of the inpatients in rural hospitals had no surgical or nonsurgical procedures performed during their inpatient stay. This contrasts with 37% of inpatients in urban hospitals who received no procedures. Urban hospital inpatients were 3 times as likely as rural hospital inpatients to have 4 or more procedures during their hospitalization (12% vs 4%).
It is noteworthy that certain procedures were largely performed in urban hospitals. Ninety-three percent of the inpatient cardiovascular procedures were performed in urban hospitals, including 95% of the coronary artery bypass graft procedures and 96% of procedures to remove coronary artery obstruction. Under the diagnostic and therapeutic procedures category, 93% of the angiocardiography or arteriography inpatient procedures were performed in urban hospitals (data not shown).
Discussion This study found that rural hospitals play a vital
role in providing inpatient hospital care in rural
Figure 2. Percent distribution of discharges from rural and urban hospitals, by age: United States, 2001.
Source: National Hospital Discharge Survey.
. . . . . Hospitals . . . . .
The Journal of Rural Health 336 Vol. 22, No. 4
communities, as evidenced by the volume of hospitalizations, inpatient procedures performed, and days of hospital care provided. For the 17% of the population who live in rural areas, the presence of a hospital is not only supportive of the local economy but also provides the option of obtaining inpatient care without traveling to urban areas.
Patients of all ages with a wide variety of medical conditions and needs are treated in rural hospitals. The fact that elderly patients comprise a larger share of the inpatients in rural hospitals makes these hospitals more dependent on Medicare payment than urban hospitals. Furthermore, proportionally more of the inpatients served by rural hospitals are in the oldest old (85 years and older) age group than those served by urban hospitals. This has important implications for the type of care they deliver. Nevertheless, half of the rural hospital inpatients are younger than 65 years and most of these have payment sources other than Medicare. The elderly ’ s longer stays increase the share of days of care paid by Medicare but, nonetheless, over a third of the days still have non-Medicare payment sources. The proportion of inpatients expecting payment from HMOs or PPOs is smaller in rural than in urban areas. This fi nding refl ects the lack of a competitive HMO market in many of these areas, which hampers efforts to achieve health plan competition in both the private and the Medicare markets. 9
Perhaps most revealing of the role of rural hospitals are the data on inpatient procedures, which clearly indicate a difference in the quantity and types of services between the 2 settings. Eleven percent of total inpatient procedures were performed in rural hospitals although 17% of the population live in rural areas. A higher percentage of rural hospital inpatients did not receive any procedures at all during their hospital stay, and those who did receive them received fewer. The data on sophisticated cardiovascular and certain other inpatient procedures show that they were overwhelmingly being performed in urban hospitals. In contrast, the current results show that there are a number of routine inpatient diagnostic and therapeutic procedures performed in rural as well as urban hospitals.
Rural hospital inpatients had shorter average hospital stays than urban hospital inpatients. A direct consequence of this is that the proportion of rural hospital days (14%) is less than what would have been expected in view of the proportion of population that is rural (17%). Estimates published by Ricketts and Heaphy 5 , based on 1996 American Hospital Association data, showed that urban hospitals had shorter average stays (5.9 days) than rural hospitals (7.4 days). It is
interesting to note this reversal in recent years. Shorter average stays in rural hospitals may be due to their greater likelihood of discharging inpatients to other facilities (short-term hospitals or long-term institutions — including swing beds) rather than directly home.
Data from the ARF 19 included in the Table show that counties where rural hospitals were located had a greater availability of short-term, general staffed hospital beds than counties where urban hospitals were located. These results are consistent with fi ndings from a 1999 MedPAC Report, which found that rural Medicare benefi ciaries have equal access to health care but that they rely more heavily on hospital care than their urban counterparts. 9
The higher rates of avoidable hospitalization reported here may be indicative of problems with availability and/or access to timely and appropriate primary care. The ARF data in the Table also show a lower supply of active physicians in rural areas. Alternatively, the rates could refl ect the presence of older, sicker patients who need short-term hospitalization periodically to stabilize their conditions and prevent potentially serious, even life-threatening, complications less likely to occur in younger patients with the same conditions. These same vulnerabilities of older patients could mean that they have more readmissions within a short period of time after their discharge. Other factors not examined in this study could also contribute to the higher avoidable hospitalization rates.
Failure to seek care at an earlier stage, when conditions are more responsive to treatment, is associated with lower socioeconomic status. 14 The data in the Table, based on the ARF, show that counties with rural hospitals had a lower percentage of college-educated residents and a lower median income than counties with urban hospitals. This could be an indication that patients in these rural counties have additional barriers to access to care, such as no or inadequate health insurance, less ability to pay for uncovered care, and more limited health knowledge (including the importance of seeking care at an early stage in an illness).
Limitations. There is considerable variability among hospitals found in rural and urban areas. Within the rural hospitals, for example, there are rural referral centers that may more closely resemble urban hospitals in their patients, their staffs, and the types of services they provide. But many rural hospitals are smaller and offer fewer services. 2 In contrast, all hospitals in urban areas are not large or teaching hospitals. To accomplish the objective of this research, which was to present as yet unreleased, descriptive,
. . . . . Hospitals . . . . .
. . . . . Hospitals . . . . .
Hall, Owings and Shinogle 337 Fall 2006
nationally representative data on the full spectrum of hospitals in both settings, we did not control for these or other variables.
Direct measures of illness severity for urban and rural hospital patients were not available in this study since NHDS does not gather clinical data on hospital inpatients. Indirect measures, such as number of diagnoses, number of procedures received, and average length of stay were used to indicate the seriousness of patients ’ conditions. The Charlson/Deyo index deals with the number of comorbidities but not the severity of the principal diagnosis. Because of the reliance mainly on indirect measures, the conclusions presented here relating to severity of patient condition are more suggestive than defi nitive.
Similarly, ICD-9-CM categories used to compare the types of diagnoses and procedures of urban and rural hospital patients were not indicative of patient severity. These categories are by nature general and include conditions and illnesses with varying levels of severity, and procedures with different levels of complexity. Disaggregation to more specifi c diagnostic and procedure categories was not possible due to large standard errors for many rural estimates. The measures used here do, however, illustrate the broad scope of medical conditions treated and the nature of the services provided in hospitals in each of these settings.
Since this study focused on inpatients only, information was not included on the number of patients who were treated in rural hospital emergency rooms and then transferred to urban hospitals, often because they presented with very complex medical conditions. Therefore, this article does not include data on the total percentage of outpatients and inpatients transferred from rural to urban hospitals.
Nor does this article analyze patterns of outmigration or bypassing of rural hospitals by patients who lived in rural areas but went directly to urban hospitals for their inpatient care, either by choice or by physician referral. The NHDS estimated that in 2001 approximately 30% of rural residents’ hospitalizations were in urban hospitals. This article ’ s focus is on the inpatient services (the supply of care) provided in rural and urban hospitals. We do not attempt here to address the range of issues related to rural residents ’ choice of where to receive their inpatient care.
Implications. Having access to inpatient medical care, including services for surgery, injuries, and acute medical conditions, is particularly important in rural areas in view of the chronic shortage of physicians. 3 Prior research has shown that the most vulnerable populations (ie, the elderly, those on Medicaid, and the
uninsured) are the least likely to travel out of the rural counties for hospital care. These groups of patients would be adversely affected by policies leading to rural hospital closures or regionalization of services. 21
Medicare payment policy decisions made at the national level can affect the fi nancial viability of rural hospitals, and therefore the access to inpatient care, for all rural residents — both those on Medicare and those with other payment sources. For that reason, the possible effects on both Medicare and non-Medicare patients of proposed changes in these policies should be considered. The nationally representative data presented in this article provide a basis for policymakers, researchers, and practitioners to assess the current role of rural hospitals in delivering inpatient care to all patients. This should also be valuable in efforts to develop a vision for the future role of rural hospitals in the 21st century.
References 1. Ormand B , Wallin SW , Golenson SM . Supporting the Rural
Health Care Safety Net . Urban Institute , ed . 2000 . Urban Institute. Occasional Paper No. 36 .
2. Moscovice I , Wholey DR , Klingner J , Knott A . Measuring rural hospital quality . J Rural Health . 2004 ; 20 : 383 - 393 .
3. Rosenblatt RA , Hartley D . Physicians and rural America . In : Ricketts TC , ed . Rural Health in the United States . New York : Oxford University Press ; 1999:38-51 .
4. United States Department of Agriculture ERS . Measuring Rurality:
What is Rural? Briefi ng Room 2005. Available at : http://www.ers.usda.gov/Briefi ng/Rurality . Accessed February 25, 2005 .
5. Ricketts TC , Heaphy PE . Hospitals in rural America . In : Ricketts TC , ed . Rural Health in the United States . New York : Oxford University Press ; 1999:101-112 .
6. Moscovice I , Stensland J . Rural hospitals: trends, challenges, and a future research and policy analysis agenda . J Rural Health . 2002 ; 18 : 197 - 210 .
7. Moscovice I , Wells KB , Stensland J . Rural Hospitals:
Accomplishments and Present Challenges . Minneapolis, Minn : University of Minnesota Rural Health Research Center ; 1996 .
8. Buczko W . Rural Medicare benefi ciaries ’ use of rural and urban hospitals . J Rural Health . 2001 ; 17 ( 1 ): 53 - 58 .
9. Medicare Payment Advisory Commission . Report to Congress:
Medicare in Rural America . Washington, DC: MedPAC ; 2001 . 10. Buczko W . Bypassing of local hospitals by rural Medicare
benefi ciaries . J Rural Health . 1994 ; 10 : 237 - 246 . 11. Department of Health and Human Services, Rural Task Force .
One department serving rural America — report to the secretary. 2002 . Available at : http://ruralhealth.hrsa.gov/publicReport.htm . Accessed February 18, 2005 .
12. Dennison C , Pokras R . Design and operation of the National Hospital Discharge Survey: 1988 redesign . Vital Health Stat 1 . 2000 ;( 39 ): 1 - 42 .
13. Offi ce of Management and Budget (OMB) . Revised Defi nitions
of Metropolitan Statistical Areas, New Defi nitions of Micropolitan
Statistical Area and Combined Statistical Areas and Guidance on Use
of Statistical Defi nitions of these Areas. OMB Bulletin 03-04. 2003. Available at http://www.whitehouse.gov/omb/bulletins/b03-04.html .
. . . . . Hospitals . . . . .
. . . . . Hospitals . . . . .
The Journal of Rural Health 338 Vol. 22, No. 4
14. Billings J , Zeitel L , Lukomnik J , Carey TS , Blank AE , Newman L . Impact of socioeconomic status on hospital use in New York City . Health Aff . 1993 ; 12 ( 1 ): 162 - 173 .
15. Bindman AB , Grumbach K , Osmond D , et al . Preventable hospitalizations and access to health care . JAMA . 1995 ; 274 : 305 - 311 .
16. Gaskin DJ , Hoffman C . Racial and ethnic differences in preventable hospitalizations across 10 states . Med Care Res Rev . 2000 ; 57 : 85 - 107 .
17. Pappas G , Hadden WC , Kozak LJ , Fisher GF . Potentially avoidable hospitalizations: inequalities in rates between US socioeconomic groups . Am J Public Health . 1997 ; 87 : 811 - 816 .
18. Deyo RA . Adapting a clinical comorbidity index for use with ICD-9-CM administrative data — a response . J Clin Epidemiol . 1993 ; 46 : 1081 - 1082 .
19. US Department of Health and Human Services, Health Resources and Services Administration, Bureau of Health Professionals . Area Resource File (ARF) . Rockville , Md ; 2003 .
20. Research Triangle Institute . SUDAAN User’s Manual, Release 7.0 . Research Triangle Park, NC : Research Triangle Institute ; 1996 .
21. Basu J , Cooper J . Out-of-area travel from rural and urban counties: a study of ambulatory care sensitive hospitalizations for New York State residents . J Rural Health . 2000 ; 16 ( 2 ): 129 - 138 .
