0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Original Investigation |

Trends in Postdischarge Mortality and Readmissions:  Has Length of Stay Declined Too Far? FREE

David W. Baker, MD, MPH; Doug Einstadter, MD, MPH; Scott S. Husak, BS; Randall D. Cebul, MD
[+] Author Affiliations

From the Center for Health Care Research and Policy (Drs Baker, Einstadter, and Cebul and Mr Husak) and Department of Medicine (Drs Baker, Einstadter, and Cebul), Case Western Reserve University at MetroHealth Medical Center; and Department of Epidemiology and Biostatistics, Case Western Reserve University (Drs Baker, Einstadter, and Cebul), Cleveland, Ohio. Dr Baker is now with the Division of General Internal Medicine, Department of Medicine, Feinberg School of Medicine, Northwestern University, Chicago, Ill. The authors have no relevant financial interest in this article.


Arch Intern Med. 2004;164(5):538-544. doi:10.1001/archinte.164.5.538.
Text Size: A A A
Published online

Background  Length of hospital stay continues to decline, but the effect on postdischarge outcomes is unclear.

Methods  We determined trends in risk-adjusted mortality rates and readmission rates for 83 445 Medicare patients discharged alive after hospitalization for myocardial infarction, heart failure, gastrointestinal hemorrhage, chronic obstructive pulmonary disease, pneumonia, or stroke. Patients were stratified into deciles of observed/expected length of stay to determine whether patients whose length of stay was much shorter than expected had higher risk-adjusted mortality and readmission rates. Analyses were stratified by whether a do-not-resuscitate (DNR) order was written within 2 days of admission (early) or later.

Results  From 1991 through 1997, risk-adjusted postdischarge mortality generally remained stable for patients without a DNR order. Postdischarge mortality increased by 21% to 72% for patients with early DNR orders and increased for 2 of 6 diagnoses for patients with late DNR orders. Markedly shorter than expected length of stay was associated with higher than expected risk-adjusted mortality for patients with early DNR orders but not for others (no DNR and late DNR). Risk-adjusted readmission rates remained stable from 1991 through 1997, except for a 15% (95% confidence interval, 3%-30%) increase for patients with congestive heart failure. Short observed/expected length of stay was not associated with higher readmission rates.

Conclusions  The dramatic decline in length of stay from 1991 through 1997 was not associated with worse postdischarge outcomes for patients without DNR orders. However, postdischarge mortality increased among patients with early DNR orders, and some of this trend may be due to patients being discharged more rapidly than previously.

Figures in this Article

During the past decade, length of stay for patients admitted to US hospitals has continued the steady decline that began with the implementation of the Prospective Payment System for Medicare in 1982.14 Our group5 previously reported that, between 1991 and 1997, mean length of stay for Medicare patients hospitalized in northeast Ohio declined dramatically for acute myocardial infarction (AMI; from 10.6 to 8.1 days), congestive heart failure (CHF; 9.2 to 6.6 days), gastrointestinal (GI) hemorrhage (8.3 to 6.2 days), chronic obstructive pulmonary disease (COPD; 7.5 to 5.4 days), pneumonia (10.3 to 7.3 days), and stroke (10.4 to 6.3 days). This inexorable decline in length of stay raises concern that increasing numbers of patients may be discharged before their condition has been fully stabilized.68 If more patients are being discharged prematurely as a result of very short lengths of stay, this could cause higher mortality and readmission rates after discharge.

Relatively few studies have examined associations between length of stay and health outcomes.3,6,7,9,10 Kosecoff and colleagues6 found that the proportion of patients discharged home in unstable condition increased from 10% immediately before Prospective Payment System implementation (1981-1982) to 15% shortly afterward (1985-1986). Patients discharged home in unstable condition were 60% more likely to die in the 90 days after discharge. However, that study did not determine whether individuals with shorter than expected length of stay were more likely to be discharged in unstable condition. Fitzgerald and colleagues11 also examined changes in outcomes after Prospective Payment System implementation and found that rates of institutionalization after hip fracture were substantially higher during the later years when length of stay was shorter. Others have reported worse outcomes among patients with shorter lengths of stay for schizophrenia9 and depression.10

In contrast, a Canadian study found that, while length of stay declined between 1989 and 1993 for the 4 conditions analyzed, readmission rates increased for only 1 of 4 conditions examined; hospitals with shorter lengths of stay did not have higher readmission rates.8 Philbin and Roerden12 did not find an adverse effect of short length of stay among patients with CHF. However, that study was problematic because it did not examine outcomes for patients whose length of stay was shorter than expected on the basis of their admission severity of illness. Failure to account for expected length of stay can produce spurious relationships between length of stay and readmissions or postdischarge mortality.13

This study examines trends in the risk of death during the 30 days after discharge for Medicare patients hospitalized in northeast Ohio hospitals between 1991 and 1997 and analyzes whether individuals whose length of stay was shorter than expected had higher readmission rates and risk-adjusted mortality after discharge.

STUDY POPULATION

This study was approved by the MetroHealth Medical Center Institutional Review Board, Cleveland, Ohio. We used data from a previous study that combined data from the Cleveland Health Quality Choice (CHQC) program14 with Medicare Provider Analysis and Review (MEDPAR) files from the Health Care Financing Administration (now the Centers for Medicare and Medicaid Services). Both the CHQC program14 and our method for merging the CHQC database with MEDPAR files have been described previously.5 Briefly, CHQC was a regional initiative designed to objectively measure and compare risk-adjusted outcomes for hospitals in northeast Ohio. All 30 nonfederal hospitals in greater metropolitan Cleveland participated from 1991 through 1999; our study uses only data from 1991 through 1997 because MEDPAR files were not available past 1997 at the time the study was begun. Every 6 months, hospitals provided CHQC with a list of all patients (excluding interhospital transfers) discharged with a principal diagnosis of AMI, CHF, GI hemorrhage, COPD, pneumonia, and stroke (see Baker et al5 for full list of International Classification of Diseases, Ninth Revision [ICD-9] codes). Chart abstraction for CHQC was performed by medical records technicians at each hospital to obtain patients' demographics; admission source (eg, home or nursing home); comorbid conditions; admission vital signs and neurologic findings; and results of laboratory, radiologic, and electrocardiographic testing. Thus, the combined database includes extremely detailed clinical data for all eligible Medicare admissions to nonfederal hospitals in northeast Ohio between 1991 and 1997 for patients 18 years and older.

Separate databases were created for the 6 conditions. As was done for our previous analyses of mortality trends,5 we selected only an individual's first admission within the CHQC disease-specific database. In addition, some patients were admitted for the same diagnosis before 1991 or were admitted to a non-CHQC hospital for the same diagnosis before their index admission in CHQC (eg, admitted to a hospital in Florida for CHF in 1994 and to a CHQC hospital in 1995). To address this, we searched MEDPAR files from 1989 to 1997 and excluded patients who had a non-CHQC admission with the same diagnosis in the 2 years before their index admission in CHQC. Patients with a qualifying first admission in more than one disease category (eg, pneumonia and stroke) were included in all databases. Finally, because we were interested in the relationship between length of stay and postdischarge mortality, we excluded any patient who died during the index hospitalization.

ADMISSION SEVERITY OF ILLNESS

Analyses were conducted with SAS software, version 8.01 (SAS Institute Inc, Cary, NC). The methods for developing our 6 disease-specific models of severity of illness at admission, the complete lists of variables included in the models, and the predictive validity and calibration of the models have been fully described previously.5 Briefly, we used data from 1991 to 1992 to develop logistic regression models with death at 30 days after admission as the dependent variable. We then used the β-coefficients from these baseline models and individuals' unique values for the variables in the models to determine an admission severity-of-illness score indicating the patients' probability of death within 30 days of admission (scale range, 0%-100%). All models showed consistent performance (eg, c-statistics and calibration)15 across all study years. We also examined separate models of predictors of death during the 30 days after discharge. Since the β-coefficients for predictor variables were similar to our original models of admission severity of illness, we use the latter for consistency with our previous work.

TRENDS IN POSTDISCHARGE MORTALITY, 1991-1997

We determined death within 30 days of discharge (yes or no) by using the date of discharge and the "date of death" variable in MEDPAR. Trends in postdischarge mortality from 1991 through 1997 were analyzed by means of logistic regression with age, sex, race, admission from a nursing home, admission severity of illness, and year of admission as independent variables. Trends were approximately linear, so year of admission was entered as a single, continuous variable.

STRATIFIED ANALYSIS BY DO-NOT-RESUSCITATE STATUS

We hypothesized a priori that mortality trends might differ for patients who had a do-not-resuscitate (DNR) order. Specifically, we hypothesized that changes in attitudes toward life-sustaining treatments could lead to increasing postdischarge mortality for patients with DNR orders while mortality remained stable for those without DNR orders.

The CHQC chart abstractors determined whether a DNR order was written and the hospital day on which it occurred. We created 2 categories of DNR orders: (1) early DNR orders, defined as occurring on the first or second hospital day, and (2) late DNR orders, defined as occurring on day 3 or later.16 An early DNR order may be a statement of general preferences in the event of a cardiopulmonary arrest or an indicator that the patient was admitted in moribund condition. In contrast, many late DNR orders are written because a patient's condition is deteriorating, and further treatment is thought to be futile.16 We repeated all analyses of mortality trends by using multivariate logistic regression with stratification by DNR status (no order, early DNR, and late DNR).

OBSERVED AND EXPECTED LENGTH OF STAY

Length of stay (in days) was determined from the discharge date minus the admission date. Patients who were admitted and discharged on the same day were assigned a length of stay of 1 day. Because of the skewed distribution of length of stay, we used a natural logarithm transformation in most analyses.

To determine "expected" length of stay, we used 1991 to 1992 as our baseline. Using only patients from 1991 to 1992, we regressed the natural logarithm of the length of stay on age, sex, race, admission from a nursing home, and all of the variables used in our admission severity-of-illness models. The r2 for the models was 0.11 for AMI, 0.08 for CHF, 0.05 for COPD, 0.07 for GI hemorrhage, 0.10 for pneumonia, and 0.07 for stroke. We then used the β-coefficients from the variables in the baseline models and each individual's data for these variables to determine the expected length of stay if the patient had been admitted in 1991 to 1992 for all patients in all study years. The observed length of stay was then used to calculate the ratio of the observed to expected length of stay.

OBSERVED/EXPECTED LENGTH OF STAY AND POSTDISCHARGE MORTALITY

We anticipated that there would be a nonlinear relationship between the observed/expected length of stay and postdischarge mortality. Individuals with lengths of stay much longer than expected are likely to have had either unmeasured comorbidity on admission or an in-hospital complication that required prolonged treatment. Such individuals would also be expected to have higher risk-adjusted postdischarge mortality relative to those with average length of stay because of these comorbities and complications. However, we postulated that individuals with markedly shorter than expected length of stay would also have higher risk-adjusted postdischarge mortality. We therefore created deciles of observed/expected length of stay and used these as our main independent variables. Specifically, we used logistic regression with "death within 30 days of discharge" as the dependent variable, deciles 1 to 4 and 7 to 10 of observed/expected length of stay as the main independent variables (with deciles 5 and 6 as the reference category), and age, sex, race, and admission severity of illness as covariates. All analyses were stratified according to whether the patient had a DNR order written within the first 2 days of hospitalization, since early discharge for a patient who has a DNR order is likely to occur for different reasons than for patients who did not have a DNR order. The results for patients with no DNR order and those with late DNR orders were similar, so the 2 groups were combined. Statistical significance was determined with 2-sided tests with a P value of .05 used as a cutoff.

THIRTY-DAY READMISSIONS

We searched the MEDPAR database to identify all readmissions within 30 days of discharge and the principal ICD-9 diagnosis code for the readmission. Readmissions were coded as condition specific (ie, readmission for CHF if the index admission was for CHF) or all-cause (ie, any readmission regardless of principal diagnosis) by means of the same set of ICD-9 codes used to classify the index admissions. The rates of condition-specific 30-day readmissions were too low to allow meaningful interpretation, and we present analyses only for all-cause readmissions. We analyzed (1) trends in readmission rates and (2) the relationship between observed/expected length of stay and the risk of readmission as described above for postdischarge mortality for mortality, stratifying by DNR status. Trends were similar regardless of DNR status, so we present results only for all patients combined.

The demographic data of patients with the 6 target medical conditions who were discharged alive from hospitals in northeast Ohio between 1991 and 1997 are shown in Table 1. The average age was approximately 77 years, slightly more than half of the patients were women, and approximately 85% were white, which is consistent with the demographics of northeast Ohio. The mean length of stay declined steadily from 1991 to 1997 (Table 2; P<.001 for all trends). Similarly, the mean observed/expected length of stay steadily declined (Table 2). Trends in length of stay were similar for patients who had DNR orders and those who did not (data not shown).

Table Graphic Jump LocationTable 1. Demographic Characteristics of 83 445 Medicare Patients Hospitalized in Northeast Ohio With Selected Conditions Between 1991 and 1997 Who Were Discharged Alive
Table Graphic Jump LocationTable 2. Mean Length of Stay and Observed/Expected Length of Stay From 1991 Through 1997*
POSTDISCHARGE MORTALITY TRENDS

Between 1991 and 1997, postdischarge mortality increased significantly for patients with AMI (adjusted relative risk, 1.65; 95% confidence interval [CI], 1.24-2.19) and stroke (adjusted relative risk, 1.59; 95% CI, 1.27-1.97). There was no significant change in postdischarge mortality for patients with CHF, GI hemorrhage, COPD, and pneumonia (Table 3), although there were upward trends for patients with GI hemorrhage and pneumonia.

Table Graphic Jump LocationTable 3. Mortality Rates in the 30 Days After Discharge According to the Presence of a Do-Not-Resuscitate Order*

Trends in postdischarge mortality differed according to whether patients had a DNR order written and the timing of the order (early vs late; Table 3). For patients who did not have a DNR order written, the risk of death during the 30 days after discharge generally remained stable. The exception was AMI, for which the adjusted relative risk of death increased from 1991 through 1997 (adjusted relative risk, 1.47; 95% CI, 1.01-2.13). In contrast, the risk-adjusted postdischarge mortality rate for patients with early DNR orders increased for all diagnoses from 1991 through 1997 (Table 3). These trends were statistically significant for patients with AMI (adjusted relative risk, 1.72; 95% CI, 1.03-2.65), pneumonia (adjusted relative risk, 1.28; 95% CI, 1.03-1.56), and stroke (adjusted relative risk, 1.60; 95% CI, 1.22-2.04). There was no consistent pattern for patients with late DNR orders (Table 3). However, the postdischarge mortality rate for patients with stroke who had a late DNR order increased dramatically from 1991 through 1997 (adjusted relative risk, 2.03; 95% CI, 1.36-2.74).

LENGTH OF STAY AND POSTDISCHARGE MORTALITY

To determine whether patients with markedly shorter than expected length of stay had an increased risk of death after discharge, we divided patients into deciles of observed/expected length of stay. For the 6 conditions, patients in the lowest decile (decile 1) had observed lengths of stay that averaged 45% to 73% shorter than expected, and patients in the second lowest decile had observed lengths of stay that were 23% to 41% lower than expected.

Overall, there was no relationship between observed/expected length of stay and postdischarge mortality. However, this differed by DNR status. Among patients who had an early DNR order, those in the lowest decile of observed/expected length of stay had a higher risk-adjusted mortality rate after discharge than patients whose observed length of stay was approximately as expected (deciles 5 and 6) for 4 of the 6 conditions (black boxes, Figure 1). However, this increased risk was only statistically significant for patients with GI hemorrhage (adjusted relative risk, 1.74; 95% CI, 1.22-2.29). When all patients with early DNR orders (ie, all 6 conditions) were combined into a single model, the group in decile 1 had a 16% higher risk of death than patients in deciles 5 and 6 (adjusted relative risk, 1.16; 95% CI, 0.98-1.38; P = .08). The relationship between shorter than expected length of stay and higher than expected mortality was present across the entire study period, and there was no change in the association over time.

Place holder to copy figure label and caption
Figure 1.

Adjusted relative risk of death within 30 days of discharge for the decile of patients with the lowest observed/expected length of stay (relative to patients whose observed length of stay was approximately as expected [deciles 5 and 6]). DNR indicates do not resuscitate; AMI, acute myocardial infarction; CHF, congestive heart failure; GIH, gastrointestinal hemorrhage; COPD, chronic obstructive pulmonary disease; PNEU, pneumonia; and STR, stroke. Limit lines indicate 95% confidence intervals.

Graphic Jump Location

In contrast, among patients without an early DNR order (ie, no DNR order or a late DNR order), those in the lowest decile of observed/expected length of stay did not have a higher risk-adjusted postdischarge mortality rate than patients whose observed length of stay was approximately as expected (deciles 5 and 6; open circles, Figure 1).

THIRTY-DAY READMISSION TRENDS

There was little change in the 30-day readmission rate from 1991 through 1997 for patients with COPD, GI hemorrhage, pneumonia, and stroke (Table 4). However, the risk of readmission increased between 1991 and 1997 for patients with CHF (adjusted relative risk, 1.15; 95% CI, 1.03-1.30). There was also an upward trend in the risk-adjusted readmission rate for patients with AMI (adjusted relative risk, 1.12; 95% CI, 0.96-1.32), although this did not reach statistical significance (P = .16). The crude readmission rates and trends in readmission rates between 1991 and 1997 were similar regardless of whether a DNR order was written or the timing of the DNR order (early vs late).

Table Graphic Jump LocationTable 4. Trends in Readmission Rates Within 30 Days of Discharge
LENGTH OF STAY AND 30-DAY READMISSION RATES

Patients in the lowest decile of observed/expected length of stay did not have a higher risk-adjusted readmission rate than patients whose observed length of stay was approximately as expected (deciles 5 and 6; Figure 2). The results were similar regardless of whether a DNR order was written or the timing of the DNR order (early vs late).

Place holder to copy figure label and caption
Figure 2.

Adjusted relative risk of readmission within 30 days of discharge for the decile of patients with the lowest observed/expected length of stay (relative to patients whose observed length of stay was approximately as expected [deciles 5 and 6]). AMI indicates acute myocardial infarction; CHF, congestive heart failure; GIH, gastrointestinal hemorrhage; COPD, chronic obstructive pulmonary disease; PNEU, pneumonia; and STR, stroke. Limit lines indicate 95% confidence intervals.

Graphic Jump Location

Despite dramatic reductions in length of stay from 1991 through 1997, we found little evidence that shorter length of stay was associated with higher mortality after discharge or higher readmission rates. These findings provide some reassurance that it is possible to reduce length of stay without jeopardizing patients' health. Improvements in therapies, more rapid administration of effective therapies, greater willingness to evaluate some medical problems in the outpatient setting, more efficient discharge practices, and expanded use of home health care could all contribute to shorter mean lengths of stay without cutting corners. For example, Sgura and colleagues4 found that use of primary reperfusion, β-blockers, and aspirin was associated with shorter length of stay for patients with AMI. Thus, more widespread use of these therapies from 1991 through 1997 could contribute to both improved outcomes and shorter length of stay.

Although our findings are encouraging overall, the consistent increase in postdischarge mortality for patients with early DNR orders (Table 3) should raise concern that this vulnerable group of patients may have been harmed by declines in length of stay. This is underscored by the fact that patients with early DNR orders whose length of stay was markedly shorter than expected (decile 1) had higher risk-adjusted postdischarge mortality rates for 4 of the 6 conditions (Figure 1). Nevertheless, these trends and associations should be interpreted cautiously and should not be construed as unequivocal evidence that shorter lengths of stay cause higher mortality after discharge. The association between observed/expected length of stay and mortality was relatively weak and seen only for patients in decile 1. Thus, declining observed/expected length of stay explains only a fraction of the increasing postdischarge mortality rate for patients with early DNR orders (Table 3).

An alternative explanation for this finding is that, between 1991 and 1997, physicians' practice styles changed and they became more likely to rapidly discharge patients who were clearly terminally ill. Thus, both shorter length of stay and higher postdischarge mortality rates may reflect less aggressive in-hospital treatment of patients with early DNR orders. This may be an entirely appropriate, even desirable, change in practice for terminally ill patients. The CHQC program did not abstract information on quality of care, stability at discharge, or patient-physician discussions about preferences for aggressive care. Thus, we cannot distinguish these 2 competing hypotheses for the increasing postdischarge mortality rate from 1991 through 1997 and the association between very short length of stay and higher postdischarge mortality among patients with early DNR orders. Additional studies are needed to examine whether in-hospital quality of care for patients with DNR orders has declined.

Another concerning finding is that the postdischarge mortality rate increased by almost 50% from 1991 to 1997 for patients with AMI who did not have a DNR order written. This finding was of borderline statistical significance. Moreover, some of this trend may be explained by the very low mortality rate in 1991 (Table 3), which could have occurred merely by chance. In addition, we did not adjust our levels of statistical significance to account for the multiple analyses conducted. Finally, patients with AMI whose observed length of stay was dramatically lower than expected did not have higher rates of postdischarge mortality; those with the shortest risk-adjusted length of stay (decile 1) actually had a lower than expected risk of death after discharge (Figure 1). These factors suggest that the increasing postdischarge mortality rate for patients with AMI who did not have a DNR order may be no more than a chance finding. However, this same group showed a trend toward an increasing 30-day readmission rate from 1991 through 1997, so it remains possible that the observed increase in risk-adjusted postdischarge mortality is real and results from declining quality of care.

There are several important limitations to our study. Our models of expected length of stay were relatively weakly predictive of observed length of stay. The clinical variables available from CHQC only include data from the time of admission. Yet, length of stay is influenced by individuals' response to therapy in addition to their clinical condition on hospitalization. The inaccuracy of our length-of-stay models would weaken our ability to detect associations between observed/expected length of stay and postdischarge mortality or readmission rates. Moreover, a lack of relationship between observed/expected length of stay and postdischarge outcomes does not mean that there are no patients who are discharged prematurely and die or are readmitted as a consequence. A patient could develop complications in the hospital that result in longer than expected length of stay based on admission characteristics but still be discharged prematurely before the complications were adequately treated. Our methods would classify this individual as having a longer than expected length of stay and not detect him or her as having been discharged prematurely.

Because our study included only clinical data from admission, we could not directly assess instability at discharge. Kosecoff and colleagues6 found that there was in increase between 1981 to 1982 and 1985 to 1986 in the number of patients discharged home in unstable condition, and those discharged in unstable condition were 60% more likely to die after discharge. That study did not report whether individuals with shorter than expected length of stay were more likely to be discharged in unstable condition. Although our study suggests that shorter length of hospital stay per se is not harmful for most patients, the financial pressures leading hospitals to reduce their length of stay could still have caused an increase in the number of patients discharged in unstable condition. Objective measures of instability at discharge are available,6,17 and they should be used to examine these issues and to determine whether patients with DNR orders are at increased risk because of premature discharge.

Corresponding author: David W. Baker, MD, MPH, Division of General Internal Medicine, Feinberg School of Medicine, Northwestern University, 676 N St Clair St, Suite 200, Chicago, IL 60611 (e-mail: dwbaker@northwestern.edu).

Accepted for publication March 31, 2003.

This study was supported by grant R01 HS09969 from the Agency for Healthcare Research and Quality, Rockville, Md.

Kominski  GFWitsberger  C Trends in length of stay for Medicare patients: 1979-87. Health Care Financ Rev. 1993;15121- 135
PubMed
Not Available, Adult average length of stay for acute care hospitals Healthc Financ Manage. 1998;5220
PubMed
Clarke  ARosen  R Length of stay: how short should hospital care be? Eur J Public Health. 2001;11166- 170
PubMed Link to Article
Sgura  FAWright  RSKopecky  SLGrill  JPReeder  GS Length of stay in myocardial infarction. Cost Qual. 2001;612- 2025
PubMed
Baker  DWEinstadter  DThomas  CHusak  SGordon  NHCebul  RD Mortality trends during a program that publicly reported hospital performance. Med Care. 2002;40879- 890
PubMed Link to Article
Kosecoff  JKahn  KLRogers  WH  et al.  Prospective payment system and impairment at discharge: the "quicker-and-sicker" story revisited. JAMA. 1990;2641980- 1983
PubMed Link to Article
Epstein  AMBogen  JDreyer  PThorpe  KE Trends in length of stay and rates of readmission in Massachusetts: implications for monitoring quality of care. Inquiry. 1991;2819- 28
PubMed
Harrison  MLGraff  LARoos  NPBrownell  MD Discharging patients earlier from Winnipeg hospitals: does it adversely affect quality of care? CMAJ. 1995;153745- 751
PubMed
Appleby  LDesai  PNLuchins  DJGibbons  RDHedeker  DR Length of stay and recidivism in schizophrenia: a study of public psychiatric hospital patients. Am J Psychiatry. 1993;15072- 76
PubMed
Lieberman  PBWiitala  SAElliott  BMcCormick  SGoyette  SB Decreasing length of stay: are there effects on outcomes of psychiatric hospitalization? Am J Psychiatry. 1998;155905- 909
PubMed
Fitzgerald  JFMoore  PSDittus  RS The care of elderly patients with hip fracture: changes since implementation of the prospective payment system. N Engl J Med. 1988;3191392- 1397
PubMed Link to Article
Philbin  EFRoerden  JB Longer hospital length of stay is not related to better clinical outcomes in congestive heart failure. Am J Manag Care. 1997;31285- 1291
PubMed
Leyland  AH Examining the relationship between length of stay and readmission rates for selected diagnoses in Scottish hospitals. IMA J Math Appl Med Biol. 1995;12175- 184
PubMed Link to Article
Rosenthal  GEHarper  DL Cleveland Health Quality Choice: a model for collaborative community-based outcomes assessment. Jt Comm J Qual Improv. 1994;20425- 442
PubMed
Ash  ASShwartz  M Evaluating the performance of risk-adjustment methods: dichotomous outcomes. Iezzoni  LIed.Risk Adjustment for Measuring Healthcare Outcomes Chicago, Ill Health Administration Press1997;444- 449
Wenger  NSPearson  MLDesmond  KABrook  RHKahn  KL Outcomes of patients with do-not-resuscitate orders: toward an understanding of what do-not-resuscitate orders mean and how they affect patients. Arch Intern Med. 1995;1552063- 2068
PubMed Link to Article
Halm  EAFine  MJKapoor  WNSinger  DEMarrie  TJSiu  AL Instability on hospital discharge and the risk of adverse outcomes in patients with pneumonia. Arch Intern Med. 2002;1621278- 1284
PubMed Link to Article

Figures

Place holder to copy figure label and caption
Figure 1.

Adjusted relative risk of death within 30 days of discharge for the decile of patients with the lowest observed/expected length of stay (relative to patients whose observed length of stay was approximately as expected [deciles 5 and 6]). DNR indicates do not resuscitate; AMI, acute myocardial infarction; CHF, congestive heart failure; GIH, gastrointestinal hemorrhage; COPD, chronic obstructive pulmonary disease; PNEU, pneumonia; and STR, stroke. Limit lines indicate 95% confidence intervals.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 2.

Adjusted relative risk of readmission within 30 days of discharge for the decile of patients with the lowest observed/expected length of stay (relative to patients whose observed length of stay was approximately as expected [deciles 5 and 6]). AMI indicates acute myocardial infarction; CHF, congestive heart failure; GIH, gastrointestinal hemorrhage; COPD, chronic obstructive pulmonary disease; PNEU, pneumonia; and STR, stroke. Limit lines indicate 95% confidence intervals.

Graphic Jump Location

Tables

Table Graphic Jump LocationTable 1. Demographic Characteristics of 83 445 Medicare Patients Hospitalized in Northeast Ohio With Selected Conditions Between 1991 and 1997 Who Were Discharged Alive
Table Graphic Jump LocationTable 2. Mean Length of Stay and Observed/Expected Length of Stay From 1991 Through 1997*
Table Graphic Jump LocationTable 3. Mortality Rates in the 30 Days After Discharge According to the Presence of a Do-Not-Resuscitate Order*
Table Graphic Jump LocationTable 4. Trends in Readmission Rates Within 30 Days of Discharge

References

Kominski  GFWitsberger  C Trends in length of stay for Medicare patients: 1979-87. Health Care Financ Rev. 1993;15121- 135
PubMed
Not Available, Adult average length of stay for acute care hospitals Healthc Financ Manage. 1998;5220
PubMed
Clarke  ARosen  R Length of stay: how short should hospital care be? Eur J Public Health. 2001;11166- 170
PubMed Link to Article
Sgura  FAWright  RSKopecky  SLGrill  JPReeder  GS Length of stay in myocardial infarction. Cost Qual. 2001;612- 2025
PubMed
Baker  DWEinstadter  DThomas  CHusak  SGordon  NHCebul  RD Mortality trends during a program that publicly reported hospital performance. Med Care. 2002;40879- 890
PubMed Link to Article
Kosecoff  JKahn  KLRogers  WH  et al.  Prospective payment system and impairment at discharge: the "quicker-and-sicker" story revisited. JAMA. 1990;2641980- 1983
PubMed Link to Article
Epstein  AMBogen  JDreyer  PThorpe  KE Trends in length of stay and rates of readmission in Massachusetts: implications for monitoring quality of care. Inquiry. 1991;2819- 28
PubMed
Harrison  MLGraff  LARoos  NPBrownell  MD Discharging patients earlier from Winnipeg hospitals: does it adversely affect quality of care? CMAJ. 1995;153745- 751
PubMed
Appleby  LDesai  PNLuchins  DJGibbons  RDHedeker  DR Length of stay and recidivism in schizophrenia: a study of public psychiatric hospital patients. Am J Psychiatry. 1993;15072- 76
PubMed
Lieberman  PBWiitala  SAElliott  BMcCormick  SGoyette  SB Decreasing length of stay: are there effects on outcomes of psychiatric hospitalization? Am J Psychiatry. 1998;155905- 909
PubMed
Fitzgerald  JFMoore  PSDittus  RS The care of elderly patients with hip fracture: changes since implementation of the prospective payment system. N Engl J Med. 1988;3191392- 1397
PubMed Link to Article
Philbin  EFRoerden  JB Longer hospital length of stay is not related to better clinical outcomes in congestive heart failure. Am J Manag Care. 1997;31285- 1291
PubMed
Leyland  AH Examining the relationship between length of stay and readmission rates for selected diagnoses in Scottish hospitals. IMA J Math Appl Med Biol. 1995;12175- 184
PubMed Link to Article
Rosenthal  GEHarper  DL Cleveland Health Quality Choice: a model for collaborative community-based outcomes assessment. Jt Comm J Qual Improv. 1994;20425- 442
PubMed
Ash  ASShwartz  M Evaluating the performance of risk-adjustment methods: dichotomous outcomes. Iezzoni  LIed.Risk Adjustment for Measuring Healthcare Outcomes Chicago, Ill Health Administration Press1997;444- 449
Wenger  NSPearson  MLDesmond  KABrook  RHKahn  KL Outcomes of patients with do-not-resuscitate orders: toward an understanding of what do-not-resuscitate orders mean and how they affect patients. Arch Intern Med. 1995;1552063- 2068
PubMed Link to Article
Halm  EAFine  MJKapoor  WNSinger  DEMarrie  TJSiu  AL Instability on hospital discharge and the risk of adverse outcomes in patients with pneumonia. Arch Intern Med. 2002;1621278- 1284
PubMed Link to Article

Correspondence

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
Submit a Comment

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 54

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Collections
PubMed Articles