Cranial neurosurgical 30-day readmissions by clinical indication

View More View Less
  • 2 Departments of Surgery and
  • 3 Neurosurgery,
  • 1 Stanford School of Medicine, Stanford, California
Full access

OBJECT

Postsurgical readmissions are common and vary by procedure. They are significant drivers of increased expenditures in the health care system. Reducing readmissions is a national priority that has summoned significant effort and resources. Before the impact of quality improvement efforts can be measured, baseline procedure-related 30-day all-cause readmission rates are needed. The objects of this study were to determine population-level, 30-day, all-cause readmission rates for cranial neurosurgery and identify factors associated with readmission.

METHODS

The authors identified patient discharge records for cranial neurosurgery and their 30-day all-cause readmissions using the Agency for Healthcare Research and Quality (AHRQ) State Inpatient Databases for California, Florida, and New York. Patients were categorized into 4 groups representing procedure indication based on ICD-9-CM diagnosis codes. Logistic regression models were developed to identify patient characteristics associated with readmissions. The main outcome measure was unplanned inpatient admission within 30 days of discharge.

RESULTS

A total of 43,356 patients underwent cranial neurosurgery for neoplasm (44.23%), seizure (2.80%), vascular conditions (26.04%), and trauma (26.93%). Inpatient mortality was highest for vascular admissions (19.30%) and lowest for neoplasm admissions (1.87%; p < 0.001). Thirty-day readmissions were 17.27% for the neoplasm group, 13.89% for the seizure group, 23.89% for the vascular group, and 19.82% for the trauma group (p < 0.001). Significant predictors of 30-day readmission for neoplasm were Medicaid payer (OR 1.33, 95% CI 1.15–1.54) and fluid/electrolyte disorder (OR 1.44, 95% CI 1.29–1.62); for seizure, male sex (OR 1.74, 95% CI 1.17–2.60) and index admission through the emergency department (OR 2.22, 95% CI 1.45–3.43); for vascular, Medicare payer (OR 1.21, 95% CI 1.05–1.39) and renal failure (OR 1.52, 95% CI 1.29–1.80); and for trauma, congestive heart failure (OR 1.44, 95% CI 1.16–1.80) and coagulopathy (OR 1.51, 95% CI 1.25–1.84). Many readmissions had primary diagnoses identified by the AHRQ as potentially preventable.

CONCLUSIONS

The frequency of 30-day readmission rates for patients undergoing cranial neurosurgery varied by diagnosis between 14% and 24%. Important patient characteristics and comorbidities that were associated with an increased readmission risk were identified. Some hospital-level characteristics appeared to be associated with a decreased readmission risk. These baseline readmission rates can be used to inform future efforts in quality improvement and readmission reduction.

ABBREVIATIONSNCHS = National Center for Health Statistics; PPACA = Patient Protection and Affordable Care Act; SID = State Inpatient Database.

OBJECT

Postsurgical readmissions are common and vary by procedure. They are significant drivers of increased expenditures in the health care system. Reducing readmissions is a national priority that has summoned significant effort and resources. Before the impact of quality improvement efforts can be measured, baseline procedure-related 30-day all-cause readmission rates are needed. The objects of this study were to determine population-level, 30-day, all-cause readmission rates for cranial neurosurgery and identify factors associated with readmission.

METHODS

The authors identified patient discharge records for cranial neurosurgery and their 30-day all-cause readmissions using the Agency for Healthcare Research and Quality (AHRQ) State Inpatient Databases for California, Florida, and New York. Patients were categorized into 4 groups representing procedure indication based on ICD-9-CM diagnosis codes. Logistic regression models were developed to identify patient characteristics associated with readmissions. The main outcome measure was unplanned inpatient admission within 30 days of discharge.

RESULTS

A total of 43,356 patients underwent cranial neurosurgery for neoplasm (44.23%), seizure (2.80%), vascular conditions (26.04%), and trauma (26.93%). Inpatient mortality was highest for vascular admissions (19.30%) and lowest for neoplasm admissions (1.87%; p < 0.001). Thirty-day readmissions were 17.27% for the neoplasm group, 13.89% for the seizure group, 23.89% for the vascular group, and 19.82% for the trauma group (p < 0.001). Significant predictors of 30-day readmission for neoplasm were Medicaid payer (OR 1.33, 95% CI 1.15–1.54) and fluid/electrolyte disorder (OR 1.44, 95% CI 1.29–1.62); for seizure, male sex (OR 1.74, 95% CI 1.17–2.60) and index admission through the emergency department (OR 2.22, 95% CI 1.45–3.43); for vascular, Medicare payer (OR 1.21, 95% CI 1.05–1.39) and renal failure (OR 1.52, 95% CI 1.29–1.80); and for trauma, congestive heart failure (OR 1.44, 95% CI 1.16–1.80) and coagulopathy (OR 1.51, 95% CI 1.25–1.84). Many readmissions had primary diagnoses identified by the AHRQ as potentially preventable.

CONCLUSIONS

The frequency of 30-day readmission rates for patients undergoing cranial neurosurgery varied by diagnosis between 14% and 24%. Important patient characteristics and comorbidities that were associated with an increased readmission risk were identified. Some hospital-level characteristics appeared to be associated with a decreased readmission risk. These baseline readmission rates can be used to inform future efforts in quality improvement and readmission reduction.

ABBREVIATIONSNCHS = National Center for Health Statistics; PPACA = Patient Protection and Affordable Care Act; SID = State Inpatient Database.

Postsurgical readmissions are common and are significant drivers of increased expenditures in the health care system.3,4,9,11,23 Early readmissions, particularly those within 30 days of discharge, have been identified as a quality indicator for hospitals.13,14 The reduction of 30-day all-cause readmissions presents an opportunity for quality improvement by bettering overall patient outcomes and reducing aggregate health care costs.17 Such a reduction is challenging since there is wide variation in condition- and surgery-specific readmission rates among hospitals because of several factors, including differences in care delivery and acuteness of disease.28,30

The Patient Protection and Affordable Care Act (PPACA) recommends using quality metrics, such as 30-day readmissions, to drive quality improvement.15 Over the next 3 years, penalties for excessive readmissions will rise from 1% to 3% of Medicare base reimbursements for inpatient services for select admitting diagnoses.5 The Hospital Readmissions Reduction Program, a project of the PPACA, has begun a pilot program to identify excess readmissions for acute myocardial infarction, heart failure, and pneumonia.7,19 Penalties for excessive readmissions may come to include all covered procedures.

While several studies have examined early readmission rates for a range of medical conditions, rates after surgical intervention remain understudied. In neurosurgery, readmissions studies have focused on spinal procedures and have been institution specific, estimating an early readmission rate below 10%.1,2,18,26,29 To our k nowledge, a comprehensive evaluation of neurosurgical readmissions for cranial procedures does not exist. In this observational study, we report rates of 30-day all-cause readmission following cranial neurosurgery and provide information about the relationship between patient and hospital characteristics, and the risk of readmission following cranial neurosurgery.

Methods

Data Source

Patient discharges between 2010 and 2011 were identified in the Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project, State Inpatient Databases (SIDs) for California, Florida, and New York.12 The SIDs contain all discharges for inpatient admissions from nonfederal hospitals in the respective states. These 3 states were chosen because of data availability, state population, and state diversity.

Hospital-level information was linked to the discharge database using the American Hospital Association's 2008 Annual Survey Database (AHA).16 City and county information in the AHA 2008 Annual Survey Database was used to determine the hospital's National Center for Health Statistics' (NCHS') 2006 Urban-Rural Classification Scheme level.21 Neurosurgical residency status was based on an online list of training programs (http://www.aans.org/Young%20Neurosurgeons/Medical%20Students/Residency%20Directory.aspx).

Cohort Definition

Patients 18 years of age or older who had undergone a cranial neurosurgical procedure were identified within the SIDs by using ICD-9-CM procedure codes (Appendix). The patient's earliest admission with a procedure code was considered the index admission. For index admissions, we used ICD-9-CM diagnosis codes to establish 4 cohorts based on procedure indication: neoplasm, seizure, vascular, and trauma. Patients who underwent cranial neurosurgery but did not fall into any of the above groups were excluded from the study. Transfers to other shortterm hospitals were considered continuations of the index visit, and disposition and discharge date were taken from the last record in the transfer sequence. Exclusion criteria included missing linkage variable, missing discharge date, missing hospital data, discharge in the last quarter of 2011, and an age < 18 years, > 90 years, or missing.

Outcome Variables

Thirty-day all-cause readmission was our primary outcome of interest. Readmissions are defined as the first unplanned inpatient admission within 30 days of discharge from the index visit. Readmissions for aftercare, follow-up examination, or administrative purposes (primary ICD-9-CM codes: V5*, V67*, or V68*) were not considered unplanned admissions. Patients who died during the index admission or during aftercare visits were not eligible for readmission. Patient race was categorized as white, Hispanic, black, and “missing or other.” Primary expected payer was categorized as private, Medicare, Medicaid, or other. Comorbidities were identified for each patient at their index visit using the Elixhauser comorbidity index.8 Nurse/patient ratio was calculated using standardized methods.27 Hospital volume was defined by totaling surgical volume for each diagnosis in each hospital in each year studied and by determining tertile groups. Patient living area was defined by the NCHS as rural (micropolitan, small, and medium counties) or urban (large central or fringe metropolitan counties).

Statistical Analysis

Fisher's exact test was used to identify patient- and hospital-level characteristics associated with 30-day readmission rates in each diagnosis group. Characteristics with p < 0.1 in the univariate analysis were included in logistic regression models developed for readmission status for each diagnosis group. Statistical analyses were performed using Stata version 11 SE (StataCorp LP). This study was determined to be exempt from institutional review board approval. A p value ≤ 0.05 was significant.

Results

Table 1 presents patient demographics. We identified 43,356 patients with a cranial neurosurgery procedure: 19,178 (44.23%) for neoplasm, 1213 (2.80%) for seizure, 11,289 (26.04%) for vascular conditions, and 11,676 (26.93%) for trauma. Variation was seen in patient sex, age, primary expected payer, and race across diagnosis groups. Patients presented with more comorbidities for vascular (mean 2.95) and traumatic (mean 2.58) indications than those presenting with neoplastic (mean 1.88) or seizure indications (mean 1.58). With some variation between diagnosis groups, California represented 43% of our cohort, Florida 26%, and New York 31%. Patients with vascular or trauma indications were more likely to be admitted as inpatients through the emergency department on their index visit (72.10% and 76.82%, respectively) than those with neoplastic or seizure indications (33.31% and 23.99%, respectively).

TABLE 1

Demographic information for cranial neurosurgical patients by diagnosis group for index admission, 2010–2011

VariableDiagnosis Group
NeoplasmSeizureVascularTrauma
No. of patients (%)19,178 (44.23)1213 (2.80)11,289 (26.04)11,676 (26.93)
Age group in yrs (% of patients)
 18–3913.5049.308.8120.50
 40–6453.3541.0641.1830.77
 65–7420.806.6821.6816.21
 75–9012.352.9728.3332.51
Female (% of patients)54.3847.1941.6233.67
Race (% of patients)
 White73.7666.3758.1966.61
 Black7.668.9615.6510.12
 Hispanic12.9820.7617.4517.21
 Other/missing5.603.908.726.25
Payer (% of patients)
 Private48.0342.4627.4725.64
 Medicare33.8027.6249.3747.79
 Medicaid12.4422.9215.3413.43
 Other5.737.017.8313.15
Mean Elixhauser Comorbidity Score (median)1.88 (2)1.58 (1)2.95 (3)2.58 (2)
State (% of patients)43.0142.8742.9144.37
 California25.8123.4129.1030.72
 Florida31.1833.7227.9924.91
 New York
Admission through ED (% of patients)33.3123.9972.1076.82

ED = emergency department.

Table 2 presents hospital characteristics. Across diagnosis groups, most patients presented to hospitals with more than 200 beds (91.24%) in mostly urban settings (80.60% NCHS Groups 1 or 2). Patients with neoplastic or seizure indications were more likely to present to hospitals with a neurosurgical residency program (47.22% and 60.02%, respectively) than patients with vascular or trauma indications (33.03% and 27.29%, respectively).

TABLE 2

Hospital-level information for cranial neurosurgical patients by diagnosis group for index admission, 2010–2011

VariableDiagnosis Group
NeoplasmSeizureVascularTrauma
No. of patients19,178121311,28911,676
Bed size group (% of patients)
 6–199 beds10.046.187.777.88
 200–499 beds41.4227.4547.2050.13
 ≥500 beds48.5466.3645.0341.99
Mean medical-surgical ICU beds43.9243.3140.9936.33
Mean nurse/patient ratio4.655.054.494.45
NCHS urban/rural classification (% of patients)
 Large central or fringe metropolitan counties (NCHS 1 & 2)83.7885.8279.3576.06
 Medium, small, & micropolitan counties (NCHS 3–6)16.2214.1820.6523.94
Certified trauma center (% of patients)53.2069.0256.7765.67
Regional resource trauma center (% of patients)35.0350.9534.9834.74
Hospital associated w/ neurosurgical residency (% of patients)47.2260.0233.0327.29
Neurology service (% of patients)95.5997.1993.7292.15
Hospital vol tertile (% caseload)
 Low vol4.1110.726.945.50
 Mid vol15.3512.0420.9019.01
 High vol80.5577.2572.1775.49

ICU = intensive care unit.

Table 3 reports on patient outcomes. Inpatient mortality was highest for vascular admissions and lowest for neoplasm admissions (19.30% vs 1.87%, respectively, p < 0.001). Thirty-day all-cause readmissions were 17.27% for neoplasm, 13.89% for seizure, 23.89% for vascular, and 19.82% for trauma (p < 0.001). In all diagnosis groups, patients returned for readmission to their index hospital less than 43% of the time. Vascular and trauma patients had longer hospital stays (13.9 and 13.0 days, respectively) and greater total charges ($199,317.40 and $188,130.30, respectively) than neoplasm patients (8.0 days and $131,136.40; p < 0.001). Vascular and trauma patients were more likely to readmit earlier (9.4 and 9.5 days, respectively) than neoplasm patients (12.18 days; p < 0.001). We investigated this issue further, finding that for vascular and trauma cases, many readmissions occurred within the first 5 days; for neoplasm cases, there was a less steep decline over time (Fig. 1).

FIG. 1.
FIG. 1.

Histogram showing the frequency of readmission by days since discharge for each of the 4 diagnosis groups. Trends demonstrate that trauma and vascular cases that readmit tend to do so earlier than neoplasm or epilepsy cases that readmit. Figure is available in color online only.

TABLE 3

Characteristics of outcomes for cranial neurosurgical patients by diagnosis group for index admission, 2010–2011

CharacteristicDiagnosis Groupp Value
NeoplasmSeizureVascularTrauma
No. of patients19,178121311,28911,676
In-hospital mortality (% of patients)1.872.0719.3013.70<0.001
30-Day all-cause readmission (% of patients)17.2713.8923.8919.82<0.001
Readmission to index hospital (% of patients)42.9236.9731.0131.97<0.001
Mean length of stay in days (SD)8.02 (10.15)9.80 (11.55)13.92 (17.20)12.95 (16.31)<0.001
Mean total charges in US$ (SD)131,136 (131,132)160,226 (174,187)199,317 (219,352)188,130 (222,722)<0.001
Mean days to first revisit (SD)12.18 (8.91)10.63 (8.96)9.36 (9.06)9.48 (8.91)
Mean revisit length of stay in days (SD)7.71 (10.35)8.64 (9.51)12.51 (20.19)11.15 (17.89)<0.001
Mean revisit total charges (SD)68,663 (93,122)78,570 (114,279)91,321 (146,807)79,674 (111,888)<0.001

SD = standard deviation; US = United States.

Performed ANOVA with Bonferroni correction. Neoplasm versus vascular (p < 0.001) and trauma (p < 0.001) significant; all other associations not significant.

For neoplastic conditions, the probability of readmission was increased for blacks (OR 1.21, 95% CI 1.03–1.42) and Hispanics (OR 1.16, 95% CI 1.01–1.32) compared with whites and was increased for Medicare payer (OR 1.24, 95% CI 1.09–1.40) and Medicaid payer (OR 1.33, 95% CI 1.15–1.54) compared with private payer (Table 4). Patient sex influenced readmissions, increasing rates for males (OR 1.19, 95% CI 1.09–1.31). Rates were also higher for patients with congestive heart failure (OR 1.40, 95% CI 1.04–1.88), paralysis (OR 1.27, 95% CI 1.11–1.46), diabetes (OR 1.15, 95% CI 1.01–1.30), metastases (OR 1.37, 95% CI 1.21–1.56), and electrolyte disorder (OR 1.44, 95% CI 1.29–1.62). Patients were at an increased risk of readmission if their index admission for the procedure came through the emergency department (OR 1.22, 95% CI 1.10–1.35). Hospital characteristics were not associated with increased or decreased rates of readmission.

TABLE 4

Predictors of 30-day all-cause readmissions for cranial neurosurgical patients by diagnosis group for index admission, 2010–2011*

CharacteristicNeoplasm (OR[95% CI])Seizure (OR[95% CI])Vascular (OR[95% CI])Trauma (OR[95% CI])
Patient
Age1.00 (1.00–1.01)1.02 (1.01–1.04)1.00 (1.00–1.01)1.01 (1.00–1.01)
Sex
 Female1 [Reference]1 [Reference]1 [Reference]1 [Reference]
 Male1.19 (1.09–1.31)1.74 (1.17–2.60)1.10 (1.00–1.22)1.12 (0.98–1.27)
Race
 White1 [Reference]1 [Reference]1 [Reference]1 [Reference]
 Black1.21 (1.03–1.42)0.91 (0.45–1.83)1.18 (1.02–1.36)1.05 (0.90–1.29)
 Hispanic1.16 (1.01–1.32)1.14 (0.67–1.96)0.98 (0.85–1.13)0.91 (0.76–1.08)
 Other/missing1.03 (0.83–1.26)0.32 (0.07–1.44)1.06 (0.88–1.27)1.01 (0.78–1.31)
Payer
 Private1 [Reference]1 [Reference]1 [Reference]1 [Reference]
 Medicare1.24 (1.09–1.40)0.92 (0.56–1.52)1.21 (1.05–1.39)1.08 (0.90–1.29)
 Medicaid1.33 (1.15–1.54)1.47 (0.87–2.50)0.87 (0.74–1.03)1.01 (0.82–1.23)
 Other1.08 (0.88–1.32)0.70 (0.29–1.69)0.79 (0.63–0.98)0.64 (0.51–0.80)
State
 California1 [Reference]1 [Reference]1 [Reference]1 [Reference]
 Florida0.96 (0.84–1.10)1.40 (0.84–2.34)1.17 (1.00–1.38)0.89 (0.75–1.05)
 New York0.96 (0.84–1.09)0.90 (0.54–1.51)0.88 (0.75–1.04)0.79 (0.66–0.96)
Initial admission from ED1.22 (1.10–1.35)2.22 (1.45–3.43)1.11 (0.99–1.25)1.13 (0.97–1.32)
Patient comorbidities
 Congestive heart failure1.40 (1.04–1.88)0.47 (0.09–2.51)1.55 (1.31–1.83)1.44 (1.16–1.80)
 Hypertension0.94 (0.85–1.04)1.49 (0.92–2.42)1.09 (0.97–1.23)1.17 (1.02–1.34)
 Paralysis1.27 (1.11–1.46)2.10 (1.28–3.46)1.29 (1.16–1.44)1.16 (0.99–1.37)
 Chronic pulmonary disease1.06 (0.93–1.21)1.10 (0.95–1.27)1.24 (1.04–1.46)
 Diabetes1.15 (1.01–1.30)1.05 (0.52–2.13)1.19 (1.06–1.34)1.13 (0.97–1.32)
 Renal failure1.52 (1.29–1.80)
 Liver disease1.15 (0.80–1.63)1.17 (0.87–1.58)1.30 (0.94–1.81)
 Metastatic cancer1.37 (1.21–1.56)
 Coagulopathy1.42 (1.19–1.68)1.51 (1.25–1.84)
 Obesity1.30 (1.09–1.54)
 Fluid & electrolyte disorders1.44 (1.29–1.62)1.34 (0.83–2.16)1.18 (1.07–1.31)1.34 (1.18–1.52)
 Alcohol abuse0.90 (0.75–1.09)
 Depression1.19 (0.97–1.46)
Hospital
Bed size group
 6–199 beds1.01 (0.77–1.34)
 200–499 beds0.96 (0.82–1.13)
 ≥500 beds1 [Reference]
Medical-surgical ICU beds**1.00 (1.00–1.00)1.00 (1.00–1.00)
High nurse/patient ratio0.99 (0.96–1.02)0.88 (0.79–0.98)0.98 (0.94–1.02)
NCHS urban/rural classification
 Large central or fringe metropolitan counties (NCHS 1 & 2)1 [Reference]
 Medium, small, & micropolitan counties (NCHS 3–6)0.83 (0.71–0.96)
Hospital associated w/ neurosurgical residency0.98 (0.87–1.11)0.95 (0.82–1.11)0.97 (0.81–1.17)
Neurology service
Certified trauma center (%)0.99 (0.82–1.19)
Regional resource trauma center (%)1.19 (0.98–1.45)
 Hospital vol tertile
 High vol1 [Reference]1 [Reference]1 [Reference]1 [Reference]
 Medium vol0.94 (0.80–1.10)1.07 (0.59–1.95)1.01 (0.87–1.17)1.01 (0.82–1.23)
 Low vol0.97 (0.75–1.26)0.53 (0.25–1.12)0.92 (0.72–1.17)0.91 (0.65–1.28)

Boldface type indicates statistical significance.

For seizure conditions, readmissions increased with male sex (OR 1.74, 95% CI 1.17–2.60) and age (OR 1.02, 95% CI 1.01–1.04; Table 4). Patients originally admitted through the emergency department were more than twice as likely to have a readmission (OR 2.22, 95% CI 1.45–3.43), as were patients with paralysis (OR 2.10, 95% CI 1.28–3.46). A high nurse/patient ratio was associated with a lower rate of readmission (OR 0.88, 95% CI 0.79–0.98) for these patients.

For vascular conditions, readmissions increased with male sex (OR 1.10, 95% CI 1.00–1.22), black race compared with white race (OR 1.18, 95% CI 1.02–1.36), and Medicare payer compared with private payer (OR 1.21, 95% CI 1.05–1.39; Table 4). Patients discharged from hospitals in Florida had a relatively increased risk of readmission compared with California (OR 1.17, 95% CI 1.00–1.38). Patients at an increased risk of readmission had comorbidities of congestive heart failure (OR 1.55, 95% CI 1.31–1.83), paralysis (OR 1.29, 95% CI 1.16–1.44), diabetes (OR 1.19, 95% CI 1.06–1.34), renal failure (OR 1.52, 95% CI 1.29–1.80), coagulopathy (OR 1.42, 95% CI 1.19–1.68), obesity (OR 1.30, 95% CI 1.09–1.54), and fluid/electrolyte disorder (OR 1.18, 95% CI 1.07–1.31). Discharges from rural hospitals had a decreased risk of readmission compared with discharges from urban hospitals (OR 0.83, 95% CI 0.71–0.96).

For trauma cases, readmissions increased with age (OR 1.01, 95% CI 1.00–1.01) and comorbidities such as congestive heart failure (OR 1.44, 95% CI 1.16–1.80), hypertension (OR 1.17, 95% CI 1.02–1.34), chronic pulmonary disease (OR 1.24, 95% CI 1.04–1.46), coagulopathy (OR 1.51, 95% CI 1.25–1.84), and fluid/electrolyte disorder (OR 1.34, 95% CI 1.18–1.52). Discharges in New York had a relatively decreased readmission risk compared with discharges in California (OR 0.79, 95% CI 0.66–0.96; Table 4).

Common diagnoses for readmission varied among diagnosis groups, but septicemia, postoperative infection, pulmonary embolism, and urinary tract infection were common diagnoses for readmission across all groups. Patients with vascular and trauma indications had a high rate of readmissions for hemorrhage (Table 5).

TABLE 5

Most common primary diagnoses for cranial neurosurgical patient readmissions by diagnosis group for index admission, 2010–2011

Neoplasm (3250 cases)Seizure (165 cases)Vascular (2180 cases)Trauma (1999 cases)
ICD-9-CM Code & Diagnosis%*ICD-9-CM Code & Diagnosis%*ICD-9-CM Code & Diagnosis%*ICD-9-CM Code & Diagnosis%*
198.3 - secondary neoplasm5.85998.59 - postop infection6.06432.1 - subdural hemorrhage12.20432.1 - subdural hemorrhage7.15
998.59 - postop infection5.54345.90 - epilepsy4.8538.9 - septicemia5.64852.20 - subdural hemor rhage4.60
997.09 - nervous system complications4.49324.0 - intracranial abscess3.64518.81 - acute respiratory failure5.37852.21 - subdural hemor rhage4.25
38.9 - septicemia3.32996.2 - complication of device3.64431 -intracerebral hemorrhage4.3138.9 - septicemia3.90
415.19 - pulmonary embolism3.11345.41 - status epilepticus3.03518.5 - pulmonary insufficiency2.29518.81 - acute respiratory failure3.35
191.1 - frontal lobe neoplasia2.25345.91 - status epilepticus2.42599.0 - urinary tract Infection2.25998.59 - postop infection2.90
486 - pneumonia2.18996.63 - infection from device2.42998.59 - postop infection2.06518.5 - pulmonary insufficiency2.90
997.01 - CNS complication1.69331.4 - obstructive hydrocephalus1.82507.0 - aspiration pneumonia1.93507.0 - aspiration pneumonia2.00
453.41 - deep vein thrombosis1.54415.19 - pulmonary embolism1.82415.19 - pulmonary embolism1.70599.0 - urinary tract infection1.90
331.4 - obstructive hydrocephalus1.54453.41 - deep vein thrombosis1.82430 - subarachnoid hemorrhage1.65780.39 - convulsions1.90

Percent of total readmissions.

Discussion

In our population-based study of cranial neurosurgical procedures, we found 30-day all-cause readmissions ranged from 13.89% in seizure cases to 23.89% in vascular cases. Patient characteristics, such as race, primary expected payer, and patient comorbidities, were often associated with 30-day readmissions, whereas hospital characteristics were generally not associated with readmissions. Among the patients readmitted, postoperative complications were among the top primary diagnoses. Our study offers evidence that neurosurgical readmission rates vary greatly by underlying procedure indication and that many readmissions are the result of surgical complications.

Placing Our Results in Context

Our reported readmission rates for cranial neurosurgical procedures are substantially higher than previously reported estimates. Previous neurosurgical studies have focused mainly on spinal procedures and have reported readmission rates below 10%.18,26,29 However, our approach to capture readmissions was more inclusive and captured patients who were not readmitted to the site of their index visit. Many of the previous studies were site specific and only captured readmissions at the site of the index procedure. Our data set allowed us to identify patients who were readmitted to any hospital within the state of their index procedure; our data showed that less than 43% of patients were readmitted to the site of their index procedure, regardless of diagnosis group. Our approach more accurately captured the real incidence of patient readmission and the accurate costs within a given state system. This improved accuracy allowed more careful identification of factors contributing to higher reported readmission rates.

Patient Characteristics

The most consistent predictive factors associated with 30-day all-cause readmissions in our models were patient comorbidities. In particular, patients with congestive heart failure, paralysis, and electrolyte/fluid disorders were far more likely to have a readmission than were patients without these comorbidities. Index admission to an inpatient bed through the emergency department was also highly predictive of readmission in most cases. Among patients with Medicare or Medicaid insurance, we found an increased readmission risk in neoplasm and vascular cases. While we do not know the mechanism for this increase, it is possible that patients with private insurance have better access to other health care facilities (for example, primary care facilities and clinics), which may reflect decreased rates of readmission to the inpatient setting. While factors such as patient comorbidities, payer, and route of admission are not modifiable, they may help clinicians to better identify patients at higher risk of readmission and take measures to decrease the likelihood of an unnecessary readmission. We identified only one modifiable factor that may decrease the risk of readmission: a high nurse/patient ratio for seizure patients. Further studies on these variables are needed to better understand their association with inpatient readmission.

Our study identified patient factors (for example, white race and private insurance payer) that were consistently associated with a decreased probability of readmission. Medicaid payer, in particular, was associated with an increased risk of readmission in some cases, lending credence to those who suggest that low socioeconomic status correlates with higher health care spending6 and those who are currently investigating that question.22 Factors, such as hospital size, surgical volume, and location, were all investigated, with no significant associations found. Furthermore, patients discharged from hospitals associated with neurosurgical residencies were not found to have a higher or lower risk of 30-day readmission; patients undergoing procedures at trauma centers similarly did not see an increased or decreased risk of 30-day readmission.

Modifiable Factors

More evidence is required to identify the factors that may lead to improved outcomes, particularly modifiable factors. Some of our data suggest that many readmissions identified in our study are potentially preventable complications: postoperative infections, urinary tract infections, pneumonia, and pulmonary emboli. These data support previously published literature on patient safety event and risk of readmission and extend this research to include site-specific procedures.20,25 Further investigation into adverse events associated with readmissions may reveal the extent of these “preventable” readmissions, which may advance progress in quality improvement.

Study Limitations

Our study has several limitations, for example, the use of administrative data to track quality improvement. As we used all-capture state data sets, we are confident that these sources provided accurate numbers on hospital readmissions within a state. A frequent limitation of administrative database studies is nuance in the coding of similar diagnoses and procedures.10 We believe this is a minor issue in our work, as the diagnosis groups we established based on primary diagnosis are fairly distinct and thus not prone to that sort of error. However, the clinical picture of each patient in our data set is limited to diagnosis and procedure codes, and the nuances of each patient's experience cannot be captured in such a study, as administrative database studies may inadequately capture secondary diagnoses.24 Our study may also be understating the risk of readmission for neurosurgical procedures, as the data do not capture patients who die outside the hospital after their procedure or seek additional treatment in another state.

Conclusions

In summary, our study is the most comprehensive look to date at 30-day all-cause readmissions for cranial neurosurgery in hospitals for both old and young adults. Our results showed that readmission rates vary greatly by procedure indication. Factors associated with an increased readmission risk included patient characteristics such as race and comorbidities; however, the characteristics most associated with an increased risk of readmission are not modifiable by hospitals or physicians. These baseline readmission rates are the beginning of standardized measures and will inform quality improvement programs in the future.

Author Contributions

Conception and design: Hernandez-Boussard. Acquisition of data: Hernandez-Boussard. Analysis and interpretation of data: all authors. Drafting the article: Moghavem. Critically revising the article: Hernandez-Boussard, Ratliff. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Hernandez-Boussard. Statistical analysis: Hernandez-Boussard, Moghavem, Morrison. Study supervision: Hernandez-Boussard.

Supplemental Information

Previous Presentation

A portion of this work was presented at the 2014 American Association of Neurological Surgeons Annual Meeting held in San Francisco, California, on April 5–9.

APPENDIX TABLE 1
Procedures
01 Incision and excision of skull, brain, and cerebral meninges
01.0 Cranial puncture
01.1 Diagnostic procedures on skull, brain, and cerebral meninges
01.2 Craniotomy and craniectomy
01.3 Incision of brain and cerebral meninges
01.4 Operations on thalamus and globus pallidus
01.5 Other excision or destruction of brain and meninges
01.6 Excision of lesion of skull
02 Other operations on skull, brain, and cerebral meninges
02.0 Cranioplasty
02.1 Repair of cerebral meninges
02.2 Ventriculostomy
02.9 Other operations on skull, brain, and cerebral meninges
04.01 Removal of acoustic neuroma
APPENDIX TABLE 2
Diagnosis Groups
Neoplasm160.0 - Malignant neoplasm of nasal cavities
170.0 - Malignant neoplasm of skull/face bones
191* - Malignant neoplasm of brain
192.1 - Malignant neoplasm of cerebral meninges
194* - Malignant neoplasm of other endocrine glands
198* - Secondary malignant neoplasm of other and unspecified sites
200.20 - Burkitt's lymphoma
200.50 - Primary nervous system lymphoma
202.80 - Other malignant lymphomas, unspecified site, extranodal/solid organ
202.81 - Other malignant lymphomas, lymph nodes of head, face, and neck
209.79 - Secondary neuroendocrine tumor
213.0 - Benign neoplasm of skull/face bone
215.0 - Benign neoplasm of soft tissue - head
225* - Benign neoplasm of brain and other parts of nervous system
227* - Benign neoplasm of other endocrine glands
237* - Neoplasm of uncertain behavior of endocrine glands and nervous system
239.6 - Brain neoplasm of unspecified nature
253* - Disorders of pituitary gland and its hypothalamic control
255* - Disorders of adrenal gland
Seizure345 - Epilepsy
Vascular349.2 - Disorder of cerebral meninges
430* - Subarachnoid hemorrhage
431* - Intracerebral hemorrhage
432* - Other intracranial hemorrhage
433* - Occlusion and stenosis of pre-cerebral arteries
434* - Occlusion of cerebral arteries
437* - Other cerebrovascular disease
438.20 - Hemiplegia/hemiparesis, late effect of cerebrovascular disease
438.89 - Other late effects of cerebrovascular disease
747.81 - Anomaly of cerebrovascular system
996.1 - Malfunctioning vascular device/graft
Trauma800* - Fracture of skull
801* - Fracture of skull base
802* - Fracture of facial bones
803* - Other skull fractures
851* - Cerebral laceration and contusion
852* - Subarachnoid, subdural, and extradural hemorrhage, following injury
853* - Other and unspecified intracranial hemorrhage, following injury
853* - Other and unspecified intracranial hemorrhage, following

References

  • 1

    Amin BY, , Tu TH, , Schairer WW, , Na L, , Takemoto S, & Berven S, : Pitfalls of calculating hospital readmission rates based on nonvalidated administrative data sets. Clinical article. J Neurosurg Spine 18:134138, 2013

    • Search Google Scholar
    • Export Citation
  • 2

    Angevine PD, & McCormick PC: Editorial. Readmissions. J Neurosurg Spine 18:132133, 2013

  • 3

    Burke RE, & Coleman EA: Interventions to decrease hospital readmissions: keys for cost-effectiveness. JAMA Intern Med 173:695698, 2013

  • 4

    Casey K, , Hernandez-Boussard T, , Mell MW, & Lee JT: Differences in readmissions after open repair versus endovascular aneurysm repair. J Vasc Surg 57:8995, 2013

    • Search Google Scholar
    • Export Citation
  • 5

    Centers for Medicare & Medicaid Services: Readmissions Reduction Program (http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/Readmissions-Reduction-Program.html) [Accessed December 2, 2014]

    • Search Google Scholar
    • Export Citation
  • 6

    Cooper RB: Inequality is at the Core of High Health Care Spending: A View From the OECD. Health Affairs Blog (http://healthaffairs.org/blog/2013/10/09/inequality-is-at-thecore-of-high-health-care-spending-a-view-from-the-oecd/) [Accessed December 2, 2014]

    • Search Google Scholar
    • Export Citation
  • 7

    Dharmarajan K, , Hsieh AF, , Lin Z, , Bueno H, , Ross JS, & Horwitz LI, : Diagnoses and timing of 30-day readmissions after hospitalization for heart failure, acute myocardial infarction, or pneumonia. JAMA 309:355363, 2013

    • Search Google Scholar
    • Export Citation
  • 8

    Elixhauser A, , Steiner C, , Harris DR, & Coffey RM: Comorbidity measures for use with administrative data. Med Care 36:827, 1998

  • 9

    Flood KL, , Maclennan PA, , McGrew D, , Green D, , Dodd C, & Brown CJ: Effects of an acute care for elders unit on costs and 30-day readmissions. JAMA Intern Med 173:981987, 2013

    • Search Google Scholar
    • Export Citation
  • 10

    Gologorsky Y, , Knightly JJ, , Chi JH, & Groff MW: The Nationwide Inpatient Sample database does not accurately reflect surgical indication for fusion. J Neurosurg Spine 21:984993, 2014

    • Search Google Scholar
    • Export Citation
  • 11

    Hannan EL, , Zhong Y, , Krumholz H, , Walford G, , Holmes DR Jr, & Stamato NJ, : 30-day readmission for patients undergoing percutaneous coronary interventions in New York state. JACC Cardiovasc Interv 4:13351342, 2011

    • Search Google Scholar
    • Export Citation
  • 12

    Healthcare Cost and Utilization Project: Overview of the State Inpatient Databases (SID) (www.hcup-us.ahrq.gov/sidoverview.jsp) [Accessed December 2, 2014]

    • Search Google Scholar
    • Export Citation
  • 13

    Hernandez AF, , Greiner MA, , Fonarow GC, , Hammill BG, , Heidenreich PA, & Yancy CW, : Relationship between early physician follow-up and 30-day readmission among Medicare beneficiaries hospitalized for heart failure. JAMA 303:17161722, 2010

    • Search Google Scholar
    • Export Citation
  • 14

    Jencks SF, , Williams MV, & Coleman EA: Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med 360:14181428, 2009

    • Search Google Scholar
    • Export Citation
  • 15

    Kocher RP, & Adashi EY: Hospital readmissions and the Affordable Care Act: paying for coordinated quality care. JAMA 306:17941795, 2011

    • Search Google Scholar
    • Export Citation
  • 16

    Kralovec PD, & Mullner R: The American Hospital Association's Annual Survey of Hospitals: continuity and change. Health Serv Res 16:351355, 1981

    • Search Google Scholar
    • Export Citation
  • 17

    Lawson EH, , Hall BL, , Louie R, , Ettner SL, , Zingmond DS, & Han L, : Association between occurrence of a postoperative complication and readmission: implications for quality improvement and cost savings. Ann Surg 258:1018, 2013

    • Search Google Scholar
    • Export Citation
  • 18

    McCormack RA, , Hunter T, , Ramos N, , Michels R, , Hutzler L, & Bosco JA: An analysis of causes of readmission after spine surgery. Spine (Phila Pa 1976) 37:12601266, 2012

    • Search Google Scholar
    • Export Citation
  • 19

    McHugh MD, & Ma C: Hospital nursing and 30-day readmissions among Medicare patients with heart failure, acute myocardial infarction, and pneumonia. Med Care 51:5259, 2013

    • Search Google Scholar
    • Export Citation
  • 20

    Mull HJ, , Borzecki AM, , Chen Q, , Shin MH, & Rosen AK: Using AHRQ patient safety indicators to detect postdischarge adverse events in the veterans health administration. Am J Med Qual 29:213219, 2014

    • Search Google Scholar
    • Export Citation
  • 21

    National Center for Health Statistics: 2006 Urban-Rural Classification Scheme for Counties. Centers for Disease Control and Prevention (http://www.cdc.gov/nchs/data_access/urban_rural.htm#counties2006) [Accessed December 2, 2014]

    • Search Google Scholar
    • Export Citation
  • 22

    National Quality Forum: Risk Adjustment for Socioeconomic Status of Other Sociodemographic Factors. Technical Report Washington DC, National Quality Forum, 2014

    • Search Google Scholar
    • Export Citation
  • 23

    Podulka J, , Barrett M, , Jiang HJ, & Steiner C: 30-Day Readmissions following hospitalizations for chronic vs. acute conditions, 2008. HCUP Statistical Brief #127. Healthcare Cost and Utilization Project (http://www.hcup-us.ahrq.gov/reports/statbriefs/sb127.pdf [Accessed December 3, 2014]

    • Search Google Scholar
    • Export Citation
  • 24

    Romano PS, & Mark DH: Bias in the coding of hospital discharge data and its implications for quality assessment. Med Care 32:8190, 1994

    • Search Google Scholar
    • Export Citation
  • 25

    Rosen AK, , Loveland S, , Shin M, , Shwartz M, , Hanchate A, & Chen Q, : Examining the impact of the AHRQ Patient Safety Indicators (PSIs) on the Veterans Health Administration: the case of readmissions. Med Care 51:3744, 2013

    • Search Google Scholar
    • Export Citation
  • 26

    Shah MN, , Stoev IT, , Sanford DE, , Gao F, , Santiago P, & Jaques DP, : Are readmission rates on a neurosurgical service indicators of quality of care?. J Neurosurg 119:10431049, 2013

    • Search Google Scholar
    • Export Citation
  • 27

    Spetz J, , Donaldson N, , Aydin C, & Brown DS: How many nurses per patient? Measurements of nurse staffing in health services research. Health Serv Res 43:16741692, 2008

    • Search Google Scholar
    • Export Citation
  • 28

    Tsai TC, , Joynt KE, , Orav EJ, , Gawande AA, & Jha AK: Variation in surgical-readmission rates and quality of hospital care. N Engl J Med 369:11341142, 2013

    • Search Google Scholar
    • Export Citation
  • 29

    Wang MC, , Shivakoti M, , Sparapani RA, , Guo C, , Laud PW, & Nattinger AB: Thirty-day readmissions after elective spine surgery for degenerative conditions among US Medicare beneficiaries. Spine J 12:902911, 2012

    • Search Google Scholar
    • Export Citation
  • 30

    Yeh RW, , Rosenfield K, , Zelevinsky K, , Mauri L, , Sakhuja R, & Shivapour DM, : Sources of hospital variation in shortterm readmission rates after percutaneous coronary intervention. Circ Cardiovasc Interv 5:227236, 2012

    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

Contributor Notes

Correspondence Tina Hernandez-Boussard, Stanford School of Medicine, Department of Surgery, 1070 Arastradero #307, Palo Alto, CA 94305. email: boussard@stanford.edu.

INCLUDE WHEN CITING Published online February 6, 2015; DOI: 10.3171/2014.12.JNS14447.

DISCLOSURE The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Mr. Moghavem was supported by the Stanford School of Medicine Medical Scholars Fund. Dr. Hernandez-Boussard was supported by Grant No. K01 HS018558 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.

  • View in gallery

    Histogram showing the frequency of readmission by days since discharge for each of the 4 diagnosis groups. Trends demonstrate that trauma and vascular cases that readmit tend to do so earlier than neoplasm or epilepsy cases that readmit. Figure is available in color online only.

  • 1

    Amin BY, , Tu TH, , Schairer WW, , Na L, , Takemoto S, & Berven S, : Pitfalls of calculating hospital readmission rates based on nonvalidated administrative data sets. Clinical article. J Neurosurg Spine 18:134138, 2013

    • Search Google Scholar
    • Export Citation
  • 2

    Angevine PD, & McCormick PC: Editorial. Readmissions. J Neurosurg Spine 18:132133, 2013

  • 3

    Burke RE, & Coleman EA: Interventions to decrease hospital readmissions: keys for cost-effectiveness. JAMA Intern Med 173:695698, 2013

  • 4

    Casey K, , Hernandez-Boussard T, , Mell MW, & Lee JT: Differences in readmissions after open repair versus endovascular aneurysm repair. J Vasc Surg 57:8995, 2013

    • Search Google Scholar
    • Export Citation
  • 5

    Centers for Medicare & Medicaid Services: Readmissions Reduction Program (http://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/Readmissions-Reduction-Program.html) [Accessed December 2, 2014]

    • Search Google Scholar
    • Export Citation
  • 6

    Cooper RB: Inequality is at the Core of High Health Care Spending: A View From the OECD. Health Affairs Blog (http://healthaffairs.org/blog/2013/10/09/inequality-is-at-thecore-of-high-health-care-spending-a-view-from-the-oecd/) [Accessed December 2, 2014]

    • Search Google Scholar
    • Export Citation
  • 7

    Dharmarajan K, , Hsieh AF, , Lin Z, , Bueno H, , Ross JS, & Horwitz LI, : Diagnoses and timing of 30-day readmissions after hospitalization for heart failure, acute myocardial infarction, or pneumonia. JAMA 309:355363, 2013

    • Search Google Scholar
    • Export Citation
  • 8

    Elixhauser A, , Steiner C, , Harris DR, & Coffey RM: Comorbidity measures for use with administrative data. Med Care 36:827, 1998

  • 9

    Flood KL, , Maclennan PA, , McGrew D, , Green D, , Dodd C, & Brown CJ: Effects of an acute care for elders unit on costs and 30-day readmissions. JAMA Intern Med 173:981987, 2013

    • Search Google Scholar
    • Export Citation
  • 10

    Gologorsky Y, , Knightly JJ, , Chi JH, & Groff MW: The Nationwide Inpatient Sample database does not accurately reflect surgical indication for fusion. J Neurosurg Spine 21:984993, 2014

    • Search Google Scholar
    • Export Citation
  • 11

    Hannan EL, , Zhong Y, , Krumholz H, , Walford G, , Holmes DR Jr, & Stamato NJ, : 30-day readmission for patients undergoing percutaneous coronary interventions in New York state. JACC Cardiovasc Interv 4:13351342, 2011

    • Search Google Scholar
    • Export Citation
  • 12

    Healthcare Cost and Utilization Project: Overview of the State Inpatient Databases (SID) (www.hcup-us.ahrq.gov/sidoverview.jsp) [Accessed December 2, 2014]

    • Search Google Scholar
    • Export Citation
  • 13

    Hernandez AF, , Greiner MA, , Fonarow GC, , Hammill BG, , Heidenreich PA, & Yancy CW, : Relationship between early physician follow-up and 30-day readmission among Medicare beneficiaries hospitalized for heart failure. JAMA 303:17161722, 2010

    • Search Google Scholar
    • Export Citation
  • 14

    Jencks SF, , Williams MV, & Coleman EA: Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med 360:14181428, 2009

    • Search Google Scholar
    • Export Citation
  • 15

    Kocher RP, & Adashi EY: Hospital readmissions and the Affordable Care Act: paying for coordinated quality care. JAMA 306:17941795, 2011

    • Search Google Scholar
    • Export Citation
  • 16

    Kralovec PD, & Mullner R: The American Hospital Association's Annual Survey of Hospitals: continuity and change. Health Serv Res 16:351355, 1981

    • Search Google Scholar
    • Export Citation
  • 17

    Lawson EH, , Hall BL, , Louie R, , Ettner SL, , Zingmond DS, & Han L, : Association between occurrence of a postoperative complication and readmission: implications for quality improvement and cost savings. Ann Surg 258:1018, 2013

    • Search Google Scholar
    • Export Citation
  • 18

    McCormack RA, , Hunter T, , Ramos N, , Michels R, , Hutzler L, & Bosco JA: An analysis of causes of readmission after spine surgery. Spine (Phila Pa 1976) 37:12601266, 2012

    • Search Google Scholar
    • Export Citation
  • 19

    McHugh MD, & Ma C: Hospital nursing and 30-day readmissions among Medicare patients with heart failure, acute myocardial infarction, and pneumonia. Med Care 51:5259, 2013

    • Search Google Scholar
    • Export Citation
  • 20

    Mull HJ, , Borzecki AM, , Chen Q, , Shin MH, & Rosen AK: Using AHRQ patient safety indicators to detect postdischarge adverse events in the veterans health administration. Am J Med Qual 29:213219, 2014

    • Search Google Scholar
    • Export Citation
  • 21

    National Center for Health Statistics: 2006 Urban-Rural Classification Scheme for Counties. Centers for Disease Control and Prevention (http://www.cdc.gov/nchs/data_access/urban_rural.htm#counties2006) [Accessed December 2, 2014]

    • Search Google Scholar
    • Export Citation
  • 22

    National Quality Forum: Risk Adjustment for Socioeconomic Status of Other Sociodemographic Factors. Technical Report Washington DC, National Quality Forum, 2014

    • Search Google Scholar
    • Export Citation
  • 23

    Podulka J, , Barrett M, , Jiang HJ, & Steiner C: 30-Day Readmissions following hospitalizations for chronic vs. acute conditions, 2008. HCUP Statistical Brief #127. Healthcare Cost and Utilization Project (http://www.hcup-us.ahrq.gov/reports/statbriefs/sb127.pdf [Accessed December 3, 2014]

    • Search Google Scholar
    • Export Citation
  • 24

    Romano PS, & Mark DH: Bias in the coding of hospital discharge data and its implications for quality assessment. Med Care 32:8190, 1994

    • Search Google Scholar
    • Export Citation
  • 25

    Rosen AK, , Loveland S, , Shin M, , Shwartz M, , Hanchate A, & Chen Q, : Examining the impact of the AHRQ Patient Safety Indicators (PSIs) on the Veterans Health Administration: the case of readmissions. Med Care 51:3744, 2013

    • Search Google Scholar
    • Export Citation
  • 26

    Shah MN, , Stoev IT, , Sanford DE, , Gao F, , Santiago P, & Jaques DP, : Are readmission rates on a neurosurgical service indicators of quality of care?. J Neurosurg 119:10431049, 2013

    • Search Google Scholar
    • Export Citation
  • 27

    Spetz J, , Donaldson N, , Aydin C, & Brown DS: How many nurses per patient? Measurements of nurse staffing in health services research. Health Serv Res 43:16741692, 2008

    • Search Google Scholar
    • Export Citation
  • 28

    Tsai TC, , Joynt KE, , Orav EJ, , Gawande AA, & Jha AK: Variation in surgical-readmission rates and quality of hospital care. N Engl J Med 369:11341142, 2013

    • Search Google Scholar
    • Export Citation
  • 29

    Wang MC, , Shivakoti M, , Sparapani RA, , Guo C, , Laud PW, & Nattinger AB: Thirty-day readmissions after elective spine surgery for degenerative conditions among US Medicare beneficiaries. Spine J 12:902911, 2012

    • Search Google Scholar
    • Export Citation
  • 30

    Yeh RW, , Rosenfield K, , Zelevinsky K, , Mauri L, , Sakhuja R, & Shivapour DM, : Sources of hospital variation in shortterm readmission rates after percutaneous coronary intervention. Circ Cardiovasc Interv 5:227236, 2012

    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 826 227 17
PDF Downloads 382 99 12
EPUB Downloads 0 0 0