Downloaded from http://jnis.bmj.com/ on May 28, 2015 - Published by group.bmj.com
Socioeconomics
ORIGINAL RESEARCH
Hospital-based financial analysis of endovascular therapy and intravenous thrombolysis for large vessel acute ischemic strokes: the ‘bottom line’ Ansaar T Rai,1 Kim Evans2 1
Department of Radiology and Neurosurgery and Neurology, West Virginia University Healthcare, Morgantown, West Virginia, USA 2 Decision Support West Virginia University Healthcare, Morgantown, West Virginia, USA Correspondence to Dr Ansaar T Rai, Department of Radiology and Neurosurgery and Neurology, West Virginia University Healthcare, Room 2278, HSCS, PO Box 9235, Morgantown, WV 26506, USA; [email protected] Received 16 December 2013 Revised 9 January 2014 Accepted 11 January 2014 Published Online First 29 January 2014
ABSTRACT Background Economic viability is important to any hospital striving to be a comprehensive stroke center. An inability to recover cost can strain sustained delivery of advanced stroke care. Objective To carry out a comparative financial analysis of intravenous (IV) recombinant tissue plasminogen activator and endovascular (EV) therapy in treating large vessel strokes from a hospital’s perspective. Methodology Actual hospital’s charges, costs, and payments were analyzed for 265 patients who received treatment for large vessel strokes. The patients were divided into an EV (n=141) and an IV group (n=124). The net gain/loss was calculated as the difference between payments received and the total cost. Results The charges, costs, and payments were significantly higher for the EV than the IV group ( p96 h or principal diagnosis except face, mouth & neck without major operation Intracranial hemorrhage or cerebral infarction with CC Intracranial hemorrhage or cerebral infarction without CC/MCC Peripheral and cranial nerve and other nervous system procedure with CC Transient ischemia Extracranial procedures with CC Extensive O.R. procedure unrelated to principal diagnosis with MCC Transient ischemia Total
MS-62 MS-63 MS-24 CMS-559 CMS-543 MS-64 MS-3 MS-4 MS-65 MS-66 CMS-7 CMS-524 CMS-533 MS-981 MS-69
IV Mean (SE) payment, $
N
Mean (SE) payment, $
N
Total
21 27 29
8253 (3453) 23 851 (3453) 43 293 (2939)
36 28 5
7907 (905) 21 800 (1027) 39 264 (2426)
57 55 34
8 6 20
14 471 (5595) 17 968 (6461) 36 393 (3539)
19 14 0
14 645 (1246) 12 011 (1452) –
27 20 20
8 8 5 1
22 553 (5595) 36 745 (5595) 18 937 (7077) 225 360 (15 825)
9 1 0 1
16 216 (1811) 46 659 (5432) – 58 633 (5432)
17 9 5 2
1
86 098 (15 825)
2
87 164 (3841)
3
2 1 1 1 0 3 1 124
4519 (3841) 7765 (5432) 0 (5432) 5420 (5432)
3 3 2 2 2 3 1 265
1 2 1 1 2 0 0 141
5571 19 803 20 048 5276 17 696 – –
(15 825) (11 190) (15 825) (15 825) (11 190)
– 26 608 (3136) 5759 (5432)
The diagnosis related groups are listed in order of frequency with the most commonly used codes listed first. CC, complications or comorbidities; ECMO, extracorporeal membrane oxygenation; MCC, major complications or comorbidities.
Rai AT, et al. J NeuroIntervent Surg 2015;7:150–156. doi:10.1136/neurintsurg-2013-011085
151
Downloaded from http://jnis.bmj.com/ on May 28, 2015 - Published by group.bmj.com
Socioeconomics Figure 1 The payer distribution for endovascular therapy (EV) and intravenous thrombolysis (IV): BC (ALL), Blue Cross, CHAMPUS, Civilian Health And Medical Program of the Uniformed Services; COMM INS, Commercial Insurance—WV; DPW-WV, Department of Public Works—WV; HMO, Health Maintenance Organization; MC A, Medicare Part A; MC HMO, Medicare-HMO; PEIA, Public Employees Insurance Agency—WV; PMA, Pennsylvania Manufacturers’ Association; PVT-PAY, Private Pay.
no relationship between financial gain/loss and baseline NIHSS score (p=0.9), patients’ age (p=0.2), or in-hospital mortality (p=0.4). The clinical outcome was significantly associated with the financial outcome. In patients with a favorable outcome, the hospital had a net financial gain of $3853 (±$21 155) as opposed to a financial loss of $2906 (±$15 088) for those with a poor outcome (p=0.003). The length of stay was associated with the clinical outcome for both groups. In the EV group there was a significantly shorter length of stay of 6.9 (±6) days for patients with a favorable outcome compared with 9.6 (±8) days for those with a poor outcome (p=0.03). Similarly, in the IV group the length of stay was 5 (±2.6) days for patients with a favorable outcome as opposed to 8.6 (±8) days for those with a poor outcome (p=0.008). Patients who were discharged home led to a hospital gain of $3113 (±$14 760) as opposed to a financial loss of $1913 (±$18 298) for patients who were discharged to another destination (p=0.04). Among the stroke risk factors, hypertension and hyperlipidemia had significant correlation with the financial outcome. In hypertensive patients there was a net financial loss of $1665 (±$14 082)
as opposed to a gain of $6079 (±$31 616) in those without hypertension (p=0.03). The presence of hyperlipidemia was associated with a net loss of $3526 (±$15 046) compared with a net gain of $2604 (±$20 443) in patients without hyperlipidemia (p=0.01). There was no significant association with other risk factors such as diabetes, atrial fibrillation, and smoking. The favorable outcome of 42.3% in the EV group was significantly higher than the 29.4% in the IV group (p=0.03). The hospital mortality was 22.7% for the EV and 23.4% for the IV groups ( p=0.9) and the length of stay was 8.5 (±7.2) days for the EV group and 7.4 (±6.8) days for the IV group ( p=0.2). The financial gain/loss was analyzed for the two treatment groups based on in-hospital mortality, clinical outcome, and discharge destination (figure 4). For patients who survived the hospitalization, the hospital had a financial gain for the EV group and a loss for the IV group ( p=0.04). The opposite was true for patients who died during their admission. Patients with a favorable outcome tended to result in a financial gain for the hospital in the EV group but not the IV group. A financial loss was seen in patients with a poor outcome regardless of the treatment
Figure 2 The payments associated with each payer separated for endovascular therapy (EV) and intravenous thrombolysis (IV). The bars represent the mean payments and the vertical lines represent one SE from the mean: BC (ALL), Blue Cross, CHAMPUS, Civilian Health And Medical Program of the Uniformed Services; COMM INS, Commercial Insurance—WV; DPW-WV, Department of Public Works—WV; HMO, Health Maintenance Organization; MC A, Medicare Part A; MC HMO, Medicare-HMO; PEIA, Public Employees Insurance Agency—WV; PMA, Pennsylvania Manufacturers’ Association; PVT-PAY, Private Pay. 152
Rai AT, et al. J NeuroIntervent Surg 2015;7:150–156. doi:10.1136/neurintsurg-2013-011085
BC N Mean (SD)
CHAMPUS N Mean (SD)
COMM INS N Mean (SD)
DPW-WV N Mean (SD)
HMO N Mean (SD)
MC A N Mean (SD)
MC HMO N Mean (SD)
MEDICAID N Mean (SD)
PEIA N Mean (SD)
CMS-14
3 6830 (5958)
0
3 9093 (6866)
4 5021 (458)
2 17 714 (13 445)
39 8202 (831)
2 10 693 (4844)
0
2 4279 (1809)
MS-61
4 34 247 (3758)
0
0
0
36 22 024 (2335)
7 23 366 (680)
5 13 983 (3129)
MS-23
4 67 520 (31 101)
1 26 260 – 0
1 80 553 – 0
0
0
6 42 607 (5315)
2 22 568 (1471)
0
0
0
20 42 034 (15 163) 24 14 806 (1953) 13 11 522 (1897)
2 15 484 (244) 2 12 108 (544)
0
1 10 502 – 1 14 030 – 1 7703 1 5618 – 2 12 481 (3245) 0
MS-62
0
MS-63
2 11 128 (15 738)
0
0
0
0
MS-24
6 63 148 (27 317)
0
1 56 316
0
0
7 27 836 (1301)
2 28 998 (580)
CMS-559
1 30 822 – 2 49 968 (437)
1 13 169 – 0
1 17 581 – 0
0
2 31 606 (7322)
12 16 799 (652)
0
1 14 859 – 0
0
6 30 037 (4706)
0
0
4 19 976 (6028)
1 14 782 – 1 58 633 – 2 86 401 (428)
CMS-543
0
0
0
0
MS-3
0
0
0
0
MS-4
0
0
1 225 360 – 0
0
0
MS-65
0
0
0
0
0
0
0
0
1 6333 – 0
0
0
0
0
0
0
MS-66
CMS-7
CMS-524
1 33 274 – 0
0
0
1 87 624 – 2 4520 (6391) 1 7765 – 1 20 048 – 2 5348 (102)
PVT PAY N Mean (SD)
1 10 553 – 0
1 0 – 1 0
0
0
34 0.02
0
0
0
0
27 0.004 20
Socioeconomics
ORIGINAL RESEARCH
Hospital-based financial analysis of endovascular therapy and intravenous thrombolysis for large vessel acute ischemic strokes: the ‘bottom line’ Ansaar T Rai,1 Kim Evans2 1
Department of Radiology and Neurosurgery and Neurology, West Virginia University Healthcare, Morgantown, West Virginia, USA 2 Decision Support West Virginia University Healthcare, Morgantown, West Virginia, USA Correspondence to Dr Ansaar T Rai, Department of Radiology and Neurosurgery and Neurology, West Virginia University Healthcare, Room 2278, HSCS, PO Box 9235, Morgantown, WV 26506, USA; [email protected] Received 16 December 2013 Revised 9 January 2014 Accepted 11 January 2014 Published Online First 29 January 2014
ABSTRACT Background Economic viability is important to any hospital striving to be a comprehensive stroke center. An inability to recover cost can strain sustained delivery of advanced stroke care. Objective To carry out a comparative financial analysis of intravenous (IV) recombinant tissue plasminogen activator and endovascular (EV) therapy in treating large vessel strokes from a hospital’s perspective. Methodology Actual hospital’s charges, costs, and payments were analyzed for 265 patients who received treatment for large vessel strokes. The patients were divided into an EV (n=141) and an IV group (n=124). The net gain/loss was calculated as the difference between payments received and the total cost. Results The charges, costs, and payments were significantly higher for the EV than the IV group ( p96 h or principal diagnosis except face, mouth & neck without major operation Intracranial hemorrhage or cerebral infarction with CC Intracranial hemorrhage or cerebral infarction without CC/MCC Peripheral and cranial nerve and other nervous system procedure with CC Transient ischemia Extracranial procedures with CC Extensive O.R. procedure unrelated to principal diagnosis with MCC Transient ischemia Total
MS-62 MS-63 MS-24 CMS-559 CMS-543 MS-64 MS-3 MS-4 MS-65 MS-66 CMS-7 CMS-524 CMS-533 MS-981 MS-69
IV Mean (SE) payment, $
N
Mean (SE) payment, $
N
Total
21 27 29
8253 (3453) 23 851 (3453) 43 293 (2939)
36 28 5
7907 (905) 21 800 (1027) 39 264 (2426)
57 55 34
8 6 20
14 471 (5595) 17 968 (6461) 36 393 (3539)
19 14 0
14 645 (1246) 12 011 (1452) –
27 20 20
8 8 5 1
22 553 (5595) 36 745 (5595) 18 937 (7077) 225 360 (15 825)
9 1 0 1
16 216 (1811) 46 659 (5432) – 58 633 (5432)
17 9 5 2
1
86 098 (15 825)
2
87 164 (3841)
3
2 1 1 1 0 3 1 124
4519 (3841) 7765 (5432) 0 (5432) 5420 (5432)
3 3 2 2 2 3 1 265
1 2 1 1 2 0 0 141
5571 19 803 20 048 5276 17 696 – –
(15 825) (11 190) (15 825) (15 825) (11 190)
– 26 608 (3136) 5759 (5432)
The diagnosis related groups are listed in order of frequency with the most commonly used codes listed first. CC, complications or comorbidities; ECMO, extracorporeal membrane oxygenation; MCC, major complications or comorbidities.
Rai AT, et al. J NeuroIntervent Surg 2015;7:150–156. doi:10.1136/neurintsurg-2013-011085
151
Downloaded from http://jnis.bmj.com/ on May 28, 2015 - Published by group.bmj.com
Socioeconomics Figure 1 The payer distribution for endovascular therapy (EV) and intravenous thrombolysis (IV): BC (ALL), Blue Cross, CHAMPUS, Civilian Health And Medical Program of the Uniformed Services; COMM INS, Commercial Insurance—WV; DPW-WV, Department of Public Works—WV; HMO, Health Maintenance Organization; MC A, Medicare Part A; MC HMO, Medicare-HMO; PEIA, Public Employees Insurance Agency—WV; PMA, Pennsylvania Manufacturers’ Association; PVT-PAY, Private Pay.
no relationship between financial gain/loss and baseline NIHSS score (p=0.9), patients’ age (p=0.2), or in-hospital mortality (p=0.4). The clinical outcome was significantly associated with the financial outcome. In patients with a favorable outcome, the hospital had a net financial gain of $3853 (±$21 155) as opposed to a financial loss of $2906 (±$15 088) for those with a poor outcome (p=0.003). The length of stay was associated with the clinical outcome for both groups. In the EV group there was a significantly shorter length of stay of 6.9 (±6) days for patients with a favorable outcome compared with 9.6 (±8) days for those with a poor outcome (p=0.03). Similarly, in the IV group the length of stay was 5 (±2.6) days for patients with a favorable outcome as opposed to 8.6 (±8) days for those with a poor outcome (p=0.008). Patients who were discharged home led to a hospital gain of $3113 (±$14 760) as opposed to a financial loss of $1913 (±$18 298) for patients who were discharged to another destination (p=0.04). Among the stroke risk factors, hypertension and hyperlipidemia had significant correlation with the financial outcome. In hypertensive patients there was a net financial loss of $1665 (±$14 082)
as opposed to a gain of $6079 (±$31 616) in those without hypertension (p=0.03). The presence of hyperlipidemia was associated with a net loss of $3526 (±$15 046) compared with a net gain of $2604 (±$20 443) in patients without hyperlipidemia (p=0.01). There was no significant association with other risk factors such as diabetes, atrial fibrillation, and smoking. The favorable outcome of 42.3% in the EV group was significantly higher than the 29.4% in the IV group (p=0.03). The hospital mortality was 22.7% for the EV and 23.4% for the IV groups ( p=0.9) and the length of stay was 8.5 (±7.2) days for the EV group and 7.4 (±6.8) days for the IV group ( p=0.2). The financial gain/loss was analyzed for the two treatment groups based on in-hospital mortality, clinical outcome, and discharge destination (figure 4). For patients who survived the hospitalization, the hospital had a financial gain for the EV group and a loss for the IV group ( p=0.04). The opposite was true for patients who died during their admission. Patients with a favorable outcome tended to result in a financial gain for the hospital in the EV group but not the IV group. A financial loss was seen in patients with a poor outcome regardless of the treatment
Figure 2 The payments associated with each payer separated for endovascular therapy (EV) and intravenous thrombolysis (IV). The bars represent the mean payments and the vertical lines represent one SE from the mean: BC (ALL), Blue Cross, CHAMPUS, Civilian Health And Medical Program of the Uniformed Services; COMM INS, Commercial Insurance—WV; DPW-WV, Department of Public Works—WV; HMO, Health Maintenance Organization; MC A, Medicare Part A; MC HMO, Medicare-HMO; PEIA, Public Employees Insurance Agency—WV; PMA, Pennsylvania Manufacturers’ Association; PVT-PAY, Private Pay. 152
Rai AT, et al. J NeuroIntervent Surg 2015;7:150–156. doi:10.1136/neurintsurg-2013-011085
BC N Mean (SD)
CHAMPUS N Mean (SD)
COMM INS N Mean (SD)
DPW-WV N Mean (SD)
HMO N Mean (SD)
MC A N Mean (SD)
MC HMO N Mean (SD)
MEDICAID N Mean (SD)
PEIA N Mean (SD)
CMS-14
3 6830 (5958)
0
3 9093 (6866)
4 5021 (458)
2 17 714 (13 445)
39 8202 (831)
2 10 693 (4844)
0
2 4279 (1809)
MS-61
4 34 247 (3758)
0
0
0
36 22 024 (2335)
7 23 366 (680)
5 13 983 (3129)
MS-23
4 67 520 (31 101)
1 26 260 – 0
1 80 553 – 0
0
0
6 42 607 (5315)
2 22 568 (1471)
0
0
0
20 42 034 (15 163) 24 14 806 (1953) 13 11 522 (1897)
2 15 484 (244) 2 12 108 (544)
0
1 10 502 – 1 14 030 – 1 7703 1 5618 – 2 12 481 (3245) 0
MS-62
0
MS-63
2 11 128 (15 738)
0
0
0
0
MS-24
6 63 148 (27 317)
0
1 56 316
0
0
7 27 836 (1301)
2 28 998 (580)
CMS-559
1 30 822 – 2 49 968 (437)
1 13 169 – 0
1 17 581 – 0
0
2 31 606 (7322)
12 16 799 (652)
0
1 14 859 – 0
0
6 30 037 (4706)
0
0
4 19 976 (6028)
1 14 782 – 1 58 633 – 2 86 401 (428)
CMS-543
0
0
0
0
MS-3
0
0
0
0
MS-4
0
0
1 225 360 – 0
0
0
MS-65
0
0
0
0
0
0
0
0
1 6333 – 0
0
0
0
0
0
0
MS-66
CMS-7
CMS-524
1 33 274 – 0
0
0
1 87 624 – 2 4520 (6391) 1 7765 – 1 20 048 – 2 5348 (102)
PVT PAY N Mean (SD)
1 10 553 – 0
1 0 – 1 0
0
0
34 0.02
0
0
0
0
27 0.004 20
