DOI: 10.3171/2014.7.JNS14236 ©AANS, 2014
Reducing costs while maintaining quality in endovascular neurosurgical procedures Clinical article Osama N. Kashlan, M.D.,1 Thomas J. Wilson, M.D.,1 Neeraj Chaudhary, M.B.B.S., M.R.C.S., F.R.C.R., 2 Joseph J. Gemmete, M.D., 2 William R. Stetler Jr., M.D.,1 N. Reed Dunnick, M.D., 2 B. Gregory Thompson, M.D.,1 and Aditya S. Pandey, M.D.1 Departments of 1Neurosurgery and 2Radiology, University of Michigan, Ann Arbor, Michigan Object. As medical costs continue to rise during a time of increasing medical resource utilization, both hospitals and physicians must attempt to limit superfluous health care expenses. Neurointerventional treatment has been shown to be costly, but it is often the best treatment available for certain neuropathologies. The authors studied the effects of 3 policy changes designed to limit the costs of performing neurointerventional procedures at the University of Michigan. Methods. The authors retrospectively analyzed the costs of performing neurointerventional procedures during the 6-month periods before and after the implementation of 3 cost-saving policies: 1) the use of an alternative, more economical contrast agent, 2) standardization of coil prices through negotiation with industry representatives to receive economies of scale, and 3) institution of a feedback method to show practitioners the costs of unused products per patient procedure. The costs during the 6-month time intervals before and after implementation were also compared with costs during the most recent 6-month time period. Results. The policy requiring use of a more economical contrast agent led to a decrease in the cost of contrast usage of $42.79 per procedure for the first 6 months after implementation, and $137.09 per procedure for the most current 6-month period, resulting in an estimated total savings of $62,924.31 for the most recent 6-month period. The standardized coil pricing system led to savings of $159.21 per coil after the policy change, and $188.07 per coil in the most recent 6-month period. This yielded total estimated savings of $76,732.56 during the most recent 6-month period. The feedback system for unused items decreased the cost of wasted products by approximately $44.36 per procedure in the 6 months directly after the policy change and by $48.20 per procedure in the most recent 6-month period, leading to total estimated savings of $22,123.80 during the most recent 6-month period. According to extrapolation over a 1-year period, the 3 policy changes decreased costs by an estimated $323,561.34. Conclusions. Simple cost-saving policies can lead to substantial reductions in costs of neurointerventional procedures while maintaining high levels of quality and growth of services. (http://thejns.org/doi/abs/10.3171/2014.7.JNS14236)
W
Key Words • contrast media • cost reduction measures • endovascular coils • endovascular procedures • inventory control • neurointerventional procedures • vascular disorders
health care reform headlining the media, both the US government and its citizens are acutely aware that funding medical care will continue to be a challenging problem as we face limited supplies and increasing demand. To this end, the cost-effectiveness of managing aneurysms using various endovascular techniques versus open microsurgical management has been much debated. Endovascular treatments are usually considered more costly, especially up-front procedures that include the use of expensive radiology equipment, catheters, and coils. Irrespective of this controversy, once a decision has been made to pursue endovascular management, measures should be undertaken to minimize costs while continuing to provide quality care. ith
J Neurosurg / August 29, 2014
Costs associated with neurointerventional procedures can broadly be thought of in 3 categories: preoperative evaluation costs, procedure-related costs, and postoperative care costs. Within each category, several possible means exist for keeping costs down. At the University of Michigan, 3 simple policy changes were made to implement cost-cutting measures for procedure-related items: 1) decreasing the cost of contrast media, 2) optimizing coil prices via negotiation with vendors, and 3) minimizing the amount of opened but unused equipment. Specifically, in October 2008, an active effort was started to allow for the preferential use of Isovue (Bracco) contrast agent over Visipaque (General Electric Healthcare) contrast medium. Next, in Au1
O. N. Kashlan et al. gust 2010, a system designed to give feedback to operating physicians regarding the cost of wasted (opened but unused) products during the performance of treatment procedures was initiated. Finally, in August 2011, negotiations between the university and representatives from manufacturers of endovascular coils used in neurointerventional procedures resulted in the standardization of coil prices from all vendors. In this study, we analyzed the impact of these 3 policy changes on the cost of neurointerventional procedures at our institution.
Methods Study Design
This cohort study was approved by the University of Michigan Institutional Review Board, and data were obtained by retrospective review of medical records and billing data. Depersonalized data from the Department of Radiology database were obtained for the 6-month period immediately preceding each specific policy change, the 6-month period immediately following each specific policy change, and the most recent 6-month period (March 2012–August 2012). All adult patients undergoing a neurointerventional procedure were included in the analysis.
Effect of Choice of Contrast Agent
In October 2008, Isovue contrast medium was made the default contrast medium for use in neurointerventional procedures instead of Visipaque contrast medium. Visipaque contrast medium remained available. Isovue is an iodinated, low-osmolar, nonionic contrast agent. Visipaque is an iodinated contrast agent that is iso-osmolar to blood and is nonionic. Ultravist (Bayer Healthcare) and Omnipaque (General Electric Healthcare) were also sporadically used. Information was obtained from the database for the periods April 2008–September 2008 (6 months prior to policy change) and October 2008–March 2009 (6 months immediately after policy change) and for the most recent 6-month period (March 2012–August 2012). Data obtained included date of the procedure, contrast type used, amount of contrast used, and total price of contrast used. The average cost of contrast agent used per procedure was then calculated by dividing the total cost of contrast used in each 6-month period by the number of procedures performed in those 6 months.
Effect of Standardization of Coil Pricing
In August 2011, negotiations between the university and representatives from multiple endovascular coil manufacturers resulted in the standardization of coil prices and the generation of shelf pricing for all vendors, using economies of scale to decrease the price per coil. Shelf pricing consisted of an average market price, depending on the type of coil (helical, 3D, or complex) as well as the length and diameter of the coil. For example, Coil A, which may cost $200 from Vendor 1, $400 from Vendor 2, $1000 from Vendor 3, and $800 from Vendor 4, would have an average shelf price of $600. Vendors were gathered and asked to either raise or reduce their price for
2
a specific coil to match that average shelf price, so that all similar coils had the same price regardless of vendor. Information was obtained from the database for February 2011–July 2011 (6 months prior to policy change), August 2011–January 2012 (6 months immediately after policy change), and March 2012–August 2012 (the most recent 6-month period). Data obtained included date of procedure, coil manufacturer, catalog number, coil description, and price of each coil used. The average cost per coil was calculated by dividing the total cost of all coils used in each 6-month period by the number of coils used. Because of agreements with the vendors, specifics regarding pricing and coils used cannot be published. Effect of Tabulating Inventory of Unused Products
In August 2010, a system to provide feedback to operating physicians regarding the cost of opened but unused products during their treatment procedures was initiated. After each procedure, the operating physician was provided with an inventory of opened but unused products and the associated cost of these products. Information was obtained from the database for February 2010–July 2010 (6 months prior to the policy change), August 2010–January 2011 (6 months immediately after the policy change), and March 2012–August 2012 (the most recent 6-month period). Data obtained included date of procedure, product manufacturer, product description, and price of the opened but unused product. The average cost of opened but unused products per procedure was calculated by dividing the total cost of unused products by the number of procedures for each 6-month period.
Measures of Quality of Care
To assess quality of care, procedure-related adverse events were recorded during each study period. The types of events used for assessment were thromboembolic events (defined as clinically evident transient ischemic attack or stroke within 30 days of the aneurysm coil embolization or vessel occlusion recognized by angiography at the time of coil embolization), intraoperative aneurysm rupture, aneurysm recurrence, and severe allergic contrast reaction.
Results Effect of Choice of Contrast Agent
Table 1 shows the total number of cases performed and choice of contrast medium used within each 6-month period. As seen by the continued predominance of Visipaque use in the period directly following the policy change, the response to the change from Visipaque to Isovue was not immediate. Hesitation on the part of operating physicians to change their contrast choice likely explains this phenomenon. However, during the most recent 6-month period, there was a nearly complete shift from the use of Visipaque to the use of Isovue. This drastic change is reflected in Fig. 1. The distribution of procedures performed remained relatively constant throughout these policy changes and into the most recent 6-month period. During the 6-month J Neurosurg / August 29, 2014
Reducing neurointerventional costs while maintaining quality TABLE 1: Number of procedures dichotomized by choice of contrast medium used during the three 6-month periods analyzed Contrast Agent Analysis Period
Visipaque
Isovue
Ultravist
Omnipaque
Multiple Agents
Total
April 2008–September 2008 (pre-change) October 2008–March 2009 (post-change) March 2012–August 2012 (most recent period)
278 200 36
6 130 370
24 0 0
3 9 1
40 60 52
351 399 459
period preceding the policy changes, 73.5% of procedures were diagnostic cerebral angiograms, 20.8% were embolization procedures, and 5.7% were other procedures, including spinal angiograms, balloon-test occlusions, thrombectomies, Wada tests, petrosal sinus sampling, and extracranial stent placements. In the most recent 6-month period, the division was 65.4%, 27.2%, and 7.4%, respectively. While resources used during each procedure over a single time interval cannot be exactly constant, consistent variation (as demonstrated above) allows for general comparison of resources used and cost of procedures over our study period. With respect to the use of contrast media, the policy change making Isovue the preferred contrast agent was found to have a profound effect on cost savings, with estimated total savings of $62,924.31 from the 6 months preceding the policy change to the most recent 6 months, and average savings of $137.09 per procedure. These changes in costs are shown in Table 2. Effect of Coil Pricing Standardization
Negotiating standardized coil pricing decreased the cost of performing neurointerventional procedures. In the 6 months after policy implementation, there was a $159.21 decrease in the average cost per coil. In the most recent 6 months (March 2012–August 2012), there was a $188.07 decrease in the average cost per coil compared with the 6 months preceding policy implementation. Overall, the cost of coils in the 6 months preceding the policy change was $569,960.70 for 393 coils; the cost for the 6 months following the policy change was $834,034.07 for 646 coils; the total cost in the most recent 6 months was
$514,981.90 for 408 coils. Thus, the total cost of coils was less during the most recent 6 months than during the 6 months preceding the policy change, despite more coils being placed in the most recent period. Interventionalist practice patterns remained similar before and after the policy change, as 6 of the 10 most frequently used coils were the same in the 6 months before the policy change and the 6 months after the policy change. Additional specifics regarding coils used and pricing cannot be published because of our agreement with vendors. Effect of Tabulating Inventory of Opened But Unused Products
After being provided with an inventory of opened but unused products, operating physicians seemed to be more conscious of unused products and more mindful of what products were opened for each case. This effect was demonstrated by the almost immediate decrease in costs that occurred with the implementation of this system, as summarized in Table 3. From the 6 months preceding the policy change to the most recent 6 months, the total cost of opened but unused products was reduced by an estimated $48.20 per procedure.
Total Effect of the 3 Policy Changes
The estimated savings for each policy change and the total savings for all 3 policy changes in the most recent 6-month period are shown in Table 4. The total estimated savings from all 3 policy changes was found to be $161,780.67, which can be extrapolated to a yearly estimated savings of $323,561.34.
Fig. 1. Percentage of Isovue use 6 months before policy implementation, 6 months after policy implementation, and during the most recent 6-month period. Vertical lines divide the 3 different time periods.
J Neurosurg / August 29, 2014
3
O. N. Kashlan et al. TABLE 2: Effect of change in choice of contrast agent on costs for the three 6-month periods analyzed Analysis Period
Total Cost of Contrast Media
Contrast Cost per Procedure
April 2008–September 2008 October 2008–March 2009 March 2012–August 2012
$67,728.39 $59,915.98 $25,643.08
$192.96 $150.17 $55.87
Measures of Quality of Care
Table 5 shows measures of the quality of care before and after the policy change and for the most recent 6-month period, including the occurrence of thromboembolic events, intraoperative rupture, aneurysm recurrence, and severe allergic contrast reaction. Quality of care was maintained after each policy change and during the most recent 6-month period. We used a preoperative prophylactic regimen consisting of steroids (hydrocortisone or prednisone) and diphenhydramine for all patients at risk for or with a history of an adverse reaction (including a minor reaction) to a contrast material. While this system was used, no severe allergic reactions to contrast materials were seen during any of the study periods. In addition, no intraoperative ruptures were observed during any of the study periods. Aneurysm recurrence and thromboembolic events were observed with similar frequencies before and after each policy change.
Discussion
As health care costs in America continue to be scrutinized, practitioners must find ways to reduce costs while maintaining high-quality medical care. We evaluated costs associated with neurointerventional procedures classified into 3 broad categories: preoperative evaluation, operative procedure, and postoperative care costs. Whereas other groups have explored cost reduction through practice-related changes in preoperative evaluation or the disposition of patients following endovascular procedures (such as comparing the cost-effectiveness of outpatient treatment versus admission to intermediate care units versus use of intensive care units), we instituted policy changes designed to reduce procedure-related costs while allowing practitioners to maintain their practice preferences while providing quality care. Three policy changes implemented at our institution led to a marked decrease in the cost of performing neu-
rointerventional procedures. In 1993, Hanwell and Haythorn3 suggested that cost reduction can be achieved only through the cooperative efforts of radiology administrators, material managers, suppliers, and clinicians. Collaboration by these same parties was necessary for the success of the policy changes at our institution. The policy changes that accounted for the largest reductions in costs were the standardization of coil pricing, followed by the preferential use of Isovue contrast medium. While each change led to moderate cost savings, the savings in cumulative cost, estimated at more than $300,000 per year, was significant. Interestingly, the rapidity with which each policy change led to cost savings varied. The effect of logging and reporting opened but unused items had an almost immediate maximal effect. In the most recent 6 months, the cost savings per procedure was around $48 versus around $44 in the 6-month period immediately following the policy change. When operating physicians were made aware of unused items, they immediately tailored their default surgical sets to what was actually being used and seemed to become mindful of what items were being opened during each procedure. Our feedback system also compared clinicians with their peers, likely creating an atmosphere in which clinicians felt compelled to avoid wasting substantially more products than their peers and thus effectively maximized cost savings. In contrast, the cost savings incurred by switching contrast agents occurred in a more delayed fashion. Initially, clinicians were reluctant to deviate from their accustomed contrast agent, but over time they became comfortable with Isovue as the default agent. This effect is shown in Fig. 1. Whereas Isovue use was only around 40%–50% in the 6-month period immediately after the policy change, this rate increased to around 90% in the most recent 6 months. As the use of Isovue increased, so did the cost savings from this policy change. Thus, getting interventionalists to acquiesce was the key to success for this policy change. The simple act of making clinicians aware of costs can have a profound effect. At a Swedish university emergency department, distribution of a price list to make clinicians aware of costs associated with diagnostic tests resulted in reduced costs compared with a group that did not receive the price list.6 All of our interventions likely had the effect of making clinicians aware of costs, but in particular, informing them of costs associated with opened but unused products not only made them cognizant of costs but gave them a specific area of waste to minimize. Hansen and Nicholson2 noted that education regarding misuse or waste of supplies is an effective
TABLE 3: Effect of tabulating inventory of opened but unused products on the cost of neurointerventional procedures for the three 6-month periods analyzed* Analysis Period
Total Cost of Unused Products
Average Cost of Unused Products per Procedure
Savings per Procedure
February 2010–July 2010 August 2010–January 2011 March 2012–August 2012
$58,863.76 $27,462.73 $19,638.46
$90.99 $46.63 $42.79
NA $44.36 $48.20
* NA = not applicable.
4
J Neurosurg / August 29, 2014
Reducing neurointerventional costs while maintaining quality TABLE 4: Estimated savings from policy changes for the most recent 6-month period Policy Change
Savings per Procedure (or coil) No. of Procedures (or coils) Total Estimated Savings
change in brand of contrast agent coil pricing standardization tabulating inventory of unused products
$137.09 $188.07 (per coil) $48.20
459 408 (coils) 459
$62,924.31 $76,732.56 $22,123.80
for each situation without consideration of price. This intervention was highly effective. Despite more coils being placed during the most recent 6-month period, the total cost for coils was less than the total cost in the 6 months preceding the policy change. Thus, despite expanding services we were not only able to decrease the cost per coil but in this instance were actually able to decrease the absolute cost. Moreover, costs were lowered without affecting practice patterns, as demonstrated by the fact that 6 of the 10 most frequently used coils were the same in the 6 months preceding the policy change and for the 6 months following the policy change. Most importantly in this regard, quality measures were similar before and after policy changes. We did not observe any change in the rate of thromboembolic events, intraoperative aneurysm rupture, aneurysm recurrence, or severe allergic contrast reaction. Thus, high-quality care was provided with the implementation of each of these policy changes. Ultimately, we believe that in an era of rising health care costs, using cost-effective methods is important, but it is of the utmost importance that this is done while maintaining the highest standard of care. This study shows only an estimate of the savings, and it has all the faults of a retrospective analysis. The assumption was made that the procedures performed in each period were of equal complexity and length and therefore required a similar amount of contrast agent per procedure. Evaluation of our data strengthened this assumption. In the 6 months preceding the policy change allowing for the preferential use of Isovue, and in the most recent 6-month period, the distribution of procedures between diagnostic cerebral angiograms, emboli-
means of reducing costs. We also found that to be the case. However, maintenance of this cost savings requires continued education of and investment by clinicians. While cost cutting is important, maintaining quality care while reducing costs is essential. Ammar1 previously demonstrated that quality care can be maintained during carotid endarterectomy while implementing effective cost-cutting strategies. One important strategy is maintaining the availability to clinicians of the gamut of options that were available prior to implementing costsaving measures. Notably, none of the policy changes in our study made previously available techniques or products unavailable. Thus, clinicians were able to continue to provide quality health care while minimizing costs. This was certainly apparent with the standardized coil pricing. Multiple trials, including the MAPS (Matrix and Platinum Science) trial and the Cerecyte trial, have failed to show a decrease in recurrence based on coil types.4,5 The HELPS (Hydrogel-coated coils versus bare platinum coils for the endovascular treatment of intracranial aneurysms) trial demonstrated that hydrogel coils did not reduce late aneurysm rupture or improve outcome compared with bare-metal coils.7 Even though one coil type has not been proven superior to another coil type, industry uses various pricing for similar types of coils. Our goal was to introduce a shelf-pricing system to reduce costs while maintaining the same practice patterns by the treating interventionalists. Though prices were standardized, all coils continued to be available. With prices standardized, price consideration was one less thing for each clinician to consider when determining which coil to place. The clinician was free to choose the optimum coil
TABLE 5: Quality of care based on occurrence of adverse events in patients undergoing neurointerventional procedures before and after policy changes and during the most recent 6-month period* Policy Change
Thromboembolic Event
Intraoperative Rupture
Aneurysm Recurrence
Severe Allergic Contrast Reaction
contrast medium change 6 months pre-change (n = 34) 6 months post-change (n = 37)
0 (0.0) 1 (2.7)
0 (0.0) 0 (0.0)
5 (14.7) 5 (13.5)
0 (0.0) 0 (0.0)
6 months pre-change (n = 39)
1 (2.6)
0 (0.0)
5 (12.8)
0 (0.0)
6 months post-change (n = 43)
1 (2.3)
0 (0.0)
3 (7.0)
0 (0.0)
6 months pre-change (n = 49)
1 (2.0)
0 (0.0)
8 (16.3)
0 (0.0)
6 months post-change (n = 41)
1 (2.4)
0 (0.0)
4 (9.8)
0 (0.0)
most recent 6 months (n = 41)
2 (4.9)
0 (0.0)
4 (9.8)
0 (0.0)
standardized coil pricing
tabulation of unused products
* Values are number (%) of patients with procedure-related adverse events.
J Neurosurg / August 29, 2014
5
O. N. Kashlan et al. zation procedures, and other procedures was similar. In fact, there was a trend toward more embolization procedures in the most recent 6-month period. Embolization procedures typically require more contrast media. Nevertheless, the total cost of contrast media for the most recent 6-month period was significantly lower than that for the 6-month period preceding the policy change. While the overall difficulty of each procedure cannot be assessed, the distribution of procedure types serves as a surrogate marker and adds to the validity of our conclusion that the policy change was effective. Our positive findings in this retrospective cohort study suggest that simple policy changes can lead to significant improvements in cost savings with little or no change in treatment procedures. A prospective trial at another institution is warranted to confirm our findings over a longer time period.
Conclusions
Simple administrative policy changes surrounding neurointerventional procedures can result in significant cost reductions without compromising patient care or changing clinician practice. Our retrospective series shows that use of economical contrast media, standardized coil pricing, and a feedback system to keep clinicians cognizant of resources used saved our institution more than $300,000 in total savings while allowing for growth of services and quality care. Acknowledgment We wish to thank Mark E. Newcom, M.B.A., for his invaluable input into the system used to provide feedback to neurosurgeons regarding unused, but opened, products. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Pandey, Chaudhary,
6
Gemmete, Dunnick, Thompson. Acquisition of data: Pandey, Kashlan, Wilson, Chaudhary, Gemmete. Analysis and interpretation of data: Kashlan, Wilson, Chaudhary, Gemmete, Stetler. Drafting the article: Kashlan, Wilson, Stetler. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Pandey. Study supervision: Pandey, Dunnick, Thompson. References 1. Ammar AD: Cost-efficient carotid surgery: a comprehensive evaluation. J Vasc Surg 24:1050–1056, 1996 2. Hansen KK, Nicholson LR: Cutting healthcare costs. Can physicians make a difference? Postgrad Med 86:91–93, 96, 98, 1989 3. Hanwell LL, Haythorn RJ: Successful purchasing strategies involve radiology administrators, materiel managers and suppliers. Radiol Manage 15:31–34, 1993 4. McDougall CG, Johnston SC, Gholkar A, Barnwell SL, Vazquez Suarez JC, Massó Romero J, et al: Bioactive versus bare platinum coils in the treatment of intracranial aneurysms: the MAPS (Matrix and Platinum Science) trial. AJNR Am J Neuroradiol 35:935–942, 2014 5. Molyneux AJ, Clarke A, Sneade M, Mehta Z, Coley S, Roy D, et al: Cerecyte coil trial: angiographic outcomes of a prospective randomized trial comparing endovascular coiling of cerebral aneurysms with either cerecyte or bare platinum coils. Stroke 43:2544–2550, 2012 6. Schilling UM: Cutting costs: the impact of price lists on the cost development at the emergency department. Eur J Emerg Med 17:337–339, 2010 7. White PM, Lewis SC, Gholkar A, Sellar RJ, Nahser H, Cognard C, et al: Hydrogel-coated coils versus bare platinum coils for the endovascular treatment of intracranial aneurysms (HELPS): a randomised controlled trial. Lancet 377:1655– 1662, 2011 Manuscript submitted January 30, 2014. Accepted July 17, 2014. Please include this information when citing this paper: published online August 29, 2014; DOI: 10.3171/2014.7.JNS14236. Address correspondence to: Aditya S. Pandey, M.D., Department of Neurosurgery, University of Michigan, 1500 E. Medical Center Dr., Rm. 3552 TC, Ann Arbor, MI 48109-5338. email: adityap@ med.umich.edu.
J Neurosurg / August 29, 2014
Reducing costs while maintaining quality in endovascular neurosurgical procedures Clinical article Osama N. Kashlan, M.D.,1 Thomas J. Wilson, M.D.,1 Neeraj Chaudhary, M.B.B.S., M.R.C.S., F.R.C.R., 2 Joseph J. Gemmete, M.D., 2 William R. Stetler Jr., M.D.,1 N. Reed Dunnick, M.D., 2 B. Gregory Thompson, M.D.,1 and Aditya S. Pandey, M.D.1 Departments of 1Neurosurgery and 2Radiology, University of Michigan, Ann Arbor, Michigan Object. As medical costs continue to rise during a time of increasing medical resource utilization, both hospitals and physicians must attempt to limit superfluous health care expenses. Neurointerventional treatment has been shown to be costly, but it is often the best treatment available for certain neuropathologies. The authors studied the effects of 3 policy changes designed to limit the costs of performing neurointerventional procedures at the University of Michigan. Methods. The authors retrospectively analyzed the costs of performing neurointerventional procedures during the 6-month periods before and after the implementation of 3 cost-saving policies: 1) the use of an alternative, more economical contrast agent, 2) standardization of coil prices through negotiation with industry representatives to receive economies of scale, and 3) institution of a feedback method to show practitioners the costs of unused products per patient procedure. The costs during the 6-month time intervals before and after implementation were also compared with costs during the most recent 6-month time period. Results. The policy requiring use of a more economical contrast agent led to a decrease in the cost of contrast usage of $42.79 per procedure for the first 6 months after implementation, and $137.09 per procedure for the most current 6-month period, resulting in an estimated total savings of $62,924.31 for the most recent 6-month period. The standardized coil pricing system led to savings of $159.21 per coil after the policy change, and $188.07 per coil in the most recent 6-month period. This yielded total estimated savings of $76,732.56 during the most recent 6-month period. The feedback system for unused items decreased the cost of wasted products by approximately $44.36 per procedure in the 6 months directly after the policy change and by $48.20 per procedure in the most recent 6-month period, leading to total estimated savings of $22,123.80 during the most recent 6-month period. According to extrapolation over a 1-year period, the 3 policy changes decreased costs by an estimated $323,561.34. Conclusions. Simple cost-saving policies can lead to substantial reductions in costs of neurointerventional procedures while maintaining high levels of quality and growth of services. (http://thejns.org/doi/abs/10.3171/2014.7.JNS14236)
W
Key Words • contrast media • cost reduction measures • endovascular coils • endovascular procedures • inventory control • neurointerventional procedures • vascular disorders
health care reform headlining the media, both the US government and its citizens are acutely aware that funding medical care will continue to be a challenging problem as we face limited supplies and increasing demand. To this end, the cost-effectiveness of managing aneurysms using various endovascular techniques versus open microsurgical management has been much debated. Endovascular treatments are usually considered more costly, especially up-front procedures that include the use of expensive radiology equipment, catheters, and coils. Irrespective of this controversy, once a decision has been made to pursue endovascular management, measures should be undertaken to minimize costs while continuing to provide quality care. ith
J Neurosurg / August 29, 2014
Costs associated with neurointerventional procedures can broadly be thought of in 3 categories: preoperative evaluation costs, procedure-related costs, and postoperative care costs. Within each category, several possible means exist for keeping costs down. At the University of Michigan, 3 simple policy changes were made to implement cost-cutting measures for procedure-related items: 1) decreasing the cost of contrast media, 2) optimizing coil prices via negotiation with vendors, and 3) minimizing the amount of opened but unused equipment. Specifically, in October 2008, an active effort was started to allow for the preferential use of Isovue (Bracco) contrast agent over Visipaque (General Electric Healthcare) contrast medium. Next, in Au1
O. N. Kashlan et al. gust 2010, a system designed to give feedback to operating physicians regarding the cost of wasted (opened but unused) products during the performance of treatment procedures was initiated. Finally, in August 2011, negotiations between the university and representatives from manufacturers of endovascular coils used in neurointerventional procedures resulted in the standardization of coil prices from all vendors. In this study, we analyzed the impact of these 3 policy changes on the cost of neurointerventional procedures at our institution.
Methods Study Design
This cohort study was approved by the University of Michigan Institutional Review Board, and data were obtained by retrospective review of medical records and billing data. Depersonalized data from the Department of Radiology database were obtained for the 6-month period immediately preceding each specific policy change, the 6-month period immediately following each specific policy change, and the most recent 6-month period (March 2012–August 2012). All adult patients undergoing a neurointerventional procedure were included in the analysis.
Effect of Choice of Contrast Agent
In October 2008, Isovue contrast medium was made the default contrast medium for use in neurointerventional procedures instead of Visipaque contrast medium. Visipaque contrast medium remained available. Isovue is an iodinated, low-osmolar, nonionic contrast agent. Visipaque is an iodinated contrast agent that is iso-osmolar to blood and is nonionic. Ultravist (Bayer Healthcare) and Omnipaque (General Electric Healthcare) were also sporadically used. Information was obtained from the database for the periods April 2008–September 2008 (6 months prior to policy change) and October 2008–March 2009 (6 months immediately after policy change) and for the most recent 6-month period (March 2012–August 2012). Data obtained included date of the procedure, contrast type used, amount of contrast used, and total price of contrast used. The average cost of contrast agent used per procedure was then calculated by dividing the total cost of contrast used in each 6-month period by the number of procedures performed in those 6 months.
Effect of Standardization of Coil Pricing
In August 2011, negotiations between the university and representatives from multiple endovascular coil manufacturers resulted in the standardization of coil prices and the generation of shelf pricing for all vendors, using economies of scale to decrease the price per coil. Shelf pricing consisted of an average market price, depending on the type of coil (helical, 3D, or complex) as well as the length and diameter of the coil. For example, Coil A, which may cost $200 from Vendor 1, $400 from Vendor 2, $1000 from Vendor 3, and $800 from Vendor 4, would have an average shelf price of $600. Vendors were gathered and asked to either raise or reduce their price for
2
a specific coil to match that average shelf price, so that all similar coils had the same price regardless of vendor. Information was obtained from the database for February 2011–July 2011 (6 months prior to policy change), August 2011–January 2012 (6 months immediately after policy change), and March 2012–August 2012 (the most recent 6-month period). Data obtained included date of procedure, coil manufacturer, catalog number, coil description, and price of each coil used. The average cost per coil was calculated by dividing the total cost of all coils used in each 6-month period by the number of coils used. Because of agreements with the vendors, specifics regarding pricing and coils used cannot be published. Effect of Tabulating Inventory of Unused Products
In August 2010, a system to provide feedback to operating physicians regarding the cost of opened but unused products during their treatment procedures was initiated. After each procedure, the operating physician was provided with an inventory of opened but unused products and the associated cost of these products. Information was obtained from the database for February 2010–July 2010 (6 months prior to the policy change), August 2010–January 2011 (6 months immediately after the policy change), and March 2012–August 2012 (the most recent 6-month period). Data obtained included date of procedure, product manufacturer, product description, and price of the opened but unused product. The average cost of opened but unused products per procedure was calculated by dividing the total cost of unused products by the number of procedures for each 6-month period.
Measures of Quality of Care
To assess quality of care, procedure-related adverse events were recorded during each study period. The types of events used for assessment were thromboembolic events (defined as clinically evident transient ischemic attack or stroke within 30 days of the aneurysm coil embolization or vessel occlusion recognized by angiography at the time of coil embolization), intraoperative aneurysm rupture, aneurysm recurrence, and severe allergic contrast reaction.
Results Effect of Choice of Contrast Agent
Table 1 shows the total number of cases performed and choice of contrast medium used within each 6-month period. As seen by the continued predominance of Visipaque use in the period directly following the policy change, the response to the change from Visipaque to Isovue was not immediate. Hesitation on the part of operating physicians to change their contrast choice likely explains this phenomenon. However, during the most recent 6-month period, there was a nearly complete shift from the use of Visipaque to the use of Isovue. This drastic change is reflected in Fig. 1. The distribution of procedures performed remained relatively constant throughout these policy changes and into the most recent 6-month period. During the 6-month J Neurosurg / August 29, 2014
Reducing neurointerventional costs while maintaining quality TABLE 1: Number of procedures dichotomized by choice of contrast medium used during the three 6-month periods analyzed Contrast Agent Analysis Period
Visipaque
Isovue
Ultravist
Omnipaque
Multiple Agents
Total
April 2008–September 2008 (pre-change) October 2008–March 2009 (post-change) March 2012–August 2012 (most recent period)
278 200 36
6 130 370
24 0 0
3 9 1
40 60 52
351 399 459
period preceding the policy changes, 73.5% of procedures were diagnostic cerebral angiograms, 20.8% were embolization procedures, and 5.7% were other procedures, including spinal angiograms, balloon-test occlusions, thrombectomies, Wada tests, petrosal sinus sampling, and extracranial stent placements. In the most recent 6-month period, the division was 65.4%, 27.2%, and 7.4%, respectively. While resources used during each procedure over a single time interval cannot be exactly constant, consistent variation (as demonstrated above) allows for general comparison of resources used and cost of procedures over our study period. With respect to the use of contrast media, the policy change making Isovue the preferred contrast agent was found to have a profound effect on cost savings, with estimated total savings of $62,924.31 from the 6 months preceding the policy change to the most recent 6 months, and average savings of $137.09 per procedure. These changes in costs are shown in Table 2. Effect of Coil Pricing Standardization
Negotiating standardized coil pricing decreased the cost of performing neurointerventional procedures. In the 6 months after policy implementation, there was a $159.21 decrease in the average cost per coil. In the most recent 6 months (March 2012–August 2012), there was a $188.07 decrease in the average cost per coil compared with the 6 months preceding policy implementation. Overall, the cost of coils in the 6 months preceding the policy change was $569,960.70 for 393 coils; the cost for the 6 months following the policy change was $834,034.07 for 646 coils; the total cost in the most recent 6 months was
$514,981.90 for 408 coils. Thus, the total cost of coils was less during the most recent 6 months than during the 6 months preceding the policy change, despite more coils being placed in the most recent period. Interventionalist practice patterns remained similar before and after the policy change, as 6 of the 10 most frequently used coils were the same in the 6 months before the policy change and the 6 months after the policy change. Additional specifics regarding coils used and pricing cannot be published because of our agreement with vendors. Effect of Tabulating Inventory of Opened But Unused Products
After being provided with an inventory of opened but unused products, operating physicians seemed to be more conscious of unused products and more mindful of what products were opened for each case. This effect was demonstrated by the almost immediate decrease in costs that occurred with the implementation of this system, as summarized in Table 3. From the 6 months preceding the policy change to the most recent 6 months, the total cost of opened but unused products was reduced by an estimated $48.20 per procedure.
Total Effect of the 3 Policy Changes
The estimated savings for each policy change and the total savings for all 3 policy changes in the most recent 6-month period are shown in Table 4. The total estimated savings from all 3 policy changes was found to be $161,780.67, which can be extrapolated to a yearly estimated savings of $323,561.34.
Fig. 1. Percentage of Isovue use 6 months before policy implementation, 6 months after policy implementation, and during the most recent 6-month period. Vertical lines divide the 3 different time periods.
J Neurosurg / August 29, 2014
3
O. N. Kashlan et al. TABLE 2: Effect of change in choice of contrast agent on costs for the three 6-month periods analyzed Analysis Period
Total Cost of Contrast Media
Contrast Cost per Procedure
April 2008–September 2008 October 2008–March 2009 March 2012–August 2012
$67,728.39 $59,915.98 $25,643.08
$192.96 $150.17 $55.87
Measures of Quality of Care
Table 5 shows measures of the quality of care before and after the policy change and for the most recent 6-month period, including the occurrence of thromboembolic events, intraoperative rupture, aneurysm recurrence, and severe allergic contrast reaction. Quality of care was maintained after each policy change and during the most recent 6-month period. We used a preoperative prophylactic regimen consisting of steroids (hydrocortisone or prednisone) and diphenhydramine for all patients at risk for or with a history of an adverse reaction (including a minor reaction) to a contrast material. While this system was used, no severe allergic reactions to contrast materials were seen during any of the study periods. In addition, no intraoperative ruptures were observed during any of the study periods. Aneurysm recurrence and thromboembolic events were observed with similar frequencies before and after each policy change.
Discussion
As health care costs in America continue to be scrutinized, practitioners must find ways to reduce costs while maintaining high-quality medical care. We evaluated costs associated with neurointerventional procedures classified into 3 broad categories: preoperative evaluation, operative procedure, and postoperative care costs. Whereas other groups have explored cost reduction through practice-related changes in preoperative evaluation or the disposition of patients following endovascular procedures (such as comparing the cost-effectiveness of outpatient treatment versus admission to intermediate care units versus use of intensive care units), we instituted policy changes designed to reduce procedure-related costs while allowing practitioners to maintain their practice preferences while providing quality care. Three policy changes implemented at our institution led to a marked decrease in the cost of performing neu-
rointerventional procedures. In 1993, Hanwell and Haythorn3 suggested that cost reduction can be achieved only through the cooperative efforts of radiology administrators, material managers, suppliers, and clinicians. Collaboration by these same parties was necessary for the success of the policy changes at our institution. The policy changes that accounted for the largest reductions in costs were the standardization of coil pricing, followed by the preferential use of Isovue contrast medium. While each change led to moderate cost savings, the savings in cumulative cost, estimated at more than $300,000 per year, was significant. Interestingly, the rapidity with which each policy change led to cost savings varied. The effect of logging and reporting opened but unused items had an almost immediate maximal effect. In the most recent 6 months, the cost savings per procedure was around $48 versus around $44 in the 6-month period immediately following the policy change. When operating physicians were made aware of unused items, they immediately tailored their default surgical sets to what was actually being used and seemed to become mindful of what items were being opened during each procedure. Our feedback system also compared clinicians with their peers, likely creating an atmosphere in which clinicians felt compelled to avoid wasting substantially more products than their peers and thus effectively maximized cost savings. In contrast, the cost savings incurred by switching contrast agents occurred in a more delayed fashion. Initially, clinicians were reluctant to deviate from their accustomed contrast agent, but over time they became comfortable with Isovue as the default agent. This effect is shown in Fig. 1. Whereas Isovue use was only around 40%–50% in the 6-month period immediately after the policy change, this rate increased to around 90% in the most recent 6 months. As the use of Isovue increased, so did the cost savings from this policy change. Thus, getting interventionalists to acquiesce was the key to success for this policy change. The simple act of making clinicians aware of costs can have a profound effect. At a Swedish university emergency department, distribution of a price list to make clinicians aware of costs associated with diagnostic tests resulted in reduced costs compared with a group that did not receive the price list.6 All of our interventions likely had the effect of making clinicians aware of costs, but in particular, informing them of costs associated with opened but unused products not only made them cognizant of costs but gave them a specific area of waste to minimize. Hansen and Nicholson2 noted that education regarding misuse or waste of supplies is an effective
TABLE 3: Effect of tabulating inventory of opened but unused products on the cost of neurointerventional procedures for the three 6-month periods analyzed* Analysis Period
Total Cost of Unused Products
Average Cost of Unused Products per Procedure
Savings per Procedure
February 2010–July 2010 August 2010–January 2011 March 2012–August 2012
$58,863.76 $27,462.73 $19,638.46
$90.99 $46.63 $42.79
NA $44.36 $48.20
* NA = not applicable.
4
J Neurosurg / August 29, 2014
Reducing neurointerventional costs while maintaining quality TABLE 4: Estimated savings from policy changes for the most recent 6-month period Policy Change
Savings per Procedure (or coil) No. of Procedures (or coils) Total Estimated Savings
change in brand of contrast agent coil pricing standardization tabulating inventory of unused products
$137.09 $188.07 (per coil) $48.20
459 408 (coils) 459
$62,924.31 $76,732.56 $22,123.80
for each situation without consideration of price. This intervention was highly effective. Despite more coils being placed during the most recent 6-month period, the total cost for coils was less than the total cost in the 6 months preceding the policy change. Thus, despite expanding services we were not only able to decrease the cost per coil but in this instance were actually able to decrease the absolute cost. Moreover, costs were lowered without affecting practice patterns, as demonstrated by the fact that 6 of the 10 most frequently used coils were the same in the 6 months preceding the policy change and for the 6 months following the policy change. Most importantly in this regard, quality measures were similar before and after policy changes. We did not observe any change in the rate of thromboembolic events, intraoperative aneurysm rupture, aneurysm recurrence, or severe allergic contrast reaction. Thus, high-quality care was provided with the implementation of each of these policy changes. Ultimately, we believe that in an era of rising health care costs, using cost-effective methods is important, but it is of the utmost importance that this is done while maintaining the highest standard of care. This study shows only an estimate of the savings, and it has all the faults of a retrospective analysis. The assumption was made that the procedures performed in each period were of equal complexity and length and therefore required a similar amount of contrast agent per procedure. Evaluation of our data strengthened this assumption. In the 6 months preceding the policy change allowing for the preferential use of Isovue, and in the most recent 6-month period, the distribution of procedures between diagnostic cerebral angiograms, emboli-
means of reducing costs. We also found that to be the case. However, maintenance of this cost savings requires continued education of and investment by clinicians. While cost cutting is important, maintaining quality care while reducing costs is essential. Ammar1 previously demonstrated that quality care can be maintained during carotid endarterectomy while implementing effective cost-cutting strategies. One important strategy is maintaining the availability to clinicians of the gamut of options that were available prior to implementing costsaving measures. Notably, none of the policy changes in our study made previously available techniques or products unavailable. Thus, clinicians were able to continue to provide quality health care while minimizing costs. This was certainly apparent with the standardized coil pricing. Multiple trials, including the MAPS (Matrix and Platinum Science) trial and the Cerecyte trial, have failed to show a decrease in recurrence based on coil types.4,5 The HELPS (Hydrogel-coated coils versus bare platinum coils for the endovascular treatment of intracranial aneurysms) trial demonstrated that hydrogel coils did not reduce late aneurysm rupture or improve outcome compared with bare-metal coils.7 Even though one coil type has not been proven superior to another coil type, industry uses various pricing for similar types of coils. Our goal was to introduce a shelf-pricing system to reduce costs while maintaining the same practice patterns by the treating interventionalists. Though prices were standardized, all coils continued to be available. With prices standardized, price consideration was one less thing for each clinician to consider when determining which coil to place. The clinician was free to choose the optimum coil
TABLE 5: Quality of care based on occurrence of adverse events in patients undergoing neurointerventional procedures before and after policy changes and during the most recent 6-month period* Policy Change
Thromboembolic Event
Intraoperative Rupture
Aneurysm Recurrence
Severe Allergic Contrast Reaction
contrast medium change 6 months pre-change (n = 34) 6 months post-change (n = 37)
0 (0.0) 1 (2.7)
0 (0.0) 0 (0.0)
5 (14.7) 5 (13.5)
0 (0.0) 0 (0.0)
6 months pre-change (n = 39)
1 (2.6)
0 (0.0)
5 (12.8)
0 (0.0)
6 months post-change (n = 43)
1 (2.3)
0 (0.0)
3 (7.0)
0 (0.0)
6 months pre-change (n = 49)
1 (2.0)
0 (0.0)
8 (16.3)
0 (0.0)
6 months post-change (n = 41)
1 (2.4)
0 (0.0)
4 (9.8)
0 (0.0)
most recent 6 months (n = 41)
2 (4.9)
0 (0.0)
4 (9.8)
0 (0.0)
standardized coil pricing
tabulation of unused products
* Values are number (%) of patients with procedure-related adverse events.
J Neurosurg / August 29, 2014
5
O. N. Kashlan et al. zation procedures, and other procedures was similar. In fact, there was a trend toward more embolization procedures in the most recent 6-month period. Embolization procedures typically require more contrast media. Nevertheless, the total cost of contrast media for the most recent 6-month period was significantly lower than that for the 6-month period preceding the policy change. While the overall difficulty of each procedure cannot be assessed, the distribution of procedure types serves as a surrogate marker and adds to the validity of our conclusion that the policy change was effective. Our positive findings in this retrospective cohort study suggest that simple policy changes can lead to significant improvements in cost savings with little or no change in treatment procedures. A prospective trial at another institution is warranted to confirm our findings over a longer time period.
Conclusions
Simple administrative policy changes surrounding neurointerventional procedures can result in significant cost reductions without compromising patient care or changing clinician practice. Our retrospective series shows that use of economical contrast media, standardized coil pricing, and a feedback system to keep clinicians cognizant of resources used saved our institution more than $300,000 in total savings while allowing for growth of services and quality care. Acknowledgment We wish to thank Mark E. Newcom, M.B.A., for his invaluable input into the system used to provide feedback to neurosurgeons regarding unused, but opened, products. Disclosure The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author contributions to the study and manuscript preparation include the following. Conception and design: Pandey, Chaudhary,
6
Gemmete, Dunnick, Thompson. Acquisition of data: Pandey, Kashlan, Wilson, Chaudhary, Gemmete. Analysis and interpretation of data: Kashlan, Wilson, Chaudhary, Gemmete, Stetler. Drafting the article: Kashlan, Wilson, Stetler. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Pandey. Study supervision: Pandey, Dunnick, Thompson. References 1. Ammar AD: Cost-efficient carotid surgery: a comprehensive evaluation. J Vasc Surg 24:1050–1056, 1996 2. Hansen KK, Nicholson LR: Cutting healthcare costs. Can physicians make a difference? Postgrad Med 86:91–93, 96, 98, 1989 3. Hanwell LL, Haythorn RJ: Successful purchasing strategies involve radiology administrators, materiel managers and suppliers. Radiol Manage 15:31–34, 1993 4. McDougall CG, Johnston SC, Gholkar A, Barnwell SL, Vazquez Suarez JC, Massó Romero J, et al: Bioactive versus bare platinum coils in the treatment of intracranial aneurysms: the MAPS (Matrix and Platinum Science) trial. AJNR Am J Neuroradiol 35:935–942, 2014 5. Molyneux AJ, Clarke A, Sneade M, Mehta Z, Coley S, Roy D, et al: Cerecyte coil trial: angiographic outcomes of a prospective randomized trial comparing endovascular coiling of cerebral aneurysms with either cerecyte or bare platinum coils. Stroke 43:2544–2550, 2012 6. Schilling UM: Cutting costs: the impact of price lists on the cost development at the emergency department. Eur J Emerg Med 17:337–339, 2010 7. White PM, Lewis SC, Gholkar A, Sellar RJ, Nahser H, Cognard C, et al: Hydrogel-coated coils versus bare platinum coils for the endovascular treatment of intracranial aneurysms (HELPS): a randomised controlled trial. Lancet 377:1655– 1662, 2011 Manuscript submitted January 30, 2014. Accepted July 17, 2014. Please include this information when citing this paper: published online August 29, 2014; DOI: 10.3171/2014.7.JNS14236. Address correspondence to: Aditya S. Pandey, M.D., Department of Neurosurgery, University of Michigan, 1500 E. Medical Center Dr., Rm. 3552 TC, Ann Arbor, MI 48109-5338. email: adityap@ med.umich.edu.
J Neurosurg / August 29, 2014
