COMMISSION ON CANCER ONCOLOGY LECTURE
Re-Engineering of Care: Surgical Leadership Glenn D Steele Jr,
MD, PhD, FACS
in addition there is significant market extension by the insurance companies into areas southeast and south central in Pennsylvania where there is no Geisingeremployed provider coverage. Most recently, the insurance company has ventured into Maine, New Jersey, Delaware and West Virginia in a variety of relationships with preferred non-Geisinger providers e attempting to recreate the Geisinger insurance - Geisinger provider “sweet spot” with select non-Geisinger delivery systems. The major Geisinger value advantage and its innovation engine is in fact the overlap between the patients that Geisinger cares for and the members that Geisinger insures. We call this our Geisinger innovation “sweet spot.” If one or another new care technique developed in the “sweet spot” creates better outcome over time for an individual or a cohort of patients, total cost of care most often goes down. This quality improvement is a direct benefit to our patients. The cost reduction represents a direct financial benefit to our insurance company’s business model. That value can then be redistributed within our single fiduciary back to the particular component that deserves the credit regardless of whether it is the insurance company, the doctor group, or the hospital platform. Some of the value must of course be given back to the buyer of our medical services as either premium or co-pay decreases. Although Geisinger is often referred to as the Kaiser Permanente (KP) of Pennsylvania, Geisinger and KP are quite different. Almost 50% of the care given to Geisinger’s insurance members is provided by the non-Geisinger physician panel in non-Geisinger owned hospitals. And all of the Geisinger owned hospitals are open to non-employed physician staff with a number of Geisinger owned hospitals including those in northeast Pennsylvania having the majority of patient care provided by non-Geisinger physicians. Therefore, when there is a significant hospital-based care redesign or population-based ambulatory care redesign that seems beneficial, the initial results most often obtained within the Geisinger sweet spot, can be tested by scaling into the more heterogeneous independent physician market. And finally, if the “sweet spot” innovation is sustained, it is scaled to all patients including those cared for by Geisinger providers but insured by nonGeisinger payers (still over 50% of our system’s payer mix). The innovative non fee-for-service reimbursement packages, for example our so-called single price “warranty,” have remained thus far restricted to our Geisinger payer-provider interaction.1,2
It is a great honor to be the invited speaker for the 2013 American College of Surgeons’ Commission on Cancer Lecture. As most of you know, I have been associated with the Commission on Cancer for over three decades. The executive and voluntary leadership of the College and the Commission have to a large extent been responsible for the major clinical improvements in hospital-based care for cancer patients over many years. Today’s lecture represents a continued commitment to that important process. I have spent the last 13 years at a remarkable organization with unusual capabilities to scale one possible care optimization model that looks to the future of healthcare reform (Figure 1). Geisinger Health System is a truly integrated health services organization encompassing every component of healthcare delivery, an unusual fiduciary structure that includes both insurance payer and clinical care enterprise, a stable patient demography that consists of large rural and post-industrial urban populations, the ability to assess new care delivery results not just over days or months but over generations, and finally, a remarkable electronic infrastructure that has been functionally refined now for almost 20 years. Geisinger services three unique market demographies. In northeast Pennsylvania there is population growth from significant migration out of the greater New York metropolitan area and a consolidation of provider entities into two major systems, Community Health System out of Brentwood, Tennessee and Geisinger Health System. These two systems have markedly different business models, the former being hospital-centric and Geisinger focusing more on population based care. The Central Susquehanna market is a static demography with older, poorer and sicker patients compared to just about any part of the United States outside of the deep South. And our western demographic (focused in Centre and Juniata counties) is a mix of the other two. On the payer side of Geisinger, there is a complete overlap in all three of the above provider markets and
Disclosure Information: Nothing to disclose. Presented at the American College of Surgeons 99th Annual Clinical Congress, Washington DC, October 2013. Received November 18, 2013; Accepted November 18, 2013. From Geisinger Health System, Danville, PA. Correspondence address: Glenn D Steele, Jr, MD, PhD, FACS, Geisinger Health System, 100 North Academy Ave, Danville, PA 17822-2201. email: [email protected]
ª 2014 by the American College of Surgeons Published by Elsevier Inc.
881
ISSN 1072-7515/14/$36.00 http://dx.doi.org/10.1016/j.jamcollsurg.2013.11.031
882
Steele
J Am Coll Surg
Re-Engineering of Care: Surgical Leadership
Figure 1. Geisinger, an integrated health services organization.
A formal set of strategic priorities drive the Geisinger operations: Quality and Innovation d Patient Center Focus B Patient activation (empowerment) B Culture of quality, safety and health d Value Re-engineering Market Leadership d Extending the GHS Brand d Scaling and Generalizing Innovation The Geisinger Family d Personal and Professional Well Being Up to 20% of total cash compensation for providers and up to 40% of total cash compensation for clinical and administrative leaders are based upon achieving the goals encompassed in our strategic priorities set.3 Geisinger’s top goal focuses on quality and fundamental innovation in caregiving. This discussion highlights the value re-engineering aspect of our innovation commitment and whether or not re-engineered care can be generalized beyond the unique Geisinger fiduciary structure, culture, and market. Geisinger care re-engineering is based upon two touchstones. One emanates from the 2003 McGlynn, et al., Rand study that postulated almost 45% of routine care provided in a variety of markets for a variety of prevalent medical problems was either too much, too little, or wrong.4,5 This initial study has been substantiated numerous times and is now widely accepted. And even
if one discounts the result because of a variety of methodological issues, the fact that a significant amount of cost in caregiving does not bring value to the people who get the care or may in fact hurt them is, in my view, a remarkable opportunity. Geisinger’s care re-engineering attempts to remove as much of this unnecessary or hurtful cost as possible while maintaining or improving near-term as well as long-term patient outcome. Our second lynchpin for re-engineering care is an extension of the knowledge that cost and quality do not correlate.6,7 This has been known for several decades and in fact our more recent care redesign road tests at Geisinger convince us that, more often than not, high cost patient populations are those that have worse healthcare quality outcomes. Thus high cost is usually a surrogate marker for bad outcome. Value re-engineering should therefore accomplish two goals simultaneously: better outcome and lower cost. Patients and professionals are no longer faced with choosing between cost reduction and quality! The Geisinger care transformation initiatives are: d
d d d d
ProvenCare for Acute Episodic Care (the “Warranty”) ProvenCare Chronic Disease ProvenHealth Navigator (Advanced Medical Home) Transitions of Care PRIDE (Proven Innovation Drive for Excellence)
Here, we will discuss several examples of what we call ProvenCare Acute, our version of hospital based or
Vol. 218, No. 5, May 2014
Table 1. ProvenCare CAB v2.0 42 ACC / AHA 2011 Class I and IIIh guidelines 25 Additional Geisinger consensus-based guidelines 67 Total guidelines adopted and translated into 120 best practices
hospital associated care re-engineering. We began this effort in 2007 as a part of ProvenCare CABG as an attempt to reduce unjustified variation in how our cardiologists and cardiac surgeons performed elective coronary artery revascularization.2,8,9 The almost 18 month “sociology” experiment arriving at default best practice for every aspect of patient care from the time of diagnosis through rehabilitation, the commitment to achieve 100% of the default care goals 100% of the time for each patient, the journey to achieving this, and the effects on both cost as well as near-term and long-term quality have been remarkable and have led recently to version 2.0 (Table 1), commercial affirmation with our Walmart Center of Excellence program, and a broadening of the ProvenCare value re-engineering portfolio to the summary listed in Table 2. Generalization of ProvenCare Acute beyond Geisinger was initiated by Matt Facktor of our system, Doug Wood at University of Washington, Seattle and Dave Winchester, medical director of cancer programs, Commission on Cancer at the American College of Surgeons in 2009.10 This multi-institutional effort has been based on an attempt to standardize the diagnostic and Table 3. ProvenCare Lung Cancer Collaborative Team Members Matthew Facktor, MD, Geisinger Health System, Danville, PA Doug Wood, MD, University of Washington Medical Center, Seattle, WA David P. Winchester, MD, FACS, Medical Director, Cancer Programs, Commission on Cancer John Howington, MD, NorthShore University Health System, Evanston, IL David White, MD, Duke Raleigh Hospital, Raleigh, NC Malcolm DeCamp, MD, Northwestern University Medical Center, Chicago, IL Jemi Olak, MD, Kern Medical Center, Bakersfield, CA Rose Ganim, MD (Primary Contact), The Baystate Medical Center, Springfield, MA John R. Handy, MD, Providence Health & Services, Portland, OR Kimberly Costas, MD (Primary Contact), Providence Regional Medical Center, Everett, WA Nikhilesh M. Korgaonkar, MD, MBA, Sinai Hospital of Baltimore, Baltimore, MD Thomas V. Bilfinger, MD, Stony Brook University Medical Center, Stony Brook, NY Syed M. Quadri, MD, UMass Memorial Health Care, Worcester, MA
Steele
Re-Engineering of Care: Surgical Leadership
883
Table 2. ProvenCare Portfolio ProvenCare: e ProvenCare Coronary Artery Bypass Graft (CABG) e ProvenCare Percutaneous Coronary Intervention (PCI) e ProvenCare Bariatric Surgery e ProvenCare Perinatal e ProvenCare Lung Cancer (CoC Collaborative) e ProvenCare Fragility Hip Fracture e ProvenCare Heart Failure e ProvenCare COPD e ProvenCare Lumbar Spine e ProvenCare Total Hip e ProvenCare Total Knee e ProvenCare Epilepsy ProvenCare Evidence-Based Guidelines (in conjunction with PRIDE): e Sepsis (ED) e Chest Pain e R/O MI (ED) e Pediatric Abdominal Pain (R/O Appendicitis [ED]) Portfolio as of September 1, 2013.
therapeutic approach to patients with potentially resectable lung cancers, to build decision aids and reminders into enabling systems whether or not they were electronic, to design the desired actions as a default, to create redundancy to assure 100% compliance, to bundle related tasks, and to encourage a teamwork approach to feedback, training, and continuous improvement. All of this was to be accomplished outside of the framework of a formal randomized clinical trial. In other words, could the Geisinger institutional value re-engineering effort be generalized nationally and applied to a compelling prevalent and sub-optimally treated cancer? The ProvenCare Lung Cancer collaborative team members are listed in Table 3 and the specific expectations for this multi-institutional effort were as follows: d d d
d
Decrease complication and mortality rates Increase the efficiency of process Increase the proficiency in providing evidence-based care Eliminate unwarranted variation of care
Table 4. Collaborative Participants 6 Diverse Institutions 4 Academic Medical Centers 2 Urban (NCI designated Comprehensive Cancer Centers) 1 Suburban 1 Rural (NCI designated Community Cancer Centers Program) 2 Community Medical Centers 1 Suburban (University affiliate) 1 County-owned Solo thoracic surgeon Minimal IT support, paper based medical record
884
Steele
J Am Coll Surg
Re-Engineering of Care: Surgical Leadership
Figure 2. Collaborative participants.
Baseline data justified the need for such an exercise. Outside of the context of randomized clinical trials (where most of us as providers and most of us as patients live) multiple studies show unjustified variation in staging as well as in surgical technique.11,12 In the STS database review published in 2008, mediastinoscopy was performed in only 21% of patients (vs. 27% in the 2005 American College of Surgery survey data), while during resection only 65% of the samples had defined mediastinal lymph nodes found (vs. 58% in the earlier American College of Surgery survey data). And finally, only 54% of the cases had what was considered by STS complete documented staging. It is our assumption that such unjustified variation leads to decreased quality and increased cost, similar to almost every other complicated task in life. The ProvenCare model in this multi-institutional generalizing effort was based on the following internal Geisinger design principles: (1) evidence based medicine was embedded into the workflow; (2) principles of reliability science and care redesign included standardization, error proofing, and failure mode redesign; (3) effective data feedback strategies were implemented; and finally, (4) patients were engaged in their own care. Institutions included in the first and most recently the second tranche of this multi-institutional experiment were intentionally diverse. Thus the question of reliably delivering default best practice was challenged across a variety of healthcare delivery platforms, across a variety of patient populations, and in systems with paper or electronic medical records. Original participating institutions are described in Table 4 and Figure 2.
Responsibilities of the clinical leaders and principle investigators are as follows: d
d d d
d d
Develop a culture that expects and insists on elimination of unwarranted variation as a safety issue Clearly describe the organizational outcome goals Set process expectations e at least 10 2 level reliability Patients should drive any variation in protocols rather than individual providers Require clinicians to communicate/document exceptions Provide resources to measure the outcomes and reasons for non-compliance
Process redesign principles included: eliminating any care steps that could be eliminated, automating any work that could be automated, delegating work that must be done to appropriately trained non-physician staff when possible, creating tools to enhance reliability of or
Table 5. Best Practice Resources 1. CMS Quality Measures (Patient Quality Reporting Initiative, PQRI) 2. ACS National Surgical Improvement Program (NSQIP) 3. ACS Infection & Ventilator Bundles 4. National Quality Forum Voluntary Consensus Standards for Cardiac Surgery 5. National Comprehensive Cancer Network (NCCN) 6. American College of Chest Physicians (ACCP) 7. American College of Cardiology (ACC) & American Heart Association (AHA) 8. Society of Thoracic Surgeons (STS) General Thoracic Surgery Database
Vol. 218, No. 5, May 2014
Steele
Re-Engineering of Care: Surgical Leadership
885
Figure 3. ProvenCare elective pulmonary resection: process flow best practices examples. (Reprinted with permission from CA Cancer J Clin 2011;61:382-396.)
efficiency in the care provided, and activating and engaging the patients. The establishment of default best practice was an important part of developing the correct clinical leadership sociology. Each potential default best practice and/ or guideline was assigned to an advocate who was expected to validate and defend through critique of the literature, advice of experts, and discussion amongst coleaders. Oftentimes general guidelines needed to be translated into specific concrete metric-based goals. In other words, ProvenCare behavior needed to be precisely established, including accountability and defining how
to track. “Opt out” provisions were always included with the caveat that they needed to be justified in real time and acceptable to co-clinical leaders. Best practice resources as in the case with ProvenCare CABG, enabled “off the shelf” evidence-based or consensus-based starting points for all compartments of the entire episode default care redesign (Table 5). A summary of the flow is detailed in Figure 3. The chronology of the first phase of the study was as follows: d August 2009 e Thoracic leader engagement and consensus on rules and evidence
Figure 4. ProvenCare lung cancer: electronic health record tools. CA, cancer; NA, not applicable; EKG, electrocardiogram; Hx, history; PET, positron emission tomography. (Reprinted with permission from CA Cancer J Clin 2011;61:382-396.)
886
Steele
J Am Coll Surg
Re-Engineering of Care: Surgical Leadership
Figure 5. Phase II collaborative participants. d
d d d d
September to December 2009 e Institutional commitment, team formation and pre-work January 2010 e Collaborative team kick-off session March 2010 e Collaborative WebEx 1 July 2010 e Go live September 2010 e Collaborative WebEx 2
An example of the ProvenCare enabling for those that had EHR is shown in Figure 4.
Thus far there have been over 1,100 patients accrued. Phase II sites went live on August 6, 2012. Collaborative participants are defined at present in Figure 5. Overall reliability of the commitment to best practice is summarized in Figure 6. At the present accrual, the secondary outcomes analyses summarized in Figure 7 are underway. Key questions that may be answered by this effort are as follows. Is this learning collaborative generalizable?
Figure 6. Overall reliability.
Steele
Vol. 218, No. 5, May 2014
Re-Engineering of Care: Surgical Leadership
887
Figure 7. Secondary outcomes study design.
Will the cost/quality metrics hold up to analysis? Is this approach one potential alternative to randomized clinical trials (particularly in the absence of any sophisticated methodology to assure lack of bias in observational studies).13 And finally can we prove that this care redesign technique is a dynamic innovation engine (not static cookbook medicine)? In the face of all of the political, financial, and transactional anxieties surrounding PPACA (ObamaCare), as professionals we need to remember the clinical care baseline at present. Our patients experience unjustified variation in quality, access and cost of care. Care variation is not just geographic in nature. It exists within given hospitals and among our various specialists even treating patients with similar medical problems in the same population cohort. We continue to have unwarranted and fragmented caregiving in most systems of care outside of unusual integrated systems like Geisinger and a few others. Our payment incentives remain perversely linked to piece-rate fee-for-service where the major discussion is price per unit increase on the provider side and decreased reimbursement per unit on the payer side. And we are facing a complex new set of payment incentives predicated on fundamentally new care delivery models in which some aspect of bundled episode or population risk taking is expected. Finally we continue interacting with patients as if they are passive recipients of care. I would submit that where we want to be includes 1) affordable coverage for all, 2) payment for value not
volume, 3) coordinated care, 4) continuous improvement in innovation even outside of randomized clinical trials, 5) much more patient activation (I find that to be a much better phrase than patient empowerment), and finally 6) well-articulated national healthcare goals, leadership, and accountability. So I believe all of this means that we as professionals and particularly we as surgical leaders should think about our pride of professionalism as we move through healthcare reform. We know that no matter what happens politically in order to get to where we want to be from where we are there will be a significant emphasis on increased value for those we care for. The only way to achieve this increased value is to extract a significant amount of cost and redistribute it. That’s not going to happen by simply focusing on price per unit. It is predicated on re-engineering care and removing unnecessary care units with an eye to assuring maintained or even improved patient quality. And this is a perfectly designed challenge for surgeons. Reengineering care is the right thing to do and I would submit that surgeons have a unique capability to do it. Acknowledgment: This lecture is a summary of many surgical innovations at Geisinger and beyond. I particularly wish to acknowledge the work of Alfred E Casale, MD, Matthew A Facktor, MD, and the leadership of the American College of Surgeons’ Commission on Cancer, David P Winchester, MD.
888
Steele
Re-Engineering of Care: Surgical Leadership
REFERENCES 1. Abelson R. Bid for Better Hospital Care, Heart Surgery With a Warranty. New York Times 2007 May 17. 2. Casale AS, Paulus RA, Selna M, et al. ProvenCare: A provider-driven pay-for-performance program for acute episodic cardiac surgical care. Ann Surg 2007;246:613e623. 3. Lee TH, Bothe A, Steele GD. How Geisinger structures its physicians’ compensation to support improvements in quality, efficiency, and volume. Health Aff 2012;31:2068e2073. 4. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Engl J Med 2003;348:2635e2645. 5. Steinberg EP. Improving the quality of care e can we practice what we preach? N Engl J Med 2003;348:2681e2683. 6. Newhouse JP. Why is there a quality chasm? Health Aff 2002; 21:13e25. 7. Lansky D, Milstein A. Quality measurement in orthopaedics: the purchasers’ view. Clin Orthop Relat Res 2009;467: 2548e2555.
J Am Coll Surg
8. Paulus RA, Davis K, Steele GD. Continuous innovation in health care: implications of the Geisinger experience. Health Aff (Millwood) 2008;27:1235e1245. 9. Berry SA, Doll MC, McKinley KE, et al. ProvenCare: quality improvement model for designing highly reliable care in cardiac surgery. Qual Saf Health Care 2009;18:360e368. 10. Katlic MR, Facktor MA, Berry SA, et al. ProvenCare lung cancer: a multi-institutional improvement collaborative. CA Cancer J Clin 2011;61:382e396. 11. Little AG, Rusch VW, Bonner JA, et al. Patterns of surgical care of lung cancer patients. Ann Thorac Surg 2005;80: 2051e2056. 12. Boffa DJ, Allen MS, Grab JD, et al. Data from the Society of Thoracic Surgeons General Thoracic Surgery database: The surgical management of primary lung tumors. J Thorac Cardiovasc Surg 2008;135:247e254. 13. Stewart WF, Shah NR, Selna MJ, et al. Bridging the inferential gap: the electronic health record and clinical evidence. Health Aff (Millwood) 2007;26:w192ew194.
Re-Engineering of Care: Surgical Leadership Glenn D Steele Jr,
MD, PhD, FACS
in addition there is significant market extension by the insurance companies into areas southeast and south central in Pennsylvania where there is no Geisingeremployed provider coverage. Most recently, the insurance company has ventured into Maine, New Jersey, Delaware and West Virginia in a variety of relationships with preferred non-Geisinger providers e attempting to recreate the Geisinger insurance - Geisinger provider “sweet spot” with select non-Geisinger delivery systems. The major Geisinger value advantage and its innovation engine is in fact the overlap between the patients that Geisinger cares for and the members that Geisinger insures. We call this our Geisinger innovation “sweet spot.” If one or another new care technique developed in the “sweet spot” creates better outcome over time for an individual or a cohort of patients, total cost of care most often goes down. This quality improvement is a direct benefit to our patients. The cost reduction represents a direct financial benefit to our insurance company’s business model. That value can then be redistributed within our single fiduciary back to the particular component that deserves the credit regardless of whether it is the insurance company, the doctor group, or the hospital platform. Some of the value must of course be given back to the buyer of our medical services as either premium or co-pay decreases. Although Geisinger is often referred to as the Kaiser Permanente (KP) of Pennsylvania, Geisinger and KP are quite different. Almost 50% of the care given to Geisinger’s insurance members is provided by the non-Geisinger physician panel in non-Geisinger owned hospitals. And all of the Geisinger owned hospitals are open to non-employed physician staff with a number of Geisinger owned hospitals including those in northeast Pennsylvania having the majority of patient care provided by non-Geisinger physicians. Therefore, when there is a significant hospital-based care redesign or population-based ambulatory care redesign that seems beneficial, the initial results most often obtained within the Geisinger sweet spot, can be tested by scaling into the more heterogeneous independent physician market. And finally, if the “sweet spot” innovation is sustained, it is scaled to all patients including those cared for by Geisinger providers but insured by nonGeisinger payers (still over 50% of our system’s payer mix). The innovative non fee-for-service reimbursement packages, for example our so-called single price “warranty,” have remained thus far restricted to our Geisinger payer-provider interaction.1,2
It is a great honor to be the invited speaker for the 2013 American College of Surgeons’ Commission on Cancer Lecture. As most of you know, I have been associated with the Commission on Cancer for over three decades. The executive and voluntary leadership of the College and the Commission have to a large extent been responsible for the major clinical improvements in hospital-based care for cancer patients over many years. Today’s lecture represents a continued commitment to that important process. I have spent the last 13 years at a remarkable organization with unusual capabilities to scale one possible care optimization model that looks to the future of healthcare reform (Figure 1). Geisinger Health System is a truly integrated health services organization encompassing every component of healthcare delivery, an unusual fiduciary structure that includes both insurance payer and clinical care enterprise, a stable patient demography that consists of large rural and post-industrial urban populations, the ability to assess new care delivery results not just over days or months but over generations, and finally, a remarkable electronic infrastructure that has been functionally refined now for almost 20 years. Geisinger services three unique market demographies. In northeast Pennsylvania there is population growth from significant migration out of the greater New York metropolitan area and a consolidation of provider entities into two major systems, Community Health System out of Brentwood, Tennessee and Geisinger Health System. These two systems have markedly different business models, the former being hospital-centric and Geisinger focusing more on population based care. The Central Susquehanna market is a static demography with older, poorer and sicker patients compared to just about any part of the United States outside of the deep South. And our western demographic (focused in Centre and Juniata counties) is a mix of the other two. On the payer side of Geisinger, there is a complete overlap in all three of the above provider markets and
Disclosure Information: Nothing to disclose. Presented at the American College of Surgeons 99th Annual Clinical Congress, Washington DC, October 2013. Received November 18, 2013; Accepted November 18, 2013. From Geisinger Health System, Danville, PA. Correspondence address: Glenn D Steele, Jr, MD, PhD, FACS, Geisinger Health System, 100 North Academy Ave, Danville, PA 17822-2201. email: [email protected]
ª 2014 by the American College of Surgeons Published by Elsevier Inc.
881
ISSN 1072-7515/14/$36.00 http://dx.doi.org/10.1016/j.jamcollsurg.2013.11.031
882
Steele
J Am Coll Surg
Re-Engineering of Care: Surgical Leadership
Figure 1. Geisinger, an integrated health services organization.
A formal set of strategic priorities drive the Geisinger operations: Quality and Innovation d Patient Center Focus B Patient activation (empowerment) B Culture of quality, safety and health d Value Re-engineering Market Leadership d Extending the GHS Brand d Scaling and Generalizing Innovation The Geisinger Family d Personal and Professional Well Being Up to 20% of total cash compensation for providers and up to 40% of total cash compensation for clinical and administrative leaders are based upon achieving the goals encompassed in our strategic priorities set.3 Geisinger’s top goal focuses on quality and fundamental innovation in caregiving. This discussion highlights the value re-engineering aspect of our innovation commitment and whether or not re-engineered care can be generalized beyond the unique Geisinger fiduciary structure, culture, and market. Geisinger care re-engineering is based upon two touchstones. One emanates from the 2003 McGlynn, et al., Rand study that postulated almost 45% of routine care provided in a variety of markets for a variety of prevalent medical problems was either too much, too little, or wrong.4,5 This initial study has been substantiated numerous times and is now widely accepted. And even
if one discounts the result because of a variety of methodological issues, the fact that a significant amount of cost in caregiving does not bring value to the people who get the care or may in fact hurt them is, in my view, a remarkable opportunity. Geisinger’s care re-engineering attempts to remove as much of this unnecessary or hurtful cost as possible while maintaining or improving near-term as well as long-term patient outcome. Our second lynchpin for re-engineering care is an extension of the knowledge that cost and quality do not correlate.6,7 This has been known for several decades and in fact our more recent care redesign road tests at Geisinger convince us that, more often than not, high cost patient populations are those that have worse healthcare quality outcomes. Thus high cost is usually a surrogate marker for bad outcome. Value re-engineering should therefore accomplish two goals simultaneously: better outcome and lower cost. Patients and professionals are no longer faced with choosing between cost reduction and quality! The Geisinger care transformation initiatives are: d
d d d d
ProvenCare for Acute Episodic Care (the “Warranty”) ProvenCare Chronic Disease ProvenHealth Navigator (Advanced Medical Home) Transitions of Care PRIDE (Proven Innovation Drive for Excellence)
Here, we will discuss several examples of what we call ProvenCare Acute, our version of hospital based or
Vol. 218, No. 5, May 2014
Table 1. ProvenCare CAB v2.0 42 ACC / AHA 2011 Class I and IIIh guidelines 25 Additional Geisinger consensus-based guidelines 67 Total guidelines adopted and translated into 120 best practices
hospital associated care re-engineering. We began this effort in 2007 as a part of ProvenCare CABG as an attempt to reduce unjustified variation in how our cardiologists and cardiac surgeons performed elective coronary artery revascularization.2,8,9 The almost 18 month “sociology” experiment arriving at default best practice for every aspect of patient care from the time of diagnosis through rehabilitation, the commitment to achieve 100% of the default care goals 100% of the time for each patient, the journey to achieving this, and the effects on both cost as well as near-term and long-term quality have been remarkable and have led recently to version 2.0 (Table 1), commercial affirmation with our Walmart Center of Excellence program, and a broadening of the ProvenCare value re-engineering portfolio to the summary listed in Table 2. Generalization of ProvenCare Acute beyond Geisinger was initiated by Matt Facktor of our system, Doug Wood at University of Washington, Seattle and Dave Winchester, medical director of cancer programs, Commission on Cancer at the American College of Surgeons in 2009.10 This multi-institutional effort has been based on an attempt to standardize the diagnostic and Table 3. ProvenCare Lung Cancer Collaborative Team Members Matthew Facktor, MD, Geisinger Health System, Danville, PA Doug Wood, MD, University of Washington Medical Center, Seattle, WA David P. Winchester, MD, FACS, Medical Director, Cancer Programs, Commission on Cancer John Howington, MD, NorthShore University Health System, Evanston, IL David White, MD, Duke Raleigh Hospital, Raleigh, NC Malcolm DeCamp, MD, Northwestern University Medical Center, Chicago, IL Jemi Olak, MD, Kern Medical Center, Bakersfield, CA Rose Ganim, MD (Primary Contact), The Baystate Medical Center, Springfield, MA John R. Handy, MD, Providence Health & Services, Portland, OR Kimberly Costas, MD (Primary Contact), Providence Regional Medical Center, Everett, WA Nikhilesh M. Korgaonkar, MD, MBA, Sinai Hospital of Baltimore, Baltimore, MD Thomas V. Bilfinger, MD, Stony Brook University Medical Center, Stony Brook, NY Syed M. Quadri, MD, UMass Memorial Health Care, Worcester, MA
Steele
Re-Engineering of Care: Surgical Leadership
883
Table 2. ProvenCare Portfolio ProvenCare: e ProvenCare Coronary Artery Bypass Graft (CABG) e ProvenCare Percutaneous Coronary Intervention (PCI) e ProvenCare Bariatric Surgery e ProvenCare Perinatal e ProvenCare Lung Cancer (CoC Collaborative) e ProvenCare Fragility Hip Fracture e ProvenCare Heart Failure e ProvenCare COPD e ProvenCare Lumbar Spine e ProvenCare Total Hip e ProvenCare Total Knee e ProvenCare Epilepsy ProvenCare Evidence-Based Guidelines (in conjunction with PRIDE): e Sepsis (ED) e Chest Pain e R/O MI (ED) e Pediatric Abdominal Pain (R/O Appendicitis [ED]) Portfolio as of September 1, 2013.
therapeutic approach to patients with potentially resectable lung cancers, to build decision aids and reminders into enabling systems whether or not they were electronic, to design the desired actions as a default, to create redundancy to assure 100% compliance, to bundle related tasks, and to encourage a teamwork approach to feedback, training, and continuous improvement. All of this was to be accomplished outside of the framework of a formal randomized clinical trial. In other words, could the Geisinger institutional value re-engineering effort be generalized nationally and applied to a compelling prevalent and sub-optimally treated cancer? The ProvenCare Lung Cancer collaborative team members are listed in Table 3 and the specific expectations for this multi-institutional effort were as follows: d d d
d
Decrease complication and mortality rates Increase the efficiency of process Increase the proficiency in providing evidence-based care Eliminate unwarranted variation of care
Table 4. Collaborative Participants 6 Diverse Institutions 4 Academic Medical Centers 2 Urban (NCI designated Comprehensive Cancer Centers) 1 Suburban 1 Rural (NCI designated Community Cancer Centers Program) 2 Community Medical Centers 1 Suburban (University affiliate) 1 County-owned Solo thoracic surgeon Minimal IT support, paper based medical record
884
Steele
J Am Coll Surg
Re-Engineering of Care: Surgical Leadership
Figure 2. Collaborative participants.
Baseline data justified the need for such an exercise. Outside of the context of randomized clinical trials (where most of us as providers and most of us as patients live) multiple studies show unjustified variation in staging as well as in surgical technique.11,12 In the STS database review published in 2008, mediastinoscopy was performed in only 21% of patients (vs. 27% in the 2005 American College of Surgery survey data), while during resection only 65% of the samples had defined mediastinal lymph nodes found (vs. 58% in the earlier American College of Surgery survey data). And finally, only 54% of the cases had what was considered by STS complete documented staging. It is our assumption that such unjustified variation leads to decreased quality and increased cost, similar to almost every other complicated task in life. The ProvenCare model in this multi-institutional generalizing effort was based on the following internal Geisinger design principles: (1) evidence based medicine was embedded into the workflow; (2) principles of reliability science and care redesign included standardization, error proofing, and failure mode redesign; (3) effective data feedback strategies were implemented; and finally, (4) patients were engaged in their own care. Institutions included in the first and most recently the second tranche of this multi-institutional experiment were intentionally diverse. Thus the question of reliably delivering default best practice was challenged across a variety of healthcare delivery platforms, across a variety of patient populations, and in systems with paper or electronic medical records. Original participating institutions are described in Table 4 and Figure 2.
Responsibilities of the clinical leaders and principle investigators are as follows: d
d d d
d d
Develop a culture that expects and insists on elimination of unwarranted variation as a safety issue Clearly describe the organizational outcome goals Set process expectations e at least 10 2 level reliability Patients should drive any variation in protocols rather than individual providers Require clinicians to communicate/document exceptions Provide resources to measure the outcomes and reasons for non-compliance
Process redesign principles included: eliminating any care steps that could be eliminated, automating any work that could be automated, delegating work that must be done to appropriately trained non-physician staff when possible, creating tools to enhance reliability of or
Table 5. Best Practice Resources 1. CMS Quality Measures (Patient Quality Reporting Initiative, PQRI) 2. ACS National Surgical Improvement Program (NSQIP) 3. ACS Infection & Ventilator Bundles 4. National Quality Forum Voluntary Consensus Standards for Cardiac Surgery 5. National Comprehensive Cancer Network (NCCN) 6. American College of Chest Physicians (ACCP) 7. American College of Cardiology (ACC) & American Heart Association (AHA) 8. Society of Thoracic Surgeons (STS) General Thoracic Surgery Database
Vol. 218, No. 5, May 2014
Steele
Re-Engineering of Care: Surgical Leadership
885
Figure 3. ProvenCare elective pulmonary resection: process flow best practices examples. (Reprinted with permission from CA Cancer J Clin 2011;61:382-396.)
efficiency in the care provided, and activating and engaging the patients. The establishment of default best practice was an important part of developing the correct clinical leadership sociology. Each potential default best practice and/ or guideline was assigned to an advocate who was expected to validate and defend through critique of the literature, advice of experts, and discussion amongst coleaders. Oftentimes general guidelines needed to be translated into specific concrete metric-based goals. In other words, ProvenCare behavior needed to be precisely established, including accountability and defining how
to track. “Opt out” provisions were always included with the caveat that they needed to be justified in real time and acceptable to co-clinical leaders. Best practice resources as in the case with ProvenCare CABG, enabled “off the shelf” evidence-based or consensus-based starting points for all compartments of the entire episode default care redesign (Table 5). A summary of the flow is detailed in Figure 3. The chronology of the first phase of the study was as follows: d August 2009 e Thoracic leader engagement and consensus on rules and evidence
Figure 4. ProvenCare lung cancer: electronic health record tools. CA, cancer; NA, not applicable; EKG, electrocardiogram; Hx, history; PET, positron emission tomography. (Reprinted with permission from CA Cancer J Clin 2011;61:382-396.)
886
Steele
J Am Coll Surg
Re-Engineering of Care: Surgical Leadership
Figure 5. Phase II collaborative participants. d
d d d d
September to December 2009 e Institutional commitment, team formation and pre-work January 2010 e Collaborative team kick-off session March 2010 e Collaborative WebEx 1 July 2010 e Go live September 2010 e Collaborative WebEx 2
An example of the ProvenCare enabling for those that had EHR is shown in Figure 4.
Thus far there have been over 1,100 patients accrued. Phase II sites went live on August 6, 2012. Collaborative participants are defined at present in Figure 5. Overall reliability of the commitment to best practice is summarized in Figure 6. At the present accrual, the secondary outcomes analyses summarized in Figure 7 are underway. Key questions that may be answered by this effort are as follows. Is this learning collaborative generalizable?
Figure 6. Overall reliability.
Steele
Vol. 218, No. 5, May 2014
Re-Engineering of Care: Surgical Leadership
887
Figure 7. Secondary outcomes study design.
Will the cost/quality metrics hold up to analysis? Is this approach one potential alternative to randomized clinical trials (particularly in the absence of any sophisticated methodology to assure lack of bias in observational studies).13 And finally can we prove that this care redesign technique is a dynamic innovation engine (not static cookbook medicine)? In the face of all of the political, financial, and transactional anxieties surrounding PPACA (ObamaCare), as professionals we need to remember the clinical care baseline at present. Our patients experience unjustified variation in quality, access and cost of care. Care variation is not just geographic in nature. It exists within given hospitals and among our various specialists even treating patients with similar medical problems in the same population cohort. We continue to have unwarranted and fragmented caregiving in most systems of care outside of unusual integrated systems like Geisinger and a few others. Our payment incentives remain perversely linked to piece-rate fee-for-service where the major discussion is price per unit increase on the provider side and decreased reimbursement per unit on the payer side. And we are facing a complex new set of payment incentives predicated on fundamentally new care delivery models in which some aspect of bundled episode or population risk taking is expected. Finally we continue interacting with patients as if they are passive recipients of care. I would submit that where we want to be includes 1) affordable coverage for all, 2) payment for value not
volume, 3) coordinated care, 4) continuous improvement in innovation even outside of randomized clinical trials, 5) much more patient activation (I find that to be a much better phrase than patient empowerment), and finally 6) well-articulated national healthcare goals, leadership, and accountability. So I believe all of this means that we as professionals and particularly we as surgical leaders should think about our pride of professionalism as we move through healthcare reform. We know that no matter what happens politically in order to get to where we want to be from where we are there will be a significant emphasis on increased value for those we care for. The only way to achieve this increased value is to extract a significant amount of cost and redistribute it. That’s not going to happen by simply focusing on price per unit. It is predicated on re-engineering care and removing unnecessary care units with an eye to assuring maintained or even improved patient quality. And this is a perfectly designed challenge for surgeons. Reengineering care is the right thing to do and I would submit that surgeons have a unique capability to do it. Acknowledgment: This lecture is a summary of many surgical innovations at Geisinger and beyond. I particularly wish to acknowledge the work of Alfred E Casale, MD, Matthew A Facktor, MD, and the leadership of the American College of Surgeons’ Commission on Cancer, David P Winchester, MD.
888
Steele
Re-Engineering of Care: Surgical Leadership
REFERENCES 1. Abelson R. Bid for Better Hospital Care, Heart Surgery With a Warranty. New York Times 2007 May 17. 2. Casale AS, Paulus RA, Selna M, et al. ProvenCare: A provider-driven pay-for-performance program for acute episodic cardiac surgical care. Ann Surg 2007;246:613e623. 3. Lee TH, Bothe A, Steele GD. How Geisinger structures its physicians’ compensation to support improvements in quality, efficiency, and volume. Health Aff 2012;31:2068e2073. 4. McGlynn EA, Asch SM, Adams J, et al. The quality of health care delivered to adults in the United States. N Engl J Med 2003;348:2635e2645. 5. Steinberg EP. Improving the quality of care e can we practice what we preach? N Engl J Med 2003;348:2681e2683. 6. Newhouse JP. Why is there a quality chasm? Health Aff 2002; 21:13e25. 7. Lansky D, Milstein A. Quality measurement in orthopaedics: the purchasers’ view. Clin Orthop Relat Res 2009;467: 2548e2555.
J Am Coll Surg
8. Paulus RA, Davis K, Steele GD. Continuous innovation in health care: implications of the Geisinger experience. Health Aff (Millwood) 2008;27:1235e1245. 9. Berry SA, Doll MC, McKinley KE, et al. ProvenCare: quality improvement model for designing highly reliable care in cardiac surgery. Qual Saf Health Care 2009;18:360e368. 10. Katlic MR, Facktor MA, Berry SA, et al. ProvenCare lung cancer: a multi-institutional improvement collaborative. CA Cancer J Clin 2011;61:382e396. 11. Little AG, Rusch VW, Bonner JA, et al. Patterns of surgical care of lung cancer patients. Ann Thorac Surg 2005;80: 2051e2056. 12. Boffa DJ, Allen MS, Grab JD, et al. Data from the Society of Thoracic Surgeons General Thoracic Surgery database: The surgical management of primary lung tumors. J Thorac Cardiovasc Surg 2008;135:247e254. 13. Stewart WF, Shah NR, Selna MJ, et al. Bridging the inferential gap: the electronic health record and clinical evidence. Health Aff (Millwood) 2007;26:w192ew194.
