- Email: [email protected]
The Siren Song of Physician Incentives: Avoiding the Rocks
Practice Management
The Siren Song of Physician Incentives: Avoiding the Rocks James T. McDeavitt, MD As the United States attempts to reform its health care system, various incentive programs are playing an increasingly important role. In this review, the primary dynamics that drive the rise of incentives in health care management are discussed. Increasingly well-designed studies on the impact of incentives on outcomes continue to yield variable and, at times, unexpected results. The incorporation of incentives into the overall process of organizational cultural change is an important tool but one with significant limitations. PM R 2013;5:970-973
He’s a one trick pony One trick is all that horse can do He does one trick only It’s the principal source of his revenue Paul Simon, One Trick Pony
INTRODUCTION The mechanics of physician reimbursement used to be relatively simple. Physicians would see patients, bill for services, collect, cover their overhead, and live on whatever was left over. Incentive programs played a role in group practices but probably more to drive citizenship than fundamental practice improvement. In the emerging and evolving reform environment, formal incentive programs, both simple and complex, are becoming the norm. Over the next few years, almost every clinically active physician in the United States will either be partially reimbursed through novel incentive programs or will be involved in the design of these programs. As the field exuberantly embraces this management tool, it will be increasingly important to understand the value and limitations of incentive programs. Incentives can be important tools, but may also result in unintended and undesired consequences.
THE MOVEMENT TOWARD AN INCENTIVE CULTURE There are 3 primary and interrelated dynamics that drive the transition to incentive payment structures: (1) the dramatic shift from an independent physician model to an employment model, (2) the national move from volume-based to value-based payment systems, and (3) the professionalization of physician group management. In 2002, for every hospital-owned practice, there were 3.5 physician-owned practices. By 2008, the ratio was essentially 1 to 1. Hospital and/or health systemeowned practice is now the predominant model, and the trend seems to be accelerating [1]. As a consequence, physicians are aggregating into larger and larger groups that consist of hundreds and, in some cases, thousands of providers. This dramatic shift appears to be driven by a variety of factors: declining revenues in independent practices, increasing regulatory complexity, mandated electronic health record implementation, access to capital, and generational shifts in worklife balance expectations, to name a few [2]. Regardless of the underlying dynamics, it is a historic demographic shift. PM&R
970
1934-1482/13/$36.00 Printed in U.S.A.
J.T.M. Division of Education and Research, Carolinas HealthCare System, PO Box 32861, Charlotte, NC 28232. Address correspondence to: J.T.M.; e-mail: james.mcdeavitt@ carolinashealthcare.org Disclosure: nothing to disclose
ª 2013 by the American Academy of Physical Medicine and Rehabilitation Vol. 5, 970-973, November 2013 http://dx.doi.org/10.1016/j.pmrj.2013.09.012
PM&R
The second important driver of consolidation is the attempted national transition to value-based purchasing. Hospitals are already experiencing adjustment to payments based on performance metrics, which affect a modest but increasing proportion of the total payment: 2% by 2017 [3]. Although this may seem like a small amount, the impact must be judged against existing hospital margins, which average between 4% and 5% [4]. For many organizations, the dollars tied to value-based purchasing programs represent their entire operating margin. For those organizations, successful implementation of value-based purchasing programs is an existential exercise. The third broad dynamic encouraging the incentive culture is the professionalization of health care management, with a shift in the leadership nidus from clinicians to business professionals. During the first half of the 20th century, physicians dominated the leadership and development of health care resources in the United States. Initial constraints were placed on the payment system by the implementation of federal funding systems (Medicare and Medicaid) in the 1960s, but the system remained flush with resources and was largely able to carry out “business as usual.” With the advent of prospective payment in the 1980s, the U.S. system entered an era of increasingly constrained resources. The need to balance limited resources with desired outcomes led to the importation of business practices from outside of health care and resulted in the nidus of leadership shifting from clinicians to management professionals [5]. Large organizations of employed (but not necessarily aligned) physicians face the critical task of changing behavior in support of value-based purchasing initiatives led by professional managers. The growing importance of rational and well-designed incentive programs is probably both inevitable and desirable. However, the allure of the perfect incentive designed to drive a specific outcome must be tempered with a clear understanding of limitations of this important tool.
INCENTIVES IN NONeHEALTH CARE SETTINGS Incentives do not always deliver promised results. At a level familiar to most parents, there is a continued debate about the role of childhood monetary allowance tied to specific chores. Proponents might argue that this practice teaches responsibility and the value of money. Detractors might counter that children should learn to do their share of chores around the house because sharing the burden is part of belonging to a family. The positive impact of teaching financial responsibility can be at odds with a negative impact on character development and family acculturation. The dynamic plays out in more concrete ways as well. Leading up to the housing bust of the early 2000s, banks made a large number of high-risk loans. It subsequently has been demonstrated that changing mortgage brokers’ payment methodology from a salary to a commission model leads
Vol. 5, Iss. 11, 2013
971
lenders to make loans that they would not otherwise make, for higher total dollar values [6]. It is likely that the architects of new broker incentive plans had good motives (increase organizational margin) or even noble ones (increase home ownership). They did not set out to increase loan default rates and crash the global economy. Education, similar to health care, is an area in which society attempts to apply business principles to improve outcomes. The No Child Left Behind Act was signed into law in 2002 and requires states to set challenging academic standards, recruit and retain highly qualified teachers, and monitor progress based on standardized testing. The goals of the program are laudable, but the results have been mixed and difficult to judge in a highly politicized environment. Some experts have suggested that implementation at the state level might result in lower academic standards, increased segregation, and movement of the most talented teachers out of the most challenging environments [7]. Others studying the impact of incentive programs have expressed concern that resources could be siphoned away from high-performing schools to support those schools that are failing [8].
THE SCIENCE OF INCENTIVES Frederick Herzberg studied employee motivation in the 1950s and 1960s, and produced an often-cited article on the topic [9]. Based on results of studies of multiple professions in different environments, he purports to have gained some insight into what drives extreme motivation and engagement in a workforce. At the core of his thesis, the factors that produce job satisfaction and motivation are separate and distinct from those that produce dissatisfaction and disengagement. He identifies a number of “hygiene factors” (eg, company policies, supervision, interpersonal relationships, working conditions, salary, status). These factors, when strongly negative (very low pay, bad relationships with coworkers) tend to drive extreme dissatisfaction. However, the converse is not true. When present in the positive, they do not drive motivation. He also identifies “intrinsic factors” (achievement, recognition, valuable work, responsibility, opportunity for growth and advancement). When absent, these factors do not necessarily lead to dissatisfaction. However, when present, they tend to be highly motivating and aligning. Note the role of monetary compensation in this work. If a worker is significantly underpaid, then he or she is likely to be dissatisfied; if overpaid, he or she will not necessarily be motivated and engaged. The same dynamic has been demonstrated related to physician performance in the United States [10] and internationally [11]. Some of Herzberg’s tenants have been popularized by the business writer Daniel Pink, both in print [12] and in a viral YouTube video (YouTube LLC, San Bruno, CA) [13]. Among his key messages: incentives can be effective for certain repetitive tasks.
972
McDeavitt
However, when attempts are made to apply incentives to highly complex cognitive or creative work, there frequently is a paradoxical effect by which the desired behavior or outcome is actually reduced.
INCENTIVES IN HEALTH CARE The central tenant of the “pay-for-performance” (PFP) transition by businesses and payers (both governmental and private) has tremendous face-value appeal. In a volumebased reimbursement system, a physician or hospital is reimbursed by increasing throughput. There is no value or reward for higher quality, better outcomes, superior communication skills, and so forth. The practitioner does better by doing more. PFP programs attempt to shift the underlying dynamic of health care economics by shifting some payment drivers to quality, outcomes, and service. Many treatments in medicine (eg, antibiotics for uncomplicated otitis media) seem to be so intuitively obvious as not to require study; however, when subjected to rigorous evaluation, they sometimes prove to be useless, if not harmful. The existing literature on the impact of incentives on outcomes is mixed. In 2003, Rosenthal and Frank [14] published a comprehensive review and analysis of the peer-reviewed research literature on the effect of performance contracting on health care quality. The review identified a total of 7 empirical studies. Four studies were randomized trials with deliberately identified pre- and poststudy quality metrics. Of those 4 studies, 2 revealed modest positive results. Most of the literature is characterized by small sample sizes, selection biases, and methodologic design flaws. The theme of the inconsistent and unpredictable impact of PFP incentives continues in more recent and better-designed research. In a 6-year study published in 2012, 252 hospitals in the Premier Healthcare Informatics program volunteered to participate in a PFP trial [15]. The volunteer hospitals were compared with 3363 Premier Healthcare Informatics hospitals not in the trial. Attempts were made to control for both patient and hospital characteristics. The hospitals provided data on 33 performance metrics for 3 medical (acute myocardial infarction, congestive heart failure, pneumonia) and 2 surgical conditions (coronary artery bypass grafting and total joint replacements). Top-performing hospitals could earn an incentive payment of up to 2% of Medicare payments, whereas the poorest performing centers could lose up to 2%. After 6 years, the mortality improved in both the study and nonstudy hospitals, and the improvement was almost uniformly statistically identical, both in absolute improvement and rate of improvement. Similar results emerged from a large-scale study of physician practices. In April of 2005, 10 physician practices volunteered to join the Medicare Physician Group Practice Demonstration (MPGPD), an accountable care organization precursor [16]. Over a 5-year period, the practices received
THE SIREN SONG OF PHYSICIAN INCENTIVES
incentive payments driven by shared savings from patients under their management. The costs were compared for the 990,117 patients in the MPGPD against 7,514,453 patients in other Medicare practices. The results showed modest savings overall, with the bulk of the savings in the dually eligible population, which contains disproportionate numbers of patients of low socioeconomic status and poor social support systems. When the patients who were dually eligible were removed from the analysis, which left the Medicare patients only, there were no statistically significant savings. In fact, based on the confidence intervals, the cost of care could have been more expensive. In both this study of physician practices and the previously referenced study of hospitals, it is perhaps significant that the participating providers were not randomly selected; they were volunteers and presumably considered themselves to be relatively highperforming organizations with a chance of succeeding in the incentive-based system. Despite the lack of compelling data indicating that incentive payment programs are effective, a significant portion of hospital reimbursement is currently tied to incentives, and many physicians will be required to participate in incentive-based programs beginning in 2015 [17]. In some ways, implementation of incentive and shared savings programs remains, in part, an experiment [18]. The unpredictable nature of incentives in health care is not limited to the United States. On the opposite end of the spectrum from the MPGPD project, the regional government of the Fengsan Township in China attempted to introduce incentives to alter physician behavior [19]. The average physician makes approximately $200 per year through a mix of fee-for-service and direct sale of pharmaceuticals. The government had goals of reducing spending at the village level by reducing unnecessary care for the young and healthy, and by reducing the prescription of unnecessary medications. The government introduced logically designed incentives (salary guarantee, a small volume incentive, and a larger quality incentive). The new program was successful in meeting the stated goals; however, referral patterns shifted so that patients who were sicker were transferred to more expensive city clinics. After 5 years, no cost savings were realized. There are many reasons why incentives may not work as desired or anticipated. As in the MPGPD, results may not apply equally to all patient demographics or diagnoses. Measurements are likely to remain imperfect. Providers may find it impossible to manage multiple incentive-based payment contracts from different payers, all with different priorities, measurements, and payment schemes. There also is a long list of potential unintended consequences from incentive programs. Setting a performance bar too high may be demotivating for lower-performing organizations, which may choose to give up rather than invest resources to fight a battle that cannot be won [20]. Safety net providers may not have adequate resources to meet payer expectations, resulting in reduced reimbursement and a potential spiral into
PM&R
insolvency [21]. Providers may exclude difficult patients from panels or may concentrate so intently on meeting a limited number of externally defined performance metrics that other important elements of a complex health care interaction are neglected. Establishing specific and limited rules for health care professionals involved in complex cognitive tasks may erode the provider’s sense of professionalism and ultimately change the way that society views physicians, from selfless professionals to mindless technocrats [22].
CONCLUSION The shifting emphasis in health care toward incentive payments is likely here to stay. The nation appears to have reached a consensus that paying for quality and outcomes, rather than volume, is the appropriate direction in which payment systems should evolve. Larger groups and health delivery systems will need to find effective mechanisms to align physicians around common organizational goals. Welldesigned and rational incentive programs are a necessary tool in this process. However, leaders should not put too much faith in the power of this particular lever. The system is complex, and fundamental improvement will be complex. Organizations should consider the judicious use of incentives but should also develop other tools in their armamentarium: building a mission-driven culture of excellence, commitment to physician and practitioner lifelong learning, engagement of patients through nontraditional modalities (group visits, teleconsultation), harnessing the potential of social media to name just a few. To drive necessary change in health care, it is not enough to design the perfect incentive, sit back, and wait for the positive results to flow.
REFERENCES 1. Kocher R, Sahni NR. Hospitals’ race to employ physicians: The logic behind a money-losing proposition. N Engl J Med 2011;364:1790-1793. 2. Pate DC. Hospital-physician relations in a post-health care reform environment. J Leg Med 2012;33:7-20. 3. VanLare JM, Conway PH. Value-based purchasing—national programs to move from volume to value. N Engl J Med 2012;4:292-295.
Vol. 5, Iss. 11, 2013
973
4. Shen JJ, Ginn GO. Financial position and adoption of electronic health records: A retrospective longitudinal study. J Health Care Finance 2012;38:61-77. 5. Payne J, Leiter J. Structuring agency: Examining healthcare management in the USA and Australia using organizational theory. J Health Organ Manag 2013;27:106-126. 6. Agarwal S, Ben-David I. Do Loan Officers’ Incentives Lead to Lax Lending Standards? Available at http://www.econ.yale.edu/wshiller/ behfin/2012-04-11/Agarwal_Ben-David.pdf. Accessed August 31, 2013. 7. Ryan JE. The perverse incentives of the No Child Left Behind Act. NYU Law Rev 2004:932-989. 8. Ahn T, Vigdor J. Were all those standardized tests for nothing? Available at: http://www.aei.org/files/2013/05/17/-vigdor-and-ahn-nclbsanctions-paper_15005080098.pdf. Accessed October 29, 2013. 9. Herzberg F. One more time: How do you motivate employees? Harvard Bus Rev 2003;81:87-96. 10. Cassel CK, Jain SH. Assessing individual physician performance: Does measurement suppress motivation? JAMA 2012;307:2595-2596. 11. Lindgren A, Baathe F, Dellve L. Why risk professional fulfillment: A grounded theory of physician engagement in healthcare development. Int J Health Plan M 2013;28:e138-e157. 12. Pink DH. Drive: The Surprising Truth About What Motivates Us. New York, NY: Riverhead Books; 2011. 13. YouTube. RSA Animate - Drive: The surprising truth about what motivates us. April 1, 2010 http://www.youtube.com/watch?v¼u6X APnuFjJc. Accessed August 31, 2013. 14. Rosenthal MB, Frank RG. What is the empirical basis for paying for quality in health care? Med Care Res Rev 2006;63:135-157. 15. Jha AK, Joynt KE, Orav EJ, Epstein AM. The long-term effect of premier pay for performance on patient outcomes. N Engl J Med 2012;366: 1606-1615. 16. Colla CH, Wennberg DE, Meara E, et al. Spending differences associated with the Medicare physician group practice demonstration. JAMA 2012;38:1015-1023. 17. VanLare JM, Blum JD, Conway PH. Linking performance with payment: Implementing the physician value-based payment modifier. JAMA 2012;38:2089-2090. 18. Berwick DM. ACOs: Promise, not panacea. JAMA 2012;308: 1038-1039. 19. Wang H, Zhang L, Yip W, Hsiao W. An experiment in payment reform for doctors in rural China reduced some unnecessary care but did not lower total costs. Health Aff 2011;30:2427-2436. 20. Dowd B, Feldman R, Nersesian W. Setting pay for performance targets: Do poor performers give up? Health Econ 2013;22:168-179. 21. Jha AK, Orav EJ, Epstein AM. The effect of financial incentives on hospitals that serve poor patients. Ann Intern Med 2010;153: 299-306. 22. Jain SH, Cassel CK. Societal perceptions of physicians: Knights, knaves or pawns? JAMA 2010;304:1009-1010.
The Siren Song of Physician Incentives: Avoiding the Rocks James T. McDeavitt, MD As the United States attempts to reform its health care system, various incentive programs are playing an increasingly important role. In this review, the primary dynamics that drive the rise of incentives in health care management are discussed. Increasingly well-designed studies on the impact of incentives on outcomes continue to yield variable and, at times, unexpected results. The incorporation of incentives into the overall process of organizational cultural change is an important tool but one with significant limitations. PM R 2013;5:970-973
He’s a one trick pony One trick is all that horse can do He does one trick only It’s the principal source of his revenue Paul Simon, One Trick Pony
INTRODUCTION The mechanics of physician reimbursement used to be relatively simple. Physicians would see patients, bill for services, collect, cover their overhead, and live on whatever was left over. Incentive programs played a role in group practices but probably more to drive citizenship than fundamental practice improvement. In the emerging and evolving reform environment, formal incentive programs, both simple and complex, are becoming the norm. Over the next few years, almost every clinically active physician in the United States will either be partially reimbursed through novel incentive programs or will be involved in the design of these programs. As the field exuberantly embraces this management tool, it will be increasingly important to understand the value and limitations of incentive programs. Incentives can be important tools, but may also result in unintended and undesired consequences.
THE MOVEMENT TOWARD AN INCENTIVE CULTURE There are 3 primary and interrelated dynamics that drive the transition to incentive payment structures: (1) the dramatic shift from an independent physician model to an employment model, (2) the national move from volume-based to value-based payment systems, and (3) the professionalization of physician group management. In 2002, for every hospital-owned practice, there were 3.5 physician-owned practices. By 2008, the ratio was essentially 1 to 1. Hospital and/or health systemeowned practice is now the predominant model, and the trend seems to be accelerating [1]. As a consequence, physicians are aggregating into larger and larger groups that consist of hundreds and, in some cases, thousands of providers. This dramatic shift appears to be driven by a variety of factors: declining revenues in independent practices, increasing regulatory complexity, mandated electronic health record implementation, access to capital, and generational shifts in worklife balance expectations, to name a few [2]. Regardless of the underlying dynamics, it is a historic demographic shift. PM&R
970
1934-1482/13/$36.00 Printed in U.S.A.
J.T.M. Division of Education and Research, Carolinas HealthCare System, PO Box 32861, Charlotte, NC 28232. Address correspondence to: J.T.M.; e-mail: james.mcdeavitt@ carolinashealthcare.org Disclosure: nothing to disclose
ª 2013 by the American Academy of Physical Medicine and Rehabilitation Vol. 5, 970-973, November 2013 http://dx.doi.org/10.1016/j.pmrj.2013.09.012
PM&R
The second important driver of consolidation is the attempted national transition to value-based purchasing. Hospitals are already experiencing adjustment to payments based on performance metrics, which affect a modest but increasing proportion of the total payment: 2% by 2017 [3]. Although this may seem like a small amount, the impact must be judged against existing hospital margins, which average between 4% and 5% [4]. For many organizations, the dollars tied to value-based purchasing programs represent their entire operating margin. For those organizations, successful implementation of value-based purchasing programs is an existential exercise. The third broad dynamic encouraging the incentive culture is the professionalization of health care management, with a shift in the leadership nidus from clinicians to business professionals. During the first half of the 20th century, physicians dominated the leadership and development of health care resources in the United States. Initial constraints were placed on the payment system by the implementation of federal funding systems (Medicare and Medicaid) in the 1960s, but the system remained flush with resources and was largely able to carry out “business as usual.” With the advent of prospective payment in the 1980s, the U.S. system entered an era of increasingly constrained resources. The need to balance limited resources with desired outcomes led to the importation of business practices from outside of health care and resulted in the nidus of leadership shifting from clinicians to management professionals [5]. Large organizations of employed (but not necessarily aligned) physicians face the critical task of changing behavior in support of value-based purchasing initiatives led by professional managers. The growing importance of rational and well-designed incentive programs is probably both inevitable and desirable. However, the allure of the perfect incentive designed to drive a specific outcome must be tempered with a clear understanding of limitations of this important tool.
INCENTIVES IN NONeHEALTH CARE SETTINGS Incentives do not always deliver promised results. At a level familiar to most parents, there is a continued debate about the role of childhood monetary allowance tied to specific chores. Proponents might argue that this practice teaches responsibility and the value of money. Detractors might counter that children should learn to do their share of chores around the house because sharing the burden is part of belonging to a family. The positive impact of teaching financial responsibility can be at odds with a negative impact on character development and family acculturation. The dynamic plays out in more concrete ways as well. Leading up to the housing bust of the early 2000s, banks made a large number of high-risk loans. It subsequently has been demonstrated that changing mortgage brokers’ payment methodology from a salary to a commission model leads
Vol. 5, Iss. 11, 2013
971
lenders to make loans that they would not otherwise make, for higher total dollar values [6]. It is likely that the architects of new broker incentive plans had good motives (increase organizational margin) or even noble ones (increase home ownership). They did not set out to increase loan default rates and crash the global economy. Education, similar to health care, is an area in which society attempts to apply business principles to improve outcomes. The No Child Left Behind Act was signed into law in 2002 and requires states to set challenging academic standards, recruit and retain highly qualified teachers, and monitor progress based on standardized testing. The goals of the program are laudable, but the results have been mixed and difficult to judge in a highly politicized environment. Some experts have suggested that implementation at the state level might result in lower academic standards, increased segregation, and movement of the most talented teachers out of the most challenging environments [7]. Others studying the impact of incentive programs have expressed concern that resources could be siphoned away from high-performing schools to support those schools that are failing [8].
THE SCIENCE OF INCENTIVES Frederick Herzberg studied employee motivation in the 1950s and 1960s, and produced an often-cited article on the topic [9]. Based on results of studies of multiple professions in different environments, he purports to have gained some insight into what drives extreme motivation and engagement in a workforce. At the core of his thesis, the factors that produce job satisfaction and motivation are separate and distinct from those that produce dissatisfaction and disengagement. He identifies a number of “hygiene factors” (eg, company policies, supervision, interpersonal relationships, working conditions, salary, status). These factors, when strongly negative (very low pay, bad relationships with coworkers) tend to drive extreme dissatisfaction. However, the converse is not true. When present in the positive, they do not drive motivation. He also identifies “intrinsic factors” (achievement, recognition, valuable work, responsibility, opportunity for growth and advancement). When absent, these factors do not necessarily lead to dissatisfaction. However, when present, they tend to be highly motivating and aligning. Note the role of monetary compensation in this work. If a worker is significantly underpaid, then he or she is likely to be dissatisfied; if overpaid, he or she will not necessarily be motivated and engaged. The same dynamic has been demonstrated related to physician performance in the United States [10] and internationally [11]. Some of Herzberg’s tenants have been popularized by the business writer Daniel Pink, both in print [12] and in a viral YouTube video (YouTube LLC, San Bruno, CA) [13]. Among his key messages: incentives can be effective for certain repetitive tasks.
972
McDeavitt
However, when attempts are made to apply incentives to highly complex cognitive or creative work, there frequently is a paradoxical effect by which the desired behavior or outcome is actually reduced.
INCENTIVES IN HEALTH CARE The central tenant of the “pay-for-performance” (PFP) transition by businesses and payers (both governmental and private) has tremendous face-value appeal. In a volumebased reimbursement system, a physician or hospital is reimbursed by increasing throughput. There is no value or reward for higher quality, better outcomes, superior communication skills, and so forth. The practitioner does better by doing more. PFP programs attempt to shift the underlying dynamic of health care economics by shifting some payment drivers to quality, outcomes, and service. Many treatments in medicine (eg, antibiotics for uncomplicated otitis media) seem to be so intuitively obvious as not to require study; however, when subjected to rigorous evaluation, they sometimes prove to be useless, if not harmful. The existing literature on the impact of incentives on outcomes is mixed. In 2003, Rosenthal and Frank [14] published a comprehensive review and analysis of the peer-reviewed research literature on the effect of performance contracting on health care quality. The review identified a total of 7 empirical studies. Four studies were randomized trials with deliberately identified pre- and poststudy quality metrics. Of those 4 studies, 2 revealed modest positive results. Most of the literature is characterized by small sample sizes, selection biases, and methodologic design flaws. The theme of the inconsistent and unpredictable impact of PFP incentives continues in more recent and better-designed research. In a 6-year study published in 2012, 252 hospitals in the Premier Healthcare Informatics program volunteered to participate in a PFP trial [15]. The volunteer hospitals were compared with 3363 Premier Healthcare Informatics hospitals not in the trial. Attempts were made to control for both patient and hospital characteristics. The hospitals provided data on 33 performance metrics for 3 medical (acute myocardial infarction, congestive heart failure, pneumonia) and 2 surgical conditions (coronary artery bypass grafting and total joint replacements). Top-performing hospitals could earn an incentive payment of up to 2% of Medicare payments, whereas the poorest performing centers could lose up to 2%. After 6 years, the mortality improved in both the study and nonstudy hospitals, and the improvement was almost uniformly statistically identical, both in absolute improvement and rate of improvement. Similar results emerged from a large-scale study of physician practices. In April of 2005, 10 physician practices volunteered to join the Medicare Physician Group Practice Demonstration (MPGPD), an accountable care organization precursor [16]. Over a 5-year period, the practices received
THE SIREN SONG OF PHYSICIAN INCENTIVES
incentive payments driven by shared savings from patients under their management. The costs were compared for the 990,117 patients in the MPGPD against 7,514,453 patients in other Medicare practices. The results showed modest savings overall, with the bulk of the savings in the dually eligible population, which contains disproportionate numbers of patients of low socioeconomic status and poor social support systems. When the patients who were dually eligible were removed from the analysis, which left the Medicare patients only, there were no statistically significant savings. In fact, based on the confidence intervals, the cost of care could have been more expensive. In both this study of physician practices and the previously referenced study of hospitals, it is perhaps significant that the participating providers were not randomly selected; they were volunteers and presumably considered themselves to be relatively highperforming organizations with a chance of succeeding in the incentive-based system. Despite the lack of compelling data indicating that incentive payment programs are effective, a significant portion of hospital reimbursement is currently tied to incentives, and many physicians will be required to participate in incentive-based programs beginning in 2015 [17]. In some ways, implementation of incentive and shared savings programs remains, in part, an experiment [18]. The unpredictable nature of incentives in health care is not limited to the United States. On the opposite end of the spectrum from the MPGPD project, the regional government of the Fengsan Township in China attempted to introduce incentives to alter physician behavior [19]. The average physician makes approximately $200 per year through a mix of fee-for-service and direct sale of pharmaceuticals. The government had goals of reducing spending at the village level by reducing unnecessary care for the young and healthy, and by reducing the prescription of unnecessary medications. The government introduced logically designed incentives (salary guarantee, a small volume incentive, and a larger quality incentive). The new program was successful in meeting the stated goals; however, referral patterns shifted so that patients who were sicker were transferred to more expensive city clinics. After 5 years, no cost savings were realized. There are many reasons why incentives may not work as desired or anticipated. As in the MPGPD, results may not apply equally to all patient demographics or diagnoses. Measurements are likely to remain imperfect. Providers may find it impossible to manage multiple incentive-based payment contracts from different payers, all with different priorities, measurements, and payment schemes. There also is a long list of potential unintended consequences from incentive programs. Setting a performance bar too high may be demotivating for lower-performing organizations, which may choose to give up rather than invest resources to fight a battle that cannot be won [20]. Safety net providers may not have adequate resources to meet payer expectations, resulting in reduced reimbursement and a potential spiral into
PM&R
insolvency [21]. Providers may exclude difficult patients from panels or may concentrate so intently on meeting a limited number of externally defined performance metrics that other important elements of a complex health care interaction are neglected. Establishing specific and limited rules for health care professionals involved in complex cognitive tasks may erode the provider’s sense of professionalism and ultimately change the way that society views physicians, from selfless professionals to mindless technocrats [22].
CONCLUSION The shifting emphasis in health care toward incentive payments is likely here to stay. The nation appears to have reached a consensus that paying for quality and outcomes, rather than volume, is the appropriate direction in which payment systems should evolve. Larger groups and health delivery systems will need to find effective mechanisms to align physicians around common organizational goals. Welldesigned and rational incentive programs are a necessary tool in this process. However, leaders should not put too much faith in the power of this particular lever. The system is complex, and fundamental improvement will be complex. Organizations should consider the judicious use of incentives but should also develop other tools in their armamentarium: building a mission-driven culture of excellence, commitment to physician and practitioner lifelong learning, engagement of patients through nontraditional modalities (group visits, teleconsultation), harnessing the potential of social media to name just a few. To drive necessary change in health care, it is not enough to design the perfect incentive, sit back, and wait for the positive results to flow.
REFERENCES 1. Kocher R, Sahni NR. Hospitals’ race to employ physicians: The logic behind a money-losing proposition. N Engl J Med 2011;364:1790-1793. 2. Pate DC. Hospital-physician relations in a post-health care reform environment. J Leg Med 2012;33:7-20. 3. VanLare JM, Conway PH. Value-based purchasing—national programs to move from volume to value. N Engl J Med 2012;4:292-295.
Vol. 5, Iss. 11, 2013
973
4. Shen JJ, Ginn GO. Financial position and adoption of electronic health records: A retrospective longitudinal study. J Health Care Finance 2012;38:61-77. 5. Payne J, Leiter J. Structuring agency: Examining healthcare management in the USA and Australia using organizational theory. J Health Organ Manag 2013;27:106-126. 6. Agarwal S, Ben-David I. Do Loan Officers’ Incentives Lead to Lax Lending Standards? Available at http://www.econ.yale.edu/wshiller/ behfin/2012-04-11/Agarwal_Ben-David.pdf. Accessed August 31, 2013. 7. Ryan JE. The perverse incentives of the No Child Left Behind Act. NYU Law Rev 2004:932-989. 8. Ahn T, Vigdor J. Were all those standardized tests for nothing? Available at: http://www.aei.org/files/2013/05/17/-vigdor-and-ahn-nclbsanctions-paper_15005080098.pdf. Accessed October 29, 2013. 9. Herzberg F. One more time: How do you motivate employees? Harvard Bus Rev 2003;81:87-96. 10. Cassel CK, Jain SH. Assessing individual physician performance: Does measurement suppress motivation? JAMA 2012;307:2595-2596. 11. Lindgren A, Baathe F, Dellve L. Why risk professional fulfillment: A grounded theory of physician engagement in healthcare development. Int J Health Plan M 2013;28:e138-e157. 12. Pink DH. Drive: The Surprising Truth About What Motivates Us. New York, NY: Riverhead Books; 2011. 13. YouTube. RSA Animate - Drive: The surprising truth about what motivates us. April 1, 2010 http://www.youtube.com/watch?v¼u6X APnuFjJc. Accessed August 31, 2013. 14. Rosenthal MB, Frank RG. What is the empirical basis for paying for quality in health care? Med Care Res Rev 2006;63:135-157. 15. Jha AK, Joynt KE, Orav EJ, Epstein AM. The long-term effect of premier pay for performance on patient outcomes. N Engl J Med 2012;366: 1606-1615. 16. Colla CH, Wennberg DE, Meara E, et al. Spending differences associated with the Medicare physician group practice demonstration. JAMA 2012;38:1015-1023. 17. VanLare JM, Blum JD, Conway PH. Linking performance with payment: Implementing the physician value-based payment modifier. JAMA 2012;38:2089-2090. 18. Berwick DM. ACOs: Promise, not panacea. JAMA 2012;308: 1038-1039. 19. Wang H, Zhang L, Yip W, Hsiao W. An experiment in payment reform for doctors in rural China reduced some unnecessary care but did not lower total costs. Health Aff 2011;30:2427-2436. 20. Dowd B, Feldman R, Nersesian W. Setting pay for performance targets: Do poor performers give up? Health Econ 2013;22:168-179. 21. Jha AK, Orav EJ, Epstein AM. The effect of financial incentives on hospitals that serve poor patients. Ann Intern Med 2010;153: 299-306. 22. Jain SH, Cassel CK. Societal perceptions of physicians: Knights, knaves or pawns? JAMA 2010;304:1009-1010.