Condition based payment: improving care of chronic illness
http://www.100md.com
《英国医生杂志》
1 HealthOregon, 503 NE Laddington Court, Portland, OR 97232, USA
Correspondence to: A DiPiero, Division of General Internal Medicine and Geriatrics, L475, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97239, USA dipieroa@ohsu.edu
Fee-for-service is more than a payment method; it defines the method of care. Fee-for-condition—a payment method that rewards superior results and encourages innovation—could greatly improve care for chronic conditions
This situation will be rectified only when providers gain control of a pool of money allocated to care for a patient's medical condition and are empowered to direct that money to achieve the best results at the best price. Would capitation or case rates (two other known payment methods) prove effective? Capitation should promote prevention and efficient care. However, since payments usually do not reflect the health status of the individual patient, capitation creates incentives to either avoid sicker patients or undertreat those patients. Case rate payment is a single, consolidated fee for a procedure or other discrete service. For example, a case rate for a hip replacement would be a single payment for preoperative evaluation through postoperative rehabilitation. Although capitation encourages the efficient use of resources, its downside is to preordain a specific treatment for a particular condition, which may or may not be the most effective and efficient treatment.6
Fee-for-condition
The condition serves as the optimal unit of payment. Fee-for-condition is the comprehensive care of a condition for a fixed fee. It has three advantages over other payment methods. Firstly, it aligns providers and patients. Fundamentally, patients want their providers to efficiently and effectively prevent, diagnose, and treat their conditions.7 In a sense, the condition is the epicentre of health care.8 Secondly, condition based outcomes are the ideal unit of measurement and comparison.9 Thirdly, conditions are more readily risk adjusted so that the provider's compensation and the provider's liability closely approximate each other—in distinct contrast to capitation.
Conditions would be defined by subdividing each disease, syndrome, or set of symptoms into categories of conditions that reflect the spectrum of clinical severity. This creates a risk stratified categorisation. For example, type 2 diabetes would be subdivided into categories of conditions that account for co-existing conditions and that reflect gradations of clinical severity.
Consider the case of a 45 year old woman with diabetes controlled by diet and a glycated haemoglobin concentration of 6.2% and no comorbidity. She would be assigned a condition category and her provider would receive a monthly fee for caring for her condition. A 60 year old man taking insulin who had hypertension and peripheral neuropathy would be assigned to a different category of diabetes and would carry a fee-for-condition that reflects his more intense resource needs. Profiles of patients' conditions would be updated at specified intervals on the basis of fee-for-condition claims, which include clinical information. Providers would be paid monthly. In practice, at least initially, the scope of care covered by the fee-for-condition may be limited to a subset of services, such as outpatient services, encompassing doctors' services and pharmacy, laboratory, and radiology services.
The condition is the optimal payment unit to align the incentives between patients, providers, and payers to stimulate sustainable improvements and innovations in chronic care.
Setting fee-for-condition payments
Let's consider how the fee would be set for caring for a condition. Under fee-for-service, providers are paid the same fee whatever the patient's outcome and whatever the patient's health. The chronic care model and other quality initiatives will succeed only when price reflects value - that is, when payment for care is better aligned with the patient's clinical complexity and with the outcomes of care. This could be accomplished in several ways.
A fee for each category of condition may be based on historical costs and clinical features that predict future resource consumption. The provider would receive a fixed fee in exchange for sharing clinical data with the payer. This is the method we are seeing implemented in our region. The negotiation thus changes from a discussion focused on the fee to a discussion about the value of the clinical results.
Pay-for-performance, a trend in provider payment, proposes to pay a portion of compensation to providers who fulfil specified processes and outcomes.10 Fee-for-condition is philosophically and administratively aligned with paying for performance. We envisage many applications. For example, the payment for performance could be a bonus on top of fee-for-condition, or alternatively a portion of the fee-for-condition could be withheld and distributed to providers when they achieve certain process and outcome targets for a given condition.
Prices set by the market, with value determined by the patient, may be used by fans of a more market oriented approach. With this method, providers set their own price and defend it on the basis of their performance. This is implemented by establishing a reference price, which is a maximum price that the insurer will pay for each condition. Reference pricing has traditionally been used for drugs;11 we advocate its use for conditions. This method works only with insurance plans that use coinsurance, where patients pay a portion of their bills until they reach a maximum. Providers set their prices above, equal to, or below the reference price. Patients who go to a provider with prices above the reference price pay the difference themselves. When providers set their own fees for caring for conditions, it is the patients and providers who determine performance and value.
The reference price may be established through methods compatible with the health system of a country. A national payer using a uniform fee schedule may set a reference price reflecting average costs; a private insurer may set the reference price at some percentile of all prices in the market for a condition.
A new role for information
Clinicians, hospitals, and other providers face a growing call for transparency—public disclosure of clinical results.12 This movement is fuelled by research that suggests that transparency can improve quality.13 14 However, providers have proceeded cautiously. They wonder how transparency will actually improve results, and many fear their disclosure will be used against them.15 16 Providers are correct in surmising that when outcomes and reimbursement are decoupled (as under fee-for-service) there is no reasonable way to interpret performance or price. Transparency by itself will not improve care; it must be paired with a reimbursement method that leaves no room for dispute over price and performance. Fee-for-condition connects price, care, and outcomes. With fee-for-condition and pricing set by providers, providers will unleash the power of information for self improvement and will use it to persuade the public of their superior value. This is a far more constructive path than the present course, which positions information as a regulatory tool and the payer as a regulator.
Who receives the fee-for-condition payment?
With fee-for-condition, the question "who will receive the payment?" arises. The chronic care model reminds us that it is important to keep an open mind regarding the optimal delivery system. Fee-for-service unintentionally defines a specific delivery system—but fee-for-condition sets a goal of superior care for the condition and aligns providers and patients with this goal. Various types of organisations can be successful under such a system. Doctors, hospitals, nursing homes, and other healthcare professionals and facilities will engage in creative relationships to deliver seamless, coordinated care at a single consolidated price: a fee for a condition. In this way fee-for-condition stimulates delivery organisations to be innovative, without specifying a specific type of structure or provider.
Providers' and patients' behaviour under fee-for-condition
Payment methods should be assessed in light of their influence on the behaviours of providers and patients. Fee-for-service disconnects payment from outcomes. Providers lobby the payer for regular increases in fees on the basis of anecdotal assessments of costs and value. Pay-for-performance links pay to fulfilment of medical processes, but it remains an arrangement between the payer and provider.
In contrast, fee-for-condition aligns provider and patient. Initially the provider will control a pool of funds corresponding to a market-average cost for a condition. In return, the provider shares quality and service information with patients and payers. Accordingly, attention focuses on the outcomes achieved for a given fee. For example, the 60 year old man taking insulin who has hypertension and peripheral neuropathy compares the services and results of each prospective provider for his specific category of illness. Patients select the provider that delivers the results that are most consistent with their goals and values. As the model evolves, fee-for-condition will combine with pay-for-performance, making a portion of the payment to providers who improve specific outcomes. A more market oriented system may combine prices set by providers and reference prices set by the market for each condition. This enables innovations in quality and service, and changes in patients' demand, to disseminate more rapidly. For example, consider the provider whose self management classes and innovative patient registry—a database used to monitor clinical data and identify high risk patients—deliver superior results at a fraction of the cost of current care. That provider moves the market by pricing below the reference price and attracting patients who respond to the value presented. Other providers must imitate, innovate, or lose business.
Providers succeed by keeping their costs below the reference price or by earning a higher price by delivering better results than other providers. Patients are rewarded for seeking the best care at the best price. Continual improvement and innovation become the provider's best path to financial success.
Unintended consequences
All methods of paying providers must confront the problems of gaming and administrative costs. Fee-for-condition is no exception.
Fee-for-service financially rewards increasing the output of tasks whereas capitation rewards the avoidance of sick patients. Fee-for-condition delivers a more balanced approach to risk by adjusting the payment for the severity of the condition and rewarding superior results. Gaming is restrained because condition based payments achieve much greater transparency of quality and cost. When conditions are linked to price and outcomes, a practice that costs more and has poorer outcomes than its competitors is unsustainable in the long run. This is especially true in health systems with greater cost sharing, where patients have a personal financial liability for the practices they choose.
Administrative costs are a prime consideration in reform of payment mechanisms. Fee-for-service requires thousands of codes representing every potential task. Pay-for-performance adds complexity since the payer must capture clinical processes, link them to a condition, and analyse them with respect to performance targets that must be updated regularly. In contrast, fee-for-condition reduces the administrative burden by unifying the unit of payment, the price, and the outcomes and letting the patient assess the value. However, developing a new condition based code for categorising diagnoses will incur costs. We foresee a staged rollout beginning with the dozen chronic conditions responsible for the greatest morbidity and costs.17 The research on adjusting payment by severity in chronic conditions will help in creating a condition based code.18 As is occurring in our region, the code would then be tested in pilot programmes and improved incrementally.
Summary points
Fee-for-condition better aligns the provider and patient than do today's payment mechanisms
It rewards providers for innovations that improve results and rewards patients for seeking the best care at the best price
By linking price and outcomes, fee-for-condition achieves greater transparency of quality and cost than do current payment mechanisms
Chronic care practitioners can effectively use their advantage in performance to lead the way to a more rational payment system
All proposed changes in provider compensation risk backlash from providers and patients, as we learned from the American experiment with managed care. Pay-for-performance risks antagonising providers and patients by imposing definitions of quality and value. Fee-for-condition has less risk of backlash because providers are paid a fee that better reflects the complexity of the patient's condition, the clinical results are more transparent, and the provider and the patient together define treatment and value.
A way forward
Bodenheimer T, Wagner EH, Grumbach K. Improving primary care for patients with chronic illness: the chronic care model, part 2. JAMA. 2002;288: 1909-14.
Olivarius NF, Beck-Nielsen H, Andreasen AH, Horder M, Pedersen PA. Randomised controlled trial of structured personal care of type 2 diabetes mellitus. BMJ 2001;323: 970-5.
Rundall TG, Shortell SM, Wang MC, Casalino L, Bodenheimer T, Gillies RR, et al. As good as it gets? Chronic care management in nine leading US physician organisations. BMJ 2002;325: 958-61.
Li R, Simon J, Bodenheimer T, Gillies RR, Casalino L, Schmittdiel J, et al. Organizational factors affecting the adoption of diabetes care management processes in physician organizations. Diabetes Care 2004;27: 2312-6.
Sheila Leatherman, Donald Berwick, Debra Iles, Lawrence S Lewin, Frank Davidoff, Thomas Nolan, et al. The business case for quality: case studies and an analysis. Health Aff (Millwood) 2003;22: 17-30.
Robinson JC. Theory and practice in the design of physician payment incentives. Milbank Q 2001;79: 149-77.
Porter ME, Teisberg EO. Redefining competition in health care. Harv Bus Rev 2004;82: 64-76, 136.
Porter R. The greatest benefit to mankind: a medical history of humanity. New York: Norton, 1999.
Brook RH, McGlynn EA, Cleary PD. Quality of health care. Part 2: measuring quality of care. N Engl J Med 1996;335: 966-70.
Epstein AM, Lee TH, Hamel MB. Paying physicians for high-quality care. N Engl J Med 2004;350: 406-10.
Kanavos P, Reinhardt U. Reference pricing for drugs: is it compatible with U.S. health care? Health Aff (Millwood) 2003;22: 16-30.
Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century. Washington, DC: National Academy Press, 2001.
Hibbard JH, Stockard J, Tusler M. Does publicizing hospital performance stimulate quality improvement efforts? Health Aff (Millwood) 2003;22: 84-94.
Marshall MN, Shekelle PG, Leatherman S, Brook RH. The public release of performance data: what do we expect to gain? A review of the evidence. JAMA 2000;283: 1866-74.
Robinson AR, Hohmann KB, Rifkin JI, Topp D, Gilroy CM, Pickard JA, et al. Physician and public opinions on quality of health care and the problem of medical errors. Arch Intern Med 2002;162: 2186-90.
Marshall MN, Hiscock J, Sibbald B. Attitudes to the public release of comparative information on the quality of general practice care: qualitative study. BMJ 2002;325: 1278.
Druss BG, Marcus SC, Olfson M, Pincus HA. The most expensive medical conditions in America. Health Aff (Millwood) 2002;21: 105-11.
Hughes JS, Averill RF, Eisenhandler J, Goldfield NI, Muldoon J, Neff JM, et al. Clinical risk groups (CRGs): a classification system for risk-adjusted capitation-based payment and health care management. Med Care 2004;42: 81-90.
Anderson GF, Poullier JP. Health spending, access, and outcomes: trends in industrialized countries. Health Aff (Millwood) 1999;18: 178-92.
Reinhardt UE, Hussey PS, Anderson GF. US health care spending in an international context. Health Aff (Millwood) 2004;23: 10-25.(Albert DiPiero, co-chair1, David G Sande)
Correspondence to: A DiPiero, Division of General Internal Medicine and Geriatrics, L475, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, Portland, OR 97239, USA dipieroa@ohsu.edu
Fee-for-service is more than a payment method; it defines the method of care. Fee-for-condition—a payment method that rewards superior results and encourages innovation—could greatly improve care for chronic conditions
This situation will be rectified only when providers gain control of a pool of money allocated to care for a patient's medical condition and are empowered to direct that money to achieve the best results at the best price. Would capitation or case rates (two other known payment methods) prove effective? Capitation should promote prevention and efficient care. However, since payments usually do not reflect the health status of the individual patient, capitation creates incentives to either avoid sicker patients or undertreat those patients. Case rate payment is a single, consolidated fee for a procedure or other discrete service. For example, a case rate for a hip replacement would be a single payment for preoperative evaluation through postoperative rehabilitation. Although capitation encourages the efficient use of resources, its downside is to preordain a specific treatment for a particular condition, which may or may not be the most effective and efficient treatment.6
Fee-for-condition
The condition serves as the optimal unit of payment. Fee-for-condition is the comprehensive care of a condition for a fixed fee. It has three advantages over other payment methods. Firstly, it aligns providers and patients. Fundamentally, patients want their providers to efficiently and effectively prevent, diagnose, and treat their conditions.7 In a sense, the condition is the epicentre of health care.8 Secondly, condition based outcomes are the ideal unit of measurement and comparison.9 Thirdly, conditions are more readily risk adjusted so that the provider's compensation and the provider's liability closely approximate each other—in distinct contrast to capitation.
Conditions would be defined by subdividing each disease, syndrome, or set of symptoms into categories of conditions that reflect the spectrum of clinical severity. This creates a risk stratified categorisation. For example, type 2 diabetes would be subdivided into categories of conditions that account for co-existing conditions and that reflect gradations of clinical severity.
Consider the case of a 45 year old woman with diabetes controlled by diet and a glycated haemoglobin concentration of 6.2% and no comorbidity. She would be assigned a condition category and her provider would receive a monthly fee for caring for her condition. A 60 year old man taking insulin who had hypertension and peripheral neuropathy would be assigned to a different category of diabetes and would carry a fee-for-condition that reflects his more intense resource needs. Profiles of patients' conditions would be updated at specified intervals on the basis of fee-for-condition claims, which include clinical information. Providers would be paid monthly. In practice, at least initially, the scope of care covered by the fee-for-condition may be limited to a subset of services, such as outpatient services, encompassing doctors' services and pharmacy, laboratory, and radiology services.
The condition is the optimal payment unit to align the incentives between patients, providers, and payers to stimulate sustainable improvements and innovations in chronic care.
Setting fee-for-condition payments
Let's consider how the fee would be set for caring for a condition. Under fee-for-service, providers are paid the same fee whatever the patient's outcome and whatever the patient's health. The chronic care model and other quality initiatives will succeed only when price reflects value - that is, when payment for care is better aligned with the patient's clinical complexity and with the outcomes of care. This could be accomplished in several ways.
A fee for each category of condition may be based on historical costs and clinical features that predict future resource consumption. The provider would receive a fixed fee in exchange for sharing clinical data with the payer. This is the method we are seeing implemented in our region. The negotiation thus changes from a discussion focused on the fee to a discussion about the value of the clinical results.
Pay-for-performance, a trend in provider payment, proposes to pay a portion of compensation to providers who fulfil specified processes and outcomes.10 Fee-for-condition is philosophically and administratively aligned with paying for performance. We envisage many applications. For example, the payment for performance could be a bonus on top of fee-for-condition, or alternatively a portion of the fee-for-condition could be withheld and distributed to providers when they achieve certain process and outcome targets for a given condition.
Prices set by the market, with value determined by the patient, may be used by fans of a more market oriented approach. With this method, providers set their own price and defend it on the basis of their performance. This is implemented by establishing a reference price, which is a maximum price that the insurer will pay for each condition. Reference pricing has traditionally been used for drugs;11 we advocate its use for conditions. This method works only with insurance plans that use coinsurance, where patients pay a portion of their bills until they reach a maximum. Providers set their prices above, equal to, or below the reference price. Patients who go to a provider with prices above the reference price pay the difference themselves. When providers set their own fees for caring for conditions, it is the patients and providers who determine performance and value.
The reference price may be established through methods compatible with the health system of a country. A national payer using a uniform fee schedule may set a reference price reflecting average costs; a private insurer may set the reference price at some percentile of all prices in the market for a condition.
A new role for information
Clinicians, hospitals, and other providers face a growing call for transparency—public disclosure of clinical results.12 This movement is fuelled by research that suggests that transparency can improve quality.13 14 However, providers have proceeded cautiously. They wonder how transparency will actually improve results, and many fear their disclosure will be used against them.15 16 Providers are correct in surmising that when outcomes and reimbursement are decoupled (as under fee-for-service) there is no reasonable way to interpret performance or price. Transparency by itself will not improve care; it must be paired with a reimbursement method that leaves no room for dispute over price and performance. Fee-for-condition connects price, care, and outcomes. With fee-for-condition and pricing set by providers, providers will unleash the power of information for self improvement and will use it to persuade the public of their superior value. This is a far more constructive path than the present course, which positions information as a regulatory tool and the payer as a regulator.
Who receives the fee-for-condition payment?
With fee-for-condition, the question "who will receive the payment?" arises. The chronic care model reminds us that it is important to keep an open mind regarding the optimal delivery system. Fee-for-service unintentionally defines a specific delivery system—but fee-for-condition sets a goal of superior care for the condition and aligns providers and patients with this goal. Various types of organisations can be successful under such a system. Doctors, hospitals, nursing homes, and other healthcare professionals and facilities will engage in creative relationships to deliver seamless, coordinated care at a single consolidated price: a fee for a condition. In this way fee-for-condition stimulates delivery organisations to be innovative, without specifying a specific type of structure or provider.
Providers' and patients' behaviour under fee-for-condition
Payment methods should be assessed in light of their influence on the behaviours of providers and patients. Fee-for-service disconnects payment from outcomes. Providers lobby the payer for regular increases in fees on the basis of anecdotal assessments of costs and value. Pay-for-performance links pay to fulfilment of medical processes, but it remains an arrangement between the payer and provider.
In contrast, fee-for-condition aligns provider and patient. Initially the provider will control a pool of funds corresponding to a market-average cost for a condition. In return, the provider shares quality and service information with patients and payers. Accordingly, attention focuses on the outcomes achieved for a given fee. For example, the 60 year old man taking insulin who has hypertension and peripheral neuropathy compares the services and results of each prospective provider for his specific category of illness. Patients select the provider that delivers the results that are most consistent with their goals and values. As the model evolves, fee-for-condition will combine with pay-for-performance, making a portion of the payment to providers who improve specific outcomes. A more market oriented system may combine prices set by providers and reference prices set by the market for each condition. This enables innovations in quality and service, and changes in patients' demand, to disseminate more rapidly. For example, consider the provider whose self management classes and innovative patient registry—a database used to monitor clinical data and identify high risk patients—deliver superior results at a fraction of the cost of current care. That provider moves the market by pricing below the reference price and attracting patients who respond to the value presented. Other providers must imitate, innovate, or lose business.
Providers succeed by keeping their costs below the reference price or by earning a higher price by delivering better results than other providers. Patients are rewarded for seeking the best care at the best price. Continual improvement and innovation become the provider's best path to financial success.
Unintended consequences
All methods of paying providers must confront the problems of gaming and administrative costs. Fee-for-condition is no exception.
Fee-for-service financially rewards increasing the output of tasks whereas capitation rewards the avoidance of sick patients. Fee-for-condition delivers a more balanced approach to risk by adjusting the payment for the severity of the condition and rewarding superior results. Gaming is restrained because condition based payments achieve much greater transparency of quality and cost. When conditions are linked to price and outcomes, a practice that costs more and has poorer outcomes than its competitors is unsustainable in the long run. This is especially true in health systems with greater cost sharing, where patients have a personal financial liability for the practices they choose.
Administrative costs are a prime consideration in reform of payment mechanisms. Fee-for-service requires thousands of codes representing every potential task. Pay-for-performance adds complexity since the payer must capture clinical processes, link them to a condition, and analyse them with respect to performance targets that must be updated regularly. In contrast, fee-for-condition reduces the administrative burden by unifying the unit of payment, the price, and the outcomes and letting the patient assess the value. However, developing a new condition based code for categorising diagnoses will incur costs. We foresee a staged rollout beginning with the dozen chronic conditions responsible for the greatest morbidity and costs.17 The research on adjusting payment by severity in chronic conditions will help in creating a condition based code.18 As is occurring in our region, the code would then be tested in pilot programmes and improved incrementally.
Summary points
Fee-for-condition better aligns the provider and patient than do today's payment mechanisms
It rewards providers for innovations that improve results and rewards patients for seeking the best care at the best price
By linking price and outcomes, fee-for-condition achieves greater transparency of quality and cost than do current payment mechanisms
Chronic care practitioners can effectively use their advantage in performance to lead the way to a more rational payment system
All proposed changes in provider compensation risk backlash from providers and patients, as we learned from the American experiment with managed care. Pay-for-performance risks antagonising providers and patients by imposing definitions of quality and value. Fee-for-condition has less risk of backlash because providers are paid a fee that better reflects the complexity of the patient's condition, the clinical results are more transparent, and the provider and the patient together define treatment and value.
A way forward
Bodenheimer T, Wagner EH, Grumbach K. Improving primary care for patients with chronic illness: the chronic care model, part 2. JAMA. 2002;288: 1909-14.
Olivarius NF, Beck-Nielsen H, Andreasen AH, Horder M, Pedersen PA. Randomised controlled trial of structured personal care of type 2 diabetes mellitus. BMJ 2001;323: 970-5.
Rundall TG, Shortell SM, Wang MC, Casalino L, Bodenheimer T, Gillies RR, et al. As good as it gets? Chronic care management in nine leading US physician organisations. BMJ 2002;325: 958-61.
Li R, Simon J, Bodenheimer T, Gillies RR, Casalino L, Schmittdiel J, et al. Organizational factors affecting the adoption of diabetes care management processes in physician organizations. Diabetes Care 2004;27: 2312-6.
Sheila Leatherman, Donald Berwick, Debra Iles, Lawrence S Lewin, Frank Davidoff, Thomas Nolan, et al. The business case for quality: case studies and an analysis. Health Aff (Millwood) 2003;22: 17-30.
Robinson JC. Theory and practice in the design of physician payment incentives. Milbank Q 2001;79: 149-77.
Porter ME, Teisberg EO. Redefining competition in health care. Harv Bus Rev 2004;82: 64-76, 136.
Porter R. The greatest benefit to mankind: a medical history of humanity. New York: Norton, 1999.
Brook RH, McGlynn EA, Cleary PD. Quality of health care. Part 2: measuring quality of care. N Engl J Med 1996;335: 966-70.
Epstein AM, Lee TH, Hamel MB. Paying physicians for high-quality care. N Engl J Med 2004;350: 406-10.
Kanavos P, Reinhardt U. Reference pricing for drugs: is it compatible with U.S. health care? Health Aff (Millwood) 2003;22: 16-30.
Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century. Washington, DC: National Academy Press, 2001.
Hibbard JH, Stockard J, Tusler M. Does publicizing hospital performance stimulate quality improvement efforts? Health Aff (Millwood) 2003;22: 84-94.
Marshall MN, Shekelle PG, Leatherman S, Brook RH. The public release of performance data: what do we expect to gain? A review of the evidence. JAMA 2000;283: 1866-74.
Robinson AR, Hohmann KB, Rifkin JI, Topp D, Gilroy CM, Pickard JA, et al. Physician and public opinions on quality of health care and the problem of medical errors. Arch Intern Med 2002;162: 2186-90.
Marshall MN, Hiscock J, Sibbald B. Attitudes to the public release of comparative information on the quality of general practice care: qualitative study. BMJ 2002;325: 1278.
Druss BG, Marcus SC, Olfson M, Pincus HA. The most expensive medical conditions in America. Health Aff (Millwood) 2002;21: 105-11.
Hughes JS, Averill RF, Eisenhandler J, Goldfield NI, Muldoon J, Neff JM, et al. Clinical risk groups (CRGs): a classification system for risk-adjusted capitation-based payment and health care management. Med Care 2004;42: 81-90.
Anderson GF, Poullier JP. Health spending, access, and outcomes: trends in industrialized countries. Health Aff (Millwood) 1999;18: 178-92.
Reinhardt UE, Hussey PS, Anderson GF. US health care spending in an international context. Health Aff (Millwood) 2004;23: 10-25.(Albert DiPiero, co-chair1, David G Sande)