Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
THIRD PARTY CARRIER PERSPECT IVE ON PHYSICIAN PAYMENT Lawrence C. Morris This discussion of insurance carrier systems for payment of physicians will address the underlying problems in physician payment system design, which are simply to determine how the carrier is going to spend other people's money for services whose quality and quantity the carrier can influence, but only to a limited extent. The goal is a system that is efficient to administer, encourages economical use of benefits, and offers distinctive values for subscribers and insurers when compared to the payment systems of competitive carriers. The analysis here does not deal with the so-called alternative, risk-sharing payment systems. This is not because they can be dismissed lightly; on the contrary, they are a significant and grow, ng segment of the health care economy. But the concern of this paper is for conventional fee-for-service payment system design. Given these goals and these limitations, the approaches to the design of a carrier's physician payment system are governed by a number of principles that may be summarized and explained here: Payment Is A Means, Not An End A Payment System Must Proceed Toward Specific Objectives In the private market there are five parties at interest, each pursuing objectives that are reasonable and legitimate, but different. Thus, payment always takes place in the context of conf licting objectives. The challenge to the carrier is to work out effective compromises to ~ a) establish attractive products or benefits, (b) differentiate those products in a highly competitive market, and {c) offer a significant variety of products to appeal to a range of philosophies, degrees of commitment to varying objectives, and abilities to pay. The parties at interest are: The Patient He is interested primarily in predictability of outcome. Ideally, he prefers full payment and free choice of physician. However, he will compromise these desires, so long as the terms of the compromise are made clear in advance of the transaction. He does not react well when he finds after the fact that his expectations have not been met. 63
The Physician He also is interested in predictability, and in consistency of results. He needs to know what to expect f ram the carrier and what to bi 11 the patient, and he is understandably intolerant of different payments for what he perceives as similar cases. He also expects that there will be exception processes for unusual cases. The Payor Eighty-five to ninety percent of people with private health insurance receive employer or union contributions to the cost of the insurance. Thus the payor and the patient are not the same, and they have different objectives. The payor (or employer or union) is interested in minimization of expenditures and maximization of value. He intends to support a good quality of care. He is interested in avoiding either extravagance or waste. If the program does not contemplate full coverage of all charges, he still wants a reasonable distribution of the available funds. Since expenditure is a product of price and utilization and intensity, he expects all three to be controlled. Thus he does not view the payment system in isolation; he sees it as part, but only part, of a financing system that must be complemented by other functions. The Pub 1 ic The publ ic expresses itself not only as a buyer of health insurance but also through regulatory and legislative processes. Regulatory activity has two thrusts: The state insurance department is interested in protecting the public from being misled, treated unfairly, overcharged for a particular set of benef its, or victimized by default. The law enforcement agencies are concerned with maintenance of an orderly, non-collusive market. As this Free market. concern has developed in recent years, it has proscribed negotiation of payments between carriers and physicians who do not share a common practice, and it is also concerned with restriction of access to the market by practitioners on the basis of license type, arbitrary qualifications, or collusive activity. Legislatures may respond to various pressures to mandate coverage of specific diseases (e.g., end stage renal disease); categories of care {e.g., mental health); or practitioners (e.g., chiropractors). The Carrier The carrier wishes to compete successfully in the market. Usually it tries to establish a niche, and a product appropriate to that niche. This implies differentiation from the objectives of others, as in price, predictability, acceptance or avoidance of poor risks, contractual relationships with physicians, or in other ways. -64-
The important point is that no system can be all things to all interests. Instead, each system represents some degree of compromise acceptable to its own parties at interest. There Are Fundamental Differences Between Institution-Carriers And Professional-Carrier Relationships The relationships between institutions and carriers in a way are like those between labor and management, in that they are interdependent, but with divergent interests. Since the patient is typically incapable of paying large hospital bills without assistance, and since risk avoidance, inherent in the purchase of coverage' typically biases toward the coverage of the more threatening hospital costs, the hospital and the carrier either reach an accommodation of their interests, or both enterprises fail. The physician, in contrast, can look to the patient for at least some payment of most of his charges. From this he gains a degree of autonomy, which varies by specialty. Furthermore, physicians are personalities, whereas institutions are organizations. Personal philosophy, likes and dislikes, habits and relationships intrude on the physician' s economic decisions to a degree not found in organizational behavior. Economic factors influencing physician behavior may include: Specialization Return on investment of time and money; education to a broader variety of treatment alternatives. Locat ion Cost of doing business; economic capability of patients; social and economic goals of the practitioner. Target Income Ability to adjust either price or utilization to meet income goal. Controls Perceived need to adjust prices in anticipation of such controls as wage-price stabilization programs and Medicare reform. Competition Price or utilization adjustments to improve competitive position. -65-
Dissatisfaction A perception of imbalance in the economic value of cognitive and technical procedures in a physician's practice, motivating to the employment of questionably useful procedures. Indirect economic factors may also influence physician behavior. For example, exposure to data on physician practices may induce a reluctance to stand out from community norms of price or utilization. Education may also influence behavior indirectly. Thus lengths of stay and use of ancillary services are influenced by patterns learned during undergraduate and postgraduate training. Fear of external control is still another influence; an example is the AMA Council on Medical Service Report D ~ June 1983 ~ suggesting that physicians deal only with patients on charge issues, and that the relationship between carrier payments and physician charges be severed. (This report was not adopted by the AMA.) Finally, location is significant to practice patterns. Different parts of the country have quite different traditions in contracting with carriers, length of stay, admissions practice, and other behaviors. A Payment System Should Be In Reasonable Synchroni zat ion With The Evolution Of The Social And Delivery Systems Health care coverage first became widespread under depression conditions categorized by scarcity of financial resources. At the same time, the level of specialization was low, resulting in relatively ongoing patient-physician relationships, with patients' economic circumstances well-known to physicians, and fees varying accordingly. The technology was limited, inflation was either negative or negligible, and the general assumption was that negotiation of fee schedules between physician groups and carriers was legally permissible. In this environment, the most common method of paying physicians was the fee schedule. The result was either indemnity--specified payment against an unspecif fed charge--or service/indemnity--full payment below a stated patient income level and indemnity above that level. Indemnity is still widely held coverage, a common mechanism for distributing a given amount of money that is inadequate for full payment. It is usually administered through fee schedules. Whereas in the 1930s and 194Os such schedules were sometimes negotiated with medical associations, the current application of the antitrust laws requi res that schedules be set unilaterally by the carrier. The -66-
implication that resultant levels of payment may not be equitable in the eyes of the physician has not been lost on the medical profession. The service/indemnity contract facilitated the concept that the fee should be consonant with the patient's ability to pay. The concept still exists, but the practice has become rare. Factors leading to its practical demise included the rise of specialization, with many physicians seeing a patient for only a single episode of care. With no ongoing relationship, and with little knowledge of the patient' s finances, many physicians have moved to a single fee for all comers. The rise of high technology also contributed to disappearance of the ~abi lity to pay. concept. The high overhead implicit in many procedures does not lend itself to signif icant discounting. Further, specialists are f requently dissatisf fed to receive the same f ee schedule as non-spec ial i sts. The evolution of the blue-collar worker into the middle class has been still another factor. There is now less perceived need to discount f ees f or this group. Attainment by many workers of wages beyond the patient income levels for fully paid benef its meant that workers in the same bargaining unit might get quite different results from the same negotiated coverage. Moreover, there were practical and potential legal difficulties in negotiating schedules in such relatively heterogeneous environments. In response to these problems, the ~ usual, customary, and reasonable charge. method was developed. It is important to note that conceptually there is no such thing as a usual and customary fee. There are, rather, a usual fee, a customary fee, and a reasonable fee. ~Usual. refers to the charge made by a specif ic physician to the majority of his patients for a specific procedure. ~Customary. refers to a range of charges for the same procedure by physicians of like qualification in the same area. The customary range is typically cut of f at the 80th to 90th percentile of physician charges. ~Reasonable. refers to a fee which is both usual and customary, or which may be neither usual nor customary, but justif fed by the specif ic clinical circumstances of the case. In practice, many carriers do not have sufficient market penetration or data to establish and maintain usual fee profiles. This can be a major weakness in purported OCR programs. UCR is based on several assumptions. First, it is assumed that differences in charge stemming from specialty, location, and overhead factors are legitimate and should be recognized. Secondly, most physicians do in fact charge one fee to all patients, and, third, UCR was originally conceived as top-of-the-line coverage and not for the majority of the insured population. Thus there existed a validating mechanism for usual charges in the charges made by self-paying and indemnif fed patients. This last assumption has been severely eroded by the adoption of a form of UCR by Medicare and its subsequent establishment as a standard in the market. 67
Usual fees can be determined either by a filing of fees by the physician or from claims data. Probing for the top of the range is common. Updates vary with respect both to the time interval and the amount allowed. It is fairly common to permit updates only annually and to limit them by some index. The Consumer Price Index is frequently used. Originally, the customary range was a pure statistical derivation. It is vulnerable to collusion, however, particularly in the smaller specialties, and to the effects of probing for the limits. The Blue Cross and Blue Shield Association, for example, has adopted a membership standard requiring controls on the customary range. Again, the rate of increase in the Consumer Price Index is a commonly used control. Medicare, in contrast, uses an index felt to be ref lect ive of overhead factors and the cost of living. Medicare also determines payment on the basis of data f ram the calendar year preceding the f iscal year, data 10 to 22 months old. This approach is not available in the private market, because given competing programs, the low acceptance rate by physicians does not satisfy the market. Some programs pay a percentage of the usual fee. The percentage is typically 80 percent, but this may vary. The purpose may be to decrease premiums to an affordable level, inhibit utilization, or both. Programs with physician participation contracts frequently limit the total charge to that permitted in a full UCR program. In the 1980s, the environment is changing again. It is now characterized by limited resources, the emergence of progressively higher technology, and a much more compet itive physician community. Furthermore, a more aggressive attitude on the part of payors has resulted in the application of more stringent cost controls. Clearly, there is wide recognition of risk-sharing systems as a desirable means of constraining expenditures. There is somewhat less recognition of risk-sharing as a means of rewarding desirable behavior. Risk, however, implies organizaion capable of organizing and managing i t. Whi le there are some programs which have applied risk principles to open-panel programs, these remain exceptions. The evolutionary course of these systems is still unclear. Payment Systems Are Designed For Specific Benefit, Administrative, And Legal Circumstances Insurers deal with large volumes of c [aims. The volume could not be accommodated without averaging prices and procedure descriptions. All coding systems embody philosophies. At one extreme are systems with multiple modifiers and detailed descriptors, intended to -68-
identify as many variables as possible in reporting cases. These are useful for medical records and the management of research data, but they are inflationary for payment purposes. Diagnosis Related Groups represent the opposite extreme, conveying little information except diagnosis and price. In current practice, carriers attempt to adopt coding sufficient to identify real differences in payment value or benefit coverage without weakening price and utilizaton prof lies more than necessary. Importantly, coding does implement both the medical policy of the carrier and the contractual requirements of the purchaser. As an example, some contracts cover abortions and sterilization, and some do not; the coding system must differentiate the purposes as well as the description of the procedure. Negotiation at one time was an accepted method of establishing payment levels. This is no longer legally permissible. Accordingly, most fee-for-service payment is made now on the basis of unilaterally established fee schedules or some variant of the UCR profile system. Finally, the data resources of the carrier are important. If UCR charge programs are to be implemented literally, the carrier must know and apply the usual charge information at the individual physician level. This implies substantial information and processing resources. Fee-For-Service Payments Are Governed Primarily By The Market In an open-panel system, the payment system will not create equity where it had not previously existed. Neither is there any known and consistent relationship between the price of a service and its quality. This market principle is true of both fully paid and indemnity programs. Indemnity programs usually strive for consistency of result. In the full payment programs, there are three fundamental alternatives: Pay the market price, change the nature of the market (HMO, Preferred Provider), or regulate either the buyer or the seller. The payment of market price does not necessarily imply unrestricted payment of all billed charges. Cutoffs establish how far a carrier will go. This is necessary to protect the average payer from extravagant expenditure (or high charges) on the part of a few. The implication, of course, is that even programs designed for full payment do not make full payment in every case. There are both exploitive and non-exploitive (e.g., the physician who gives extraordinary attention and charges for it) charging patterns which cannot be absorbed collectively. An important means of improving the program's performance is for the carrier to contract directly with physicians to accept its payments as payment in full. There are some important non-price incentives that can be used to make such arrangements attractive to the physician. In these circumstances, -69-
real differentials may arise between what the physician will accept from the contracting carrier and his charges to others. It is unfortunate that some types of practice and some services command lower payment than others requiring equivalent training and time. However, the carrier has no means of changing this pattern except to raise payments for the first group, incurring unnecessary costs, or lower them for the second, perhaps sacrificing full payment. In some specialties, particularly anesthesiology and psychiatry, a fee-for-time approach has been adopted. This provides a means of paying for treatment in which it is difficult to specify exactly what was done and exactly how it was accomplished. The method is conceptually available to other specialties, but most have not accepted it. Surgeons come closest, with a Global fee. concept embracing - preoperative examination, the operation itself, and postoperative care. Negotiation of the trade-offs required to implement comparable payment systems in medicine are significantly inhibited by the antitrust laws. Most alternatives to a market price system depend on the existence of organizations, with organizational objectives. For example, an HMO needing a particular specialty may have to negotiate salary and benefits to secure a specialist. The same thing occurs in the division of income in fee-for-service group practices. This ability to make internal trade-of fs requires formal organization, and is not sufficiently widespread to support a predominant payment system at this point. It is possible, however, to modify the payment system in order to provide incentives to ef ficient care. An example is ambulatory surgery, for which some carriers pay snore than they would if the procedure were done on an inpatient basis. It is important to differentiate procedures, so that incentive payments are not wasted on surgery that should never justify admission. Once the carrier establishes its payment levels, it may vigorously defend them. Blue Shield Plan participating contracts, for example, obligate the physician to accept Plan payments as payment in full. Several carriers use a Hold harmless. approach. The hold harmless arrangement arises from the carrier's obligation to pay a reasonable fee. In the event that carrier and physician cannot agree on a reasonable fee, the carrier will ordinarily make its best offer. If the physician accepts it, nothing more happens. If the physician continues to bill the patient, and the patient does not pay, still nothing more happens. But if the physician brings suit against the patient, the determination of what is reasonable has been transferred to the courts. The carrier is still obligated, and will present its case in court. If the physician prevails, the carrier will make the additional payment. In such circumstances, the carrier may assume the -70-
patient's costs of litigation, on the ground that its own liability was being litigated. Some participating contracts include hold harmless provisions for utilization judged to be medically unnecessary. The principles here are exactly the same: The carrier defends its determination and remains liable for the outcome. All Payment Systems Carry Incentives For Specif to Types Of Behavior Any payment system must be complemented by programs to monitor and control those incentives. As is now rather widely understood, fee-for-service payment systems carry an incentive to increase utilization. Thus to be effective in the overall financing of care, such a payment system must be supported by a variety of programs intended to counter the ·perverse incentives. and thus constrain total expenditures. The major components of a total system, excluding alternative delivery systems, are: The Payment System Administrative procedures must be in place to assure that the payment system is executed precisely and that payments are made for covered services, and only covered services. Physicians can be motivated to enter into contracts with the carrier through non-price incentives. Ordinarily, Blue Shield Plans, unlike Medicare, do not permit assignment on a per case basis. The physician contracts to accept assignments on all cases, or he receives assignment on none. Non-price incentives include direct payment to the physician, simplifying his billing; prompt payment; avoidance of bad debts on covered services; a predictable cash flow; access, in some Plans, to paperless processing; an improved competitive position or referrals; services of a field staff in the handling of problem claims, and improved patient relations. The Utilization Control System Utilization review historically has dealt with retrospective pattern analysis. An individual physician's patterns of practice are studied for significant deviations from the norm. Sometimes these are quite justif ted, somet imes not . If necessary, education and even recoupment of funds may be pursued. More recently, the emphasis in utilization control has shifted to concurrent review and preadmission certification. -71-
Benefit Design Most basic to benefit design is the need to provide outpatient alternatives to inpatient procedures. Carriers frequently draw a distinction between services which are physician-controlled and services which are patient-controlled. Thus first dollar coverage for inpatient medical care is common, whereas outpatient care is frequently covered only with coinsurance. Other aspects of a company's total benefit package may have to be examined. For example, it is extremely difficult to have a patient discharged in six days if the company' s disability program requires seven days of hospitalization for eligibility. Medical Policy Refusal to pay for routine laboratory and x-ray work, without specific indication, upon admission to a hospital, will, for example, have a perceptible effect on service charged to the carr ier. Contracts typically exclude experimental and investigational procedures, in the interest of overall community cost and quality of care. Physician Education Physicians generally do not wan~c to stand out f ram the norm as thigh chargers. ~ Frequently they are unaware that they do. Demonstrating to a physician that he is in the uppermost ranges of community charges will often result in a reduction of his charges. Claims Analys is and Intervention Claims analysis calls for a careful examination of claims experience to guide the application of education, benef it change, and occasionally (if the analysis is conducted by a contracting carrier) cancellation of the physician's contract. Solutions to specific problems such as mandatory second surgical opinions or preadmission certification also may be recommended. It is generally agreed that risk systems, involving the physician in the economic consequences of his decisions, can be successful in constraining cost. However, it is critical that the risk a physician accepts should be within his resources and within his control. A primary physician, for example, can manage his own time on a capitated basis. But he usually cannot by himself assume the risks of hospitalization and referred care for his patients. Thus an effective risk system requires either a group of physicians or at least a grouping of individual physicians together on paper for accounting purposes. -72-
There has been at least one pilot program in payment for medical, as distinct f ram surgical, care by diagnosis. The program encountered substant tat dif f iculty with secondary and tertiary diagnoses, and somewhat less difficulty with consultations and concurrent care. Some programs currently link the payment level for physicians with their utilization experience. If utilizaton is at or below a target level, per service payments rise more than they would if the targets were not met. If the utilization target is missed by a wide margin, payments do not r ise at al 1. There are also successful programs capitating primary care and paying for referred care only on demonstration of actual referral. This Gatekeeper function can be quite effective in reducing total expenditure. Referral care is difficult if not impossible to capitate in a f ree-choice system. Payment Systems Are Constrained By The Objectives Of The Buyer And By His Phi losophy And The Degree Of Committment To Those Ob jectives The usual demands made of a payment system are f ree-choice of physician, full payment of charges, and premium containment. Essentially, any payment system is capable of delivering two of these three. Indemnity offers free-choice and premium containment, although it does not necessarily contain costs for the individual patient. Most full payment programs offer both freedom of choice and payment in full, although this may be achieved by paying billed charges. Full payment and premium containment are available through HMOs and PPOs, although at sacrif ice of the unfettered choice of physician. It is possible to provide full payment, f reedom of choice, and premium containment within the same program. However, this requires exceptional design and administration, including hold harmless agreements and account-specific programs. It should be clear that payment determination is only one component of a financing system, and not necessarily the most important component. In current practice, most of the others temper the incentives of a payment system that tends to increase utilization. Some carriers are actively involved in capitation, risk, and selective contracting systems, not only to provide alternative methods of f inancing, but on the hypothesis that changes in practice patterns (admission criteria, lengths of stay) developed in such programs will carry over to conventional practice. Most buyers of coverage place high value on employee satisfaction and are reluctant to intrude on the patient-physician relationship. This attitude requires that the carrier develop programs that yield predictable results in the majority of cases and accommodate the delivery system as it is. Hold harmless programs, mandatory second opinions, 7~
) preadmission certification, and similar approaches can be implemented only with the cooperation and support of the buyer. 74
