« ПредыдущаяПродолжить »
[Dr. Rucker's prepared statement appears on p. 325.] Mr. NELLIS. Please proceed, Dr. Rucker.
Dr. RUCKER. Although the following illustration is now a decade old, Becker and associates found that general practitioners, on the average, used 270 different drug entities during the course of a single year (the range extended from 100 to almost 500). While the optimum number of drugs that can be employed effectively by a given physician will vary due to type of practice, years of experience, et cetera, the statistical probability of achieving this objective would seem to fall sharply once the limit of 60 to 100 is exceeded. Indeed, the hospital formularly study cited revealed that an unusually strong correlation0.94-obtained between the increased size of formularies and the proportion of inferior preparations carried. While the relationship between number of unique drugs ordered per year by a practitioner and the quality of those selections remains unknown, it seems foolish to disregard the negative relationship observed in the hospital survey.
Mr. NELLIS. Dr. Rucker, if I may interrupt, these 270 drug entities that you're speaking of, were they prescribed by general practitioners ?
Dr. RUCKER. They were prescribed by general practitioners, not by specialists. The study was confined to GP's. Mr. NELLIS. I see. Thank
you. Dr. RUCKER. Because of the time constraint, my analysis of prescribing practices has been restricted to the top of our therapeutic iceberg. Let me conclude this descriptive section by noting that I am unaware of a single study in the literature which has been able to document that prescriptions for psychoactive agents corresponded with the standards of rational use established by the investigator. Indeed, we could extend that generalization to all drugs. Since most surveys of prescribing behavior have been implemented within medical school environments, where care supposedly is superior to that provided in alternative settings, it seems reasonable to conclude that the general situation can only be worse than perceived. It would seem to be more productive, therefore, if the legislative bodies and researchers would devote less resources to the task of problem measurement and more to the development of constructive techniques that are likely to minimize it.
Consequently, I would like to turn to several recommendations. In my view, efforts to mitigate the irrational prescribing of psychoactive drugs are unlikely to be very effective (except perhaps in the imposition of production controls on amphetamines) if the problem is approached as a special case. Thus it is necessary for society to establish a comprehensive, coordinated program that is designed to support the rational prescribing of all marketed pharmaceuticals.
If this basic strategy is sound, it becomes essential to identify those impediments that inhibit the attainment of that goal. While the causes of suboptimum prescribing are numerous and their relative importance subject to disagreement, the three most significant, along with the corresponding recommendation, are:
(1) The absence of a comprehensive, current, and unbiased system of drug information that is readily available to support practitioner decisionmaking. In this connection, I have proposed the establishment of an integrated model in the form of a National Drug Education Foundation whose major objective is to help insure that prescribing
decisions are made in an environment of enlightened neutrality. In this connection, I would like to introduce into the record a description of this model information system that appeared in “Medical Care” in February 1976.
Mr. NELLIS. I'm sure the chairman would, without objection, admit it in the record and if you will give a copy to the reporter we would appreciate it. Dr. RUCKER. Fine. [The information referred to follows:]
DRUG INFORMATION FOR PRESCRIBERS AND DISPENSERS : TOWARD A MODEL SYSTEM
(Reprinted from Medical Care, Vol. 14, No. 2, February 1976]
(By T. Donald Rucker, Ph. D.)? Four alternative ways of providing health care professionals with drug information are examined in terms medical and social values. It is assumed that maximum patient benefit will result when prescribing decisions can be made in an environment of enlightened neutrality.
The alterative of maintaining the status quo is discarded because it does little to provide practitioners with complete data. The second alternative notes major reforms that would further regulate promotional efforts and upgrade the qualifications of detailmen. Because this approach yields a suboptimum solution, a third alternative introduces the concept of a National Drug Education Foundation. This countervailing force would disseminate objective drug information to clinicians through 2,000 therapeutic consultants, subsidize professional journals and schools, develop regional drug information centers, and conduct a variety of related functions. Foundation operations are estimated to cost $167 million per
ear or about one-seventh of current expenditures for drug promotion and information. Since commercial inputs would not only be redundant but also tend to negate Foundation efforts, the fourth model contends that social benefits can be optimized only when such outlays are terminated.
It is estimated that clinicians in the United States initiate over 1.4 billion new drug orders annually. The scientific basis for arriving at these decisions, although ultimately tempered by professional judgment,” stems essentially from two sources : (1) undergraduate training and (2) post-graduate education and information. In order to examine the medical and social utility associated with the impementation of this second function, we shall compare four alternative ways of providing such services. Our objective is to determine if one of these models appears to portend greater therapeutic and economic advantages for patients.
This question arises because many physicians (1, 6, 19] have identified limitations in the nature of drug information currently available to health care experts. Although three-fourths of the prescribers who responded to one local survey reported that present sources were either "efficient” or “highly efficient,"  critics from diverse backgrounds have stressed that contemporary methods
1 Professor of Pharmacy Administration, Division of Administrative and Social Sciences, College of Pharmacy, The Ohio State University, Columbus, Ohio 43210.
2 Some practitioners feel that medical school and actual practice training are the appropriate inputs for scientific knowledge in prescribing. If these are the key variables, then we deny the influence of promotion and advertising by pharmaceutical firms. Such a posture means that these outlays must be classified as an expenditure lacking medical value and hence economic justification.
More importantly, reliance on experience in actual practice seldom permits the control of participant bias as under the double-blind, crossover study, nor does it facilitate comparative assessment of alternative therapies. Further, in ambulatory care, the prescriber usually has no means of controlling intervening variables. This problem, coupled with the frequency of self-limiting conditions and polypharmacy, reduces the probability that the nature of patient response can be attributed to a given pharmaceutical agent.
The fact that FDA withdrew marketing approval for more than 400 ineffective drug products illustrates that these theoretical constraints have validity in the real world. In short, the prevalence of this general philosophy of casual empiricism among clinicians probably ranks among the top two factors leading to irrational prescribing. The other is the subject of this paper.
often yield results that are biased,  costly,  and incomplete.  Paradoxically, practitioners aso must contend with reuundancy in drug advertising and a plethora of studies that appear in the scientific literature. While avstracting services may mitigate this burden for some cinicians, the abiltiy of most to follow medical developments and to discriminate in terms of important therapeutic innovations would seem to be a very difficult task. Indeed, one specialist in neuropharmacology, who noted that a major part of his job responsibility was to monitor published contributions, has described the existence of this problem even in his narrow field. 
The direct connection between prescribing that fails to meet sound therapeutic criteria and the shortcomings of our system of drug information may not be apparent to nor accepted by all who are concerned about ways to improve patient care when prescription therapy may be indicated. The possibility that such a relationship may exist, though, was alluded to recently by Dr. Alexander M. Schmidt when he observed that "it would seem not unreasonable to believe that the regular barrage of messages encouraging drug use, without concomitant strong messages advising prudence, contributes to overuse." 
In order to proceed with our critique, this analysis assumes that the construction of a better system for furnishing practitioners with information on pharmaceutical preparations stands as a fundamental prerequisite for sin the effectiveness witn which therapeutic agents are used. Because prescribing practices are influenced by multiple factors, however, the impact of a model system on improving drug use relative to other measures remains indeterminate.
ASSUMPTIONS AND SYSTEM OBJECTIVES
In most instances, the ultimate objective in ordering prescription therapy is to help patients regain their normal health status or at least achieve some minimum level of functional ability. It is acknowledged, therefore, that certain pharmaceutical preparations represent a legitimate and essential form of treatment for many medical conditions. Thus, the necessity for attempting to define and develop a better information system is derived from the salient role that appropriate drug therapy can play in ameliorating illness conditions. Since the use of prescription products also may subject individuals to certain risks, especially when multiple agents are employed and/or labeling parameters are exceeded, it seems equally desirable that practitioners follow systematic steps to ensure that incidence of untoward results is minimized.
This review of alternative drug information systems takes as given the existing knowledge base governing medical and pharmaceutical science and our understanding of human behavior in dealing with complex situations. Within this framework, an optimum level of prescribing will be achieved when no other combination of medical and pharmaceutical data, delivery mechanism and public policy yields a higher probability that authorized treatment regimens can lead to a greater improvement in patient health. Among others, this condition implies that an efficient model would strive to minimize redundancy by supplying professionals with guidance on therapeutic matters only when such information seems essential for strengthening their performance.
Further, we assume that maximum patient benefit is most likely to occur when professional interpretations are based exclusively on the following factors. 1) an accurate evaluation of the patient's problems; 8 and 2) an objective evaluation of alternative courses of intervention. If drug therapy is suggested. the intrinsic properties of available medications, as established by official labeling, sanctioned by expert opinion, and conveyed by impartial means, should form the basis for decision making. Practitioner experience with particular agents, therefore, becomes important once these fundamental requirements have been satisfied. By definition, any impediments that tend to vitiate the operational integrity of this protocol-whether generated by insurance programs, pharmaceutical manufacturers, patients, or tradition-should be eschewed. Although it may be difficult to exclude subjective factors completely from this process, the best treatment plan for patients is most likely to result when the system can ensure that professional decisions are made in an environment of enlightened neutrality.
3 The selection of rational therapy may be hampered because the complexity of the patient's condition precludes a definitive or even accurate diagnosis.
• Drug products range in action from specific to placebo and each type may represent a bona fide form of chemical intervention.
The desirability of developing such a model might receive the concurrence of leading drug producers. This premise seems tenable in light of a public statement made on behalf of the National Pharmaceutical Council. In testimony before the U.S. Senate Subcommittee on Monopoly, the President of NPC held that any system failing “to respect the physician's sole responsibility for the therapy his patients receive and that permits outside parties to interfere with the physician-patient relationship ultimately endangers the health and safety of the patients.”  Since pharmaceutical manufacturers meet the criterion of “outside parties,” the conclusion may be drawn that promotionally oriented activities carried out by these firms could constitute a hazard that should be avoided.
If the first level of optimization pertains to the quality of therapeutic care, the second involves economic criteria. In order to achieve this objective, selective information on comparative product prices could be made available to prescribers. It is acknowledged, of course, that lower cost agents should be preferred only when scientific data indicate that the therapeutic response of the patient is unlikely to be compromised. Moreover, patient and dispensers might share in this responsibility if pharmacists would utilize and post a dispensing fee  and if states would relax restrictions contained in their antisubstitution laws. Finally, outlays for patient care related to drug therapy will be minimized when prescribing generally is consistent with the standards of appropriate use. In addition, it is necessary for the drug information system to reflect not only maximum cost effectiveness relative to alternative systems but a high level of operational efficiency as well. The former problem, as discussed below, appears to be solvable while the latter cannot be assessed until the model system has been specified in greater detail.
THE CONTEMPORARY SYSTEM
Our first option in providing professionals with drug information can be viewed as one of maintaining the status quo. If this position is accepted, any alterations would be confined to minor adjustments within the system. The major components of our current approach include:
(1) Persuasion and educational activities, direct and indirect, sponsored by more than 300 pharmaceutical firms which promote their products to the professions, largely through the itinerant visits of more than 23,990 detailmen. 
(2) Medical-pharmaceutical journals and news publications that appear at periodic intervals. It has been estimated that the typical physician receives 50 to 60 such mailings per month. Very few of these publications exist without the support of advertising, most of which is financed by drug producers. In addition, manufacturers spend significant sums on direct-mail advertisements. Industry support also underwrites the cost of compendial-type publications such as the popular Physicians' Desk Reference and Merck Manual.
(3) Communication with colleagues, many of whom however, may have obtained their drug expertise from reliance on the aforementioned sources. Further, prescribers also rely upon the experience of their patients in assessing the therapeutic properties of particular drugs.
(4) Continuing education programs sponsored by professional schools and societies. Many if not most of these symposia receive partial support through "educational grants” tendered by large drug firms.
(5) Independent references such as Walter Modell’s Drugs of Choice, the AMA's Drug Evaluations, the Medical Letter, and the FDA Drug Bulletin.
(6) The regulatory activities of FDA and FTC governing the therapeutic claims that appear in labeling and advertising.
The current system thus relies largely on the effectiveness of (5) and (6), coupled with some support from (4), to counter-balance and supplement the influence of commercial endeavors. Since this configuration is believed to manifest serious deficiencies, we turn to an intermediate solution.
5 The question of how to provide prescribers and purchasers with data on drug product costs and prescription prices, especially under the model outlined below, is a complex topic that necessitates separate discussion. It is essential that this general problem be approached from a total systems point of view. Otherwise, we run the risk, among others, that a number of clinicians could become so preoccupied with the economic aspects of prescription services that the time required for arriving at sound therapeutic decisions would be compromised.
6 Although many grants may appear to be given with "no strings attached," the former medical director of E. R. Squibb & Sons, Dr. Dale Console, has cautioned that drug company executives “.
make investments, not_gifts." In Hearings Before the Subcommittee on Monopoly. Competitive Problems in the Drug Industry, Part II, 1969, p. 4480.
This model holds that our basic problem is one of inadequate functional performance rather than one that necessitates complete restructuring of the total drug information system. Thus, it reflects the general principle that pharmaceutical manufacturers should continue to play a predominant role in the postgraduate education of prescribers and dispensers. Should excesses accompany promotional activities, they can be contained, it is held, by more stringent controls than are available through traditional regulatory mechanisms.
The current legislative posture seems to be moving in this direction and re. forms such as the following have been proposed by Senators Nelson and Kennedy: establish certification programs for detailmen; set strict limitations on the provision of drug samples; prohibit the use of gifts, products, and prizes as inducements to practitioners; forbid the review of prescription files by representatives of pharmaceutical manufacturers; strengthen Federal controls over drug advertising; and prepare a comprehensive drug compendium. It should be noted that several leading producers and the Pharmaceutical Manufacturers Association apparently support some of these measures.
Table 1.–National Drug Education Foundation: Functions and
proposed annual budget Functions
In millions Therapeutic consultants—2,000 consultants at average total cost per specialist of $30,000.
$60 Professional journals-subsidy
Medical, dental, and podiatric---
Drug compendium -
grams, ad hoc publications, materials for consultants, et cetera---
Determination of professional needs_
the operational dimensions of relative drug safety and efficacy,
of alternative communications techniques-
Executive staff, board support, personnel training center, rent, et cetera
167 1 Publication costs can be expected to fall significantly because of the elimination of pages formerly devoted to advertising and related solicitation.
Compendium expenses will be reduced after the initial edition has been developed. 3 Most of the proposed budget for patient education should probably be allocated to research. If cost-effectiveness studies suggest positive results, the relative emphasis can be changed to stress service programs.
THE NATIONAL DRUG EDUCATION FOUNDATION
Although the changes suggested above are predicated upon both logic and experience, a critical evaluation of our contemporary drug information system indicates that the perceived difficulties have deeper roots. Since the limitations of incomplete or biased data and excessive cost are believed to be indigenous to a communication network dominated by commercial sources, it may be useful to consider a more systematic means for furnishing heatlh care professionals with appropriate information about pharmaceutical products.