0277-9536/92 $5.00 + 0.00 Copyright 0 1992 Pergamon Press Ltd
Sot. Sci. Med. Vol. 35, No. 11, pp. 1397-1404, 1992 Printed in Great Britain. All rights reserved
BALANCING RELEVANCE AND EXCELLENCE: ORGANIZATIONAL RESPONSES TO LINK RESEARCH WITH DECISION MAKING* JULIO FRENK
National
Institute
of Public Health,
Av. Universidad 655, Col. Sta. Ma. AhuacatitlLn, Morelos, Mexico
62508 Cuernavaca,
faces the challenge of balancing relevance to decision making and excellence in the strict adherence to the norms of scientific inquiry. This paper examines the organizational responses that can be undertaken to promote integration of these potentially conflicting goals. We posit that there seem to be structural barriers to effective communication between researchers and decision makers, such as differences in priorities, time management, language, means of communication, integration of findings and definition of the final product of research. These barriers must be overcome through solutions aimed at the organization of research. In this respect, there are three possible models to approach the tension between excellence and relevance: academic subordination, segregation and integration. Only the latter makes it possible to reconcile the advantages of proximity to decision making with the procedures to assure
Abstract-Research
academic quality. In addition to organizational design and institutional development, a strategy to promote research must include a set of incentives to prevent the ‘internal brain drain’, that is, the tendency of researchers to move to managerial positions. There are four guiding principles to address this problem: parallel careers, academic autonomy, administrative sacrifice and inverted incentives. The complexities of health problems demand that we create new organizational formulas to finally balance relevance and excellence in research. Key
words-health
research, organizational design, brain drain, public policy
THE VALUES OF RESEARCH
When scientific research stopped being an activity carried out by isolated individuals working with more or less improvised tools, the problem of its organization became prominent. Just like many other areas of modern life, scientific research is undertaken in complex institutions that have a formal structure, an elaborate division of labour and a series of explicit procedures. As Blau pointed out [1, p. 11, Academe, the olive grove where Plato taught his philosophy, is still the ideal small community of learned people, but it in no way reflects the reality of research and higher education. Actually, it would be paradoxical that science, the ultimate expression of rationality, should escape from what Max Weber [2] saw as the great tendency towards the rationalization of social life; it is due to this tendency that a growing number
of human beings are born, spend most and die within complex organizations. Not all organizations carry out their equal quality, nor do they encourage the of the same values. Thus, the issue of
of their lives mission with development the organiz-
*This paper is based in part on a presentation at the International Symposium “New Directions in Health Care: From Research to Action”, Mexico City, 4 July 1986. A modified Spanish version of that presentation appeared in Ciencia 38, (3), 169-178, 1987.
ational design that is better adapted to scientific activity is of utmost importance. This issue, however, is not simple or direct. As Hake1 et al. indicate [3], there is a history, a sociology and even a psychology of science, but there is no science of science. The identification of the institutional niche that is best suited for something as complex as scientific research thus becomes a process of trial and error, charged with intuition, experience and even art. No matter whether one conceives of research as a utilitarian means for economic development or as an end in itself-as part of something more general called ‘culture’ [4]-one of the central problems of research organizations is to balance two fundamental values: excellence and relevance. Excellence means the strict adherence to a series of research rules that give objective validity to the results. It is not our purpose to analyze these rules. Moreover, there is disagreement among diverse epistemological schools regarding this issue. Nevertheless, most scientists have agreed on several theoretical and methodological attributes that allow them to judge the quality of a project. Evidence of consensus on the attributes of excellence is found in the many instances of peer review that give form to the scientific community. The second value, relevance to decision making, is the ability of research to take on problems that require a solution. For research to be relevant to decision making, more than one possible solution 1397
1398
JULIO FRENK
must exist, each with a different degree of effectiveness [5]; also, there must be some uncertainty as to the nature and effectiveness of the solutions. Thus, research produces knowledge, which reduces uncertainty. The decision maker may then utilize such knowledge, together with other considerations, to address the problem. We must state, from the outset, that the value of research relevance to decision making is subordinate to a superior value, which is the relevance of those decisions to the needs of the population. This last value assumes that the decision maker derives his or her power from a legitimate base, the ultimate source of which is to be found in the democratic process.* In the absence of such a legitimate base, the researcher and the decision maker could work in perfect harmony and yet tackle problems that are foreign or even contrary to the needs of the population. In other words, the value of relevance does not imply that the researcher accepts the decision maker’s goals or ideology uncritically. Part of this value lies in the capacity of research to identify the real needs of the population and, hence, the problems for which legitimate decisions must be made. Even if researchers and decision makers pursue similar legitimate goals, there still exists a tension between excellence and relevance. Excellence is emphasized by scientists, since it constitutes the basis for progress within a discipline. On the other hand, relevance tends to be emphasized by decision makers. The potential conflict between these values exists in almost all research settings, but is more clearly manifested in public institutions and in applied fields like public health. This paper analyzes the nature of such conflict, criticizing some common solutions and suggesting several organizational responses to ensure a synthesis of relevance and excellence in research, as well as a constructive dialogue between scientists and decision makers. SOURCES
OF POTENTIAL
CONFLICT
A fundamental process in the contemporary world has been the growing complexity of organizations, both public and private, which entails the need to make multiple decisions, each having important repercussions. A parallel process to the growth of organizations has been the explosion of scientific activity and knowledge. Given the coexistence of these two expansive processes, it is interesting to note that their paths do not always cross, especially in the public sector. Of particular concern is the
*This source of legitimacy applies mostly to decision makers in the public sector. Many of the most important decision makers are indeed government officials, although the concept comprises of a wider set of people, such as leaders of community organizations or even individual providers of services who must chose among alternative courses of action.
lack of government efforts to improve the quality of its decision making by means of a basis of systematic research. The result is that more knowledge is available than what is utilized for the pursuit of progress. In health services, for example, the deficiency is obvious [6]. Despite governmental efforts to expand population coverage and assume a growing proportion of the cost of care [7], needs and demands are still greater than available resources. Therefore, explicit or implicit decisions made by the government to allocate these resources affect equity and quality of life. In spite of all this, and of the fact that the health field has achieved substantial scientific development. many decisions are still made, in most countries, without an in-depth analysis of the real effects that programs have on health needs [8]. A major factor contributing to this gap lies in the differences between researchers and decision makers, particularly the relative value each group places on excellence and relevance. The literature on this topic is highly polarized. On the one hand, there are researchers who protest any attempt on the part of decision makers to violate the borders of the Republic of Science mentioned by Polanyi [9]; on the other hand, there are decision makers who criticize scientists for their lack of commitment to ‘real problems’ [6]. This kind of unproductive exchange fails to recognize the structural obstacles that hinder an effective integration of research and decision. These obstacles have little to do with the personalities of the individuals involved in the conflict; their roots are the different logic and demands characterizing the respective spheres of action of researchers and decision makers. The first potential conflict refers to the definition of priorities. The perception of decision makers about the most important problems may not coincide with the topics of greater scientific interest to researchers. A second obstacle has to do with real differences between political time and scientific time. In general terms, the decision maker is chronophobic, that is, a person for whom time is an enemy to be conquered as he or she is subject to the pressures of “policy formation, program administration, and problem solution” [lo, p. 191. The researcher, in contrast, is chronophilic-he or she considers time to be a central ingredient of research, since it allows for the full manifestation of the processes under study. There are also important differences in language and means of communication. The researcher is mostly concerned with communicating results in a precise manner, in order to enrich the paradigms of his or her field. This means that scientific language is often esoteric and communication occurs through specialized publications which can be found only through a detailed bibliographic search. For the decision maker, it is important that results be expressed in comprehensible language and be easily accessible.
Relevance and excellence in research A related problem is the integration of results from different sources. Growing specialization demands that researchers address clearly defined questions, even if this means a fragmentation of the object of analysis. In contrast, decision makers require comprehensive answers to a whole problem and not just to one of its researchable parcels. While research is often guided by an analytical impulse, sound decision making depends on the synthesis of evidence from various studies. Finally, there is a difference in the perception about the nature of the final product of research. For the scientific community, the product is the published article and the influence this might have on the ideas of others (measured, for example, by the number of bibliographic citations). For the decision maker, research does not end until it has had influence on a decision. To a great extent, this difference is due to dissimilar expectations about the certainty of research results. As May has pointed out [ll], the purpose of research is not so much to satisfy the demand for knowledge as to increase it, since good projects tend to give rise to new questions. The aforementioned conflicts would not be of great consequence if it were not for the fact that they have negative effects on everyone involved. The inadequate use of research findings in decision making not only impoverishes the latter, but also leads to reduced support for research. This, in turn, generates low scientific production, which completes the vicious cycle towards low utilization of research [12]. It is clear that this problem affects all types of research. However, due to its proximity to the decision making process, applied research manifests the contradictions described earlier in a more distinct way and demonstrates the need to develop creative solutions. STRATEGIES
FOR SOLUTION
Orientation of research Most discussions on the application of research results to decision making assign a central role to education, both of researchers, so that they may learn to understand the needs of decision makers, and of managers, so that they may learn to appreciate and utilize research [I l-131. It is difficult to doubt the urgent need for effective educational strategies. We must point out, however, that most obstacles encountered by decision makers and investigators are of a structural nature. Thus, overcoming them requires solutions directed towards the organization of research. The main strategy in this sense is to abandon traditional positions of confrontation and find formulas for integration. In the search for such formulas it is important to make a distinction with respect to the primary orientation of research. Broadly speaking, we can identify two major orientations. On the one hand, there are projects where the main purpose
1399
is to solve a problem. In the health field, we can call this activity research for health. On the other hand, we also find research oriented primarily towards the advancement of knowledge; we can call it research on health [ 141. This distinction corresponds closely to the more conventional division between basic and applied research, but it tends to emphasize the primary intention of the researcher. In any event, its purpose is simply to synthesize differences that are never so clear-cut in real life. Indeed, many research projects attempt both to advance knowledge and to solve problems. Nevertheless, differences in primary orientation are important on two dimensions: prospects for generalization and prospects for immediate usefulness. Research that is primarily oriented to problem solving attempts to answer a question in a way that is relevant to a particular situation and in the short term. When research aims to find results that are generalizable, it can be said to contribute to the general fund of knowledge. The report of the Commission on Health Research for Development makes a similar distinction, when it calls for two complementary research approaches: countryspecific health research, which looks for shortterm solutions to national problems, and global health research, which seeks new and generalizable knowledge [ 15, pp. 20-221. Alternative models The importance of determining the primary orientation of a project is that it will probably influence the effectiveness of different strategies to integrate relevance and excellence. In analyzing such strategies, it is useful to consider the range of possible responses. In principle there are three possible models to address the tension between these two values. Although each one is an attempt to adapt to different circumstances, they do not all have the same ability to integrate relevance and excellence. The first model can be called ‘academic subordination’. As its name indicates, this option implies that the research process is totally dependent on the requirements of decision makers. They define the topics, times and products of research, while the investigators merely carry out their specifications. This model is taken from research and development departments in industry. Although it leads to maximum relevance, the model of academic subordination presents enormous obstacles to excellence, since it violates several principles that will be discussed later on. Demeny has cogently analyzed, for the specific case of social science and population policy, the shortcomings of academic subordination, among them the risk that research may serve merely to legitimize decisions which have already been made [16]. The second option could be called the ‘segregation solution.’ In this case, the tension between excellence and relevance is not eliminated but institutionalized.
1400
JULIO FRENK
Specifically, the research center is divided into two subunits: on the one hand, a nucleus of excellence, which carries out long-term investigations oriented towards the advancement of knowledge; on the other hand, there are peripheral short-term projects oriented towards the investigation of problems that are relevant to the decision makers. This model was applied, among other cases, to the organization of the first economic research unit within the U.S. Department of Health, Education and Welfare in the 1960s [17]. The explicit purpose of this solution is for the organization as a whole to attempt reconciling excellence and relevance. Its enormous disadvantage is that it stratifies projects and researchers, running the risk of dividing them into first and second class categories. The last option is the ‘integrative solution.’ Instead of segregating relevant from excellent projects, this model seeks an integration of both values within each project. This implies a systematic effort to have each project solve a problem as well as advance knowledge. No doubt, this is the most difficult and complex option to implement. To be successful, it requires an organizational design that brings together the advantages of proximity to decision making and the structures, procedures and incentives developed by universities to ensure academic quality. Formulas for integration A design of this sort must build explicit bridges between decision makers and researchers, in order to prevent the potential conflicts that we referred to before. An example of such a bridge is the proposal developed by the World Health Organization, called decision-linked research [12]. The essence of this approach is to achieve interactions between decision makers and investigators, in order to increase the chance of decisions being made on the basis of trustworthy information, so as to benefit all of society, including, of course, the researchers themselves. The interaction between decision makers and researchers can happen at different points in the course of a reseach project. According to WHO, there is a ‘primary link’ between research and decision when the interaction occurs before the project begins. Thus, the starting point is a diagnosis of the needs of the decision making process. Primary links are characteristic of projects oriented towards the solution of problems. On the other hand, WHO speaks of a ‘secondary link’ when the starting point is a project that is initiated by the researcher. This type of link is more frequent in projects that are oriented towards the advancement of knowledge. Yet even in this case it should be possible to establish a significant interaction with decision makers, which can happen when the project is in the planning stage of formulating a protocol, when it is in the data collection or analysis phase, or when it has been completed. It is obvious that the sooner the link is established, the
more easily the project can be modified according to the needs of decision makers. If a research center wants to maintain a high level of relevance, it is necessary that all its projects establish effective links with the decision making process. The primary orientation of the projects and the nature of the research topic will determine the phase at which the link may be established. In any case, what matters is that the research unit be designed so as to encourage interaction between researchers and decision makers by overcoming the structural obstacles that were identified earlier. Figure 1 proposes a series of means to supersede each of those obstacles. Thus, one possible solution to the discrepancies in the definition of priorities is to have decision makers in the governing or consulting bodies of the research center, so that their needs can be expressed and opportunities can be identified within ongoing projects. Likewise, the time management differences we discussed before could be minimized by assuring the collaboration of researchers and decision makers since the earliest planning stages of a research project. This would facilitate the negotiation of deadlines required in order to produce opportune results. Another way of dealing with these differences is for researchers to offer intermediate products such as bibliographic revisions and progress reports that may help decision making even before the project is completed. Communication barriers can be reduced if, besides specialized articles, research projects produce ‘executive syntheses’ that outline results relevant to decision making in a comprehensible way. This solution would imply developing a system of dual reporting of results: academic papers for scientific colleagues and executive syntheses for decision makers. A complement to such syntheses would be the organization of participatory seminars where decision makers and researchers would jointly discuss results. To encourage this kind of exchange, May [1 I] suggests the training of ‘translators’ who can offer an interface between the two groups, taking into account the needs, values and priorities of each one. The idea is not far-fetched if one considers that this kind of translator already exists for the general public in the form of the science writer. In the same way, there should be a professional who could facilitate communication between scientists and decision makers, translating research results into policy and action recommendations. The problem of integrating research results can be addressed through two approaches. The first one is a detailed review of evidence, including the use of quantitative techniques of data synthesis such as meta-analysis [18-211. The second approach can be called ‘mission-oriented research’ [22,23]. This concept refers to research that aims at integrating not only different disciplines, but also different levels of analysis around a specific problem. For example,
Relevance
and excellence
in research
1401
MEANSTO SUPERSEDETHE BARRIERS
SOURCESOF POTENTIALBARRIERS BETWEENRESEARCHERS AND DECISIONMAKERS
Definition
of
- Presence governing
priorities
of decision makers in or consulting bodies
of research
-
centers
Collaboration
between
researchers Time
makers
management
since
stages
-
Language of
and accesibility
results
and
decision
the
early
planning
of a project
Identification
of
intermediate
products
of research
- Executive
syntheses
-
“Translators” policy
-
Joint
of research
seminars
into
for analysis
of
results
Integration about
of
different
the same
findings
problem
-
Meta-analysis
-
Mission-oriented
-
Definition objectives
Perceptions product vs
Fig. 1. Sources
about
the
of research:
final
scientific
utilization
in addition objectives
to of
research
discovery
-
decision
and solutions
of
research
of potential
barriers
in order to fully understand and act upon a health problem it may be necessary to integrate information on the subindividual level (through biomedical research), the individual level (through clinical research), the population level (through epidemiologic research) and the societal level (through health systems research). Finally, the discrepancy about the final product of research can be solved by specifying, since the initial formulation of every research project, a series of precise objectives for the application of results, together with the strictly scientific objectives [18]. In this way, the utilization of knowledge could stop being a haphazard event and become instead a programmed phase of the research process. This purpose would require that the scientific community give specific weight to application efforts in its evaluation of researchers’ performance, something that is not current practice. To a large extent, the debate on the utilization of knowledge leaves in the hands of the individual actors the implementation of solutions like the ones that have been suggested here. However, the present complexity in the production of knowledge demands
Greater weight to relevance evaluation of researchers
between
researchers
in
and decision
makers.
organizational solutions. It is necessary for research institutions to have differentiated structures that will facilitate the dynamic and creative linkage with their environment, especially the capacity to project scientific knowledge towards decision making. The structures and procedures proposed until now have the purpose of preserving the value of relevance. In fact, many of them have been tried out with success first at the Center for Public Health Research and later at the National Institute of Public Health in Mexico [24]. However, they are not enough for the ‘integrative solution’. In order to also preserve the value of excellence, it is necessary to adopt structures that promote academic quality, such as peer-review bodies. Another requirement for excellence is that, once the needs of the decision makers have been determined and a set of pertinent projects planned to satisfy those needs, then the responsibility for carrying out the projects should be left entirely to the researchers. That is, there should be no interference in the selection and execution of methods, and much less in the analysis of results. Thus, to the values of relevance and excellence it is necessary to add independence in the execution of projects. Actually,
1402
JULIOFRENK
this value is indispensable for the other two. It is obvious that, for an investigation to have scientific validity, it cannot be manipulated in order to reinforce the interests or ideology of decision makers [ 161. Guiding principles *
The careful design of structures and procedures would be useless if incentives were not also modified to stimulate research development. Such incentives could diminish, at least in part, the risk of “internal brain drain” [25]. This phenomenon, which is probably greater than the loss of researchers to foreign countries, refers to the abandonment of scientific activities for administrative ones. This happens when researchers find themselves in a system that gives greater privileges to those in the latter sphere of activity. Nobody doubts that an investigator who has achieved a certain stage of maturity in his or her career can occupy positions of managerial leadership, especially within the research field. Likewise, the temporary appointment, on a rotating basis, of applied researchers to decision making positions can enrich their understanding of reality. The problem arises when economic and power rewards overcome the researcher’s preferences for academic work and force him or her to shift to an administrative career. It is obvious that the internal brain drain affects all research institutions, particularly in underdeveloped countries. However, the problem is especially acute in applied research centers, since they deal with problems that are close to the concerns of the same administration towards which the drain occurs. In order to preserve the positive aspects of this proximity and at the same time reduce the risk of brain drain, the design of research organizations must include adherence to four guiding principles: 1. 2. 3. 4.
The The The The
principle principle principle principle
of of of of
parallel careers. academic autonomy. administrative sacrifice. inverted incentives.
The principle of parallel careers. As a result of a tendency to generalize vertical structures of organization, a hierarchical order is often imposed within research organizations. This hierarchy is formed, in descending order, by the director, associate directors, heads of departments and, finally, the researchers. In contrast, the first principle being proposed requires that the research career be considered parallelneither superior nor inferior-to the administrative career. Undoubtedly, investigators work within an administrative unit and, hence, are subject to its discipline. What we wish to emphasize here is that the researcher must have a clear career line, including distinct levels of assistant, associate and full re*This discussion is based, with modifications, on a portion of the paper ‘A conceptual model for public health research’ (PAHO Bullerin 22, (l), 60-71, 1988).
searcher. This line should be clearly differentiated from the administrative line, so that mobility is towards higher levels within research and not towards management positions. Actually, both careers operate under very different organizational assumptions. In administration, authority is exerted hierarchically and derives primarily from the position in the organization. In the academic career, authority comes mainly from knowledge and scientific productivity; it is exerted in a horizontal manner and is meant to inspire, guide and lead rather than command a colleague. The principle of academic autonomy. In theory, one way to follow the principle of parallel careers would be to differentiate the two lines not only functionally, but also professionally. This would imply that, just as researchers occupy academic positions, only professional administrators would be in management positions. Experience has demonstrated, however, that this approach is not satisfactory. Academic work has some characteristics that are not easily appreciated by those who have not been directly involved in it. One of those special characteristics is the fact that research requires a high degree of autonomy, since most of the intellectual process that leads to a discovery is unknown and, therefore, cannot be codified nor reduced to routines. Perhaps due to the complexity of academic work, researchers usually recognize, first and foremost, the authority derived from knowledge and experience. As a consequence, when a nonscientist directs a center of scientific production, there is often a high degree of mutual distrust between researchers and managers, which may lead to conflict. The only way to avoid this conflict is to respect the academic autonomy of investigators, and this necessarily implies that management positions in research institutions should be occupied by researchers. The principle of administrative sacrifice. There is an obvious contradiction between the two previous principles. While one suggests that research and management careers must be kept separate, the other points to the need for management positions at research centers to be occupied by researchers. The solution to this contradiction lies in the principle of administrative sacrifice, which states that occupying a management position should not be a reward for the researcher, but a sacrifice done for his colleagues-and for the scientific community in general-in order to preserve the principle of academic autonomy. The principle of inverted incentives. If the principle of administrative sacrifice is not to be a mere rhetorical statement, it must have a practical corollary. This corollary is the principle of inverted incentives. Its premise is that the present system of incentives continuously impels researchers to abandon their academic careers and transfer to administrative ones, thus contributing to the internal brain drain. At its most basic level, this phenomenon is expressed in the
Relevance
and excellence
considerable salary differences favoring managerial over research positions in many developing countries. The principle of inverted incentives consists of reversing this state of affairs so that, at each level in the research career line, there will be a greater incentive to advance to the next level than to move into a managerial position. A practical example of this principle is offered by the National Researchers System in Mexico [IS, p. 481. This publicly funded yet independent organization was created in 1984 in order to contain the brain drain that the economic crisis had exacerbated. Through a very strict peer-review process, active researchers are classified into one of four levels, each of which is accompanied by a tax-free stipend consisting of a certain number of minimum wages. This stipend is added to the regular salary of the researcher. Apart from offering higher remuneration for researchers who work outside Mexico City, the system also creates an incentive not to abandon research for managerial positions. Thus, researchers who are working in management may still be accredited by the system but receive no remuneration; once they return fully to research, they may benefit from the stipend. The final aim of the proposed guiding principles is to break the vicious circle-so common in developing countries-between the lack of adequate incentives for researchers and the weakness of research organizations. Strengthening its human and institutional infrastructure is a precondition for research to enrich decision making. CONCLUSIONS
In an ideal situation, the dilemma between relevance and excellence should be false. Even in our imperfect conditions, no reasonable decision maker rejects research as a contribution to increase the quality of his or her decisions. And even the most basic researcher will feel great satisfaction when seeing the results of his or her work applied to the solution of a problem. In this respect, Donabedian has written: I‘. the world of ideas and the world of action are not separate . . ., but inseparable parts of each other. Ideas, in particular, are the truly potent forces that shape the tangible world. The man and woman of action have no less responsibility to know and understand than does the scholar” [26]. Nevertheless, the tendency towards institutional fragmentation determines that sometimes research becomes artificially separated from decision making. Because of this, it is necessary to make an explicit effort to bring together both worlds, while allowing each one to maintain its autonomy. Research is a highly complex undertaking that requires specific conditions to flourish, including institutions and incentives to pursue the highest standards of excellence in order to create knowledge that decision makers may actively demand and utilize as informed con-
in research
1403
sumers. These are the only circumstances under which research will be able to make its distinct and irreplaceable contribution to human well-being [16]. In the health field, the growing gap between aspirations and accomplishments has brought research to the forefront of development strategies. The Commission on Health Research for Development embodies the international perception that research has become an essential component of any future advance. In its highly influential report-which will hopefully help to mobilize international resources towards research-the Commission has made an explicit call for promoting institution building and capacity strengthening in developing countries [15]. In this process, every nation must create the organizational arrangements that will make it possible to finally integrate relevance and excellence. Acknowledgements-I
am grateful to James Trestle, Avedis Donabedian and Harvey Fineberg for their valuable com-
ments to a previous version of this paper. Several of the concepts proposed here were enriched by discussions with Guillermo Sober&, Jest Laguna, Jaime Martuscelli, Jest Luis Bobadilla, Jaime Sephlveda, Enrique Ruelas and Miguel Angel Gonzhlez-Block. The international symposium where this work was first presented had the financial support of the Ministry of Health of Mexico, the Pan American Health Organization, the United States Agency for International Development and the British Council. REFERENCES
I. Blau P. M. The Organization of Academic Work. Wiley, New York, 1973. 2. Weber M. The Theory of Social and Economic Organi:ation. Oxford University Press, New York, 1947. 3. Hake1 M. D., Sorcher M., Beer M. and Moses J. L. Making it Happen: Designing Research wilh Implemen ration in Mind. Sage, Beverly Hills. 1982. R. La crisis perpetua. In M&sico Maliana 4. Ptrez-Tamayo (Edited by Aguilar-Camin H.). p. 215. O&no-Nexos, Mexico, 1986. 5. Ackoff R. L. The Arr of Problem Solving. Wiley. New York, 1978. health 6. McEwan E. D. A case for government-sponsored care research and development in the formulation of health policy and an account of early experience of government-sponsored health care research in one jurisdiction. Int. J. Hlth Serv. 3, 45, 1973. A. State intervention in 7. Frenk J. and Donabedian medical care: types, trends and variables. H//h PO/. Plan. 2, 17, 1987. A. L. f$tic~iL~eness and &ficiettcy: Random 8. Cochrane Reflections on Health Setvices. Nuffield Provincial Hospitals Trust, London, 1972. 9. Polanyi M. The republic of science. In Crireria /itt Scienrific Development (Edited by Shils E.), p. 9. MIT Press,.Cambridge, Mass., 1968. _ IO. Hauser P. M. and Duncan. 0. D. (Eds) The Studs of Popularion: An Itnrenrory and Appraisal: University di Chicago Press, Chicago, 1959. The policy uses of research. II. May J. J. Symposium: Introduction. Inquiry 12, 228. 1975. Improving health care 12. World Health Organization. through decision-linked research: Application in health systems and manpower development. Part II: Options for implementation. Document No. HMD/86.4.2. World Health Organization, Geneva, 1986.
1404
JULIO FRENK
13. Brownlee A. T. Applied research as a problem-solving tool: Strengthening-the interface between health management and research. J. Hlth Admin. Ed. 4, 31, 1986. 14. Fienk J., Bobadilla J. L., Sepulveda J., Rosenthal J. and Ruelas E. A conceptual model for public health research. Bull. Pan Am. Hlth Org. 22, 60, 1988. 15. Commission on Health Research for Development. Health Research: Essential Link to Equity in Development. Oxford University Press, New York, 1990. 16. Demeny P. Social Science and Population Policy. Center for Policy Studies, Working Papers Series No. 138. The Population Council, New York, 1988. 17. Grosse R. N. Personal communication. 1985 18. Yach D. and Dick J. Implementation of research: The key to closing the gap between public health knowledge and action. Unpublished document. 19. Glass G. V. Primary, secondary, and meta-analysis of research. Ed. Res. 5, 3, 1976. 20. Louis T. A., Fineberg H. V. and Mosteller F. Findings for public health from meta-analysis. Ann. Rev. Publ. Hlth 6, 1, 1985.
21. Thacker S. B. Meta-analysis: A quantitative approach to research integration. J.A.M.A. 259, 1685, 1988. 22. Bobadilla J. L., Frenk J., Martinez-Palomo A. and Sepilveda J. Health research in Mexico: Strengths, weaknesses and gaps. Background paper prepared for the Commission on Health Research for Development, 1988. 23. Frenk J. La salud perinatal coma modelo para la investigation con mision. In Memorias de/ Primer Connreso National sobre Defeclos al Nacimiento. Grupo _ de Estudios al Nacimiento; Mexico City, 1990. 24. Frenk J., Bobadilla J. L., Septilveda J., Rosenthal J., Ruelas E., Gonzalez-Block M. A. and Urrusti J. An innovative approach to public health research: the case of a new center in Mexico. J. Hlth Admin. Ed. 4. 461. 1986. 25. Frenk J. Los medicos y la politica: fuga de cerebros. Mundo MPd. (Mex) 3, (31), I, 1976. 26. Donabedian A. The Baxter American Foundation Prize Address. J. Hlth Admin. Ed. 4, 611, 1986.
Sot. Sci. Med. Vol. 35, No. 11, pp. 1397-1404, 1992 Printed in Great Britain. All rights reserved
BALANCING RELEVANCE AND EXCELLENCE: ORGANIZATIONAL RESPONSES TO LINK RESEARCH WITH DECISION MAKING* JULIO FRENK
National
Institute
of Public Health,
Av. Universidad 655, Col. Sta. Ma. AhuacatitlLn, Morelos, Mexico
62508 Cuernavaca,
faces the challenge of balancing relevance to decision making and excellence in the strict adherence to the norms of scientific inquiry. This paper examines the organizational responses that can be undertaken to promote integration of these potentially conflicting goals. We posit that there seem to be structural barriers to effective communication between researchers and decision makers, such as differences in priorities, time management, language, means of communication, integration of findings and definition of the final product of research. These barriers must be overcome through solutions aimed at the organization of research. In this respect, there are three possible models to approach the tension between excellence and relevance: academic subordination, segregation and integration. Only the latter makes it possible to reconcile the advantages of proximity to decision making with the procedures to assure
Abstract-Research
academic quality. In addition to organizational design and institutional development, a strategy to promote research must include a set of incentives to prevent the ‘internal brain drain’, that is, the tendency of researchers to move to managerial positions. There are four guiding principles to address this problem: parallel careers, academic autonomy, administrative sacrifice and inverted incentives. The complexities of health problems demand that we create new organizational formulas to finally balance relevance and excellence in research. Key
words-health
research, organizational design, brain drain, public policy
THE VALUES OF RESEARCH
When scientific research stopped being an activity carried out by isolated individuals working with more or less improvised tools, the problem of its organization became prominent. Just like many other areas of modern life, scientific research is undertaken in complex institutions that have a formal structure, an elaborate division of labour and a series of explicit procedures. As Blau pointed out [1, p. 11, Academe, the olive grove where Plato taught his philosophy, is still the ideal small community of learned people, but it in no way reflects the reality of research and higher education. Actually, it would be paradoxical that science, the ultimate expression of rationality, should escape from what Max Weber [2] saw as the great tendency towards the rationalization of social life; it is due to this tendency that a growing number
of human beings are born, spend most and die within complex organizations. Not all organizations carry out their equal quality, nor do they encourage the of the same values. Thus, the issue of
of their lives mission with development the organiz-
*This paper is based in part on a presentation at the International Symposium “New Directions in Health Care: From Research to Action”, Mexico City, 4 July 1986. A modified Spanish version of that presentation appeared in Ciencia 38, (3), 169-178, 1987.
ational design that is better adapted to scientific activity is of utmost importance. This issue, however, is not simple or direct. As Hake1 et al. indicate [3], there is a history, a sociology and even a psychology of science, but there is no science of science. The identification of the institutional niche that is best suited for something as complex as scientific research thus becomes a process of trial and error, charged with intuition, experience and even art. No matter whether one conceives of research as a utilitarian means for economic development or as an end in itself-as part of something more general called ‘culture’ [4]-one of the central problems of research organizations is to balance two fundamental values: excellence and relevance. Excellence means the strict adherence to a series of research rules that give objective validity to the results. It is not our purpose to analyze these rules. Moreover, there is disagreement among diverse epistemological schools regarding this issue. Nevertheless, most scientists have agreed on several theoretical and methodological attributes that allow them to judge the quality of a project. Evidence of consensus on the attributes of excellence is found in the many instances of peer review that give form to the scientific community. The second value, relevance to decision making, is the ability of research to take on problems that require a solution. For research to be relevant to decision making, more than one possible solution 1397
1398
JULIO FRENK
must exist, each with a different degree of effectiveness [5]; also, there must be some uncertainty as to the nature and effectiveness of the solutions. Thus, research produces knowledge, which reduces uncertainty. The decision maker may then utilize such knowledge, together with other considerations, to address the problem. We must state, from the outset, that the value of research relevance to decision making is subordinate to a superior value, which is the relevance of those decisions to the needs of the population. This last value assumes that the decision maker derives his or her power from a legitimate base, the ultimate source of which is to be found in the democratic process.* In the absence of such a legitimate base, the researcher and the decision maker could work in perfect harmony and yet tackle problems that are foreign or even contrary to the needs of the population. In other words, the value of relevance does not imply that the researcher accepts the decision maker’s goals or ideology uncritically. Part of this value lies in the capacity of research to identify the real needs of the population and, hence, the problems for which legitimate decisions must be made. Even if researchers and decision makers pursue similar legitimate goals, there still exists a tension between excellence and relevance. Excellence is emphasized by scientists, since it constitutes the basis for progress within a discipline. On the other hand, relevance tends to be emphasized by decision makers. The potential conflict between these values exists in almost all research settings, but is more clearly manifested in public institutions and in applied fields like public health. This paper analyzes the nature of such conflict, criticizing some common solutions and suggesting several organizational responses to ensure a synthesis of relevance and excellence in research, as well as a constructive dialogue between scientists and decision makers. SOURCES
OF POTENTIAL
CONFLICT
A fundamental process in the contemporary world has been the growing complexity of organizations, both public and private, which entails the need to make multiple decisions, each having important repercussions. A parallel process to the growth of organizations has been the explosion of scientific activity and knowledge. Given the coexistence of these two expansive processes, it is interesting to note that their paths do not always cross, especially in the public sector. Of particular concern is the
*This source of legitimacy applies mostly to decision makers in the public sector. Many of the most important decision makers are indeed government officials, although the concept comprises of a wider set of people, such as leaders of community organizations or even individual providers of services who must chose among alternative courses of action.
lack of government efforts to improve the quality of its decision making by means of a basis of systematic research. The result is that more knowledge is available than what is utilized for the pursuit of progress. In health services, for example, the deficiency is obvious [6]. Despite governmental efforts to expand population coverage and assume a growing proportion of the cost of care [7], needs and demands are still greater than available resources. Therefore, explicit or implicit decisions made by the government to allocate these resources affect equity and quality of life. In spite of all this, and of the fact that the health field has achieved substantial scientific development. many decisions are still made, in most countries, without an in-depth analysis of the real effects that programs have on health needs [8]. A major factor contributing to this gap lies in the differences between researchers and decision makers, particularly the relative value each group places on excellence and relevance. The literature on this topic is highly polarized. On the one hand, there are researchers who protest any attempt on the part of decision makers to violate the borders of the Republic of Science mentioned by Polanyi [9]; on the other hand, there are decision makers who criticize scientists for their lack of commitment to ‘real problems’ [6]. This kind of unproductive exchange fails to recognize the structural obstacles that hinder an effective integration of research and decision. These obstacles have little to do with the personalities of the individuals involved in the conflict; their roots are the different logic and demands characterizing the respective spheres of action of researchers and decision makers. The first potential conflict refers to the definition of priorities. The perception of decision makers about the most important problems may not coincide with the topics of greater scientific interest to researchers. A second obstacle has to do with real differences between political time and scientific time. In general terms, the decision maker is chronophobic, that is, a person for whom time is an enemy to be conquered as he or she is subject to the pressures of “policy formation, program administration, and problem solution” [lo, p. 191. The researcher, in contrast, is chronophilic-he or she considers time to be a central ingredient of research, since it allows for the full manifestation of the processes under study. There are also important differences in language and means of communication. The researcher is mostly concerned with communicating results in a precise manner, in order to enrich the paradigms of his or her field. This means that scientific language is often esoteric and communication occurs through specialized publications which can be found only through a detailed bibliographic search. For the decision maker, it is important that results be expressed in comprehensible language and be easily accessible.
Relevance and excellence in research A related problem is the integration of results from different sources. Growing specialization demands that researchers address clearly defined questions, even if this means a fragmentation of the object of analysis. In contrast, decision makers require comprehensive answers to a whole problem and not just to one of its researchable parcels. While research is often guided by an analytical impulse, sound decision making depends on the synthesis of evidence from various studies. Finally, there is a difference in the perception about the nature of the final product of research. For the scientific community, the product is the published article and the influence this might have on the ideas of others (measured, for example, by the number of bibliographic citations). For the decision maker, research does not end until it has had influence on a decision. To a great extent, this difference is due to dissimilar expectations about the certainty of research results. As May has pointed out [ll], the purpose of research is not so much to satisfy the demand for knowledge as to increase it, since good projects tend to give rise to new questions. The aforementioned conflicts would not be of great consequence if it were not for the fact that they have negative effects on everyone involved. The inadequate use of research findings in decision making not only impoverishes the latter, but also leads to reduced support for research. This, in turn, generates low scientific production, which completes the vicious cycle towards low utilization of research [12]. It is clear that this problem affects all types of research. However, due to its proximity to the decision making process, applied research manifests the contradictions described earlier in a more distinct way and demonstrates the need to develop creative solutions. STRATEGIES
FOR SOLUTION
Orientation of research Most discussions on the application of research results to decision making assign a central role to education, both of researchers, so that they may learn to understand the needs of decision makers, and of managers, so that they may learn to appreciate and utilize research [I l-131. It is difficult to doubt the urgent need for effective educational strategies. We must point out, however, that most obstacles encountered by decision makers and investigators are of a structural nature. Thus, overcoming them requires solutions directed towards the organization of research. The main strategy in this sense is to abandon traditional positions of confrontation and find formulas for integration. In the search for such formulas it is important to make a distinction with respect to the primary orientation of research. Broadly speaking, we can identify two major orientations. On the one hand, there are projects where the main purpose
1399
is to solve a problem. In the health field, we can call this activity research for health. On the other hand, we also find research oriented primarily towards the advancement of knowledge; we can call it research on health [ 141. This distinction corresponds closely to the more conventional division between basic and applied research, but it tends to emphasize the primary intention of the researcher. In any event, its purpose is simply to synthesize differences that are never so clear-cut in real life. Indeed, many research projects attempt both to advance knowledge and to solve problems. Nevertheless, differences in primary orientation are important on two dimensions: prospects for generalization and prospects for immediate usefulness. Research that is primarily oriented to problem solving attempts to answer a question in a way that is relevant to a particular situation and in the short term. When research aims to find results that are generalizable, it can be said to contribute to the general fund of knowledge. The report of the Commission on Health Research for Development makes a similar distinction, when it calls for two complementary research approaches: countryspecific health research, which looks for shortterm solutions to national problems, and global health research, which seeks new and generalizable knowledge [ 15, pp. 20-221. Alternative models The importance of determining the primary orientation of a project is that it will probably influence the effectiveness of different strategies to integrate relevance and excellence. In analyzing such strategies, it is useful to consider the range of possible responses. In principle there are three possible models to address the tension between these two values. Although each one is an attempt to adapt to different circumstances, they do not all have the same ability to integrate relevance and excellence. The first model can be called ‘academic subordination’. As its name indicates, this option implies that the research process is totally dependent on the requirements of decision makers. They define the topics, times and products of research, while the investigators merely carry out their specifications. This model is taken from research and development departments in industry. Although it leads to maximum relevance, the model of academic subordination presents enormous obstacles to excellence, since it violates several principles that will be discussed later on. Demeny has cogently analyzed, for the specific case of social science and population policy, the shortcomings of academic subordination, among them the risk that research may serve merely to legitimize decisions which have already been made [16]. The second option could be called the ‘segregation solution.’ In this case, the tension between excellence and relevance is not eliminated but institutionalized.
1400
JULIO FRENK
Specifically, the research center is divided into two subunits: on the one hand, a nucleus of excellence, which carries out long-term investigations oriented towards the advancement of knowledge; on the other hand, there are peripheral short-term projects oriented towards the investigation of problems that are relevant to the decision makers. This model was applied, among other cases, to the organization of the first economic research unit within the U.S. Department of Health, Education and Welfare in the 1960s [17]. The explicit purpose of this solution is for the organization as a whole to attempt reconciling excellence and relevance. Its enormous disadvantage is that it stratifies projects and researchers, running the risk of dividing them into first and second class categories. The last option is the ‘integrative solution.’ Instead of segregating relevant from excellent projects, this model seeks an integration of both values within each project. This implies a systematic effort to have each project solve a problem as well as advance knowledge. No doubt, this is the most difficult and complex option to implement. To be successful, it requires an organizational design that brings together the advantages of proximity to decision making and the structures, procedures and incentives developed by universities to ensure academic quality. Formulas for integration A design of this sort must build explicit bridges between decision makers and researchers, in order to prevent the potential conflicts that we referred to before. An example of such a bridge is the proposal developed by the World Health Organization, called decision-linked research [12]. The essence of this approach is to achieve interactions between decision makers and investigators, in order to increase the chance of decisions being made on the basis of trustworthy information, so as to benefit all of society, including, of course, the researchers themselves. The interaction between decision makers and researchers can happen at different points in the course of a reseach project. According to WHO, there is a ‘primary link’ between research and decision when the interaction occurs before the project begins. Thus, the starting point is a diagnosis of the needs of the decision making process. Primary links are characteristic of projects oriented towards the solution of problems. On the other hand, WHO speaks of a ‘secondary link’ when the starting point is a project that is initiated by the researcher. This type of link is more frequent in projects that are oriented towards the advancement of knowledge. Yet even in this case it should be possible to establish a significant interaction with decision makers, which can happen when the project is in the planning stage of formulating a protocol, when it is in the data collection or analysis phase, or when it has been completed. It is obvious that the sooner the link is established, the
more easily the project can be modified according to the needs of decision makers. If a research center wants to maintain a high level of relevance, it is necessary that all its projects establish effective links with the decision making process. The primary orientation of the projects and the nature of the research topic will determine the phase at which the link may be established. In any case, what matters is that the research unit be designed so as to encourage interaction between researchers and decision makers by overcoming the structural obstacles that were identified earlier. Figure 1 proposes a series of means to supersede each of those obstacles. Thus, one possible solution to the discrepancies in the definition of priorities is to have decision makers in the governing or consulting bodies of the research center, so that their needs can be expressed and opportunities can be identified within ongoing projects. Likewise, the time management differences we discussed before could be minimized by assuring the collaboration of researchers and decision makers since the earliest planning stages of a research project. This would facilitate the negotiation of deadlines required in order to produce opportune results. Another way of dealing with these differences is for researchers to offer intermediate products such as bibliographic revisions and progress reports that may help decision making even before the project is completed. Communication barriers can be reduced if, besides specialized articles, research projects produce ‘executive syntheses’ that outline results relevant to decision making in a comprehensible way. This solution would imply developing a system of dual reporting of results: academic papers for scientific colleagues and executive syntheses for decision makers. A complement to such syntheses would be the organization of participatory seminars where decision makers and researchers would jointly discuss results. To encourage this kind of exchange, May [1 I] suggests the training of ‘translators’ who can offer an interface between the two groups, taking into account the needs, values and priorities of each one. The idea is not far-fetched if one considers that this kind of translator already exists for the general public in the form of the science writer. In the same way, there should be a professional who could facilitate communication between scientists and decision makers, translating research results into policy and action recommendations. The problem of integrating research results can be addressed through two approaches. The first one is a detailed review of evidence, including the use of quantitative techniques of data synthesis such as meta-analysis [18-211. The second approach can be called ‘mission-oriented research’ [22,23]. This concept refers to research that aims at integrating not only different disciplines, but also different levels of analysis around a specific problem. For example,
Relevance
and excellence
in research
1401
MEANSTO SUPERSEDETHE BARRIERS
SOURCESOF POTENTIALBARRIERS BETWEENRESEARCHERS AND DECISIONMAKERS
Definition
of
- Presence governing
priorities
of decision makers in or consulting bodies
of research
-
centers
Collaboration
between
researchers Time
makers
management
since
stages
-
Language of
and accesibility
results
and
decision
the
early
planning
of a project
Identification
of
intermediate
products
of research
- Executive
syntheses
-
“Translators” policy
-
Joint
of research
seminars
into
for analysis
of
results
Integration about
of
different
the same
findings
problem
-
Meta-analysis
-
Mission-oriented
-
Definition objectives
Perceptions product vs
Fig. 1. Sources
about
the
of research:
final
scientific
utilization
in addition objectives
to of
research
discovery
-
decision
and solutions
of
research
of potential
barriers
in order to fully understand and act upon a health problem it may be necessary to integrate information on the subindividual level (through biomedical research), the individual level (through clinical research), the population level (through epidemiologic research) and the societal level (through health systems research). Finally, the discrepancy about the final product of research can be solved by specifying, since the initial formulation of every research project, a series of precise objectives for the application of results, together with the strictly scientific objectives [18]. In this way, the utilization of knowledge could stop being a haphazard event and become instead a programmed phase of the research process. This purpose would require that the scientific community give specific weight to application efforts in its evaluation of researchers’ performance, something that is not current practice. To a large extent, the debate on the utilization of knowledge leaves in the hands of the individual actors the implementation of solutions like the ones that have been suggested here. However, the present complexity in the production of knowledge demands
Greater weight to relevance evaluation of researchers
between
researchers
in
and decision
makers.
organizational solutions. It is necessary for research institutions to have differentiated structures that will facilitate the dynamic and creative linkage with their environment, especially the capacity to project scientific knowledge towards decision making. The structures and procedures proposed until now have the purpose of preserving the value of relevance. In fact, many of them have been tried out with success first at the Center for Public Health Research and later at the National Institute of Public Health in Mexico [24]. However, they are not enough for the ‘integrative solution’. In order to also preserve the value of excellence, it is necessary to adopt structures that promote academic quality, such as peer-review bodies. Another requirement for excellence is that, once the needs of the decision makers have been determined and a set of pertinent projects planned to satisfy those needs, then the responsibility for carrying out the projects should be left entirely to the researchers. That is, there should be no interference in the selection and execution of methods, and much less in the analysis of results. Thus, to the values of relevance and excellence it is necessary to add independence in the execution of projects. Actually,
1402
JULIOFRENK
this value is indispensable for the other two. It is obvious that, for an investigation to have scientific validity, it cannot be manipulated in order to reinforce the interests or ideology of decision makers [ 161. Guiding principles *
The careful design of structures and procedures would be useless if incentives were not also modified to stimulate research development. Such incentives could diminish, at least in part, the risk of “internal brain drain” [25]. This phenomenon, which is probably greater than the loss of researchers to foreign countries, refers to the abandonment of scientific activities for administrative ones. This happens when researchers find themselves in a system that gives greater privileges to those in the latter sphere of activity. Nobody doubts that an investigator who has achieved a certain stage of maturity in his or her career can occupy positions of managerial leadership, especially within the research field. Likewise, the temporary appointment, on a rotating basis, of applied researchers to decision making positions can enrich their understanding of reality. The problem arises when economic and power rewards overcome the researcher’s preferences for academic work and force him or her to shift to an administrative career. It is obvious that the internal brain drain affects all research institutions, particularly in underdeveloped countries. However, the problem is especially acute in applied research centers, since they deal with problems that are close to the concerns of the same administration towards which the drain occurs. In order to preserve the positive aspects of this proximity and at the same time reduce the risk of brain drain, the design of research organizations must include adherence to four guiding principles: 1. 2. 3. 4.
The The The The
principle principle principle principle
of of of of
parallel careers. academic autonomy. administrative sacrifice. inverted incentives.
The principle of parallel careers. As a result of a tendency to generalize vertical structures of organization, a hierarchical order is often imposed within research organizations. This hierarchy is formed, in descending order, by the director, associate directors, heads of departments and, finally, the researchers. In contrast, the first principle being proposed requires that the research career be considered parallelneither superior nor inferior-to the administrative career. Undoubtedly, investigators work within an administrative unit and, hence, are subject to its discipline. What we wish to emphasize here is that the researcher must have a clear career line, including distinct levels of assistant, associate and full re*This discussion is based, with modifications, on a portion of the paper ‘A conceptual model for public health research’ (PAHO Bullerin 22, (l), 60-71, 1988).
searcher. This line should be clearly differentiated from the administrative line, so that mobility is towards higher levels within research and not towards management positions. Actually, both careers operate under very different organizational assumptions. In administration, authority is exerted hierarchically and derives primarily from the position in the organization. In the academic career, authority comes mainly from knowledge and scientific productivity; it is exerted in a horizontal manner and is meant to inspire, guide and lead rather than command a colleague. The principle of academic autonomy. In theory, one way to follow the principle of parallel careers would be to differentiate the two lines not only functionally, but also professionally. This would imply that, just as researchers occupy academic positions, only professional administrators would be in management positions. Experience has demonstrated, however, that this approach is not satisfactory. Academic work has some characteristics that are not easily appreciated by those who have not been directly involved in it. One of those special characteristics is the fact that research requires a high degree of autonomy, since most of the intellectual process that leads to a discovery is unknown and, therefore, cannot be codified nor reduced to routines. Perhaps due to the complexity of academic work, researchers usually recognize, first and foremost, the authority derived from knowledge and experience. As a consequence, when a nonscientist directs a center of scientific production, there is often a high degree of mutual distrust between researchers and managers, which may lead to conflict. The only way to avoid this conflict is to respect the academic autonomy of investigators, and this necessarily implies that management positions in research institutions should be occupied by researchers. The principle of administrative sacrifice. There is an obvious contradiction between the two previous principles. While one suggests that research and management careers must be kept separate, the other points to the need for management positions at research centers to be occupied by researchers. The solution to this contradiction lies in the principle of administrative sacrifice, which states that occupying a management position should not be a reward for the researcher, but a sacrifice done for his colleagues-and for the scientific community in general-in order to preserve the principle of academic autonomy. The principle of inverted incentives. If the principle of administrative sacrifice is not to be a mere rhetorical statement, it must have a practical corollary. This corollary is the principle of inverted incentives. Its premise is that the present system of incentives continuously impels researchers to abandon their academic careers and transfer to administrative ones, thus contributing to the internal brain drain. At its most basic level, this phenomenon is expressed in the
Relevance
and excellence
considerable salary differences favoring managerial over research positions in many developing countries. The principle of inverted incentives consists of reversing this state of affairs so that, at each level in the research career line, there will be a greater incentive to advance to the next level than to move into a managerial position. A practical example of this principle is offered by the National Researchers System in Mexico [IS, p. 481. This publicly funded yet independent organization was created in 1984 in order to contain the brain drain that the economic crisis had exacerbated. Through a very strict peer-review process, active researchers are classified into one of four levels, each of which is accompanied by a tax-free stipend consisting of a certain number of minimum wages. This stipend is added to the regular salary of the researcher. Apart from offering higher remuneration for researchers who work outside Mexico City, the system also creates an incentive not to abandon research for managerial positions. Thus, researchers who are working in management may still be accredited by the system but receive no remuneration; once they return fully to research, they may benefit from the stipend. The final aim of the proposed guiding principles is to break the vicious circle-so common in developing countries-between the lack of adequate incentives for researchers and the weakness of research organizations. Strengthening its human and institutional infrastructure is a precondition for research to enrich decision making. CONCLUSIONS
In an ideal situation, the dilemma between relevance and excellence should be false. Even in our imperfect conditions, no reasonable decision maker rejects research as a contribution to increase the quality of his or her decisions. And even the most basic researcher will feel great satisfaction when seeing the results of his or her work applied to the solution of a problem. In this respect, Donabedian has written: I‘. the world of ideas and the world of action are not separate . . ., but inseparable parts of each other. Ideas, in particular, are the truly potent forces that shape the tangible world. The man and woman of action have no less responsibility to know and understand than does the scholar” [26]. Nevertheless, the tendency towards institutional fragmentation determines that sometimes research becomes artificially separated from decision making. Because of this, it is necessary to make an explicit effort to bring together both worlds, while allowing each one to maintain its autonomy. Research is a highly complex undertaking that requires specific conditions to flourish, including institutions and incentives to pursue the highest standards of excellence in order to create knowledge that decision makers may actively demand and utilize as informed con-
in research
1403
sumers. These are the only circumstances under which research will be able to make its distinct and irreplaceable contribution to human well-being [16]. In the health field, the growing gap between aspirations and accomplishments has brought research to the forefront of development strategies. The Commission on Health Research for Development embodies the international perception that research has become an essential component of any future advance. In its highly influential report-which will hopefully help to mobilize international resources towards research-the Commission has made an explicit call for promoting institution building and capacity strengthening in developing countries [15]. In this process, every nation must create the organizational arrangements that will make it possible to finally integrate relevance and excellence. Acknowledgements-I
am grateful to James Trestle, Avedis Donabedian and Harvey Fineberg for their valuable com-
ments to a previous version of this paper. Several of the concepts proposed here were enriched by discussions with Guillermo Sober&, Jest Laguna, Jaime Martuscelli, Jest Luis Bobadilla, Jaime Sephlveda, Enrique Ruelas and Miguel Angel Gonzhlez-Block. The international symposium where this work was first presented had the financial support of the Ministry of Health of Mexico, the Pan American Health Organization, the United States Agency for International Development and the British Council. REFERENCES
I. Blau P. M. The Organization of Academic Work. Wiley, New York, 1973. 2. Weber M. The Theory of Social and Economic Organi:ation. Oxford University Press, New York, 1947. 3. Hake1 M. D., Sorcher M., Beer M. and Moses J. L. Making it Happen: Designing Research wilh Implemen ration in Mind. Sage, Beverly Hills. 1982. R. La crisis perpetua. In M&sico Maliana 4. Ptrez-Tamayo (Edited by Aguilar-Camin H.). p. 215. O&no-Nexos, Mexico, 1986. 5. Ackoff R. L. The Arr of Problem Solving. Wiley. New York, 1978. health 6. McEwan E. D. A case for government-sponsored care research and development in the formulation of health policy and an account of early experience of government-sponsored health care research in one jurisdiction. Int. J. Hlth Serv. 3, 45, 1973. A. State intervention in 7. Frenk J. and Donabedian medical care: types, trends and variables. H//h PO/. Plan. 2, 17, 1987. A. L. f$tic~iL~eness and &ficiettcy: Random 8. Cochrane Reflections on Health Setvices. Nuffield Provincial Hospitals Trust, London, 1972. 9. Polanyi M. The republic of science. In Crireria /itt Scienrific Development (Edited by Shils E.), p. 9. MIT Press,.Cambridge, Mass., 1968. _ IO. Hauser P. M. and Duncan. 0. D. (Eds) The Studs of Popularion: An Itnrenrory and Appraisal: University di Chicago Press, Chicago, 1959. The policy uses of research. II. May J. J. Symposium: Introduction. Inquiry 12, 228. 1975. Improving health care 12. World Health Organization. through decision-linked research: Application in health systems and manpower development. Part II: Options for implementation. Document No. HMD/86.4.2. World Health Organization, Geneva, 1986.
1404
JULIO FRENK
13. Brownlee A. T. Applied research as a problem-solving tool: Strengthening-the interface between health management and research. J. Hlth Admin. Ed. 4, 31, 1986. 14. Fienk J., Bobadilla J. L., Sepulveda J., Rosenthal J. and Ruelas E. A conceptual model for public health research. Bull. Pan Am. Hlth Org. 22, 60, 1988. 15. Commission on Health Research for Development. Health Research: Essential Link to Equity in Development. Oxford University Press, New York, 1990. 16. Demeny P. Social Science and Population Policy. Center for Policy Studies, Working Papers Series No. 138. The Population Council, New York, 1988. 17. Grosse R. N. Personal communication. 1985 18. Yach D. and Dick J. Implementation of research: The key to closing the gap between public health knowledge and action. Unpublished document. 19. Glass G. V. Primary, secondary, and meta-analysis of research. Ed. Res. 5, 3, 1976. 20. Louis T. A., Fineberg H. V. and Mosteller F. Findings for public health from meta-analysis. Ann. Rev. Publ. Hlth 6, 1, 1985.
21. Thacker S. B. Meta-analysis: A quantitative approach to research integration. J.A.M.A. 259, 1685, 1988. 22. Bobadilla J. L., Frenk J., Martinez-Palomo A. and Sepilveda J. Health research in Mexico: Strengths, weaknesses and gaps. Background paper prepared for the Commission on Health Research for Development, 1988. 23. Frenk J. La salud perinatal coma modelo para la investigation con mision. In Memorias de/ Primer Connreso National sobre Defeclos al Nacimiento. Grupo _ de Estudios al Nacimiento; Mexico City, 1990. 24. Frenk J., Bobadilla J. L., Septilveda J., Rosenthal J., Ruelas E., Gonzalez-Block M. A. and Urrusti J. An innovative approach to public health research: the case of a new center in Mexico. J. Hlth Admin. Ed. 4. 461. 1986. 25. Frenk J. Los medicos y la politica: fuga de cerebros. Mundo MPd. (Mex) 3, (31), I, 1976. 26. Donabedian A. The Baxter American Foundation Prize Address. J. Hlth Admin. Ed. 4, 611, 1986.
