Guidelines for health care practice in relation to cost-effectiveness: report on a WHO workshop, Munich, 1–4 December 1980

~-........,~s for health care practice 1n relation to cost-effectiveness Report on a WHO Workshop REGIONAL OFFICE FOR EUROPE World Health Organization COPENHAGEN Rl<. C z_ EU RO Reports and Studies 53 Guidelines for health care practice in relation to cost- effectiveness Report on a WHO Workshop Munich 1 -4 December 1980 REGIONAL OFFICE FOR EUROPE World Health Organization COPENHAGEN 1981 ICP/RPD 802 ISBN 92 890 1219 6 © World Health Organization 1981 Publications of the World Health Organization enjoy copyright protection in accordance with the provisions of Protocol 2 of the Universal Copyright Convention. For rights of reproduction or translation, in part or in toto, of publications issued by the WHO Regional Office for Europe application should be made to the Regional Office for Europe, Scherfigsvej 8, DK-2100 Copen- hagen </) , Denmark. The Regional Office welcomes such applications. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the Secretariat of the World Health Organization concerning the legal status of any country , territory, city or area or of its authorities, or con- cerning the delimitation of its frontiers or boundaries. The mention of specific companies or of certain manufacturers' products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not men- tioned . Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. The views expressed in this publication are those of the participants in the Workshop and do not necessarily represent the decisions or the stated policy of the World Health Organization. PRINTED IN DENMARK ISSN 0250-8710 CONTENTS Page Introduction . . . . . . . . . . . . . . Existing guidelines in long-term care . 2 What constitutes a guideline? . . 2 What do guidelines contain? . . 4 Who develops guidelines and for what purpose? 5 Do guidelines have to be followed? . . . . . . . 6 Link bet ween guidelines and the health care system as a whole 7 A taxonomy of existing guidelines in the participating countries . 8 Some examples of existing guidelines . . . . . . . . . . . . . . . . . . . 9 Guidelines for the management of patients with obstructive lung disease in Israel . . . . . . . . . . . . . . . . . . . . . . . . . . 9 A medical care programme for hypertension in Sweden . . . . . . . . . . 11 Standardization of cardiovascular care in Poland . . . . . . . . . . . . . . I 8 National Institutes of Health consensus development conferences : screening for breast cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Professional Standards Review Organizations . . . . . . . . . . . . . . . 23 Guidelines on cancer control in the Federal Republic of Germany and the Netherlands . . . . . . . . . . 27 Cost-effectiveness analysis in health care 3 2 The application of economic appraisal . 33 Some examples of economic appraisal in health care 34 Some conclusions on methodology . . . . . . . . . . 36 Using the results of economic appraisals . . . . . . . . 38 Economic appraisal and guidelines on patterns of long-term care 39 Factors in recommending guidelines for long-term care . . . . . . . . 40 Relevant factors in recommending guidelines for long-term care 41 Contents of guidelines for long-term care . . . . 44 Implementation of guidelines for long-term care 45 Flexibility of guidelines . . . . . . . . . . . . . . 45 Using guidelines to achieve cost-effectiveness in long-term care 4 7 In which situation should guidelines be used as opposed to other cost-effectiveness measures? . . . . . . . . . . . . . . . . . . 4 7 How should guidelines be developed and implemented? . . . . . . 49 How should guidelines be evaluated from a cost-effectiveness standpoint? 50 Full cost-effectiveness studies. . . . . . . . . . . . . . . . . . . . . . . . . . 5 I The role of education and training and of research in development and application of cost-effective guidelines in long-term care Initiatives in research . . . . . . . . Initiatives in education and training Recommendations References . . . . Annex List of participants . 53 54 54 56 57 59 INTRODUCTION Health care systems in many developed countries face nsmg costs, the problems posed by rising expectations on the part of consumers and pro- viders, and increasing demands for the quality of medical practice to be evaluated. A central question is whether it is possible to reach a consensus among health care professionals, consumers , and funding agencies (such as governments or insurance agencies) on the range and type of care to be provided. Although most health care systems operate with certain struc- tural guidelines (e.g. in hospital building, medical education , licensing, staff- ing, equipment, and safety) , guidelines for treatment are more contentious since they may be thought to infringe clinical freedom. To discuss this question a Workshop on the Cost-Effectiveness of Rec- ommending Standard Patterns of Long-Term Care was held in Munich from 1 to 4 December 1980. The Workshop reviewed European and North Amer- ican experience with such guidelines in the field of chronic degenerative disease , oaying special attention to their impact on the cost and effective- ness of the care provided. The Workshop was preceded by a study phase in which multidisciplinary teams were set up in most of the participating countries to review the situation . The teams were then visited by Mr M.F. Drummond and Dr S.O. Isacsson, consultants from the WHO Regional Office for Europe , who discussed the aims of the Workshop and the guidelines from each country . The Workshop was sponsored by the Federal Republic of Germany , which hosted the meeting in Munich. Dr F . Unz , Programme Director, Ger- man Aviation and Aerospace Research Institute, assisted in the study and the Institute for Medical Informatics and Health Services Research of the GSF Research Centre (MEDIS), Munich , helped with the organization of the Workshop. Guidelines for long-term care , especially care for patients with can- cer , cardiovascular diseases , diabetes , rheumatic diseases, and mental health problems, exist in most of the participating countries. As a rule they were recommended by governments , health professional groups , insurance agen- cies, or combinations of these . They include planning norms or policies , clinical "best practice", educational standards, environmental health standards, and process guidelines of the type found in model medical care programmes. They are either advisory or obligatory , and they inevitably reflect the in- terests of the groups proposing them and the purpose for which they were devised. The questions examined by the Workshop were : - How do guidelines link up with the overall organization of health care in the various countries? - Does it matter who recommends the guidelines? - Should guidelines be aimed primarily at prevention , diagnosis, treat- ment , rehabilitation, or at all phases of care? - Should guidelines be aimed at particular diseases? - How should guidelines be implemented ; should incentives be offered or sanctions imposed? - Should consumer groups be involved in the derivation of guidelines? - What research in association with guidelines is needed into the effec- tiveness and costs of alternative patterns of care? - By what criteria should guidelines be evaluated? - How effective are guidelines in encouraging more efficient health care practices? - Is a special body needed with responsibility for coordinating the devel- opment, implementation, and evaluation of guidelines for the delivery of health care? - Can guidelines be of use in the education and training of health care professionals? - Do guidelines represent an unnecessary infringement on clinical freedom or do they , on the contrary, create new possibilities for improving the quality of care and the use of health care resources? EXISTING GUIDELINES IN LONG-TERM CARE What constitutes a guideline? In this report the term "guideline" is taken to mean a recommendation made explicit by central or local authorities or by associations of health care 2 professionals. A wide variety of guidelines were found to exist. These cover, inter a/ia: - the type of health care system (whether publicly or privately provided) ; - the system of health service planning; - the system of medical education and licensing of professionals ; - schemes for accreditation of hospital and other facilities; - arrangements for local or consumer participation in health care ; - methods of commissioning medical and health service research ; - cost containment measures (such as cash limits on government expend- iture or insurance company reimbursement limits). The guidelines of greater interest to the Workshop were those concerned with health care practice itself, consisting of: (a) Policy documents issued by central or local government. An exam- ple of such a guideline from the United Kingdom was presented at the Work- shop (J). It sets out the principles upon which services should be provided , reviews the nature and level of services existing at the time of publication , and specifies the improvements that should take place. (b) Model medical care programmes. Examples of such programmes, which originate mainly from Scandinavia, were presented from Finland and Sweden. The object of such programmes is to bring about more rational and purposeful care by stimulating groups of professionals to undertake a wide- ranging review of the nature of the disease in question and the facilities and processes required to tackle it. (c) Methodological letters. These are essentially circulars dealing with aspects of treatment methods issued by national institutes in the eastern European countries. They contain recommendations on standard patterns of diagnosis, treatment, and care and aim at developing the medically most effec- tive standards that are also economically acceptable. From time to time they reconsider topics in the light of the latest developments in medical sciences. (d) Audit and peer review schemes. The use of medical audit and of professional standards review organizations (PSROs) as a way of ensuring high clinical standards was discussed. The emphasis in this approach is on develop- ing a system of retrospective review ( usually by medical peers) of the operation of institutions and medical practitioners , rather than on deriving standards of operation per se. 3 (e) Physician protocols or algorithms. These are advisory schemes of practice (frequently concerned with diagnosis) for use predominantly by ju- nior physicians or non-medical staff, and are common in a number of coun- tries. They are sometimes associated with innovations in medical records and some evidence (particularly from the USA) suggests that the performance of junior medical staff is improved when algorithms are used. (f) Recommended resource lists. These consist mainly of lists of rec- ommended medicines or information to practitioners concerning medicines with differing costs. The same notion has also been applied to medical equip- ment and technology. Advice on the use of medicines is common in many countries and an example from the United Kingdom is the British national formulary. These are examples of the kinds of guidelines considered at the Workshop. Many more, not necessarily corresponding to these types, are outlined later in this chapter. What do guidelines contain? Judgements about patterns or types of care can be made in terms of the structure (or inputs) used to provide care, the processes (or activities) carried out in providing care , and the outcomes (or results) obtained. Judgements of structure are usually concerned with the range and type of facilities available , the grades of manpower required, and the training of that manpower. Judge- ments of process are typically concerned with the type of diagnostic , treat- ment, and rehabilitation procedures carried out , the order in which they are carried out , and the indications ( usually clinical) adopted as prerequisites for particular courses of action. Judgements of outcome relate to the objectives of the care being given. In the clinical sphere it may be clinical results such as case fatality rates, recurrence rates, and complications. In the planning sphere it may be accessibility of care and more general client satisfaction. Most of the guidelines discussed at the Workshop make recommendations about structure or process. Some, such as medical care programmes, discuss a comprehensive range of recommendations dealing with both structure and pro- cess. Obviously a comprehensive assessment of a health care programme or treatment requires the study of the interrelationship between all three ele- ments , as indicated schematically below. Inputs 4 Treatment process ---~►• Outcome However, it is important to note that guidelines dealing with structure alone are not necessarily inappropriate. It may well be that seemingly rigid struc- tural guidelines have been derived from some detailed analysis of the inter- relationship between particular inputs, the prevailing processes , and outcomes. A structural guideline to place a warning device on a piece of medical equip- ment may be based on knowledge of the effect on outcomes that such a device might have, given the safety procedures currently in operation. Who develops guidelines and for what purpose? The source of guidelines differs from country to country, but profes- sional groups themselves are very prominent in their development. In some countries the gove rnment has encouraged the professions' interest , perhaps by making available the resources to produce and distribute the guidelines. Some governments have decided to keep a low profile ; for example , in Finland the National Board of Health, in initiating the medical care programmes , ap- proached the Finnish Hospital League for help in order "to avoid any author- itative labelling of the project". Occasionally , as in Israel and Yugoslavia, fund- ing agencies such as sick funds have taken an interest , but even in those coun- tries it is largely the professions that have an interest in guidelines. It is per- haps worth noting, however, that althougl1 the definition of guidelines given earlier refers to explicit recommendations, there are many implicit guidelines operating in all countries. In one sense every health practitioner follows his own set of informal guidelines and occasionally attempts to pass them on to others through publication of details of his practice, research , and experience. Medical textbooks might also be considered as consisting of sets of guidelines for care. Given the strong professional involvement in the development of guide- lines , it is not surprising that their main purpose is to improve or maintain the quality of care. This may sometimes be at odds with cost-effectiveness, the main element of the impact of guidelines under consideration at the Work- shop. However , although quality of care is the main objective , many of the originators of guidelines recognize that the search for higher quality has its limits because of resource constraints. For example, in Hungary the aim of methodological letters is to develop those medically most effective standards which are economically also acceptable . Similarly , the objectives of medical care programmes in Finland are : - to standardize medical practice; - to improve cooperation between institutions ; - to increase the safety of patients ; - to give aid for the education of medical care personnel ; - to act as a tool for improving cooperation between different groups of personnel and to clarify the division of labour between them. 5 In addition, it is stressed that the care programmes should be constructed in principle within the limits of existing resources. A similar principle under- lies the development of medical care programmes in Sweden. In the USA three uses for clinical algorithms or protocols have been put forward: (a) as guides to non-physician providers, particularly physician assistants; (b) as the basis of quality assurance programmes; they can be used to check whether information has been collected, tests carried out, referrals made, and treatment provided in an appropriate way; and (c) as a representative of cost-effective care and as a standard for the efficient use of medical resources. In the USA, too, the PSRO programme considered below is charged with reviewing the medical necessity, quality, and appropriateness of care. Qual- ity is defined as "the meeting of professionally recognized standards of care", and appropriateness as "the possibility that services could not be provided at a lower cost in an alternative setting". Do guidelines have to be followed? A fairly prevalent view at the Workshop was that not only are guide- lines usually advisory rather than obligatory, but that they are also to be in- terpreted flexibly in the light of local circumstances. This is the main rea- son for adopting the generic term "guidelines" in this report rather than "standard patterns of care". For example , the Canadian series of guide- lines state in their introduction that: The Guidelines should not be regarded as rigid standards. Their application may vary throughout Canada and it is the prerogative of each province to accept, mod- ify or reject the Guidelines according to provincial priorities. The guidelines produced in Scandinavia reflect what the Finnish report calls "the traditions in the country, in which formal independence and individ- ualism are highly valued qualities". Both the Finnish and Swedish guidelines place much emphasis on problems of implementation and the need for local involvement. However, the extent to which guidelines, advisory or obligatory, are fol- lowed rigidly is uncertain. A monitoring system such as the PSRO in the USA can in principle operate with flexibility and sensitivity to local problems. On the other hand, despite assurances in other countries that guidelines are for guidance only, they may in fact be followed more rigidly than the originators intend. Hence the importance of implementation. 6 .------------------------------------ - ' Link between guidelines and the health care system as a whole One of the conclusions of the Workshop was that , because of the great differences in the ways in which health care is organized in different coun- tries, it is difficult to make many recommendations in respect of guide- lines; a particular recommendation might be appropriate in one country but not in another. The main aim was therefore to pool experience from the various countries in order to identify key factors that all concerned with guidelines should consider. It emerged from the discussions that within particular countries there is a failure to consider how guidelines might fit into the organization of health care as a whole . For example , a government may propose guidelines for , say, improvements in care for a neglected group , yet not provide the money to enable the aims to be achieved , e.g. by ear-marking development funds expressly so that the guidelines could be followed. Alternatively, a group of professionals may develop guidelines whose resource require- ments could never be met. Or a government may encourage professionals to develop guidelines yet not exploit the opportunity to incorporate them in an existing system of health care provision. So, in the USA, the Govern- ment may encourage the development of guidelines for health care prac- tice by professional groups (such as the National Institutes of Health Con- sensus Development Programme), yet these may not be explicitly used in the accreditation of hospitals or in hospital or physician reimbursement schemes. This does not occur only in the USA. In many countries the link be- tween the production of guidelines and , say , health care planning is ten- uous. In the United Kingdom clear priorities for health care have been laid down by the Government , but there is very little suggestion that the fund- ing of health authorities should be directly or exclusively linked to the extent to which the priorities are adhered to. In Finland the Government subsidizes local authorities to the extent of between 39% and 70% and this gives the central authorities an opportunity to guide the system towards rational goals such as a proper balance between hospital and ambulatory care. However, in both Finland and the United Kingdom there is a tradition of devolution of decision-making and responsibility , and negotiations with the local or regional authorities, as in discussions of plans submitted by lower tiers of organization , take the place of orders transmitted from above . Therefore, from the point of view of using guidelines to bring about more cost-effectiveness in health care practice, other planning and budgetary arrangements are often seen as alternatives to guidelines. It was mentioned from time to time during the discussions that to control costs it was nec- essary to exercise budgetary restraint. It may be that such action should more appropriately be viewed as complementary to guidelines. 7 A taxonomy of existing guidelines in the participating countries There are many ways in which existing guidelines can be classified. How- ever, because of the variations in the details given by each country a few cri- teria only have been adopted : the broad disease group to which they refer ; the country of origin ; the group or groups recommending them; and whether they consider elements of structure, process, or outcome. 8 Disease (or disease group) Cancer Cardiovascular diseases Diabetes me ll itus Rheumatoid diseases Menial disorders Country France Federal Republic of Germany Hungary Netherlands Yugoslavia Finland Hungary Israe l Poland Sweden Canada Hunga ry Israel Sweden Yugoslavia Federal Republic of Germany Yugoslavia (1.:cchoslovak ia United Kingdom 0 \bml y process-o rient ed . h Struc1ure-,1r1l'nted . on·gtn of guidelines Cancer Institute Gustave Roussy Professional associa tions National Institute, Ministry of Heal!h Profess ional groups Professional groups Ministry of Health and Social Affairs, National Board of Health . Hospital League, professional groups National Institute of Cardi• ology. Ministry of Health Professional groups and Sick Fund Institute of Occupational Medicine and other sources The National Board of Health and Welfare, the Sweetish Institute for Planning and Rationalization of the Health Services (SPR I), professional groups Department of National Health and Welfare National lnstitu1e. Ministry of Health Profess ional groups National Board of Pro- fessio nal Croups Professional groups Professional groups Professional groups Ministry of I lealth Sec retary of Stale for Social Services Remarks Mainly trea tment protocols'l Different cancer si1 es'1·h Cancer of uterine cervix (screening}° Acute lymphocyt ic leukaemia in ch ildren, diagnosis and treatment0 Breas t cancet° St roke, all phases of care'7 Comprehensive community control of cardiovascular disease'7 Hypertension° Cardiovascular care in the in - dustrial health services"·h Hypertension, medical care programme° ·h Diabetic Care Unith Diabetes dispcnsary'1 ·h All phases of ca rcD ·h All phases of ca re , medical care programme All phases of care" Rheumatic diseases (under prcparation'f' ·h Rheumatic diseases'l Ccrontopsfchiatric project0 • Be tt er services for mentally ill and handicapped0 , h Other guidelines presented but not included in the table were : - Guidelines for hypertension, drug treatment, and cancer (Czechos- lovakia); - Medical care programme for fracture of the neck of the femur (Finland); - Hereditary blood diseases (France); - Chronic obstructive lung diseases (Israel). In Canada the Department of National Health and Welfare regularly issues guidelines which are both structure- and process-oriented. Information about these guidelines is given in the publication Information Bulletin, Standards. The contents of one of these Bulletins are given below (Information Bulletin, Standards, March 1980). I Special focus: electrocardiography II Guidelines and standards for private practice III Hospital and laboratory guidelines and standards IV Guidelines and standards for community services V Catalogues of guidelines and standards VI Journal references VII Addresses where guidelines may be obtained In the USA guidelines exist on primary health care and are supposed to be of special value for people other than physicians. The National Institute of Health Consensus Development Conferences are special types that will be discussed later in connexion with the PSRO programme. SOME EXAMPLES OF EXISTING GUIDELINES Most of the guidelines discussed in this report are process-oriented. They may be rich in details or merely provide a few updated recommendations. The following examples of guidelines will give an idea of their different aspects. Guidelines for the management of patients with obstructive lung disease in Israel In the past few years there have been a number of experiments conducted in Israel to provide standards of medical care for specific disease entities. The 9 source of most of these standards is professional groups. The National Sick Fund and the Ministry of Health have also issued shorter versions of guidelines, and these originate from the National Sick Fund. The patient should be seen at regular intervals. 1. Smoking. Should be prohibited, including presence of the patient in places where smoking is allowed. 2. Humidity. Water is the best expectorant and patients should be en- couraged to drink at least three litres a day. In case of cardiac insufficiency the patient's weight should be monitored daily. 3. Bronchodilator drugs . (a) Inhalation up to eight times a day (ventolin), two inspirations each time. It is important to teach the proper use of inhalators. (b) Theophylline-based drugs may be prescribed, either short- or long-acting. It is important to give a sufficient dosage (e.g. two tablets of glyphylline, three to four times daily), since smaller doses are ineffective. Pay attention: (a) Avoid drugs containing ephedrine since they may cause car- diac excitability. (b) There is no justification for using arninophylline suppositories. (c) Add steroid components only if necessary to the inhalations in addition to ventolin. 4. Infection control. Respiratory infections are common in such pa- tients and should be promptly treated. Increase in quantity, density, and change in colour of expectorate, usually accompanied by dyspnoea, fatigue , or fever , usually indicates an acute bacterial infection . Patients usually re- spond well to treatment with ampicillin , tetracycline , or erythomycin for 10-14 days. Patients should be given influenza vaccination yearly. 5. Cardiac insufficiency. In case of oedema or hepatomegaly diuretics should be used. Digitalis should be used with caution since arrhythmias are common in severe chronic obstructive lung disease. Oral corticosteroids should not be used routinely. They are useful when given for short periods during an acute episode of spastic bronchitis. They should then be given in high dose for about one to two weeks (1 mg/kg/day of prednisone ). Pay attention: Sedatives, tranquillizers, hypnotic drugs, antitussives, and narcotics (such as Valium and codeine) should be avoided since they may in- crease respiratory insufficiency by depressing the respiratory centre and the cough reflex. 6. Postural drainage. Drainage and vibration of the chest by a trained family member or a physiotherapist are highly helpful in draining the expec- torate. Correct procedure is important. Bronchodilator inhalation prior to postural drainage will improve results. 7. Diaphragmatic breathing should be taught. The patients should exer- cise breathing against closed lips while exercising his abdominal muscles. 8. General physical exercises and occupational therapy should be encouraged. These guidelines on the management of patients with obstructive lung disease are intended to help primary care teams in their work and are not in- tended to be rigid standards. A medical care programme for hypertension in Sweden Some 500 000 people in Sweden have high blood pressure. The medical care authorities in Skaraborg County have, in cooperation with the Swedish Planning and Rationalization Institute for the Health Services (SPRI), devel- oped a medical care programme to combat high blood pressure. The idea of such a programme took shape gradually among health professionals in the county. Local studies on the prevalence and treatment of hypertension in some of the municipalities gradually led to an increasing interest in establishing a collective programme. From the start the programme was planned for only a certain number of the municipalities in the county. The main reason for limiting the scope of the programme was that other development work being carried out under the auspices of the public health unit indicated the advisabil- ity of gradual implementation. The need for continuous evaluation was also clear from the start. Since hypertension is mainly handled on an outpatient basis, the intention at first was to implement the programme entirely within primary care. However, it was established, on the basis of discussions, that a medical care programme for hypertension was a matter of concern in both primary care and hospitals. Internal medicine departments would be able to assist by special investigations, treatment of complex cases, training, and follow- up. Assistance was also recruited from the regional hospital (the Sahlgrenska Hospital in Gothenburg), where the head of the hypertension clinic was al- ready involved in a hypertension care programme. 11 With support from the SPRI and the Skaraborg County Council, the chief public health officer, the chief cardiologist, and the head of the hypertension clinic at the regional hospital undertook to develop a medical care programme. The programme was to be implemented in 6 of the 16 municipalities, which contain about half of the county's population of 260 000 inhabitants. A large-scale effort was begun to establish the new care programme throughout the municipalities and at the hospitals concerned. All personnel at health centres were informed and information was also disseminated to representatives of occupational health services as well as to private practitioners. Representa- tives of the district medical care officers and nurses participated in the formu- lation of different parts of the programme. The guiding principle was to work with and not for the persons who would be involved in and implement the care programme. The programme was also given approval at the regional hospital and university level. After being circulated for comments, the programme was ready in final form six months later and was then adopted by the County Medical Care Board. This decision was considered important in ensuring that the programme would be implemented. The programme is based on recommendations by the SPRI and the National Board of Health and Welfare. It begins with the epidemiological back- ground and the social consequences of the disease. It was emphasized that it should be regarded as a trial programme and it was limited to 6 of the county's 16 municipalities on a trial basis. Hypertension is defined and its epidemiology described in this particular medical care programme. It covers techniques for blood pressure measurement, criteria for follow-up and diagnostic investigations - the medical history, physical examination, electrocardiography, laboratory tests, heart and lung X-rays , urography and renography, and annual check-ups. It contains indica- tions for the initiation of antihypertensive therapy, information to the patient , drug treatment alternatives, therapy for unresponsive hypertension, other therapeutic measures, subsequent check-ups, objectives of treatment , choice of the type and level of medical care, organization of local and central hypertension units, structure , definition of work duties , personnel , premises, and work routines. Diagnosis and treatment are described in detail with recommendations, for example , of types and doses of preparations. This is followed by a descrip- tion of the medical care organization. A central hypertension unit has been created for the county, consisting of the chief public health officer, the chief cardiologist at the central hospital's department of medicine, a district medical officer, a physician with independent medical responsibility from one of the hospital departments of medicine in the county, and two nurses, one from a hospital department of medicine and one from a medical care centre. The central hypertension unit is responsible for , among other things, continuous revision of the care programme, basic and further training of 12 personnel, coordination of hypertension care, and the continuous collection of data from various units. Central hypertension clinics for referrals are estab- lished at the departments of medicine of the hospitals at the primary level of the municipalities. Hypertension clinics are also established at each medical care centre within the six municipalities. The work of detection and follow-up of hypertension cases is handled by a nurse, the diagnostic and therapeutic work by a physician. The programme is based heavily on nurse-centred care with the goal of optimizing care, mini- mizing the load on the physician, increasing the responsibility of the nurse, and permitting good follow-up and evaluation. The implementation of the programme is in stages. The first stage was from 1977 to 1978, when the quality of the care given to already known hyper- tension patients was to be improved. During this period, known but untreated hypertensives were traced and care improved. At the same time, an intensive training and information programme was implemented . During the second stage (1979-1981), an attempt was made to trace as yet undetected hyper- tensives. This is done by taking the blood pressure of all persons aged 40-69 years who visit outpatient clinics, regardless of the reason for consultation. During the third stage (I 982-1983), the hypertensives detected during the second stage will be provided with continued treatment. The third stage aims at as complete a survey of hypertension throughout the trial area as possible. The programme covers some 140 000 people, of whom roughly 45 000 are in the target age group 40-69 years. The other half of the county serves as a control area for the evaluation of the programme, the supply of health services in both areas being similar. Study of a sample of 8% (8 000) of all the people in the county aged 25-75 years revealed no major differences in health status or in the prevalence of hypertension or hypertension-related disease. Evaluation Great emphasis has been placed on the evaluation of this rather large- scale medical care programme. A preliminary estimation of the number of hypertensives requiring care in the age group 40-69 years has established the number at around 11 000, of whom about half live within the area of the programme and the other half in the control area. The evaluation is aimed at answering the following questions: 1. Is it possible to implement the care programme as proposed ( feasibility)? 2. Is the programme effective in preventing complications of high blood pressure? . 3. Is the programme defensible from the humanitarian point of view (i.e., does it reduce suffering resulting from death and disease, and does this reduction of suffering outweigh any harmful effects of the programme)? 13 4. Is the programme defensible from an economical point of view (i.e ., do economic gains resulting from reduced sickness and death outweigh the costs of the programme)? Various data are used to evaluate the programme. Comparisons will be made between the six municipalities included in the programme and other municipalities in the county. Feasibility of the programme The feasibility of the programme is evaluated by the use of standardized records for the first visit and the annual check-up and also by continuous dis- cussions and interviews with the health personnel involved in the programme. Much of this information is collected when people from the central hyper- tension unit visit the different municipalities, by regular telephone contact, and in connexion with regular educational activities. From baseline findings in a cross-sectional study of 8000 randomly selected individuals 25-75 years old carried out a few months before the programme started , the observed number of hypertensives, as compared with expected number, can be monitored in each of the six municipalities in the programme area. The annual check-up for hypertensives automatically gives information on the decrease in blood pressure in the population observed. So far no comparisons have been made with data from the communities in the control area of the county. For the programme area it is possible already to observe an improvement in the quality of care and a great increase in the number of observed hyper- tensives included in the programme. In this area there is a marked decrease in the blood-pressure level for people with hypertension. Nurses seem to appreciate their new role of frontline responsibility for their hypertensive patients and they also seem to carry out this duty better than physicians did in the previous organization. However, in some of the municipalities there is less interest in the programme because of attitudes among physicians. The patients included in the programme have been interviewed on their opinion of care before and after implementation of the programme. They seem to appreciate the care received through the programme more than conventional care and are especially satisfied with the continuity of care. For example, pa- tients are actually recalled for regular check-ups by their respective nurses . A more detailed evaluation of feasibility is being prepared and will be carried out by additional procedures and more structured interviews for the health personnel involved as well as for the patients. Medical effectiveness The effectiveness of the medical care programme will be established through a comparison of mortality and morbidity rates in the programme and 14 the control area, using special registration forms for myocardial infarction and stroke and ICD death certificates for total mortality and cause-specific mortality covering the whole population in the county. Statistical analy- sis predicts the possibility of observing significant differences between the programme area and the control area in relation to the variables used in the measurement of medical effectiveness within a period of six to seven years from the start of the programme. Quality of life The question of quality as opposed to length of life is extremely im- portant in the evaluation of health programmes. For people with hyper- tension diagnosis and treatment may bring both positive and negative changes in the quality of life. The diagnosis of hypertension may well make an in- dividual anxious and worried about the future. He may start avoiding situ- ations thought to increase the blood pressure, lose interest in work, re- duce physical activity , avoid sexual activity , and so forth. The medical treatment and the regular check-ups for several years will probably make the patient aware of having a disease, which may decrease his feeling of wellbeing. Changes in attitudes may also occur. For example , the patient may have regarded his work or family as the most important part of his life, but after learning that he has hypertension he may become increasing- ly concerned about his health. Most antihypertensive drugs, which should be taken for the rest of life , have side effects that may reduce the qual- ity of life. Detection, treatment, and follow-up of hypertension may also improve the quality of life. Regular check-ups may be appreciated and contact with the medical care system may result in an increased interest in a healthy lifestyle; better dietary and drinking habits, less smoking, and an improved physical and mental condition lead to greater wellbeing. The diagnosis of hypertension probably has different meaning for different people. For somebody experiencing symptoms the diagnosis may be a relief from anxiety about its cause; for most hypertensives, however, since hypertension is rarely accompanied by symptoms, it means an awareness of illness. In the programme, evaluation will include an analysis of the possible ef- fects on the quality of life through interviews of patients in the programme area. These interviews will be both structured and unstructured and in- clude some quantitative variables such as the number of days of sick leave before and after the detection of hypertension . Preliminary investigations so far - by means of an instrument for the measurement of quality of life developed by the WHO laboratory for stress research in Stockholm - show no evidence of a decrease in the quality of life following the detection or treatment of hypertension. 15 Economic appraisal The economic evaluation of the medical care programme for hypertension may be illustrated as in the following scheme, where as many relevant input and output variables as possible need to be identified and as far as possible also quantified and assessed: Resource inputs - cost of the health care organi zati on - patient's sacrifi ces Hypertension care in programme area I I I ... Hype rtension care in control ·ea Outcome - cost consequences due to changes in: - morbidity - qual ity of life Outcome The total input of resources for the care of hypertensives is ofless interest in this evaluation since the question is not of the "all or nothing" type but rather "a little more or a little less". The evaluation focuses on the two differ- ent ways of administering care for hypertensives: conventional care and struc- tured care according to explicit professional standards, i.e., the medical care programme. The following items constitute a reasonable basis for comparison of the resource inputs in the control and programme area : - blood pressure measurement (personnel time); - diagnostic procedures (personnel time, laboratory and X-ray costs) ; - follow-up (personnel time); - information (personnel time, production of brochures, etc.); - treatment: - cost of drugs; - cost of other forms of treatment; - cost of office premises, materials, etc.; - indirect costs; - time spent by the patient ( expressed in hours). 16 A questionnaire that was prepared covering most of these data includes the following different fractions of patient time that may serve as indicators of the organizational changes induced by the medical care programme: - time from booking appointment to day for visit ; - time from departure from home or place of work to arrival at the clinic; - total time at clinic (with physician and nurse); - time from departure from clinic to arrival home or at place of work; - total time for visit to clinic. Evaluation in economic terms of the outcome of the programme as com- pared with that of conventional care has to rely on medical evaluation of changes in mortality and morbidity from stroke and myocardial infarction. The different types of resource input for stroke or myocardial infarction are: - hospital care (emergency care); - care for chronic disease ; - rehabilitation; - home care; - technical aids; - social assistance (home nursing) ; - disability pensions ; - sick leave. It is not possible to give economic expression to all the possible conse- quences of stroke or myocardial infarction. The economic evaluation of the programme should, however, aim at clarifying the total marginal resource re- quirements and the subsequent effects of efforts to prevent early death and different stages of myocardial infarction and stroke. This means that not only the sacrifices made by the health care organization but also those made by the patients and their relatives should be taken into account as far as possible . From the preliminary results of the economic assessment of the pro- gramme there are no significant differences between the control area and the programme area as to the overall cost for treating hypertensives. A more de- tailed analysis being undertaken will reveal specific differences, if any. The medical care programme for hypertension is an example of a model health care programme implemented in half of a county in Sweden. Almost all the physicians in the county are salaried. The health service system is socialized and the guidelines are implemented in the routine work of the health service 17 system, i.e ., mainly in primary health care. The results from the evaluation of this programme may provide an answer to several questions regarding cost- effectiveness and quality of care using quite detailed guidelines. This example also shows that guidelines, called medical care programmes, can be developed in countries like Finland and Sweden with a decentralized socialized health service system. Standardization of cardiovascular care in Poland In many socialist countries national institutes are responsible under the ministry of health for working out standard patterns to control diseases, rec- ommending the most suitable methods for diagnosis, treatment, and care. For the standardization of cardiovascular care in the Polish Industrial Health Ser- vices (IHS) two types of standards are employed : organizational (mainly struc- tural) and medical (mainly process-oriented). Structural standards are import- ant within the Industrial Health Services. They have a general character but may also relate to specific diseases or health risks - the example here is cardio- vascular disease, with special reference to coronary heart disease. The structural standards include: - principles for the operation of the basic IHS unit providing medical care to patients and workers ; - principles governing the subordination of basic IHS units to the system; - principles of cooperation between basic IHS units and industrial plants ; - principles for conducting prophylactic examinations; - principles of medical cooperation within the framework of the health care system ; - principles for identifying and taking care of high-risk groups. Detailed standards on the health care of high-risk groups, not enforceable by law, set out in addition principles for keeping a high-risk group docu- mentation register. They call for : (a) records of active care as well as medical records; (b) technical details for the register; (c) a special record of active care; and (d) obligatory appointment for a next visit. The medical principles for achieving active medical care of high-risk groups for circulatory diseases link active medical care with the total system of health care of the working population. The medical standards laid down for coronary heart disease are as follows: 18 Diagnosis Physical examination - characteristic retrosternal pain connected with effort or emotional state (a feeling of fear) . Accessory examinations: obligatory - ECG, serum glutamic-oxaloacetic transaminase, leuko- cytosis, erythrocyte sedimentation; suggested - level of cholesterol and lipids (especially (3-lipoprotein). Interpretation of results of accessory examinations. Elevation of the ST segment and an inverted T wave are evidence of past myocardial infarction or myocardial ischaemia. A normal ECG does not, however, exclude coronary disease, especially during a painless period. Indications for hospitalization or specialist consultation in case of diag- nostic difficulties - not stated. Principles of differentiation - not stated. Comments on difficulties of diagnosis. Special concern should be given to men over 40 years of age. Doubtful cases should be analysed in those groups where longer productive activity allows for extra benefits. Active health care of high-risk groups Indications for including a patient in a high-risk group. Chronic coronary disease needs active medical care. Special recommendations for active medical care : patients with ECG changes indicating disturbances of coronary circu- lation should be included in the high-risk group even if they do not complain of illness. Treatment Drug treatment : antilipemic agents affecting lipid metabolism, vasodila- tors, and sedatives. If diabetes or hypertension is concomitant, drug treat- ment of those diseases is necessary. Psychotherapeutic and informative procedures: the treatment starts by explaining to the patient how he can prevent pain and further disorders by avoiding negative psychic stimuli; he should stop smoking; he should limit his activities and rest in the open air ; and, if overweight, he should observe proper habits of nutrition and dietary restraints. Other procedures include 1-anatorium treatment to remove the patient from harmful factors in his work and home environment, and rest in the open 19 air in healthy climatic conditions. Patients should not be sent to sanatoria far from their home or in the mountains. Main problems in treatment - not stated. Work ability Contraindicated are heavy work, severe physical effort, heat stress, heavy protective clothing, and all jobs where acute pain or syncope may cause an accident ( e.g. a job as truck driver). Recommended is light, quiet work with moderate physical effort. Control of work ability. When, in spite of proper work conditions, rest, work breaks, and drug treatment , retrosternal pain and ECG changes grow worse, the patient is not able to do permanent work and must be considered as disabled. Monitoring Frequency. Monitoring should be considered for each individual patient , depending on the severity of his condition, the frequency of pain, and the effectiveness of care ; it is obligatory after sanatorium treatment. Inquiry into symptoms - physical examination. Scope of accessory examinations - ECG, periodic examination of the levels of serum cholesterol, lipids, and ~-lipoproteins; in the case of obesity and diabetes that run in the patient's family, the glucose tolerance curve should be analysed after a dose of glucose. National Institutes of Health consensus development conferences: screening for breast cancer The National Institutes of Health consensus development programme is a formal process of assessment that serves as an essential step before new tech- nical knowledge enters general use. The process focuses primarily on clinically relevant research, with a view to putting important findings to use in health care or regulation. It assesses research into efficacy and safety and, as appro- priate, the ethical , legal, and cost implications as well. The method of assess- ment is by a public hearing at which all the pertinent viewpoints are heard. The record of the meeting includes a summary of the recommendations and conclusions and is disseminated to research and practising health workers and the public. The summaries also include differing views, in order to present the whole picture. The following describes the process in more detail for breast cancer screening. 20 In 1973-74 demonstration projects were initiated at 29 locations around the country under the auspices of the National Cancer Institute and the Ameri- can Cancer Society. Their purpose was to demonstrate the feasibility of periodic screening of large numbers of women for breast cancer, using the clinical history , physical examination, mammography, and thermography. Following the inception of this programme, however, questions were raised about the possible risks of ionizing radiation from mammography. In October 1975 the National Cancer Institute established three study groups to lead investigations into various aspects of these issues, by the analysis of data generated by a randomized control study conducted by the Health In- surance Plan (HIP) of Greater New York, which began in 1963. One group examined the benefits of screening; a second considered and estimated radi- ation risks; and a third re-examined the pathology of cancers. Subsequently - in January 1977 - a working group was charged with reviewing in depth the findings of the demonstration projects. In September 1977 the National Institutes of Health and its National Cancer Institute sponsored the first consensus development conference on mass breast cancer screening, including mammography. A panel of scientists, practitioners, lawyers, ethicists, and consumers was asked to produce a list of conclusions and recommendations about breast cancer screening, including mammography. The panel reviewed the reports of the four study groups mentioned above, and heard testimony from interested professionals, associations, demonstration project directors , and members of the public. Throughout its deliberations the panel repeatedly emphasized the distinction between mammography used for diagnosis - the value of which was not in question - and mammographic screening to detect possible disease in women who have no symptoms or physi- cal findings whatsoever. The following summarizes some of the findings of the panel, a number of which were later adopted by the National Cancer Institute. The data from the HIP study indicate that periodic breast cancer screening can decrease the number of deaths due to breast cancer by about 40% in women over 50 years of age, but thus far show no decrease in breast cancer mortality attributable to screening women below the age of 50. The benefit of the screening programme rests on the use of physical examination and mam- mography in combination. Neither physical examination nor techniques such as thermography or ultrasound are known to have harmful effects upon the body. The use of mammography, however, is associated with the risk of radiation expo- sure; and breast tissue is particularly susceptible to radiation damage . Cur- rent evidence suggests that risk increases linearly with increasing dose. It is estimated that every rad of exposure raises a woman's risk of breast cancer by about 1 %. This puts a limit on the advisability of repeated screening of the same population. 21 New diagnostic screening techniques are needed. The panel recom- mended greater emphasis on research with non-invasive techniques such as thermography, ultrasound, and biological markers, and recommended that thermography be discontinued as a routine part of the screening dem- onstration programme. The panel deplored the lack of clear-cut data on the efficacy and the risk-benefit ratio of screening for women under 50 , but did not come to an agreement about the feasibility and logistics of randomized clinical trials to resolve such issues. The panel recommended that the programme's informed consent form indicate the radiation dosage to be delivered to the patient and that all infor- mation gained through the programme be disclosed to the person screened as well as to her physician, so that they are able to discuss the proposed procedure . The panel recommended that the histology of any lesion smaller than one centimetre in diameter , or any papillary or intraductal proliferation originally interpreted as malignant , should be reviewed by at least two pathologists prior to definitive therapy. Any new experimental study should take into consideration a variety of issues, e.g. its justification from a cost-benefit point of view, the informed consent process, the way in which research subjects are selected, and the de- velopment of guidelines for the compensation of individual participants who are injured in the course of the study. Furthermore, more women, both pro- fessional and consumer representatives, should be included in the design and planning of any future studies. The panel recommended that screening using mammography and physical examination in combination should be continued for those women 50 years and older who are currently enrolled. Regardless of the location for mam- mographic screening, upper limits should be set on radiation exposure consist- ent with the best current data. Women subjected to mammography should ask for such information and be urged to maintain their own personal exposure records. The panel found no convincing justification for routine mammographic screening for women under the age of 50. This does not imply, however , that physical examination and breast self-examination are not important for women of any age. The panel recommended that routine mammography for women aged 40 - 49 should be restricted to those with a personal history of breast cancer or whose mothers or sisters have a history of breast cancer. Mammographic screening of women below the age of 40 should be limited to women with a personal history of breast cancer. A large number of other consensus development conferences were later held on subjects ranging from the surgical treatment of obese patients to the avoidability of insect stings. The Office for Medical Applications of Research was formally established in October 1978 in order to aid and coordinate the efforts of the constituent institutes in the process of consensus development 22 and to help translate the scientific results of research and development into information that can be used effectively in the practice of medicine and public health. In addition, a National Center for Health Care Technology was created within the Department of Health, Education, and Welfare. The Center's role is to undertake and support comprehensive technology assessments taking into account the safety, effectiveness, cost-effectiveness, and the social, ethical, and economic impact of health care technologies. The Congressional Office of Technology Assessment collaborates in collecting and disseminating informa- tion about various technologies. In addition, standards have been widely used in the review and certifi- cation of hospitals in the USA. They have been linked to guarantee high quality in Medicaid (health care for the poor) and Medicare (health care for the old) programmes. The main programme for setting health care standards and reviewing performance is conducted by the Professional Standards Review Organizations (PSROs). Professional Standards Review Organizations The PSRO programme was established by federal legislation in 1972 to ensure that health care provided to patients under Medicaid and Medicare met appropriate standards in terms of the need and level of care. A PSRO is essentially a group of local physicians operating under contract with the Government to help ensure that institutional health services provided to beneficiaries of government insurance programmes are necessary and appro- priate. Since nursing homes were included under Medicare and Medicaid, they also became part of the PSRO programme. While it is also concerned with quality control, it is clear that the major purpose of the PSRO programme is to eliminate unnecessary services, cut waste , and thereby reduce costs. Of particular concern are the costs associated with surgery and hospitalization, and most PSRO efforts have been directed toward reducing these. In general the programme is charged with reviewing the necessity, quality, and appro- priateness of care. Quality is defined as the meeting of professionally recognized standards of care, and appropriateness as the possibility that services could not be provided at a lower cost in an alternative setting. Study of PSROs in long-term care In recognition of the special problems involved in the review of long-term care services, the Federal Government in 1976 funded 15 demonstration pro- jects in which PSROs were to review the care provided for patients in long-term care facilities. In 1978 the Rand Corporation began a study of ten of these PSROs to determine whether they were achieving their objectives and, generally, to assess the usefulness of long-term care standards and review. 23 This study represents the most comprehensive assessment in the United States of the standard-setting and review process for long-term care facilities. The demonstration projects were located in vastly different areas - rural, suburban, and urban - and went about their assignments in very different ways. This, of course, complicated the study, in addition to the lack of useful baseline data, the freedom each PSRO had to determine its own objectives, and other such differences. To help develop a framework for evaluating the ten PSROs, a series of questions was developed, questions that were central to the research project but are important to the review of standards oflong-term care in general. They cover the relative emphasis that should be given in the review to particular kinds of patient - new admissions, long-stay patients, various diagnostic groups, etc. What should the review of a long-term care facility focus on and what should it achieve? What, in fact, are the appropriate out- comes for long-term care and what kinds of indicator can be developed for them? Are nursing home mortality rates and discharge rates, for example, indi- cators of quality of care? What about drug regimens or perhaps functional measures such as patient happiness and satisfaction? In regard to standards for outcomes, the researchers felt that it was not possible to use acute care experience in establishing criteria for long-term care; for one thing, diagnosis-based criteria are of limited utility because of multiple diagnoses. Should criteria be based on health problem clusters (e.g. skin problems or sight problems) or on functional status (e.g. ambulation, self- care, or mental orientation)? How does the PSRO develop criteria for such psychosocial problems as loneliness or depression, which may be a by-product of admission to a nursing home? Other questions had to do with the level of care and with costs. Should standards of care reflect norms of practice (even when they are quite low) or the ideal? How much should the PSRO be concerned about whether high standards inflate costs or even reduce the quality of life by driving out of business small family-operated facilities that may not be able to meet the standarr.s? In addressing all these difficult questions the researchers felt that PSROs need to consider expanding the review group beyond phys- icians to include nurses, physiotherapists, social workers , pharmacists, and administrators. There were also questions about data collection. What are the best sources of information for the PSRO to use in its review - staff interviews, clinical re- cords, patient interviews, or other sources? To what extent should the staff of the facility assist with data collection and what effect is it likely to have on the validity and reliability of the data? How complex should the data be? Findings Despite the methodological problems the study, conducted through a materials review, site visits, cross-site comparisons and analytic processes, 24 produced valid and important findings. The findings are highly detailed and as such difficult to summarize. However, those that appear most important are summarized below. 1. Explicit standards that can be used to determine the appropriate level of care before the admission of patients were generally not developed by the ten demonstration programmes. Some beginnings were made, however, and several of the PSROs were able, through a pre-admission review, to identify patients who could have remained in the community if community resources were available. With regard to this, the study recommends that work be done towards developing a "reliable instrument for determining need for nursing home placement and level of care". 2. A programme of pre-admission review using explicit criteria is much more desirable than review after admission because it is less expensive and less disruptive to the patient. "There is some evidence that an aggressive pre- admission review program can reduce nursing home days under Medicare - by 20%." However, it was also clear that patients denied admission will require and receive services from other sources. 3. For continued stay review (a review of patients at regular inter- vals to determine whether the care and length of stay is appropriate), ex- plicit standards have not yet been developed and are needed. It did appear, however, that a number of nursing home patients were receiving inappro- priate levels of care. For example, in one area the PSRO found that 25% of the chronic disease hospital population did not require this level of care and that three quarters of those placed at the skilled level in nursing homes should not have been. 4. The review process identified many deficiencies in the quality of nurs- ing home care. The researchers concluded that many quality-of-care problems can be detected through a review of the medical records. They also found, however , that the great majority of personnel working in long-term care facil- ities felt that reviews to determine quality are best done through direct patient interviews. Here again the study found that the standards used were not ad- equate and "continued work is necessary to develop explicit criteria for quality. Particularly in the psychosocial sphere, criteria are rudimentary and need extensive work". The study suggests that the need to develop standards and criteria is common to all the areas studied and that work on this should be done at national rather than local level. Such criteria could then be made available to individual PSROs to be modified as appropriate. The study also pointed out that "at a number of sites, a problem developed because of the lack of norms for the care of the nursing home patient. In the absence of these norms, standard setting and standard enforcing become difficult". 25 5. Data collection was often a problem and "the reliability of data collection on measures such as functional status cannot be assumed". 6. The cost of the long-term care review varied considerably between sites because of differences in definitions, data collection, cost allocations - the review activities themselves. Until there is greater standardization of these processes, no useful cost estimates can be made. 7. The PSROs were not able to follow up their recommendations to assess the degree to which they were adopted and brought about posi- tive changes. As a result, most of the "evidence" presented to demonstrate that they had some effect was anecdotal. In general , the study concluded that "the experiences of 10 demonstra- tion projects in long-term care give cause for some optimism about the potential for the PSRO in long-term care review. Despite the problems in demonstrating impact . . . innovative programs were developed, quality of care problems were identified . . . The potential of the PSRO to affect quality of care seems great . In summary, . . . review programs in long-term care , as exemplified by the 10 demonstration projects, can serve an import- ant function ensuring that the large sums of public money spent on nursing homes are purchasing better care than is currently the case. We expect , how- ever, that PSRO activity in long-term care review will not bring about dramatic reductions in the total public dollars expended in nursing homes, although it should ensure more appropriate expenditures. The PSROs show great promise in the difficult task of further defining and monitoring the quality of nursing home care and ensuring that nursing home residents receive both humane and technically sound care." In its own assessment of the use of standards in its accreditation pro- gramme, the Joint Commission on Accreditation of Hospitals indicates that a number of positive changes have taken place. These include a shift in emphasis from custodial care to active treatment , a greater regard for the rights of patients, improvement in the quality of clinical staff, a de- cline in the length of stay of patients, and an increase in the number and variety of vocational, educational, and rehabilitation programmes for patients. It is clear from the work done so far that the principal vehicle through which standards are implemented and enforced is the survey team. The expert- ise and composition of such a team are extremely important , as well as the availability of training for the team members. Competent people are too frequently not available and the training is inadequate . For example , the Rand study found that there was difficulty in "engaging physicians in criteria development and in vigorous application of peer review around quality issues". The extent to which highly trained and competent pro- fessionals are willing to participate in the process of developing standards 26 and monitoring their effectiveness will have a major impact on any procedures to review the quality and cost of long-term care. The PSRO programme is one possible way of evaluating the implementa- tion of guidelines within the health service system. Guidelines on cancer control in the Federal Republic of Germany and the Netherlands Much attention has been devoted to the control of cancer, and many countries have produced guidelines on prevention , screening, diagnosis, treatment, and follow-up. Both in the Federal Republic of Germany and in the Netherlands professional groups are engaged in producing guide- lines to help improve the quality of care within a health service system where most physicians work on a fee-for-service basis. In the Federal Republic of Germany comprehensive concepts and re- search programmes have been developed, models tested, and laws passed with the aim of setting up standards for the improvement of (1) interdis- ciplinary collaboration and (2) centralized management of patients with cancer. Standards for improving early detection of cancer with a legislat- ive basis have been implemented, and preventive medical examinations have been introduced. Every woman over 30 and every man over 45 has the opportunity to undergo an annual medical examination. The standards prescribed so far aim at early diagnosis in women of cancer of the cervix and uterus (cytological examination) and of the breast {palpation, inspec- tion), and in men of deep-seated carcinomas of the rectum and prostate gland by digital examination. Standards are also set for follow-up when suspicion appears to be well founded. So far there are no binding regulations for standardized (interdisciplinary) cancer treatment. In recent years guidelines on treatment have been established by the German Society of Surgery for 1 7 types of organic tumour. Preliminary therapeutic standards have also been put forward by the German Society of Gynaecology. So far all these recommendations have been based on inter- disciplinary discussions and/or compromises. None of the standards has been reliably and scientifically substantiated and verified by reproducible results. For cancer follow-up care there are at present only non-binding guidelines. In order to make reintegration into social, and especially professional , life easier for the cancer patient, and to initiate the rehabilitation measures necessary for this purpose, legal standards have been established. The Social Security Regulations (RVO) and the Employee Insurance Act {AVG), among others, entitle every cancer patient to several inpatient treatments, each lasting up to four weeks. This treatment is carried out under the supervision of a physician in numerous hospitals set up expressly for the purpose. Standards for providing long-term or limited invalid status and social aid to patients suffering from malignant tumours have been in existence since 1966. 27 The standards now available in the Federal Republic of Germany have found wide acceptance among physicians and patients. They also serve to pre- vent any adverse effects resulting from therapy. Some evaluation results exist. In the Netherlands guidelines on the diagnosis, treatment, and follow-up of children with leukaemia have existed for a few years . The Dutch Childhood Leukaemia Study Group (SNWLKf was set up in 1972, a coordinating com- mittee established by the Dutch Paediatric Association having been in oper- ation since 1969. Two factors stimulated this initiative: the significant improve- ment abroad in the treatment of childhood leukaemia, especially of childhood acute lymphocytic leukaemia, and the complexity of diagnosis and treatment, which require experience and specialized skills . Because of the relatively low incidence of childhood leukaemia, most paediatricians in general hospitals have only limited experience with the diagnosis and treatment of children suffering from these disorders . The objective of the SNWLK was to achieve the best treatment possible and to provide the maximum chances of survival and cure for every child with leukaemia in the Netherlands. Within the SNWLKknowledge about and experi- ence with this rapidly changing field of medicine is available and kept up to date for all paediatricians by cooperation, coordination, and centralization. The SNWLK also offers an opportunity for the mutual exchange of experi- ence and knowledge and a basis for sound statistical evaluation . In the Netherlands children with leukaemia are treated in general hospitals as well as in specialized centres. They are admitted for initial treatment (in- duction treatment), but during the subsequent maintenance treatment (24 months) are usually treated as outpatients. Approximately 95% of all Dutch paediatricians cooperate with the SNWLK, which has a medical board and a central office. The task of the board is to promote curative treatment for children with leukaemia , to promote the development of diagnostic procedures, treatment regimens, and basic research, and to bring about acceptance of these developments by paediatricians. The members of the board are paediatric haematologist-oncologists from special- ized centres and two paediatricians representing paediatricians in general hospitals. All activities in the field are concentrated in and directed from the central office. Activities in the field (Fig. I) cover recommendations to paediatricians on diagnostic procedures and treatment regimens (protocols), and diagnosis of the type of leukaemia and assessment of the stage. In the central laboratory of the SNWLK blood and bone marrow smears of all patients are examined morphologically and cytochemically to determine the type of leukaemia. Im- munological investigation of cell and membrane characteristics of leukaemic a Stichting Nederlandse Werkgroep Leukemie bij Kinderen (SNWLK), Nederlandse Vereniging voor Kindergeneeskunde. 28 Fig. 1. Activities of the Central Office of the Dutch Childhood Leukaemia Study Group (SNWLK)8 ~ No leukaemia~ ~ 1nvest1gat1on ~ Leu kaemia suspected ~SNWLK Diagnosis uncertain ◄ ► :urth~r _ p.,dianicia" L'"kaemia ----------- No<""'•: SNWL~ Fal,e ""'""' diag"oo/ Missed diagnosis ------. SNWLK --~"'---------1• Diagnosis confirmed ~ Assessment of the type of leukaemia ~-----==31 ~ Treatment ◄ Epidemiological studies By protocols --------. By protocols confirmed by / not confirmed by the SNWLK the SNWLK board ~ board Follow-up • • Evaluation and statistical analysis .. Revision and refinement of diagnosis and treatment t5 8 By courtesy of Dr E.R. van Wering (2). cells from newly diagnosed patients is performed in cooperation with the central laboratory of the Netherlands Red Cross blood transfusion service. These examinations are standardized and carried out by at least two ex- perienced independent investigators. Childhood leukaemia is a heterogeneous disease. Accurate determination of the type and stage of the disease is de- cisive for the choice of mode of treatment and it also reduces the risk of false positives. Protocols designed by the SNWLK, giving guidelines for diagnostic pro- cedures, treatment, and follow-up of patients with different types of leu- kaemia, are sent to paediatricians at their request. In this way the SNWLK has performed several prospective comparative studies ( controlled clinical trials) on childhood acute lymphocytic leukaemia. In addition, protocols with guidelines for the investigation of long-term and iatrogenic side effects of the disease and the regimen applied are distributed for the use of paediatricians who have pa- tients with long-standing remission after cessation of therapy. The course of the disease, the results obtained, and the complications are monitored by regular examination of blood and bone marrow smears in the SNWLK's central laboratory, and by registration forms, filled in at regular intervals by paediatricians during treatment and after cessation of therapy. In relation to epidemiological studies on the incidence and geographical distribution of leukaemia as possible indicators in the etiology and pathogen- esis of leukaemia, all the data and results are computerized and analysed stat- istically in close cooperation with well-known institutes. On the basis of the results of SNWLK studies and of studies performed elsewhere, new protocols are designed for diagnostic procedures and treatment regimens. The ultimate goal is to increase the chance of cure, due account being taken of the quality of life. The feasibility of electron microscope analysis ofleukaemic cells, to refine further the identification of the type of leukaemia, is now under investigation in the SNWLK. Childhood leukaemia is a group of diseases of unknown cause. Its treat- ment is neither uniform nor standardized; rather, there is a continuous search for more effective treatment based on growing knowledge and the develop- ment of therapeutic regimens built on insight into differences within the group. This has already resulted in a remarkable increase in survival. A considerable proportion (50%) of patients, especially those with acute lymphocytic leu- kaemia, have a chance of cure (Fig. 2). Recently the board decided to extend SNWLK activities to include other myeloproliferative and lymphoproliferative diseases. So far cost-effectiveness analysis has not been considered to be a task for this organization. The financial resources of the SNWLK have been provided by the Ministry of Health and Environmental Hygiene (1972-1975) on a grant basis, the Queen Wilhelmina Cancer Foundation (1975-1980) from private funds, and social health insurance funds. 30 w Fig. 2. Survival of children with acute lymphocytic leukaemia in the Netherlands % survival 100 80 ·.~:••::• .... 60 40 20 •• •• :.::•·•:·•···.. H ·.,.... ... -- \.\ ... ;,:··.:::::::::· . ----------- - •• ........... ··• .. :,. ::::·········-=------- ••• ::::::~;;: :~::.:::::;::::,::'.:::E:::'.:.~,~:::;;;;;:;J:i::ii:::::::::.:t: ::::;;::::::::c,,, 0 ---..----.---..-----.---........ ---,---,----,,----r---..-----.---..----.-----, 0 4 A = 1965 (N = 45) B = 1966 (N = 29) C = 1967(N = 55) 8 12 16 Retrospective study Source : van der Does-van der Berg, A . (3). 20 24 28 Months D = 1968 (N = 57) E = 1969(N = 56) F = 1970 (N = 76) 32 36 40 44 48 52 56 Prospective study G = 1972-1973 (N = 70) H = 1973-1975 (N = 215) COST-EFFECTNENESS ANALYSIS IN HEALTH CARE Cost-effectiveness analysis is one of a family of techniques known collec- tively as the cost-benefit approach or economic appraisal. These techniques , the other main one being cost-benefit analysis, are concerned with appraising alternatives in health care from the perspective of economic efficiency. Econ- omic efficiency is a relevant perspective because the resources used in health treatments or programmes are scarce. That is, since there are not and never will be enough resources to satisfy human wants completely , their use in one beneficial activity means that the community automatically forgoes the oppor- tunity to use them in another beneficial activity. Consequently, in the appraisal of alternative patterns of care not only their relative effectiveness but also their relative cost (in terms of resource use) should be considered , since extra costs will imply benefits lost elsewhere. This can be seen within the health systems of many countries ; pressure for high-technology medicine often means that less spectacular services, such as preventive measures or community care for the elderly, do not expand as much as could be desired. To look on resource costs as lost opportunities rather than merely finan- cial outlays is a useful approach, since more cost-effective or more efficient methods of delivery of services do not necessarily imply cuts in expenditure or cash savings but just as much the obtaining of more benefits from the existing budget. For example, if it were found that short-stay or day-only surgery was a more cost-effective way of treating patients with some conditions, increased adoption of such a method would mean that resources (in this case hospital bed days) were freed for other uses. Although in the long term it might be possible to close some wards, thereby cutting expenditure, it is more likely that the beds and medical and nursing time saved would be utilized to treat other cases. Thus it is important to distinguish between searching for a more cost- effective or efficient use of resources and simple cost cutting.0 The economic approach requires explicit consideration both of costs and of the extent to which alternative strategies meet the chosen objectives (i.e. , their effective- ness). There is no single way of measuring the effectiveness of alternatives in a cost-effectiveness study, and typically such studies build on the work of medical evaluators. Effectiveness may be a multidimensional concept including a range of indicators of quality of life or health status, not limited to narrow a Some authors find it convenient to specify an intermediate level of analysis, sometimes known as cost-productivity analysis. This is concerned with producing units of output at least cost without considering whether the output is of value. For example, such an analysis might investigate whether brain scans were being undertaken at the low- est possible cost, without explicitly considering whether they had a beneficial effect on the outcome of treatment. 32 economic indicators such as ability to return to work (although it might be argued that if a patient is able to return to work this is an indicator about his general improvement in health). The fact that economic efficiency can be justified as a legitimate perspec- tive does not mean that the cost-benefit approach, including cost-effectiveness analysis, is all that is required. Cost-effectiveness analysis requires good infor- mation about the effectiveness of therapies, and consequently more medical evaluation is r,~quired. There may also be other legitimate objectives in health care , such as equity . Equity and efficiency may sometimes go hand in hand ; for example , if a given condition could be treated with no loss of effectiveness but with fewer resources , the service could be made more widely available . Sometimes, however , the two objectives may be in conflict ; for example, it might be cost-effective to centralize some specialist services in a few geographical locations, but equitable to provide them in more locations on the ground of equality of access . Similarly it might cost more to attract some disadvantaged groups to a screening programme, but be equitable to commit extra resources in order to give them an equal opportunity to receive the service. There is no easy resolution of such conflicts, but one way to pro- ceed would be to agree on the boundaries of what is acceptable on grounds of equity and then pursue the objectives in the most efficient or cost-effective manner. The application of economic appraisal A recent review ( 4) of the literature in English has revealed over 100 stu- dies of economic appraisal in health care , and there are others in other languages. Some of the studies merely examine the costs of alternative patterns of care , thereby providing important information on the use of resources to decision-makers. Others examine the cost and effectiveness of alternative ways of meeting the same objective . Finally, some studies tackle the more complex question of whether given objectives are worth meeting at all or of how much they should be pursued. This last set of studies is inherently much more com- plex and some of them (known as full cost-benefit analysis) try to put values on the outputs or benefits of particular health programmes (Fig. 3). Since some of the valuations placed on benefits in cost-benefit studies are the subject of much debate , it is not surprising that the more limited cost- effectiveness approach, which reveals the costs of alternatives in order to compare them with relative effectiveness, has gained more widespread recog- nition. Indeed it can be demonstrated in some situations that both the alter- natives being examined have the same effectiveness and thus the alternative with lower costs can be chosen. Although evaluation of the benefits of health programmes has ultimately to be made, the emphasis here is on the use of economic appraisal as a systematic way of thinking and of providing extra information to those who will ultimately take decisions about patterns of 33 Fig. 3. Distinction between cost-effectiveness and cost-benefit analysis consumer resources Costs: What is forgone? Cost-benefit analysis Treatment objective can be questioned costs and benefits in money terms Source : Drummond, M.F. (5) . Health programme or treatment gives improvements in health Cost-benefit approach (economic appraisal) Benefits: What is obtained? benefits not in money terms Cost-effectiveness analysis Treatment objective not questioned care . It is not to be thought of as a way of giving the decision-maker the answer, since any answer inevitably incorporates important value judgements. The object is to make the decision-maker more aware , not less aware, of such judgements . Some examples of economic appraisal in health care Economic appraisal has been used in many fields of health care . In this section it is proposed to give a general description of the kinds of issues that have been explored rather than a detailed description of a limited number of case studies. Those wishing more detail of English-language studies should con- sult the summaries of existing work given in Drummond (4) or the case studies (in elective surgery, chronic renal failure , screening for cancer control and care of the elderly) developed in Drummond (5) . Other useful sources are Bunker et al. (6) and Mooney et al. (7). 34 Prevention There have been a number of studies of the cost-effectiveness of immuniz- ation programmes. In the main they compare the cost of mounting the pro- gramme with the savings arising from prevention, and set the net cost (if any) against the humanitarian benefits. A similar approach is adapted in a large number of studies of water fluoridation in the dental field. The largest group of studies is coucerned with screening: the costs and benefits of multiphasic screening; screening for breast cancer and cancer of the cervix ; chest X-rays for tuberculosis control; screening for phenylketonuria; and screening of the un- born. Again these studies concern themselves in the main with the resource consequences, i.e., the cost of mounting the screening programme and the an- ticipated savings in the cost of caring for those who would otherwise need treatment. These are then set against the medical effects, say in terms of the net cost of extending life. In many of the studies the major problem is the re- liability of the medical evidence, as there are few controlled prospective stu- ies of screening programmes. Studies of particular interest to those working in the field of chronic diseases are those by Schweizer (8) and Dickinson (9) on cervical cancer, Kodlin (JO) on breast cancer, and Steiner & Smith (I 1) on phenylketonuria. Diagnosis There have been a number of studies comparing the relative costs and ef- fectiveness of alternative methods of diagnosis . Those of particular relevance to the field of chronic diseases are by Bartlett et al. (12) and Jonsson & Marke (13) on CAT scanners, and McNeil et al. (14) on the diagnosis of hypertensive renovascular disease . The most common approach is to ascertain whether new diagnostic technology leads to cost savings by reducing the use of existing procedures, thus justifying its adoption on resource grounds alone. More sophisticated work examines the sensitivity of the cost-benefit equation to the type of caseload and the clinical indications taken, say, for brain scanning. An important issue is whether the new technology actually brings about benefits in terms of improved care , hence in outcome. Evidence that it does is often lacking, some studies merely presenting any cost differential to decision- makers for them to take into account alongside medical opinion on the effectiveness of the new technology. Treatment and rehabilitation The largest group of studies falls into this category. Some of the most in- fluential work has been in elective surgery, where economic analysis has been linked with randomized controlled trials , e.g., Russell et al. (15) on the cost- effectiveness of day-case surgery. Other important work has concerned 35 balance of care for the elderly, e.g., Wager (16) and Mooney (J 7), where econ- omists have obtained more detailed infonnation about the costs (to the health sector and to other agencies and the family) of domiciliary care. Examples of studies dealing with alternatives in the chronic diseases field are those by Carter et al. (J 8) on the relative costs of home and hospital care for haemo- philia, Bryant et al. (19) on a comparison of care and cost outcomes for stroke patients with and without home care, Cretin (20) and Sissouras &Moores (21) on coronary care units, and Weinstein et al. (22) on a comparison of coronary by-pass surgery with medical management. Some conclusions on methodology This review of studies in English suggests that a methodology is beginning to emerge for appraising choices in health care from an economic perspective. Broadly speaking, the range of costs and benefits taken into account can be depicted as changes in health service resource use , use of other community resources, and changes in health state brought about by the alternatives being appraised (Fig. 4). There are two points in particular to note about this framework. First, it is usual for economists to consider resource use outside the health sector, al- though the extent to which individual analysts do so depends on the particular alternatives being appraised ; in principle economists are interested in all re- source use associated with health programmes, no matter whose budget the expenditure falls on. Second, the range of factors typically considered includes changes in productive output brought about by , say, treatment allowing pa• tients or their families to return to work. Because of the difficulty of estimat- ing the humanitarian benefits of programmes or treatments_(i.e. , the value of changes in health state per se shown in Fig. 4) it could be argued that these production benefits have been given too much emphasis. Those considering the results of economic appraisals should take note of this , as exclusion of hu- manitarian benefits could lead to programmes for the elderly or mentally handicapped showing a poor return. This reinforces the case for a cost- effectiveness rather than a cost-benefit approach, the object being to obtain good estimates of the resource savings brought about by programmes and to combine them with good data on programme effectiveness (in tenns of the quality of life). The question of whether production savings should be taken out of the cost of programmes in cost-effectiveness studies is still open to de- bate (23). Many of the methodological issues are still being discussed by economists, but a number of comments may be made about the research currently being carried out. The quality of the medical evidence on which studies are based var- ies greatly. The better studies are based upon controlled medical evaluation and , where it is lacking or there is disagreement about the objectives of treatment or programmes (and thus about the measurement of success) , the economic 36 w -..I Fig. 4. Relevant changes in a comparison of the economic efficiency of treatments Hospital Community <::are Source: Drummond, M.F . (5). Changes in health service resource use Economic efficiency of treatments Changes in other community resource use Money Time Lost time may be work time Value of changes in health state per se Chanc:ies in health state Changes in productive output Patient Family_ ... t --~ I I ---------------- - Patient Family, friends, other individuals appraisal will be correspondingly uncertain. Although the cost side of the equation is less fraught with difficulties than the estimation of benefits, it is not without its problems. For example, some studies omit important costs, especially those falling outside the health sector. Finally, all appraisals are subject to uncertainty about the values of key medical and economic variables, the better studies incorporating a sensitivity analysis, i.e., an ex- amination of the way in which the result changes when key variables such as the size of the programme and compliance with the programme change. A sensitivity analysis is particularly important where the medical evidence is incomplete. Despite these problems economic appraisals can generate useful in- formation if interpreted intelligently. Using the results of economic appraisals The first point to remember is that there is no single simple basis for making choices in the health care field. The key to better decisions lies in the use of systematic techniques, such as economic appraisal, as an aid to decision-making rather than as a substitute for judgement. It is there- fore important for decision-makers to understand these techniques and to avoid committing themselves to value judgements unknowingly by ac- cepting results without question. This apart , an important distinction to make is between the decision in the planning context and the decision in the clinical context. For example, a planning decision might be on whether to expand the facilities for day-case surgery or to change the balance of provision between institutional and domiciliary care for the elderly . By con- trast, a decision in the clinical context would concern what is best for an individual patient or client. Obviously the two contexts are linked; the treatment of one patient is probably never seen by the clinician in isolation from the needs of other patients he is responsible for, and clinicians, in de- veloping their own services, are in essence making planning decisions. It seems that the studies carried out have generated useful information for decisions in the planning context. They have given an indication of whether on average one pattern of care is more cost-effective than another or whether it would be cost-effective, say, to mount or extend a screening programme. Nevertheless, the results of the studies do not apply in every situation. For example , the cost of alternative patterns of care may depend upon what is currently available in given localities; the question whether day- case surgery is likely to be cost-effective will depend on whether a day- case unit has to be constructed, the amount of community nurse support available, and the size of the expected caseload. So whether a computer- assisted tomography (CAT) scanner is likely to be cost-effective in a given locality may depend in part upon the number of cases with the appropriate clinical indications. 38 The interpretation of the results of economic appraisals is more com- plex in the clinical context. Here decisions are taken on the basis of infor- mation about the individual patient rather than the study population as a whole. Nevertheless there are a number of ways in which cost-effectiveness information can be of use. It may highlight the costs of therapies that have never been demonstrated to be effective (this is basically an epidemiologi- cal point , rather than an economic one, and is made admirably by Cochrane (24)). It may encourage the increased adoption of therapies that are at least as effective and less costly for most of the patient population. Where the more effective alternative is also the more costly , the appraisal may stimulate a search for sections of the population for which the less costly alternative would be equally effective . Such a section would be defined in terms of clinical (and social) indications such as severity of the condi- tion , previous history, and home circumstances. Failing this, economic appraisal can make explicit the cost, in terms either of money or of ser- vices forgone elsewhere , of opting for the more effective treatment. Neutra (25), in a study of the indications for surgical treatment of suspected acute appendicitis, demonstrated that to save one more life by changing the symp- tom severity used as an indication for operation would involve the perform- ance of so many operations on normal patients that the incremental cost of saving the life would be US$ 43 million and 2053 person-years of convalescence. In short , in the clinical context it would appear that economic appraisals do not give hard and fast rules for individual clinical decisions. They should, however , give due emphasis to one criterion that is important when deciding upon clinical policy. Economic appraisal and guidelines on patterns of long-term care Comparatively few studies have been made of the cost-effectiveness of alternative patterns of long-term care. The discussion on economic ap- praisal raises the following questions in relation to guidelines on such care. Are guidelines on patterns of long-term care based at all on economic appraisal of the alternatives concerned? If so , what is the quality of that appraisal? Are guidelines likely to encourage providers to think more about cost- effectiveness? Do they , for example , mention resource use as well as ef- fectiveness? Do they discuss therapies or practices that should be discon- tinued as well as those which are to be preferred? If studies are to be carried out of the impact. in terms of cost-effectiveness, of guidelines on long- term care, how can it be ensured that they meet appropriate methodologi- cal criteria on study design , medical evidence , assessment of costs , and sensitivity analysis? 39 FACTORS IN RECOMMENDING GUIDELINES FOR LONG-TERM CARE The central question for the Workshop was whether it is worthwhile, in terms of cost-effectiveness, to recommend standard patterns of long-term care. However, although it was the central question, it is important to note that there are a number of possible objectives for guidelines or standards, for example to attain minimum levels of safety or of quality irrespective of cost. In some countries, too, guidelines are thought to be an essential feature in the operation of a centralized health system. The study phase showed that these other objectives, not the pursuit of cost-effectiveness, are the most common reasons for recommending guidelines. An import- ant first step therefore in assessing guidelines for care is to clarify the pur- pose for which they are devised. Guidelines devised with other objectives in mind will not always work against the objective of cost-effectiveness. For example, standards aimed at greater medical effectiveness may result in higher cost-effectiveness, particularly if they lead to the discontinuation of dangerous or ineffec- tive therapies at little extra cost. Again, some standards may be justified in cost-effectiveness terms if the gains in terms of effectiveness or safety more than compensate for the extra resources consumed; standards for control of infection in operating theatres might be regarded as falling into this category. Obviously any assessment of whether such standards are worthwhile in terms of cost-effectiveness will be much more difficult than in the simpler case of a guideline that guarantees the same level of medical effectiveness while reducing costs (as when the same drug is prescribed by the generic rather than the proprietary name). Although it would be wise to tackle these simpler cases first , it is likely that the more complex questions would soon arise . Finally, there may be circumstances where the cost-effectiveness objective is explicitly not being pursued when setting standards; in the United Kingdom certain mini- mum levels of service provision may be set on the grounds of geographical equality throughout the country, irrespective of the cost-effectiveness implications. It should also be recognized that guidelines are not the only method of obtaining more cost-effectiveness in care. Other methods include budgetary measures (i.e. , either fixed financial limits on the amount of resources available for care or budgetary incentives for providers linked to the adoption of cost- effective patterns of care), other non-financial incentives to providers, research into the cost-effectiveness of alternative patterns of care , and the education of providers in cost-effectiveness considerations. These ·methods also deserve note as they could be used in association with guidelines, depending on the health system under consideration. 40 Relevant factors in recommending guidelines for long-term care The Workshop considered a number of factors likely to affect the impact , in terms of cost-effectiveness, of recommending guidelines. It was suggested that , taken together, they constitute an a priori assessment of whether a particular guideline is likely to lead to more cost-effectiveness or not. As far as possible , an opinion was expressed on whether the influence of various factors is likely to be positive (that is, likely to lead to cost-effectiveness), nega- tive , or neutral. The factors were grouped under three headings: the back- ground to the guidelines , the content of the guidelines, and the implementa- tion of the guidelines. Background to the guidelines The main objectives in recommending guidelines in the participating countries are (I) the maintenance or improvement of the quality of care and safety, (2) the stimulation of interest by the improved education of providers of care, (3) the prevention of malpractice , (4) cost containment and/or cost- effectiveness, and (5) the improvement of organizational performance. Since very few recommendations have been explicitly aimed at bringing about more cost-effectiveness, it is unlikely that the objectives behind most guidelines have had a direct influence to this effect. However, they may possibly have led to cost-effectiveness indirectly. For example, if standards set in the pursuit of cost containment lead the providers of care to be more considerate in their use of resources and to make economies without sacrifices in quality of care , they may lead to cost-effectiveness. Similarly , if when aimed at maintaining or improving the quality of care they lead to improvement without large in- creases in resource use , they could lead to cost-effectiveness. In this connexion it is particularly relevant to note the approach in Sweden and Finland, where the objective of medical care programmes is to increase the quality of care within existing resources . Source of the guidelines The main sources in the various countries are the medical and other health professions, governments or their health planning agencies (central or local), insurance agencies and sick funds , or a combination of the professions and one of the others. It seems fairly clear that a combination of the professions and the providers of resources gives the best chance of a proper consideration of effectiveness and cost. Some professional bodies , e.g. , the Royal Colleges in the United Kingdom, have made it clear that , although they understand the resource problems, their concern is with quality . Obviously guidelines likely to lead to cost-effectiveness will depend on the closeness of the association be- tween the professions and the resource providers. This appears to be good in 41 Sweden, and one set of standards in Yugoslavia was developed through such an association. The Workshop felt fairly strongly that the factors involved in the develop- ment of guidelines should not be purely medical. This is particularly important in long-term care , with its emphasis on self-reliance and independence of the individual and on coordination between health care and other services. There- fore, although resource providers may turn to the medical profession as a key group in recommending guidelines, they should aim at the widest possible in- volvement of other professions and consumer organizations. The involvement of consumer groups has so far been fairly small . Examples exist in Sweden, where the Diabetes Federation has participated in the development of diabetes guidelines, and in the United Kingdom , where community health councils are involved in health care planning. There is a strong case for giving more education to consumers in aspects of health care. Finally, it was recognized that in most countries academic institutions, with their scientific skills, have a key role in the development of guidelines. Organization of the health care system The health care systems of the participating countries vary greatly. For example, Hungary, Poland, and the United Kingdom have centralized systems, whereas the systems in the Federal Republic of Germany and Yugoslavia are very decentralized. It was felt that the type of health care system is a very important factor to take into account when recommending guidelines, particu- larly with respect to what method of implementation is to be employed and whether the guidelines are to be mandatory or advisory. In a centralized system it may be easier to develop guidelines with respect to , say, the location ofhigh- technology diagnostic aids such as CAT scanners. Also , if a particular service is organized nationally , like that for acute leukaemia in the Netherlands, it is easier to produce national guidelines. However , in most countries with more decentralized services it is likely that national guidelines, if these are developed, will need to be adapted to local circumstances if they are to command the support of providers of care . In general it was felt that it would be difficult to make international recommendations on the shape and form of guidelines and that the objective of the Workshop should be the exchange of ideas and the provision of examples from different countries. Knowledge of the epidemiology of the disease and the results of interventions In some areas epidemiological knowledge is lacking, particularly of the results of interventions. There was much discussion on whether it is better to recommend guidelines in areas where the knowledge is greatest (since recom- mendations can then be made with some confidence), or where it is least (since 42 there may be great variations in practice that have an impact on effectiveness , safety, and costs). The extent of knowledge of the disease and its treatment is a factor taken into account in Finland when deciding upon the conditions for which medical care programmes should be developed. The general conclusion of the Workshop was that guidelines should as far as possible be backed by scientific research. If the objective is cost-effectiveness they should therefore be backed by research into the cost and effectiveness of alternative treatments or programmes. Guidelines could nevertheless have a favourable impact irrespective of the state of knowledge. The key influence of knowledge may be on the approach. If knowledge is extensive , firmer recom- mendations might be made in order to reduce unsafe , ineffective , or wasteful practices . If it is lacking, guidelines might be developed in a way that would stimulate interest in education and research on disease and practices. In any case , it would be worthwhile to have a profile of guidelines, some in areas where knowled ge is greatest and where a quick impact could be demonstrated , others in areas where knowledge is lacking. Existence of bodies with a responsibility for standcrd setting The development , implementation , and evaluation of guidelines require much time and effort, particularly if they are to be adequately researched . Although many countries have professional bodies or national institutes whose function is the maintenance of standards of effectiveness , only in a few coun- tries are there any whose function is to consider issues of cost-effectiveness, such as the Swedish Planning and Rationalization Institute for the Health Services. The existence of such agencies. especially if they have a budget for evaluating alternative patterns of care. is likely to have a positive influence on the cost-effectiveness of guidelines. However , it is impossible to prescribe one universal model for such an o rganization , and it is important to be clear which of the three roles of development , implementation , or evaluation of guidelines such an organization would have a part in . level uf resource input ta the disease group ur treatments in question In principle this factor is likely to be neutral with respect to the cost- effectiveness of guidelines . If the level of resource input is low , no doubt there is an interest in cost-effectiveness . either to stretch the budget further or to ask for more money . On the other hand , if the budget is large there is more interest in cost containment, particularly on the part of the resource providers . It was pointed out that diseases with high levels of resource input are likely to be those placing a high burden on the community in terms of mortality and morbidity. Some countries have developed guidelines for com- mon conditions since they affect :i large proportion of the population in one way or another. 43 From the cost-effectiveness point of view a decisive factor is likely to be the cost of operating the guidelines themselves. If it is high the guidelines need to produce considerable benefits , either reduced costs or improved quality of health practice . This would tend to confine them to diseases with a high bur- den or high resource input. The participants had no strong views on this point. Contents of guidelines for long-term care Phase of care to which the guidelines relate In principle it should not matter whether the guidelines relate to preven- tion, diagnosis , treatment , or rehabilitation. However , the most radical im- provements in the quality of care or cost-effectiveness may be obtained in shifts from a curative to a preventive approach. Therefore, if all phases of care are to be considered , the possibility that guidelines may overemphasize the role of conventional medical treatment should be reduced to a minimum. There may be an important role for guidelines dealing with one aspect of care if that aspect can be dealt with thoroughly. Structure, process, or outcome Medical outcome standards explicitly ignore costs, so they can have a favourable impact on cost-effectiveness only indirectly, as for example if they improve effectiveness without putting up costs. This should not be taken as an argument against outcome standards per se, as it is clear that there are other legitimate objectives in setting standards. However , from the cost-effectiveness viewpoint the cost of meeting the outcome standards should at the very mini- mum be identified. This is done in France for guidelines for the management of haemophilia. The choice between structure and process guidelines is difficult; both are almost certainly required. For example , not only are structural guidelines needed on, say, the location of CAT scanners but also process guidelines on when and how to use them. It is likely to be harder to obtain agreement on process guidelines , and they should perhaps concentrate on situations where costs can be reduced without making sacrifices in effectiveness rather than on those where cost savings imply reduction in effectiveness . It is nevertheless important that process guidelines should include advice on what not to do because it may be costly and unnecessary , as well as on the preferred type of care . Research basis for recommendations Obviously any recommendation is likely to gain more respect, and so have more chance of having an effect , if it is adequately researched. Examples of 44 research in association with guidelines are to be found in Sweden in connexion with hypertension and in France in connexion with cancer. Most of the research is into the effectiveness of the patterns of care recommended in the guidelines rather than into cost-effectiveness. If recommended guidelines are to have an impact on cost -effectiveness, more research into that aspect is required. Implementation of guidelines for long-term care Research and education In implen: ~nting guidelines, the guidelines are likely to have a greater im- pact if the research on which they are based is published widely . This has been the approach in Sweden for some of its medical care programmes ; a symposium is arranged where the research background to the programme is discussed . A related point is the use of such guidelines as educational material them- selves. In Finland some medical care guidelines have been adopted on courses in training centres. A number of examples of the training of health pro- fessionals in cost-effectiveness were brought to the attention of the Workshop , e.g. in France, Poland , and the United Kingdom. Another aspect is that of tailoring general guidelines to the needs of spe- cific localities . It is probable that the pattern of care most preferable overall , from the viewpoint of either quality or cost-effectiveness, will not be the most preferable in all localities. The pattern of care may depend , for example , on the existing facilities for care and the method of organizing services. Day-case surgery is more attractive from the cost-effectiveness viewpoint if a day-case unit exists for the purpose and there is community nursing support for the patients. However , not all localities have these. The issue of local implementation has been taken up to some extent in Sweden , where ideally a testing ground is found for each medical care pro- gramme (e .g. , Skaraborg for the hypertension project) . The emphasis is very much on the involvement of the local people concerned . In Finland , those recommending medical care programmes have produced a booklet explaining how programmes can be interpreted in the light of local circumstances. This was thought by participants in the Workshop to be a welcome development , as it is quite possible that guidelines are followed mare rigidly than even those devising them anticipate, with possible bad effects. In Canada and Finland emphasis is placed on field visits to obtain views on the use of guidelines and to explain how they should be interpreted. Flexibility of guidelines In almost all the participating countries the dangers of unthinking imple- mentation of guidelines have been stressed. It is quite likely that the same form of care is not the most effective for each patient. Guidelines that discuss the 45 clinical and social indications to be taken as a basis for various courses of action are more flexible than those which recommend the same course of action for each patient. On the cost side , too, it is unlikely that the same pattern of care is to be preferred every time. What is least costly may depend on existing facil- ities and on characteristics of the patients such as their home circumstances . It is therefore likely that guidelines that incorporate flexibility will have more impact than those with a more rigid approach. It was suggested that the level of flexibility required depends on know- ledge of the epidemiology of the disease and the results of intervention, the experience of the providers using the guidelines, and the subject of the guide- lines (e.g., treatment or organization). In the Swedish diabetes project no alter- natives are given in the guidelines for the diagnosis of diabetes, but in the hypertension project two alternatives are given for the blood pressure level defining people as hypertensives and no guidance is given on the organization of services, merely a statement to the effect that most patients can be treated in the primary care sector. It was also noted that a fine line has to be drawn in the use of clinical al- gorithms in diagnosis. Although these may be fairly inflexible for inexperi- enced physician assistants, they should not necessarily be so for more ex- perienced physicians. System for review of guidelines Even if guidelines are designed in a flexible manner, they need reviewing after a certain amount of time. Experience in Europe differs in respect of systems of review. In an ideal world review would be a continuous process as the guideline was implemented, and in some countries steps have been taken to set up working groups on the operation of guidelines and to carry out surveys of users by questionnaire. Although the Workshop did not feel able to commit itself to any general statement on what the periodicity of review should be, it did feel that the development of guidelines should not be seen as a one-time process. Out-of-date guidelines soon fall into disrepute and have no impact or are even counterproductive. Incentives for adopting guidelines The Workshop felt that incentives are important, although it is difficult to generalize about them because of the differences in the organization of health care from country to country. Examples were given of various ap- proaches to influencing behaviour, such as allowing providers to redeploy some of the resources they save, changing reimbursement schemes, offering opportunities of postgraduate education, judging performance against the guidelines, and offering funds for evaluation of the impact of guidelines. A fairly frequently expressed view was that it is desirable to search for positive, 46 as opposed to negative, incentives and that in developing guidelines important aspects to stress are their potential for improving the quality of care and the involvement of everyone in bringing about change. USING GUIDELINES TO ACHIEVE COST-EFFECTIVENESS IN LONG-TERM CARE The following factors- _were seen by the Workshop to have a favourable influence on the cost-effectiveness of guidelines: - if the objective of the guidelines is more cost-effectiveness; - if the source of the guidelines is a combination of health professionals and resource providers; - if bodies exist with a responsibility for standard setting {development, implementation , or evaluation); - if the guidelines deal with the structure for care, the medical or other processes carried out , and their relationship with outcome; - if the guidelines are based on research into the costs and effectiveness of alternative practices; - if some effort is put into the education of providers in the use of the guidelines; - if the guidelines are flexible ; - if the guidelines are reviewed frequently; - if inc.,entives are given to providers to adopt the guidelines. Taken together, these conditions may be considered as constituting an a priori assessment of whether guidelines are likely to bring about a favourable impact on cost-effectiveness. A number of questions arise. In which situation should guidelines be used as opposed to other cost- effectiveness measures? It is important for those with an interest in cost-effectiveness to see guide- lines as one among a range of possible measures. Other possibilities are open. 47 First, that of offering incentives to providers for more cost-effective behaviour. In a publicly provided health care system incentives might be opportunities for further education and research or resources to redeploy. In a private health care system they might be in the form of increases in income. A second possibility is education of providers in notions of economics and cost-effectiveness. A third, cost containment measures, might include policies governing the total amount to be spent on health care, limits on the levels of reimbursement for inpatient hospital care, equipment purchases or drug expen- diture, or imposition of charges on consumers of health care. All these measures, including guidelines, are likely to have their advan- tages and disadvantages. Guidelines can involve much time, effort and expense, particularly if they are national guidelines. Hospital-based guidelines such as restricted drugs lists and physician protocols can obviously be developed much more cheaply. Spectacular successes have been reported for other ap- proaches. On the other hand, it can be argued that the effect of incentives is short-lived and that guidelines may have a longer-lasting effect. There is also evidence that some cost containment measures can easily be circumvented. The main difference between guidelines and the other approaches to cost- effectiveness is that with guidelines the main emphasis is usually on improving the quality of care, not containing costs. This suggests that the most produc- tive use of guidelines would be to see them as a way of achieving more cost- effectiveness by increasing quality without putting up costs. In most coun- tries surveys are necessary to confirm or belie that suggestion. The kind of in- dications for their use to be sought are the existence of known ineffective or dangerous practices that could be discontinued; concern about quality among health professionals; concern about undue emphasis on curative measures and the institutionalization of patients (which are costly as well as regarded as unsatisfactory by many health care providers); and lack of coordination be- tween medical and other health care, self care, and social services (on the grounds that there may be unnecessary duplication or inadequate examination of possibilities of substitution). Following such surveys it might be desirable to start a debate by making public some of the findings . It may be possible to point out large variations in practice or a worsening situation over time. In some countries the health pro- fessions themselves have often expressed the need for improvement and re- source providers have, at a later stage, participated in the development of guide- lines. Nevertheless , the participants in the Workshop stressed the need for a wider debate involving consumers even if the health care professionals are will- ing to participate in the production of guidelines. This would be valuable at least in relation to chronic diseases as a whole, but perhaps less so for high technology . Here the professions could be expected to argue for yet more in- vestment in relatively unproven modern techniques. Perhaps the approach to guidelines in this case is not so much concerned with influencing current practice as with calling for adequate research, the relevant guidelines then 48 being for research and development. In particular there might be guidelines for adding cost-effectiveness analysis to the medical evaluation of new practices or for restricting the spread of new practices while research is being carried out. Alternatively other cost containment measures could be used , such as banning the purchase of such equipment. These approaches may be more feasible in some countries than others. Countries with a decentralized health care system based on private care might need to look at other possibilities also , such as not giving reimbursements for certain diagnostic tests until they were shown to be of value, although in most countries the government has control over a large proportion of research funds and can therefore influence medical research and development. How should guidelines be developed and implemented? The Workshop suggested that those seeking to develop guidelines for cost- effectiveness purposes should bear the following considerations in mind . 1. The cost-effectiveness objective should be made clear as well as that of improving the quality of care . The best way to accomplish this is to be clear about the resources ; there should either be a stipulation that the guidelines must be implemented within existing resources or an increase in the resources should be agreed , as for a particularly neglected condition or where there is great concern among professionals about the quality of services. 2. The potential users of the guidelines (i.e. , the health professions) should be involved from the start. Their main motivation should be the possi- bility of improving patient care , although this should be supplemented by education in economic issues and in the potential benefits (and dangers) of using guidelines. 3. As far as possible the guidelines should be based on adequate research, in particular relevant cost-effectiveness research. One way of achieving this would be to attach an economist to the group preparing the guidelines. If know- ledge of co.,ts and effectiveness is lacking it may be of value to commission such research as part of the process of developing the guidelines. It may be possible to draft the guidelines in such a way as to identify areas for further research. 4. Resources will be required to bring together the group that will draft the guidelines, print and distribute them, and devise the associated edu- cational programmes. Consideration should be given to the likely benefits of the guidelines and the potential resource commitment in that context. This is where assessments of the burden of disease and the potential for change are important. 49 5. In devising guidelines, the organization of the health system as a whole and of the care of the disease in question should be considered. For ex- ample, if guidelines are being drawn up nationally, is it the government or the national bodies of health professionals that has the potential for change? In a decentralized system it may be better to have the bare outlines of guide- lines nationally and invest more resources in local implementation. The general rule should be that those who become involved in the development of guide- lines (and hence become committed to them) should be able to bring about the required changes in practice. 6. Flexibility and review are likely to be very important for the guide- lines, particularly in relation to the kinds of cost-effectiveness assessment on which the guidelines are based. Assessment should incorporate sensitivity analy- sis of the results to key medical and economic parameters. 7. Some funds should be set aside for evaluation of the impact of guidelines. How should guidelines be evaluated from a cost-effectiveness standpoint? Although most participants in the Workshop could give examples of guidelines in their countries, there are very few examples of evaluation. In Finland there are plans to evaluate the impact of medical care programmes and some evaluations are under way in Sweden (26). In Hungary there is an example of a comparison of costs and effectiveness with and without a particular set of guidelines on the management of diabetes; this was carried out by a clinician with an interest in the subject and not by those devising the guidelines. The methods for the assessment of the cost-effectiveness of alternative strategies in health care described earlier in this report could be used to assess the impact of guidelines on cost-effectiveness, the two alternatives being the delivery of care with the guidelines and the delivery of care without the guide- lines. However, it can be time-consuming and costly to carry out such a full cost-effectiveness study . A less costly approach is the use of intermediate indicators. Intermediate indicators The main indicator is the amount of cost-effectiveness introduced into the guidelines. For example: 50 - do the guidelines suggest that some practices should not be carried out since they are dangerous or ineffective? - do the guidelines recommend particular structural arrangements for care or particular medical procedures because they have been shown to be more cost-effective than alternative approaches? - do the guidelines suggest further possibilities for cost-effective ap- proaches, such as the replacement of doctors by nurses or other health care professionals? - do the guidelines make frequent reference to resource problems? - did those who drafted the guidelines call on economic advice or com- mission cost-effectiveness assessments? It may well be that the main contribution of economic appraisal is in the de- velopment of the guidelines rather than in later assessment of their impact. The second group of intermediate indicators is related to the adoption of the guidelines. Surveys need to be made of the extent to which the guideline has been adopted; the satisfaction of providers and consumers , and suggestions for improvement ; and evidence of changes in practice , particularly those with sizeable consequences for resources. Such information could be obtained by questionnaires, visits , and the examination of statistics on , say , resource utilization such as hospital ad- missions and length of inpatient stay , range and number of investigations carried out , and utilization of drugs . Full cost-effectiveness studies From time to time governments, or those recommending guidelines, may wish to carry out a full cost-effectiveness study . It was suggested at the Workshop that such studies are of most benefit when the guidelines are experimental and an assessment of their impact on the quality of care is also being carried out; when sizeable amounts of resources are at stake, thereby justifying the extra costs of economic appraisal (which would be small if a medical or qualityof-care assessment is already being carried out); and when decision-makers have a good understanding of cost-effectiveness analysis and the underlying value assumptions and so are able to interpret the results intelligently . If those conditions are met, the analysis could take the form shown in Fig. 5. The basic idea is to compare the costs of management of the disease with and without the guidelines and to relate them to outcomes, either in terms of improved quality of care or reduction in resource use arising from improved quality. The reduction in resource use could be in the health sector, other public agencies, or the family. 51 As with normal epidemiological research, it is important that the study should be well controlled. Ideally it should establish that any differences observed are attributable to the guidelines and not to other factors. Any doubts about this would reduce the validity of the study, hence the dis- advantage of undertaking a study before and after in one locality. How- ever, this is basically an epidemiological rather than an economic point. It is also apparent that controlled studies of organizational guidelines are inherently more difficult than studies of medical care guidelines. Alternatives Fig. 5. Cost-effectiveness assessment of the impact of guidelines Locality with guidelines Locality without guidelines Costs (al Resources required for devising, imp• lementing, and moni• toring the guide- lin (bl Cost of management of patients with the guidelines (take a sample of patients) Costs Cost of management of patients with - out the guidelines (take a sample of patients) Hospital care Ambulatory care Outcome Differences in mortality, mor- bidity, quality of life, and patient satisfaction Differences in health sector and family resource use arising from the above Hospital care I Ambulatory care I Agreement on the appropriate time span of such a study is also needed . It will depend on the disease in question and on whether the guidelines deal mainly with curative or preventive aspects. For example, evaluation of guidelines on the management of hypertension may be able to shL w some evidence of changes in coronary risk factors after a relatively short period, but longer-term outcome data may not be available for a number of years. Results of such studies could be used as follows. 52 I. They could be presented to the policy-maker in the following form. Extra costs of implementing the guidelines plus Extra costs of care with guide- lines (if any) less Resource savings to the health sector and family from implementing the guidelines versus Improvements in quality of care These would be stated in the most convenient units, e.g., improved survival , improved social functioning, return to work, general patient satisfaction 2. If the total of the left-hand column is zero or negative the policy- maker need look no further , as there would have been a reduction in costs as well as improvement in quality (although in an ideal world he would con- tinue to look for other guidelines that would be even more cost-effective). If it is positive, i.e. , if there is a net increase in costs from implementing the guideline, he can compare the returns in terms of quality gained with those to be obtained in other activities . THE ROLE OF EDUCATION AND TRAINING AND OF RESEARCH IN DEVELOPMENT AND APPLICATION OF COST-EFFECTIVE GUIDELINES IN LONG-TERM CARE It was frequently suggested during the visits to participating countries that recommending guidelines for care is difficult in that medicine is nearer to an art than a science. Furthermore, the medical profession as a whole is unsure about the effect of health economics on clinical freedom of action and on the health care budget. It was repeatedly stressed that in no way can guide- lines be imposed ; the way forward lies through encouraging health pro- fessionals to adopt them. These considerations indicate the importance of 53 research both into the establishment of the guidelines and into their implemen- tation. They also show the importance both of general education of the health professions in aspects of economics and of specific education in the develop- ment of guidelines for care. Initiatives in research Some of the literature in English on the cost-effectiveness of alternative patterns of care was reviewed above. Other reviews of such literature have been made in Sweden (26) and France {27). In the United Kingdom, too, several studies of effectiveness have been initiated by the Royal Colleges, and some of the guidelines cited by a number of the participating countries are based on research. In general, however , research on the costs and effectiveness of ther- apies in long-term care is somewhat patchy, although there is more research on effectiveness than on cost-effectiveness.There is even less research on the im- pact of guidelines, especially from the cost-effectiveness viewpoint. The most advanced examples are from Hungary and Sweden. The Workshop did not arrive at a strict ordering of research needs by priority, although the following topics were mentioned . 1. The methodology of cost-effectiveness and cost-benefit analysis and the values that underlie such methodology. 2. The impact of recommending guidelines and the relationship of guide- lines to other strategies for achieving cost-effectiveness. 3. The impact of cost-effectiveness studies themselves - do they in- fluence policy-makers? 4. The compliance with guidelines by providers and the ways in which it is influenced by the interests of the profession. 5. The impact of guidelines on innovation - do they encourage it or stifle it? The Workshop felt that WHO has an important role to play in pooling ex- perience of the use of guidelines. Initiatives in education and training General education of health professionals in aspects of economics In a number of countries (e.g., France, Poland, and the United Kingdom) there is some experience of general courses for health professionals in aspects 54 of administration and economics. The questions arising are : what should the content of such education and training be ; should they be organized by the health professionals themselves or who else needs to be involved; what is known about the attitudes of health professionals to such courses and their impact; and are health professionals likely to attend in sufficient numbers and what incentives, if any, should they be given to attend? Education associated with the recommendation of guidelines for care A general view expressed by most participants was that health pro- fessionals could not be expected to respond to recommended guidelines with- out some encouragement. An important aspect of such encouragement is close involvement of professional groups in the whole process of drafting and imple- mentation. Another important aspect is the education of professionals in the rationale for guidelines, education that should be linked with their continuing education. There are a number of initiatives along these lines in the participat- ing countries. In Finland and Sweden, for example , medical care programmes containing guidelines for care are published and circulated . In eastern European countries the national institutes are active and methodological letters are issued to physicians. In the United Kingdom the Royal Colleges and postgraduate centres of medical education have embarked on such initiatives. In the Federal Re- public of Germany there are plans for the publication of a book discussing clinical innovations to be distributed to physicians. In Sweden, too , at the county level a book giving the suggested best and cheapest drugs has been published. In order to implement locally medical care programmes established by central authorities , and to promote care programming in general , the following methods have been used in Finland. (a) A booklet explaining how to interpet medical care programmes in the local situation has been published . (b) Hospital regions have arranged one-day meetings in their areas on one medical care programme at a time. All service units functioning in the region are represented and the meetings are aimed at medical , nursing, and auxiliary personnel. Some regions have, in addition , arranged meetings for individual hospitals and health centres. The purpose of such meetings is to achieve practi- cal consensus on organization of the care of the disease or syndrome in the region. (c) Some regions have established special working groups to promote the development of medical care programmes. 55 (d) Teams set up locally have occasionally modified the programmes markedly . Because the main objective is educational , such activities indicate that the approach is correct. (e) After the initial phase a permanent project group has been estab- lished. The Finnish Hospital League acts as the national coordinating organ. The main responsibilities of the group are to support the regional teams on methodological issues and to keep a file on regional and local medical care programmes and applications. It also cooperates with other general care pro- grammes. The Finnish Hospital League invites the original programme teams to meetings in which the need for revision is considered at two- or three- year intervals . RECOMMENDATIONS I . In general , guidelines for the delivery of health care, especially long-term care , should be stimulated, since they create new possibilities for improving the quality of care, cost-effectiveness and the quality of life . 2. Governments should consider either setting up appropriate multidisci- plinary bodies with responsibility for coordinating the development, imple- mentation and evaluation of guidelines for the delivery of health care , especially long-term care , or modifying existing bodies. 3. Governments should attempt to coordinate efforts to develop guide- lines and standards for the process of health care delivery with other ef- forts on guidelines and standards for organization and resource allocation, cost containment, quality assurance, evaluation and budgeting. 4 . Associations and agencies developing guidelines for long-term care should consider carefully the considerations associated with their development and implementation (see p. 49) . 5. Governments, in association with professional and consumer groups, re- search institutes, insurance agencies and other appropriate bodies, should con- sider developing model guidelines for all major patient and client groups in long-term care. Special consideration should be given to prevention and the primary health care field , involving a wide range of health workers as well as lay care providers, in order to avoid attention being concentrated solely upon medical procedures. 56 6. Governments and research-funding bodies should stimulate research into the cost and effectiveness of alternative patterns of long-term care . This work should not only be in association with the development of guidelines. 7. Governments, educational institutions and professional groups should extend the training of health care providers, consumers and responsible decision-makers , both in general aspects of economics and administration and in the use of guidelines in long-term care and other fields . 8. As experience in the European Region is still limited , WHO and other appropriate international bodies should continue to pool experience on the use of guidelines in long-term care and related fields and to stimulate an exchange of ideas. 9 . WHO should stimulate the development of a shared methodology and practice of cost-effectiveness analysis and related topics in health care. 10. WHO should encourage research and discussion on the relative impact , in terms of cost-effectiveness, of guidelines for care and alternative strategies for bringing about a more rational use of health care resources. REFERENCES l. Better services for the mentally handicapped. London, Department of Health and Social Security , 1971 . 2. van Wering, E.R. Laboratorium diagnostiek van acute leukemie bij kinderen in Nederland . Nederlandsch tijdschrift voor geneeskunde, 124: 1627 (I 980). 3. van der Does-van den Berg, A. Acute lymfatische leukemie bij kinderen. In: Het medisch Jaar. Utrecht, Bohn, 1978 . 4. Drummond, M.F. Studies in economic appraisal in health care. Oxford, Oxford Medical Publications, 1981. 5. Drummond, M.F. Principles of economic appraisal in health care. Oxford, Oxford Medical Publications, 1980. 6. Bunker, T.P. et al. Costs, risks and benefits of surgery. New York, Oxford University Press, 1977. 7. Mooney , G.H. et al. Choices for health care. London , Macmillan, 1980. 8. Schweizer, S.O. Cost-effectiveness of early detection of disease. Health services research, 10: 22-32 (I 974). 9. Dickinson, L. Evaluation of the effectiveness of cytologic screening for cervical cancer - III Cost-benefit analysis. Mayo Clinic proceedings, 41 : 550-555 (1972). 57 10. Kodlin, D. A note on the cost-benefit problem in screening for breast cancer. Methods of information in medicine, 11(4): 242-247 (1972). 11. Steiner, K.C. & Smith, H.A. Application of cost-benefit analysis to aPKU screening program. Inquiry, 10: 34-40 (1973). 12. Bartlett, J .R. et al. Evaluating cost-effectiveness of diagnostic equip- ment: the brain scanner case. British medical journal, 2(40): 815-820, (1978). 13. Jonsson , E. & Marke, L. CAT scanners : the Swedish experience. Health care management review, 2(2): 37-53 (1977). 14. McNeil , B.K. et al. Cost-effectiveness calculations in the diagnosis and treatment of hypertensive renovascular disease . New England journal of medicine, 293 : 126-221 (1975). 15. Russell, I.T. et al. Day-case surgery for hernias and haemorrhoids: a clinical , social and economic evaluation. Lancet, 16: 844-847 (1977). 16. Wager, R. Care of the elderly - an exercise in cost benefit analysis com- missioned by Essex County Council. London , IMTA (now the Chartered Institute of Public Finance and Accountancy), 1972 . 17 . Mooney , G .H. Planning for balance of care of the elderly. Scottish journal of political economy, 25(2): 149-164 (1978). 18. Carter, E. et al. Cost of management of patients with haemophilia.British medical journal, 2(33): 465-537 (1976). 19. Bryant , N.H. et al. Comparison of care and cost outcomes for stroke patients with and without home care. Stroke, 5(1 ): 54-59 (1972). 20. Cretin, S. Cost/benefit analysis of treatment and prevention of myo- cardial infarction. Health services research, 12: 174-189 (1977). 21. Sis.souras, A.A. & Moores, B. Cost analysis of a system of intensive cor- onary care services. The hospital and health services review, 71(1 ) : 12-16 (1975). 22. Weinstein, M.C. et al. Coronary artery by-pass surgery: decision and policy analysis. In: Bunker, J.B. et al. Costs, risks and benefits of surgery. New York , Oxford University Press , 1977 . 23. Drummond, M.F. Welfare economies and cost benefit analysis in health care. Scottish journal of political economy, 28(2) : 125-145 (1981). 24. Cochrane, A.L. Effectiveness and efficiency. London , Nuffield Provincial Hospitals Trust , 1972. 25. Neutra , R. Indications for the surgical treatment of suspected acute appendicitis: a cost-effectiveness approach. In: Bunker, J.P. et al. Costs, risks and benefits of surgery. New York, Oxford University Press, 1977 . 26 . Jonsson, B. Cost-benefit analysis in public health and medical care. Uni- versity of Lund, Lund Economic Studies, 1976. 27. EVA. Bulletin d'analyse documentaire. Paris, Centre de Documentation pour la Medicine preventive et sociale, 1980. 58 Annex LIST OF PARTICIPANTS Temporary advisers Mr D. Affeld, Chief, Section of Health Economics and Health Planning, Federal Ministry of Labour and Social Affairs, Bonn , Federal Republic of Germany Dr S. Alsen, Assistant Director-General, National Board of Health and Welfare , Stockholm, Sweden Professor S. Barnoon, Chairman , Division of Health and Welfare Economics and Administration, Ben-Gurion University Centre for Health Sciences, Beer Sheva , Israel Ms J.M. Berry, Principal Administrative Assistant , Medical Informa- tion Section, West Midlands Regional Health Authority , Birmingham, United Kingdom Dr A.F. Correia de Campos, Lecturer in Health Economics, National School of Public Health, Lisbon, Portugal Dr A. van der Does-van den Berg, Dutch Childhood Leukaemia Study Group , The Hague, Netherlands Mr J.-P. Dupinay, Assistance publique , Public Hospitals of Paris , France Professor S. Feldman, Director , Staff College, National Institute of Men- tal Health, Rockville , MD, USA Mr P. Granicki , Research Worker , Institute of Occupational Medicine, Lodz, Poland Mr A. Griffiths , Head, Health Management and Economics , Sandoz Insti- tute for Health and Socio-economic Studies , Geneva, Switzerland Dr E. de Groot , Faculty of Medicine, State University of Limburg, Maastricht , Netherlands 59 60 Professor L.M.J . Groot , State University of Limburg, Roermond , Netherlands Dr I. Gyarfas , Chief, Department of Health Organization and Method- ology, National Institute of Cardiology, Budapest , Hungary Dr A.S. Haro , Director , Department of Planning and Evaluation, National Board of Health , Helsinki, Finland Professor J.A. Indulski, Director, Institute of Occupational Medicine, Lodz , Poland Dr J . John, Institute for Medical Informatics and Health Services Research of the GSF Research Centre (MEDIS), Munich , Federal Republic of Germany Dr P.E. Jonsson, Institute for the Planning and Rationalization of Health and Social Services (SPRI) , Stockholm, Sweden Dr K.J. Kokkola , Administrative Medical Officer , Finnish Hospital League, Helsinki, Finland Dr E. Kovesi , Head of Division , Organization , Planning and Information Centre , Ministry of Health , Budapest , Hungary Dr Q. Kiimpel , Chief, Psychogeriatric Unit , Psychiatric Hospital , Opava, Czechozlovakia Dr S. Lang, Assistant Professor of Social Medicine , Medical School , University of Zagreb , Yugoslavia Mr C. Lindmark , Head of Section , National Board of Health and Welfare, Stockholm, Sweden Mr E. Linnakko, Planning Officer , Finnish Hospital League , Helsinki , Finland Dr H.P. Lorenzen , Head, Medical Research and Technology Section, Federal Ministry for Research and Technology , Bonn, Federal Republic of Germany Professor L. von Manger-Koenig, Special Consultant on Medical Affairs, Federal Ministry for Youth, Family Affairs and Health, Bonn, Federal Republic of Germany (Chairman) Professor L. Naggan , Head, Epidemiology and Health Service Evaluation Unit , Ben-Gurion University Centre for Health Sciences, Beer Sheva, Israel Professor D. Neuhauser , Department of Community Health , Medical School, Case Western Reserve University , Cleveland , Ohio , USA Professor G.H. Ott , Evangelisches Krankenhaus , Bad Godesberg, Federal Republic of Germany Professor B. Pissarro, Department of Preventive and Social Medicine, St Antoine Faculty of Medicine , Paris , France Mr J .-P. Poullier, Head, OECD Social Indicators Programme , Directorate for Social Affairs, Manpower and Education , Organisation for Econ- omic Co-operation and Development, Paris , France Professor T.A. Ramsay , Area Director of Postgraduate Medical Education for Coventry and Warwickshire, Warwickshire Postgraduate Medical Centre , Coventry , United Kingdom Dr E. Reerink , Director , National Organization for Quality Assurance in Hospitals (CBO), Utrecht , Netherlands Dr J .L. Roberts , Regional General Administrator , West Midlands Regional Health Authority, Birmingham, United Kingdom Dr D. Schwefel , Deputy Director , Institute for Medical Informatics and Health Services Research of the GSF Research Centre (MEDIS), Munich, Federal Republic of Germany Ms I. Simon , Nursing Consultant , Munich , Federal Republic of Germany Dr H.P. Sintonen , Senior Research Officer , Research Department. Ministry of Social Affairs and Health , Helsinki , Finland Dr A. Smolek, Chief, Department of Health Economics, Institute for Social Medicine and the Organization of Health Services, Prague, Czechoslovakia Dr H. Stein , Health Policy Officer , Federal Ministry for Youth , Family Affairs and Health , Bonn, Federal Republic of Germany 61 Dr R. Stevens, Ministry of Public Health and Family Welfare , Brussels, Belgium Dr H. Strobel, Secretary General , Federal Association for Rehabilitation, Frankfurt, Federal Republic of Germany Mrs J.E. Stow, Editor , Standards Information Bulletin, Health Services Directorate , Health Services and Promotion Branch, Department of National Health and Welfare , Ottawa, Canada Dr F. Unz, Programme Director, German Aviation and Aerospace Re- search Institute , Cologne, Federal Republic of Germany Dr W.C. Wlodarczyk, Head, Organization and Management Section, Insti- tute of Occupational Health , Lodz, Poland Dr T. Zakia, Chief Medical Adviser , National Health Insurance Fund for Salaried Workers, Paris, France Consultants Mr M.F. Drummond , Lecturer in Health Economics, Health Services Management Centre , University of Birmingham, United Kingdom (fl..apporteur) Dr S.O. Isacsson , Assistant Professor , Department of Epidemiology and Health Promotion, County of Skaraborg, Sweden (Rapporteur) WHO Regional Office for Europe Dr H. Vuori , Regional Officer for Primary Health Care Dr H. Zollner, Regional Officer for Health Economics (Secretary) 62 .VHO publications may be obtained, direct or through booksellers, from. .\LGERIA .\RGENTINA \LSTRALIA AUSTRIA BANGLADESH BELGIUM BRAZIL BLRMA CANADA CHINA COLOMBIA CYPRL,S CZECHO- SLOVAKIA DE"IMARK ECUADOR EGYPT EL SALVADOR FIJI FINLAND FRANCE GERMAN DEMOCRATIC REPUBLIC GERMASY. FEDERAL REPUBLIC OF GHANA GREECE HAITI HO"IG KONG HL,NGARY ICELAND INDIA l'IDONESIA IRAN IRAQ IRELAND ISRAEL ITALY JAPAN KOREA REPUBLIC OF KL. WAIT LAO PEOPLE'S DEMOCRATIC REPLBLIC LEBANON Soc1etc Nauonalc d'Ed1uon et Jc D1ffus1on. 3 bd Zirout Youccr. ALGIERS Carlos Horsch SRL Florida 165. Galenas Giiemcs. Escrotoroo 453/465. 8UE1'05 AIRES \fa,/ OrrJ,r Sa/,s Au.s1rahan Government Publishing Service. PO Box 84. CA'IIE""A A.C.T. 2600 : or ,Jl'tr tht a 11n11tr /rom Australian Go11crnment Publishing Sen-1a- Bookshops ar 70 Alinga Street. CANIIERRA C1TY A C.T 1600 : 194 Adelaide Su-eet. 801s1A,E. Quttnsland 4000 . 347 SwanS1on Street. MEL· IOURNE. VIC 3000 : .l09 Pou Street. SYDNEY. N.S.W 1000; Ml Newman House. 100 St. Gcorse·s Terrace. PERTH. WA 6000 : Industry House. 12 Pinc Street. ADELAIDE. SA 5000: 15'>-162 Macquarie Street. H09ART. T AS 7000 - Hunter Pubhcauons. 58A G,pps Street. COLUNGWOOO, VIC 3066 - R. Holl & Son ud .. 608 St . Kolda Road. MELIOURNE. VIC 3004; Lawson House. 1~12 Clark Strttt. CRow's NEST. NSW 2065 Gerold & Co .. Graben 31. 1011 VIENNA I The WHO Programme Coordinator. G.P.O. Box 250. DACCA 5 - The Assoaauon or Voluntary Asenc,es. P 0. Box 5045. DACCA 5 Ofrlcc mtcrnauonal de L1bramc. 30 avenue Mam1x. IOSO Bausseu - Subscnpnons to World Htallh m,'1 Jean de Lannoy. 202 ucnuc du Ro, . 1060 BlluSSELS ~;~::~n:.~~1:; 0 ~~~~~~MS/OPS. Unodadc de Veroda de Publica¢es. Caixa Postal 20381 . Vil.i ,,, India. WHO Regional Office S,n,lt and hulk cop1ts of 1nd1r1duol pubhca11on.s tnot subscnpt1onsJ: Canadian Publte Health A.ssoc1at1on. 1335 Carhns Avenue. Suuc !10. OTTAWA. Ont. KIZ 8N8. Subscnp110,u ; Subscnpllon ord,rs. accompani,d h, clr,qu, mad, out 10th, Royal Bank of Canada. Ottawa. Acmunl World HC3hh Organizauon. should M ""'10th, World Health OrganizatM>n. PO Box 1800. Postal Stauon 8. OTTAWA. Ont. KIP 5RS. Co,.,,.. vpondtnct fonctm,ng subscnpnons should 1H addrts.ud 10 1ht World Health Orgamzauon. Dtstnbuuon Jnd Sales. 1211 GENEVA 27. Sw,uerl.ind China Nauonal Publicauons lmpon Corporation. PO Box 88. 8Eu1NG (PEKJNG) D1S1rohbros l.td .. Pio Alfonso Gama. Carrera 4a. Nos 36-119. CARTAGENA Publishers D,strobuton Cyprus. 30 Democrauas Ave Ay,os Dhomctoous. P.O. Box 4165. NICOSIA Anoa. Ve Smeckach 30. 111 27 f>uGUE I Munkspard Expon and Suoscnp<oon Servooe. Nem: Sepde 35. 1370 COP£NHAGEN K Lobrena Coenulica S.A .. P 0 . Box 362. Luque 22.J. GUAYAQUIL O<,ros Oflice lor Books and Reviews. 50 Kasr El Nil Street. CAlRO Lobrena ESlud,anul. Edolicoo Comerc1al B No 3. Avenoda Lobcnad. SAN S..LVAOOlt The WHO Pr011ramme Coordinator. P.O. Box 113. SuvA Akaleemonen Kor;akauppa. Keskuskalu 2, 00101 HELSINl<J 10 Lobnuroe Arnette. 2 rue Cas1mor-Del.iv11ne. 75006 PARIS Buchhaus Le1pzog. Postlach 140. 701 LE1l'ZIG Gov,.Vcrlas GmbH. Ginnhc,merstnwe 20. Postlach 5360. 6236 EsaflllllN - W E. Saarbach. Postlach 101 610. Follerstrasse 2. 5000 COLOGNE I - Alex. Hom. Sp,egelpsse 9. Postlach 3340. 6200 W1ES1AOEN Fodcs Enterprises. PO Box 1628. Acau G .C Elellhcroudakos S.A .. L1bra1nc: ontcrnationale. rue Nikos 4. ATHENS (T 1261 Max Bouchcreau. Lobraoroe "A la Caravelle", Boite post.ale 111-8. PottT-AU·PIUNCE Hons Kong Government Information Scrvooes. lleaconsloeld House. 61h Floor. Queen's Road. Central. VIC"To«IA Kultura. PO 8 149. BUDA.EST 62 - Akadcmoa, Konyvcsbolt. Vaci Ulca 22. 8uDAl'EST V Snacbiern Jonsson & Co . PO Box 1131. Ha.lnars1ract1 9. REYKJAVIK WHO Regional Oflice lor South-East Asoa. World HC31th Howe. lndrapraSlha ESlatc. Rons Road. New DELHI 110002 - Oxford Book & Stationery Co .. Sc,ndaa Howe. 'IEW DELHI 110001 . 17 Park Street. CALCUTTA 700016 tSub-.,g,nt) 11.1 /s Kalman Book Scmce ud .. Kwnans Raya No. 11 . PO Box 3105/Jkt. JAKARTA Iranian Amalpmated Dostnbuuon Asency. 151 Khoaban Soraya. TEHERAN \.t1nistry or lnformauon. Nauonal House for Publtsh1ng. D1stnbuung and Ad\lcntsing. BAGHDAD The Stauoncry OHoce. DuouN 4 Hc,hser & Co .. 3 Nathan Strauss Street. JERUSALEM Ed1z1on1 Minerva Med,ca. Corso Bramante 83-85. 10126 TURIN: Voa Lamarmora 3. 20100 MILAN Maruzcn Co. l.td .. PO Box 5050. TOKYO lntcrnauonal. 1~31 The WHO Programme Coordinator. Central P 0 . Box 540. Seoul The Kuwait Boobhops Co. l.td .. Thunayan AI-Ghancm Bldg. PO Box 2942. KLWAIT The WHO Programme Coordinator. PO Box 343. V1ENTIA~E The Levant Dostrobulon Co. S.A.R.L. Box 1181 . Makda.sso Street. Hanna Bldg. BEIRLT A/1/81 WHO publications may be ol LUXEMBOURG MALAWI MALAYSIA MEXICO MONGOLIA MOROCCO MOZAMBIQUE NEPAL NETHERLANDS NEW ZEALAND NIGERIA NORWAY PAKISTAN PAPUA NEW GUINEA PHILIPPINES POLAND PORTUGAL SIERRA LEONE SINGAPORE SOUlli AFRICA SPAIN SRI LANKA SWEDEN SWITZERLAND SYRIAN ARAB REPUBLIC THAILAND TUNISIA TURKEY UNITED KINGDOM UNITED STATES OF AMERICA VE~EZUELA YUGOSLAVIA ZAIRE ~!::.ie ~ 11\\111 \\Ill \\Ill \\II\ 1\ \11\1\111\11 \\Ill \\II \\\I T11c WHO .. " O O z g l n o .. Ra,a Chulan, KUALA LUMl'\JR OS- ll ,_w, ... omman. P.O. Bo, 629. KUALA luMl'\JR 01 ~ - - · , . ~•~enter. K. L Hilton Hotel. Jin. Trachcr. P.O. Bo• 960. KuAI.A Luwl'\Ja La Prensa Medica M .. icana. Ediciones C~1ir.c:u. Pua, de lu Facuhades 26. Apt. Poul 20-413. MEXICO CITY 20, D.F. 1tt India. WHO Rep,nal Office Edi1ions La Pone, 281 avenue Mohammed V, RAAAT INLD. Cai.u PosaaJ 4030, MAl'\JtO sn India. WHO Rep,nal Ofr.oe Medical Books Europe BV. Noonlerwal 38. 72' I BL l.oaaM Government Prin1i111 Ofrioe. Pubbc:a&ions Section. Mularave Street. Private ... WIUJNOlON I ; Walter Street. WnuNGTON; World Trade Buildi111, Cumcadc. Cui. Street. WIILUNOlON. Go_, Booltsliops at: Hannaford Bunon Buildin1, Ru1land Street. Private Bq. AUCIU.AND; 159 HeRford Street. Priva&e Bq. CHIIIS10fUIICH; Aleundra Street. P.O. Bo1 857. HAMIi.TON ; T .t G Buildilll, Princes Street. P.O. Bo1 1104, DuNDIIN - R. Hill .t Son Ud. Ideal H-. Cnr Gillia Avenue .t Eden Str.t. Newmarket. AUCIUAND I Univemly Booullop Niaeria Ud. Unoventty of Ibadan. IIIAIMN J. G . Tanum A/S. P.O. Bo• I In Sen1nam, OILo I Mirza Book Atfflcy, 65 ShallrahwE,..Quaid-E--Aam, P.O. Bo• 729. LAHou 3 Tbc WHO l'ropamme CoonlinMor. P.O. 8ol 5896, 8oaoKo World Heallh Orpnization. Rqjonal Ofr.oe for lhe Western Pacific. P.O. Bo• 2932, MANIU - Tbc Modem Book Company Inc .. P.O. Bo• 632. 922 Rizal Avenue. MANIU 2800 Skladnica Ks,qanb. ul Muowiecu 9, 00052 WAUAw (uapt pmodkabJ - BKWZ Ruell. ul Wronia 23, 00840 WAUAW (pmod,cob o,,Jy/ Livraril Rodn..,.._ 186 Rua do Ouro. LlsaoN 2 Njala Univenny Collete Bookshop (Univem1y of Sierra Leone~ Priv11e Mail Baa. FunowN Tbc WHO "'--tnme Coonlinator, 144 Moulmein Road, G .P.O. Bo1 3457, SI~ I -Select Books (Pie) Ud. 215 fan,lin Shoppi111 CentR, 2/F. 19 Tanalin Road. StNGA-. 10 Van Schaik's Bookstore (Ply) Ud. P.O. Bo1 724. 268 Church Street. PuToaaA 0001 Corneraal Atheneum S.A .. Consejo de C~to 130wl36. 8.ulcnoNA 15 ; General M....-clo 29. MADRID 20 - LibRria Diaz de Sanim. lapa:a 95 y Maldonado 6. MADRID 6 ; Baima 417 y 419. 8.ulcnoNA 22 Jtt India. WHO Rqional Ofr.oe Akliebolqot C.E. Fritzes KunaJ. Hovbokhandel Reaennppw, 12. 10327 SlOCIUtOUI Medazinischer Verlaa Hans Huber. u.,.... St.- 76, 3012 llnN! 9 M. Farras Kekhia. P.O. Bo1 No. 5221. Au"O 1tt lndoa. WHO Regional Orrioe Societe Tun111Cnne de Difrusion. 5 avenue de CanhaF, TUNIS Huet K11apev1. 469 laiklal Cadciesi. ~u. lnANMIL H.M. Sta&oonery Ofr.oe : 49 Hiah Holbom. LONDON WCIV 6HB; Ila Caale Street. EDINIUIIGH EH2 )AR ; 41 Tbc Hayes. CAIIDIFF CFI 1/W: 80 Ch,c:hester Slreet. Bn.f..n BTI 41Y: Brazcnnooc Street. MANCHDTD M60 BAS; 258 Broad Street. lltDIINGHAM 81 2HE : Soulhey House. Wine Street. BatSTOL IISI 2BQ. All -,/ ortkn fflOfMli ~ s,11110 P.O. Bo• 569. LONDON SEI 9NH Sillfk and bu/le rop,1s of ,ndi.,duol publica11o,u (no, sub-pllo,uJ: WHO Publications Cen1re USA. 49 Shendan Avenue. Al.aANY, N.Y. 12210. SubJfnpllOIU: Subscnp11on onkn. a«ompa1111d b, rlwc/c ~ out 10 1/t1 Chemical Bank. New York. Accoun1 World Heallh Orpniza110n. sltou/d ~ w,1 to tltl World Heallh Orpnazalion. P.O. Bo• 5284. Church Slreet Sta1ion. Nrw YDIIK. N.Y. 10249: Com,poruk11u ron- ffflltllf subJfnpllo,u s/tould ~ addtt:wd 10 IN World Heallh Orpnization. Dianbution and Sales. 1211 GENEVA 27. Swn.zcrland. l'llb/ia,1io,u ar, also aW11/abk fro,n 1/t1 United Na110ns Booullop. Nrw YDIIK. N.Y. 1001 7 1r,1a1/ o,r/,·1 For r,o,kn 111 IN USSR r,q,,1ri11f Ru.u,011 lditiOlu: Kornsornolsluj prospekl 18, Medicinska,a Knip. Moscow - Fo, r,o,kn outsttk IN USSR r,q,,trillf RWSUM Hitio,u: Kuznec:kij fflOII 18. Mddunarod- naJa Knip. MOICOW G-200 Edilorial ln1eramencana de Venezuela C.A .. Apanado 50.785. CAAACAS 105 - lobRna del Eae. Apar- lado 60.337, CAAACAS 106 - LibRril Medica Pans. Apanado 60.681 . CAAACAS 106 Juaoslovenska Knjip. Tera.z,ie 27/11. 11000 IIELGa.wr Librairie unoveni1aire, avenue de la Pa11 NO 167. 8.P. 1682. KINSHASA I Special terms for developing COW'ltriea are obtainable on application to the WHO Programme CoordinatOB or WHO Regional Offices listed above or to the World Health Organization, Distribution and Sales Service, 1211 Geneva 27. Switzerland. Orders from ccuntriea where sales agents have not yet been appointed may also be sent to the Geneva address, but must be paid for in pounds sterling, US dollars, or Swiss francs. Price: Sw. fr. s.- Prices are subject to change without notice.