The antibacterial paradox: essential drugs, effectiveness, and cost.

Research The antibacterial paradox: essential drugs, effectiveness, and cost F. Fasehun 1 The concept proposed byWHO of an essential drugs list that should comprise drugs corresponding to the health needs of the majority of the people has been embraced by countries, which have adapted it to their needs. In this study, the essential antibacterial drug lists of 16 countries chosen from the sixWHO regions are reviewed.Most of these countries include 73% of WHO-recommended essential antibacterials on their lists. However, most are lacking reserve antibacterials, and even some main list antibacterials, which are essential when empirical therapy fails in cases of bacterial resistance. Many factors that may be responsible for the lack of selection of these drugs, not least cost considerations, are discussed. Voir page 215 le re sume en francËais. En la pa gina 215 figura un resumen en espanÄ ol. Introduction Infectious diseases, particularly sexually transmitted diseases (STDs), acute respiratory infections and diarrhoea, are the leading cause of mortality and morbidity in the developing world. In the past decade, human immunodeficiency virus (HIV) infections and acquired immunodeficiency syn- drome (AIDS)-related complex may have led to a greater use of antibiotics, especially in children. Large amounts of money have been invested in research on anti-infective drugs, accounting for 16% of the total investment on drugs that reached the market place between 1972 and 1992, and placing this category of drugs in third place behind cardiovascular and nervous system drugs (1). Antibiotics account for the highest proportion of the drug budget in many countries and constitute the largest group of drugs purchased in developing countries, where financial resources are scarce (2). Slow economic growth and budgetary constraints in these countries have resulted in cuts in their health budgets. The emergence of antibacterial-resistant strains of Neisseria gonorrhoeae, Streptococcus pneumoniae, Shigella spp., and Salmonella spp., caused by chromosomal mutations, plasmids, or transposons that can transfer resistance determinants in diverse bacteria species faster than new drugs can be developed to fight such resistance, may drive the cost of antibacterial therapy even higher. The traditional N. gonorrhoeae treatment with penicillins, which are safe and affordable, is fast becoming a thing of the past with the widespread appearance of penicillinase-producing N. gonorrhoeae (PPNG) strains. In Africa, 30±81% of patients with gonococcal infection carry PPNG strains (3); and in South-east Asia, about 35% (4). Strains have emerged that are highly resistant either singly or multiply to penicillins, tetracyclines, spectinomycin, erythromy- cin, and thiamphenicol (5). Chromosomal resistance has rapidly made sulfonamides obsolete as anti- gonococcals, whereas plasmid resistance has com- promised the antigonococcal value of penicillins and tetracyclines (5). Thus, the emergence of N. gonor- rhoeae resistance has narrowed down the choice of antibacterial treatment to ceftriaxone and the newer fluoroquinolones, which are expensive. A total of 120 countries have adapted and implemented the list in their national drug policy (6). This article reviews the antibacterial drugs in the essential drugs list of 16 countries chosen from the six WHO regions to assess their adequacy for coping with alternative therapy in the face of antibacterial resistance. Methods The current essential drugs lists of 16 countries in the six WHO regions, namely the African Region, the Region of the Americas, the Eastern Mediterranean Region, the European Region, the South-East Asian Region, and the Western Pacific Region, were reviewed. The lists were obtained from the WHO Action Programme on Essential Drugs. Individual countries included in this study are shown in Table 1. Most of the countries selected for this study, regardless of regional classification, include 165± 1 9, chemin des Rocailles, 1292 Chambe sy, Switzerland. Requests for reprints should be sent to Dr Fasehun at the following address: c/o Essential Drugs and Medicines, World Health Organization, 1211 Geneva 27, Switzerland. Reprint No. 5767 211Bulletin of the World Health Organization, 1999, 77 (3) # World Health Organization 1999 592 (median, 249) drugs on their lists, which are meant for use at various levels of health care ranging from dispensaries to specialist hospitals (see Table 1) (7±22). The WHO essential drugs list is divided into two parts, main list drugs and complementary drugs. Main list and complementary drugs usually consist of therapeutic groups from which countries can select essential drugs. In the antibacterial category, there is a third group (reserve antibacterial agents); drugs in this group have no therapeutic alternatives, and their use is restricted to reducing the risk of developing resistance. The study did not differentiate between main list and complementary list drugs, since only two drugs (clindamycin and chloramphenicol oily suspension) are complementary. Substitutes were made for model list drugs in the same therapeutic categories; however, where there is more than one substitute, the substitutes are listed in the individual country lists only, and these lists were also reviewed. Analysis of cost per course of treatment. The analysis of the cost per course of treatment was based onWHO treatment guidelines (5, 23, 24), and did not include administrative costs or equipment and laboratory costs. Prices of antibacterial drugs, which did not include shipping and handling charges, were obtained from the International drug price indicator guide Ð management science for health, 1996, which provides an indication of the prices of generic drugs offered on the international market by nonprofit drug suppliers and procurement agencies. Prices of ceftriaxone and ciprofloxacin, whichwere not listed in the guide, were obtained from WHO. Results Essential antibacterial drugs. On average, the countries chosen for the study selected 73% of the WHO main list and complementary antibacterial drugs. Ampicillin, penicillinase-resistant cloxacillin, erythromycin, gentamicin, chloramphenicol, and sulfamethoxazole + trimethoprim appeared on all the lists (100% selection) of the 16 countries studied (see Table 2). In contrast to the WHO recommenda- tion of using ampicillin injection only, different ampicillin formulations appeared onmost of the lists. Ampicillin was mostly available as 250-mg capsules or tablets and as a 125-mg/ml suspension, as well as an injection formulation in some cases. However, amoxicillin capsules and tablets appeared on the lists of 75% of the countries studied. Among the tetracyclines, the WHO model list includes only doxycycline. However, doxycycline or tetracycline appeared on most countries' lists, i.e. 75% (Table 2) and 68.8%, respectively. Both these tetracyclines were included on the lists of 31.2% of the countries studied, whereas the remaining 68.8% included either one or the other. Ecuador, Honduras, Mali, Nigeria, and Sri Lanka did not include doxycycline on their lists. The quinolone, nalidixic acid, was selected by only 3 of 16 countries (18.8%) Ð Botswana, Philippines, and Yemen included it on their lists. The newer quinolone, ciprofloxacin, was listed by 62.5% of countries, but only Philippines and Yemen included both quinolones on their lists. Trimethoprim was the least-selected antibac- terial drug. The combination drug sulfamethoxazole + trimethoprim was preferred to the single drug trimethoprim, since only 2 of the 16 countries Ð Botswana and ZimbabweÐ included it on their lists. These countries listed both trimethoprim alone and in combination with sulfamethoxazole, whereas the other countries listed the combination. Sulfadimidine appeared in 31.3% (5 out of 16) of the lists, and metronidazole in 81.3%. Although erythromycin was included in every list, clindamycin appeared in 43.8% of the lists. Spectinomycin was listed by 37.5% of the countries (see Table 2). Only 43.6% of the countries studied listed the reserve antibacterial third-generation cefalosporins, ceftria- xone, or ceftazidime (Table 2). Although first- and Table 1: Number of essential drugs, level of health care use, and coverage, per country Region/country (ref.) No. of Level of health essential drugs care use a African Anglophone Botswana (7) 350 NA Kenya (8) 195 RH, PH, DH, SDH, HC Malawi (9) 242 HC, DH, C'H Nigeria (10) 409 GH, PHI Zimbabwe (11) 592 HC, DH, PH, SH Francophone Benin (12) 187 HC, DH, PH, UH Guinea (13) 165 U, RH, PH, HC Mali (14) 198 NA Americas Ecuador (15) 225 HC, DH, RH, SH Honduras (16) 351 AH, R'H, NH, HC Eastern Mediterranean Yemen (17) 256 4 levelsb European Estonia (18) 173 NA South-East Asian Bhutan (19) 312 RH, DH, B, D Sri Lanka (20) 220 HC, DH, PH, SH Thailand (21) 366 NA Western Pacific Philippines (22) 536 NA a HC = health centre, D = dispensary, GH = government hospital, PHI = public health institute, RH = referral hospital, DH = district hospital, SDH = subdistrict hospital, PH = provincial hospital, UH or U = university hospital, SH = specialist hospital, AH = area hospital, CH = cantonal hospital, R'H = regional hospital, B = basic health units, C'H = central hospital, NA = not available. b Not specified. 212 Bulletin of the World Health Organization, 1999, 77 (3) Research second-generation cefalosporins do not appear on the WHO list, Benin, Bhutan, Botswana, Malawi, and Thailand listed them, whereas Mali, Sri Lanka, and Yemen did not. The most commonly listed first- and second-generation cefalosporins were cefalexin (31.3%) and cefoxitin (25%). Vancomycin, another reserve antibacterial, was only included in the essential drugs lists of Botswana andHonduras. The frequency of selectionwas 12.5% (see Table 2). The data showed that 38±88% of the countries studied listed neither vancomycin, ceftriax- one, spectinomycin, nor quinolones. Other antibacterials included on the countries' lists, but which differed from those mentioned above, were piperacillin (31.3%), aminoglycoside alternatives to gentamicin Ð neomycin, amikacin, and kanamycin (each on 25% of lists) Ð and ampicillin combinations (18.8%). Costs per course of treatment of empirical therapy and in cases of bacterial resistance. Table 3 shows costs per course of treatment for STDs; highlighted are increases in treatment costs resulting from bacterial resistance. For cases of nonresistant N. gonorrhoeae ano-genital infections that respond to penicillin, tetracyclines, or sulfonamides, the cost per course of treatment is less than US$ 1, whereas in resistant cases treatedwith ceftriaxone, ciprofloxacin, or spectinomycin, the cost is over US$ 2 and may be as high as US$ 7 depending on the drug. Cost comparisons of drugs used to combat drug-resistant strains, shown in Table 4, indicate that, on the one hand, the most expensive of these drugs, spectinomycin, is 15 times more costly than doxycycline and at least 10 times more so than procaine benzylpenicillin. On the other hand, cipro- floxacin, the least expensive of the drugs, is 4±6 times more expensive than procaine benzylpenicillin, amoxicillin, or doxycycline. Syphilis, an STD with no known resistant strains, still responds well to penicillin treatment. It costs less to cure than antibacterial-resistant gonor- rhoea. Treatment of syphilis costs under US$ 0.50 with benzathine benzylpenicillin (Table 3), whereas with procaine benzylpenicillin the cost is a little over US$ 1. Similarly, empirical treatment of nonresistant cases of pneumonia with sulfamethoxazole + trimethoprim, procaine benzylpenicillin, amoxicillin, and ampicillin costs US$ 0.08, 0.21, 0.26, and 0.66, respectively, whereas a day's treatment ofmethicillin- resistant S. pneumoniae with vancomycin costs US$ 7.32, 11±90 times more than nonresistant cases. For cases of adult shigellosis that are resistant to ampicillin, chloramphenicol, sulfamethoxazole + trimethoprim, tetracycline, and nalidixic acid, the cost of a 5-day course of treatment with ciprofloxacin is US$ 26.70. This is 86 times more expensive than treatment with sulfamethoxazole + trimethoprim (US$ 0.31) and 10±11 times more so than treatment with ampicillin (US$ 2.61) or nalidixic acid (US$ 2.36). Discussion The review of the essential drugs lists of 16 countries in the WHO African Region, the Region of the Americas, South-East Asia Region, Western Pacific Region, Eastern Mediterranean Region, and Eur- opean Region showed that these countries have adopted over 70% of the WHO model list antibacterials. Essential drugs lists are intended to aid decision-making on drug procurement and supply to serve the health care needs of the majority of the population. In addition, essential drugs lists reduce duplication, help to make health care budgeting and spending more efficient, and promote rational prescribing. However, the review showed that 33±88% of the countries do not include cefalosporins, fluoro- quinolones, vancomycin, spectinomycin, trimetho- Table 2: Frequency of selection of WHO model list antibacterials by 16 countries selected from all six WHO regions Antibacterial Frequency of selection (%) Penicillins Ampicillin a 100 Cloxacillin 100 Benzathine benzylpenicillin 93.8 Benzylpenicillin 87.5 Phenylmethoxypenicillin 87.5 Procaine benzylpenicillin 81.3 Amoxicillin 75 Chloramphenicols Chloramphenicol 100 Aminoglycosides Gentamicin 100 Tetracyclines Doxycycline 75 Macrolides Erythromycin 100 Lincosamides Clindamycin 43.8 Sulfonamides Sulfamethoxazole + trimethoprim 100 Sulfadimidine 31.3 Trimethoprim 12.5 Furans Nitrofurantoin 75 Quinolones Ciprofloxacin 62.5 Nalidixic acid 18.8 Others Spectinomycin 37.5 Metronidazole 81.3 Restricted antibacterials Third-generation cefalosporins Ceftriaxone 37.3 Ceftazidime 6.3 Others Vancomycin 12.5 a Available mainly as capsules and tablets. 213Bulletin of the World Health Organization, 1999, 77 (3) The antibacterial paradox: essential drugs, effectiveness, and cost prim, nalidixic acid, or piperacillin on their lists. Of particular concern is the absence of cefalosporins, fluoroquinolones, and spectinomycin, which are useful in the treatment of resistant cases of meningitis, gonorrhoea, Salmonella infection, and shigellosis, all of which are of epidemiological importance. Cefalosporins, especially third genera- tion, are useful for treating penicillin- or chloram- phenicol-resistant Haemophilus influenzae type b meningitis, penicillin- or spectinomycin-resistant gonorrhoea, or tetracycline- or sulfamethoxazole + trimethoprim-resistant H. ducreyi chancroid. Fluor- oquinolones are indicated in shigellosis and Salmo- nella-resistant cases when initial empirical therapy fails (6). They are also alternatives to cefalosporins in cases of gonorrhoea and chancroid when oral therapy is necessary, or in hospital-acquired infections caused by Gram-negative bacilli such as Escherichia coli, Klebsiella spp., and Pseudomonas aeruginosa, or to vancomycin in cases of meticillin-resistant Staphylo- coccus aureus (6). What factors might contribute to the failure of countries to select the above-mentioned drugs? First, cost considerations may be an important factor. For infectious diseases caused by bacterial strains resistant to initial empirical therapy, newer antibac- terial treatments are needed, but they come at a price. Expensive antibacterial drugs may not be affordable to all in developing countries, for example in sub- Saharan Africa, where out-of-pocket expenses make up 40% of the health care costs (25). Patients have been known to drop out of treatment because of its cost (26, 27). The paradox is that people who can least afford treatment are those most vulnerable to infections. Other factors that might play a role in not selecting reserve antibacterials for resistant cases are the absence of laboratory facilities to test for bacterial susceptibility; lack of prescriber education on anti- microbial treatment and resistance; promotional activities by pharmaceutical companies; and the failure of countries to review their essential drugs list to reflect the results of current antibacterial susceptibility testing. More important, the failure to properly select reserve antibacterial agents has health care implica- tions. An increase in bacterial resistance would entail increases in direct and indirect health care costs that most developing countries can ill afford, for example increased morbidity and mortality, particularly in immunocompromised patients, and decreased pro- ductivity due to absence from work. Preventive measures to control the spread of resistant strains would in the long term be the most cost-effective way to reduce the burden of bacterial resistance on the health care system. The emergence of resistance to antimicrobial agents may be delayed by reducing use of such drugs. Education of all parties involved is necessary. In this regard, educating prescribers, patients, and pharmacists Ð especially community pharmacists, from whommost over-the- counter antibiotics are purchased Ð is essential. Other preventive measures include arresting the dissemination of resistant strains through good hygienic practices and avoiding agents that select for resistance genes (28). Such measures would delay emergence of bacterial resistance and reduce reliance on newer and expensive antimicrobial agents. Computerized surveillance networks such as the WHONET provide data on regional anti- bacterial susceptibility patterns, which if regularly disseminated to prescribers would improve the empirical selection of antimicrobial agents, help to develop treatment guidelines on antimicrobial agents useful for national programmes, and provide prompt treatment with newer agents to combat bacterial resistance (29). In conclusion, the fight against antimicrobial resistance can be likened to a revolving door, in which the optimal treatment for the patient with a resistant strain has the effect of increasing use of reserve drugs, and thereby speeding up selection in the population of strains resistant to these drugs as well, Table 3: Cost per treatment course of nonresistant and resistant N. gonorrhoeae infections and syphilis STD Treatment Cost per treatment course (US$) Gonorrhoea Ano-genital infection Ceftriaxone (250-mg IM single dose) OR spectinomycin (2 g single dose) OR ciprofloxacin (500 mg orally single dose) 4.98 6.87 2.67 Not penicillin-, tetracycline-, or sulfonamide- resistant Amoxicillin (3 g orally) plus probenecid 1 g OR procaine benzylpenicillin (4.8 mIU, IM) plus probenecid OR doxycycline (100 mg orally twice daily) OR tetracycline (500 mg orally thrice daily x 7 days) OR sulfamethoxazole (400 mg) + trimethoprim (80 mg) (10 tablets twice daily x 3 days) 0.57 0.68 0.45 0.58 0.47 Syphilis Early syphilis Benzathine benzylpenicillin (2.4 mIU single dose) OR procaine benzylpenicillin (1.2 mIU four times daily x 7 days) 0.36 1.27 Table 4: Cost ratio of drugs used to combat N. gonorrhoeae- resistant cases to those used in nonresistant cases Drug Ceftriaxone Spectinomycin Ciprofloxacin Doxycycline 11.1 15.3 5.9 Procaine benzylpenicillin 7.3 10.1 3.9 Tetracycline 8.6 11.8 4.6 Amoxicillin 8.7 12.1 4.7 Sulfamethoxazole + trimethoprim 10.6 14.6 5.7 214 Bulletin of the World Health Organization, 1999, 77 (3) Research compromising them for future patients. Strains of S. aureus resistant to vancomycin are already appearing (30), as well as strains resistant to fluoroquinolones, whose introduction 10 years ago for wide-spectrum activity led to their extensive use and misuse. n Acknowledgements Dr J. Quick, Director, Essential Drugs and Medi- cines, is thanked for his cooperation and for making readily available documents for the preparation of thismanuscript.Ms P. Brudon andDrR.Williams are thanked for reviewing the manuscript and for their useful comments. Re sume Le paradoxe antibacte rien : me dicaments essentiels, efficacite et couà t L'article passe en revue les listes d'antibacte riens essentiels e tablies par 16 pays choisis dans les six Re gions de l'OMS. Parmi les antibacte riens essentiels retenus par la plupart des pays, figurent les pe nicillines, les te tracyclines, les aminoglycosides, les sulfamides et les macrolides. Ces produits sont adapte s au traitement des affections respiratoires aigueÈ s, des maladies diarrhe iques et des MST pour lesquelles il n'existe pas de pharmacore sistance multiple. Ce sont ces maladies qui sont responsables de l'augmentation de la morbidite et de la mortalite dans les pays en de veloppement. Pour inverser cette tendance, on fait un usage croissant des antibacte riens. Ces produits repre sentent les deux tiers des me dicaments utilise s dans les pays en de veloppe- ment. Cependant, la re sistance croissante oppose e par les bacte ries aux antibiotiques limite l'utilite de certains de ceux qui figurent sur les listes e tablies par ces pays. Une liste de me dicaments essentiels ouÁ ne figurent ni les ce phalosporines de troisieÁme ge ne ration, ni les fluoro- quinolones, ni la spectinomycine, ni la vancomycine peut-elle encore servir aÁ quelque chose en cas de polypharmacore sistance ? On a pu montrer que l'augmentation du couà t de certains traitements conse cutive aÁ ce probleÁme de pharmacore sistance conduit aÁ une moins bonne observance. On a proce de aÁ une analyse du couà t par cure pour de terminer l'effet de la pharmacore sistance sur le prix de revient du traitement. Les re sultats obtenus montrent que le recours aÁ des produits nouveaux comme les fluoroquinolones ou les ce phalosporines par suite de l'e chec des pe nicillines, des sulfamides ou des te tra- cyclines a eu pour effet d'augmenter le couà t d'une cure dans des proportions astronomiques. Cet e tat de choses vient encore accroõÃtre la charge que les soins de sante repre sentent pour les pays en de veloppement, charge que certains d'entre eux ne sont gueÁ re en mesure de supporter. L'auteur discute les implications sanitaires de la pharmacore sistance des bacte ries et les mesures aÁ prendre pour ralentir la propagation des souches re sistantes. Resumen La paradoja de los antibacterianos: medicamentos esenciales, eficacia y costo Se procedio a revisar las listas de medicamentos antibacterianos esenciales de 16 paõÂses seleccionados entre las seis regiones de la OMS. Entre los antibacte- rianos esenciales de la mayorõÂa de esos paõÂses figuran las penicilinas, las tetraciclinas, los aminogluco cidos, las sulfamidas y los macro lidos. En ausencia de polifarma- corresistencia esos antibio ticos son adecuados para tratar las infecciones respiratorias agudas, las enferme- dades diarreicas y las ETS. Estas enfermedades son las que ma s contribuyen a la mayor morbilidad y mortalidad observada en los paõÂses en desarrollo. A fin de invertir esa tendencia, se recurre cada vez ma s a medicamentos antibacterianos, los cuales representan las dos terceras partes de los fa rmacos empleados en esos paõÂses. No obstante, el aumento de la resistencia bacteriana limita la utilidad de algunos de los antibacterianos que figuran en las listas de esos paõÂses. En caso de polifarmacorresistencia, ¿hasta que punto puede ser u til una lista de medicamentos esenciales que omita las cefalosporinas de tercera generacio n, las fluoroquinolonas, la espectinomicina o la vancomicina? Hay pruebas de que el aumento del costo de algunos tratamientos que ocasiona la resistencia bacteriana puede traducirse en una mala observancia del re gimen prescrito. Se llevo a cabo un ana lisis del costo por tratamiento para determinar el efecto de la farmacorresistencia en los costos terape uticos. Los resultadosmostraron que la administracio n de productos ma s recientes, como fluoroquinolonas o cefalosporinas, tras el fracaso de un tratamiento previo con penicilinas, sulfonamidas o tetraciclinas aumentaba el costo del tratamiento de manera astronoÂmica. Ello grava au n ma s la asistencia sanitaria en la mayorõÂa de los paõÂses en desarrollo, cosa que mal pueden permitirse. Se analizan las repercusiones sanitarias de la resistencia bacteriana y las medidas que permiten reducir la propagacio n de las cepas resistentes. 215Bulletin of the World Health Organization, 1999, 77 (3) The antibacterial paradox: essential drugs, effectiveness, and cost References 1. MacInnes R, Lumley CE, Walker SR. New chemical entity output of the international pharmaceutical industry from 1970 to 1992. Clinical pharmacology and therapeutics, 1994, 56: 339±349. 2. The world drug situation. Geneva, World Health Organization, 1988. 3. AntalGM,MehusA. Sexually transmitted diseases in developing countries. Current opinion in infectious diseases, 1988, 1: 26±32. 4. Tomasz A. Multiply resistant pathogenic bacteria: a report on the Rockefeller University workshop. New England journal of medicine, 1994, 330: 1247±1251. 5. Model prescriber information on sexually transmitted diseases and HIV. Geneva, World Health Organization, 1995. 6. The use of essential drugs. Model list of essential drugs. Seventh report of the WHO Expert Committee. Geneva, World Health Organization, 1997 (WHO Technical Report Series, No. 867). 7. Botswana drug catalogue and treatment guide. Gaborone, Botswana National Standing Committee on Drugs, 1995. 8. The Kenyan essential drugs list. Nairobi, Ministry of Health, 1993. 9. Malawi standard drug list. Lilongwe, Ministry of Health/Malawi Essential Drugs Programme, 1991. 10. Federal Republic of Nigeria essential drugs list. Abuja, Federal Ministry of Health, 1989. 11. Essential drugs list for Zimbabwe including guidelines for treatment of medical conditions common in Zimbabwe. Harare, Ministry of Health, 1989. 12. Re publique du Be nin liste nationale des me dicaments essentiels sous noms ge ne riques. Porto-Novo, MinisteÁ re de la Sante , 1994. 13. Liste des me dicaments essentiels en de nomination de Mali. Bamako, MinisteÁ re de la Sante Publique et des Affaires Sociales, 1990. 14. Liste nationale des me dicaments essentiels. In: Formulaire national des me dicaments. Conakry, Re publique de Guine e, MinisteÁ re de la Sante , 1992. 15. Cuadro nacional de medicamentos ba sicos, 2nd ed. Consejo Nacional de Salud, Comisio n Nacional de FarmacologõÂa, 1992 (in Spanish). 16. Lista de medicamentos ba sicos. Tegucigalpa, Repu blica de Honduras, SecretarõÂa de Salud Pu blica. Ministerio de Salud Pu blica in collaboration with WHO/PAHO, 1993 (in Spanish). 17. People's Democratic Republic of Yemen essential drugs list and Eastern Mediterranean drug information sheets. Ministry of Health, 1990. 18. List of essential drugs for Estonia. Tallinn, Ministry of Health, 1992. 19. National list of essential drugs of Bhutan. Thimphu, Royal Government Directorate of Health Services, 1995. 20. List of essential drugs, first revision. Colombo,Ministry of Health of Sri Lanka, 1988. 21. Thailand national list of essential drugs. Bangkok, National Drug Committee, Ministry of Health, 1992. 22. Philippines national drug formulary, 2nd ed, Essential drugs list. Manila, National Drug Committee Department of Health, 1990. 23. The management of acute respiratory infections in children. Practical guidelines for outpatient care. Geneva, World Health Organization, 1995. 24. The management and prevention of diarrhoea. Practical guide- lines, 3rd ed. Geneva, World Health Organization, 1993. 25. Shaw PR, Griffin CC. Financing health care in sub-Saharan Africa through user fees and insurance. Washington, DC, World Bank, 1995. 26. La-Ruche G, Lorougnon F, Digbeu N. Therapeutic algorithms for the management of sexually transmitted diseases at the peripheral level in Coà te d'Ivoire: assessment of efficacy and cost. Bulletin of the World Health Organization, 1995, 73: 305±313. 27. Gomo E et al. In vitro activity of several antimicrobial agents against Neisseria gonorrhoeae and comparison of cost of treatment. Central African journal of medicine, 1995, 41: 83±86. 28. Williams JD. Quo vadis with bacterial resistance. Drugs, 1995, 49: 1±9. 29. Vandepitte J et al. Guidelines for antimicrobial resistance surveillance. Alexandria, WHO Regional Office for the Eastern Mediterranean, 1996. 30. MichelM, Gutmann L. Staphylococcus aureus and vancomycin- resistant enterococci: therapeutic realities and possibilities. Lancet, 1997, 349: 1901±1906. 216 Bulletin of the World Health Organization, 1999, 77 (3) Research