Skip Navigation

Pathways for Inappropriate Dispensing of Antibiotics for Rhinosinusitis: A Randomized Trial

  1. Despina G. Contopoulos-Ioannidis1,2,
  2. Ioanna D. Koliofoti1,
  3. Ioanna C. Koutroumpa1,
  4. Ioannis A. Giannakakis1, and
  5. John P. A. Ioannidis1,3
  1. 1 Clinical Trials and Evidence-Based Medicine Unit, Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina, Greece
  2. 2 Department of Pediatrics, George Washington University School of Medicine, Washington, DC
  3. 3 Department of Medicine, Tufts University School of Medicine, Boston
  1. Reprints or correspondence: Dr. John P. A. Ioannidis, Dept. of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina 45110, Greece (jioannid{at}cc.uoi.gr).
 
Next Section

Abstract

We evaluated the extent of and factors that determine the inappropriate use of antibiotics that are obtained without a physician's prescription. Ninety-eight Greek pharmacists were visited by actress-researchers whoplayed clients requesting antibiotics without a physician's prescription. Pharmacists were randomly challenged in a scenario that involved simulated cases of acute uncomplicated rhinosinusitis with either low fever (38.5°C) or high fever (40°C). Antibiotics were offered by 34 (69%) of 49 pharmacists who were presented with the high-fever scenario and by 42 (86%) of 49 pharmacists who were presented with the low-fever scenario (risk difference, 16.3%;). Thirty-two (65%) and 35 (71%) pharmacists in the high- and low-fever study arms, P=.05 respectively, agreed to sell the actress-researchers broad-spectrum antibiotics. Only 28 (57%) and 17 (35%) pharmacists, respectively, recommended that the patient visit a physician (P=.03). Inappropriate recommendations regarding antibiotic use were very common in the studied setting. Antibiotics were more likely to be offered to persons who did not have a prescription when they were less likely to be clinically indicated.

Inappropriate antibiotic use for treatment of patients with common infections is a major problem worldwide, with great implications with regard to cost of treatment and development of resistance to antimicrobial agents [1]. Out-of-hospital community use accounts for almost two-thirds of the antibiotic market worldwide, and therapy for respiratory tract infection accounts for up to three-quarters of the ∼1 billion annual community prescriptions [2]. Antibiotics may often be dispensed without a clear clinical indication, and they are high on the list of drugs used for self-medication [3, 4]. The social and ecological mechanisms that have led to inappropriate use of antibiotics have not been rigorously studied. Physicians may sometimes prescribe antibiotics without a clear clinical indication [5, 6]. However, an alternative mechanism is that patients sometimes use antibiotics that they have purchased directly from dispensaries without a physician's recommendation. Most developed countries require a physician's prescription for systematic use of antimicrobial agents, but it is not known whether this requirement is adhered to at the community level.

There are high rates of bacterial resistance to antibiotics in Greece [7, 8]. According to state regulations, antibiotics should be dispensed only with a physician's prescription, but there have been no pertinent accusations or court decisions that might lead to activation of such laws. We were anecdotally aware of the frequent sale of antibiotics by pharmacists to patients who do not have a prescription or medical justification. The following 2 important questions arose: (1) What is the extent of this phenomenon? (2) Are antibiotics dispensed more easily in the community when symptoms are less likely to warrant medical evaluation and antimicrobial coverage? To answer these questions, we performed a randomized trial. In clinical scenarios, instructed actress-researchers asked pharmacists to supply antibiotics for the treatment of patients with simulated cases of acute rhinosinusitis, both with and without high-grade fever. We compared the 2 scenarios with regard to the rate at which antibiotics were offered, the likelihood that broad-spectrum agents were recommended, and the number of referrals made to medical care.

Previous SectionNext Section

Methods

Simulated clinical presentation and setting. In the Pathways for Inappropriate Dispensing of Antibiotics for Rhinosinusitis (PINDAR) study, we attempted to discover whether pharmacists would provide patients with antibiotics for management of symptoms that were suggestive of acute uncomplicated rhinosinusitis of 2-days' duration. Two actress-researchers pretended to be customers at pharmacies; they requested antibiotics, allegedly on behalf of a sick sister. The pharmacists were asked to provide the actress-researchers with antibiotics, even though the actress-researchers did not have a physician's prescription and even though the pharmacists did not see the “patient” concerned; the pharmacists were unaware of the experimental nature of the encounter. By means of a randomized design, we also examined whether the alleged presence of high-grade fever would alter the behavior of the pharmacists. Acute rhinosinusitis is often a self-limited viral syndrome. There is no good evidence that antibiotics offer any advantage in the management of this condition, when the probability of bacterial infection is low [9, 10]. Treatment with antibiotics is generally not recommended to patients with symptoms of 2-days' duration, and the probability of bacterial infection is even lower in the absence of high-grade fever. If antibiotics are to be prescribed, amoxicillin or trimethoprim-sulfamethoxazole are as good as expensive, broad-spectrum antibiotics [9].

Selection of pharmacies. We evaluated pharmacies in Ioannina, a university city in northwestern Greece with a population of 100,000 people. Because it is a university city, Ioannina has a generally high level of medical services and health awareness, and there is emphasis on graduate and postgraduate training and seminars for health professionals. Pharmacists in Greece obtain their degree after 5 years of university training; they also take postgraduate courses.

All 105 licensed pharmacists who practice in the area were eligible for the study. The names and addresses were obtained from the public list of licensed pharmacists. Pharmacists who knew an actress-researcher were excluded. In cases in which 2 pharmacists worked in the same office, only 1 was selected. There was no potential for harm to anyone who was involved in our randomized study, because the patients were fictitious and the pharmacists were not exposed to any risk.

Trial design. The trial was a randomized study. A computer random-number generator randomized pharmacies in 2 study arms that consisted of clinical scenarios of uncomplicated acute rhinosinusitis with either low fever (38.5°C) or high fever (40°C). Pharmacies were also stratified to be visited by 1 of the 2 actress-researchers.

Protocol. The actress-researchers who visited the pharmacies were medical students who were also key members of the theatrical group of the University of Ioannina. We used a predefined algorithm for the dialogue between the actress-researcher and the pharmacist. The actresses practiced their roles extensively during several hypothetical encounters. Two independent observers verified that the actresses knew their roles well and that they were following the algorithm consistently before they commenced the actual pharmacy visits. The pharmacy visits were made in early spring. Pharmacies were mapped on a city map for efficiency.

The algorithm was as follows. Initially the actress-researcher said to the pharmacist, “My sister has a bad cold with a fever of 38.5°C [or 40°C, depending on the arm to which the particular pharmacy was randomized] and a headache right here [the actress-researcher pointed over her sinus area while she said this]. Could you give me some antibiotics?” If the pharmacist suggested an antibiotic, the actress-researcher would ask for a stronger one: “Is this strong enough? Can you give me something stronger?” If the pharmacist did not offer an antibiotic or had offered a nonantibiotic medication, the actress-researcher would insist, “She is having a bad time; please give me something.”

We had assembled some predefined, additional information that could be provided to the pharmacist only if the pharmacist were to ask specific questions. Specific answers to any questions asked by the pharmacist were also predefined, to ensure that medical terminology was avoided. The additional information that could be provided included the patient's age (“20 years”); the absence of other medical problems; the frequency of similar episodes (“not very often”); the duration of symptoms (“2 days”); the absence of otitis, pharyngitis, cough, or dyspnea; and the presence of rhinorrhea (“runny nose”). The patient was supposed to have taken “some” antibiotics in the distant past, but the actress-researcher was not supposed to know which antibiotics. The patient was not being treated with any drugs regularly, was receiving only antipyretics currently, had not visited a doctor, and had not had a sinus radiograph performed. If the actress-researchers were asked questions about allergies, they would answer, “I don't think she has any”; if the pharmacist asked further questions specifically about allergies to penicillin, the actress-researcher would answer, “I don't know about that.” The actress-researcher was not to show concern regarding the cost of expensive antibiotics; however, at the end of the encounter, she would realize that she had unfortunately forgotten her wallet, and that she or someone else would come back to get the medications.

After the actress-researcher left each pharmacy, she completed a protocol form on which she recorded the time of the encounter, the number of other clients present in the pharmacy during the encounter, and precoded information regarding study end points. The information was cross-validated by a third independent investigator who subsequently interpreted the accumulated information.

End points. The primary end point was the dispensation of at least 1 antibiotic by the pharmacist. We also recorded the type of antibiotics that the pharmacist offered and any reasons the pharmacist had for not agreeing to dispense antibiotics. Secondary end points included the following: an immediate decision to dispense an antibiotic without any need to insist on the part of the actress-researcher, a decision to dispense an antibiotic that was “stronger” than the antibiotic that was initially suggested (after the insistence of the actress-researcher), a decision to dispense other medications (alone or in addition to antibiotics), a recommendation for evaluation by a physician (regardless of whether antibiotics were finally offered or not), questions asked by the pharmacist regarding drug allergies and prespecified items of information, and provision of information regarding drug interactions, treatment dose, and duration of treatment for the offered antibiotics.

Analysis. The study had a power of 80% to detect a difference of 20% for the primary end point between the 2 study arms, anticipating that 75% of the pharmacists might have offered antibiotics in the high-fever scenario (α=.05). Categorical end points were analyzed by use of Fisher's exact test, and risk differences were estimated. There was no heterogeneity between the results obtained by each actress; actress-stratified results were similar. Analyses were conducted by use of SPSS, version 9.0 (SPSS). All P values were 2-tailed.

Previous SectionNext Section

Results

Of 105 city pharmacies, 102 were randomized, 51 to each of the 2 clinical scenarios. Forty-nine pharmacies in each study arm could be evaluated (figure 1). By evaluating the collected information, we found very good agreement between the interpretations given by the actresses and those given by the independent observer regarding the primary and secondary end points. There was full agreement for all primary end points. Characteristics of the pharmacies that were included in the 2 study arms are shown in table 1. There were no significant differences with regard to location of the pharmacy, time of the encounter, or number of clients present in the pharmacy during the encounter.

Figure 1
View larger version:
Figure 1

CONSORT flow diagram for study of dispensing behaviors of pharmacists.

Table 1
View larger version:
Table 1

Characteristics of Greek pharmacies investigated for dispensing practices.

The majority of the pharmacists in both study arms agreed to dispense antibiotics, and more pharmacists in the low-fever arm agreed to dispense antibiotics than did pharmacists in the high-fever arm (table 2). The difference between the 2 arms was on the borderline of statistical significance (86% of the pharmacists in the low-fever arm offered antibiotics vs. 69% of the pharmacists in the high-fever arm; risk difference, 16.3%; 95% CI, 0.1–32.5;). The 2 actresses were offered P=.05 antibiotics at similar rates (33 of 46 for actress 1 vs. 43 of 52 for actress 2; P=.23). Of the pharmacists who did not offer antibiotics, all did so because the actress-researcher did not have a physician's prescription. The additional reasons that pharmacists refused to offer antibiotics are recorded in table 2.

Table 2
View larger version:
Table 2

End points in study of inappropriate dispensation of antibiotics.

Initially, the following antibiotics were offered (data are number of instances in which the antibiotic was offered in the high- and low-fever arms, respectively): amoxicillin (12, 12), ampicillin (1, 1), amoxicillin-clavulanate (15, 21), erythromycin (0, 2), clarithromycin (4, 2), roxithromycin (1, 4), a nonspecified macrolide (0, 1), vibramycin (1, 0), cefaclor (3, 3), cefuroxime (0, 2), cefatrizine (0, 1), and rifampicin (1, 0). Seven pharmacists suggested several possible antibiotics initially. After the insistence of the actress, the following “stronger” antibiotics were offered (data are number of instances in which the antibiotic was offered in the high- and low-fever arms, respectively): ampicillin (1, 0), amoxicillin-clavulanate (5, 4), clarithromycin (2, 3), roxithromycin (2, 1), azithromycin (1, 1), vibramycin (1, 2), clindamycin (0, 1), cefadroxil (1, 0), cefaclor (3, 1), cefuroxime (1, 0), cefprozil (3, 3), a nonspecified third-generation cephalosporin (0, 1), and ciprofloxacin (1, 0). The actress-researcher's second request typically led the pharmacist to offer more expensive and/or more broad-spectrum antibiotics. In total, 32 and 35 pharmacists in the high-fever and low-fever arms, respectively, suggested at least 1 expensive broad-spectrum antibiotic (i.e., an antibiotic other than the inexpensive amoxicillin, ampicillin, erythromycin, or vibramycin).

No statistically significant differences were seen between the 2 arms with regard to any secondary end points (table 2), with the exception of the recommendation that the patient should be seen by a physician (recommended by 57% of the pharmacists in the high-fever arm vs. 35% in the low-fever arm; P=.03). Several pharmacists in both study arms offered nonantibiotic medications, including antipyretics and anti-inflammatory agents (30 pharmacists total), nasal decongestants (13 pharmacists total), and vitamin C (2 pharmacists total). Some pharmacists recommended >1 of these medications.

Most pharmacists asked some additional questions. The 4 most frequently asked questions focused on associated upper or lower respiratory tract infections, presence of nasal discharge, and location of headache, and whether the patient had been seen by a physician (table 3). However, at least half of the pharmacists asked no questions regarding the 14 precoded items of information. Only 22 pharmacists asked ⩾5 questions and no pharmacist asked >8 questions. On average, pharmacists asked questions regarding 2.6 of the 14 precoded items of information, with no difference between the 2 study arms (P = .93). Only 17 pharmacists inquired about P=.93 drug allergies. The majority of the pharmacists provided adequate information on treatment dose and duration without discussing potential drug interactions.

Table 3
View larger version:
Table 3

Information requested by pharmacists.

Previous SectionNext Section

Discussion

This randomized trial examined the problem of inappropriate dispensation of antibiotics by pharmacists for management of a common community infection for which there is no evidence that antibiotics offer clear advantages [9, 10]. Seventy-seven percent of the pharmacists we studied offered antibiotics without a medical prescription and without even seeing the patient. Most of the pharmacists offered expensive broad-spectrum antibiotics. Antibiotics were more frequently offered for treatment of patients with symptoms that were suggestive of a common cold with low-grade fever than they were for cases with high-grade fever. Therefore, there was no evidence that pharmacists offered antibiotics while taking into account the severity of symptoms as an indication for treatment. On the contrary, it is conceivable that for patients with high-grade fever, several pharmacists did not wish to risk the possibility that the patient would not see a physician if she were provided with antibiotics. As a result, the offering of antibiotics was most facile when such antibiotics would be remotely likely to be of any benefit.

Certain aspects of the study design and the ethics thereof should be discussed. The use of actors to assess the health care situation should ensure that the encounters replicated real practice as closely as possible. Several measures were taken to ensure that these encounters resembled real encounters, including use of structured material, careful rehearsal of the roles of the actress-researchers that involved several hypothetical variations on the encounter before the study began, independent interpretation of the collected information by separate assessors, and a protocol that separated the immediate dispensation of antibiotics from the dispensation of antibiotics after a second request. Moreover, we aimed to have the delivery of the requests for antibiotics be consistently “low tone,” rather than histrionic, provocative, or manipulative. Finally, the questions and responses were written in a way to represent actual encounters that we had witnessed during informal visits to Greek pharmacies in different parts of the country. The purpose of the study was not to entrap pharmacists but to study their dispensing behavior under real-life settings. We did not ask for the collaboration and involvement of professional pharmacist associations, because alerting pharmacists beforehand and communication of the study purpose and design would have affected their behavior, thus undermining the purposes of the trial.

In contrast to other studies that have used actors to play standardized patients to assess quality of care [11, 12], our actors played the roles of relatives of patient and not the patients themselves. The proportion of pharmacists who agreed to dispense antibiotics is even more surprising when we consider that the pharmacists did not even see the patients concerned. There were no risks to the participants involved. Some studies that have sent standardized patients to physicians have asked for physician consent, because there is an opportunity cost from the time lost consulting with fake patients that may have been devoted to real patients. There was no such opportunity cost in our study, because the interviews took a minimum amount of time and the encounters did not affect the quality of care that other clients received.

Our study design is limited by the fact that the trial was performed in a specific city in Greece. The extent to which this dispensing behavior may apply in other regions and other countries is unknown but worth studying. When we launched the trial, we had anecdotal information about the high prevalence of this phenomenon from several parts of Greece, including Athens and other cities, and colleagues had suggested that these practices may occur in other countries as well. We preferred to target all pharmacies in a large university city, because this design would allow the most controlled execution of the study plan. We also selected a university city with a health sciences faculty, where pharmacists are likely to be, if anything, more sensitized to issues regarding inappropriate dispensing of drugs.

Several European studies have documented rising use of antimicrobial agents in the community [13, 14]; this is also true in Greece. From 1980 through 1990, the number of prescriptions for antibiotics increased by 46% in England [14], whereas in France, the number of prescriptions for antibiotics for management of upper respiratory infection increased by 86% for adults and by 115% for children [13]. This increase in antibiotic use has been more rapid for recently introduced drugs and for drugs that have been heavily advertised [14–16]. Besides overprescription by physicians, antibiotics may also be misused by the patients themselves. Self-medication is a major form of self-care and may be the most common treatment response to symptoms [17, 18]. A surveillance study in Taiwan showed that 55% of patients who arrived at an emergency department were already receiving antibiotics [19]. Even in the United States, where physician prescription is more strictly enforced by law, a recent study in an ethnically diverse community in the Bronx showed that 26% of adults had obtained antibiotics for management of upper respiratory tract infections by means other than a physician's prescription, including dispensing from pharmacists and from suppliers located outside the United States; 31% of the interviewed subjects believed that antibiotics should be routinely available over the counter [20]. Therefore, even in the United States, prescriptions provided by pharmacists and self-medication with antibiotics may occur sometimes; determination of the extent of this phenomenon requires further study. Furthermore, unnecessary overprescription by physicians in the United States is also well documented [21, 22].

Use of antimicrobial agents correlates with the prevalence of bacterial resistance in the community [23–25]. A nationwide reduction in the outpatient use of macrolides in Finland led to a decrease in erythromycin resistance among isolates of group A streptococci, from 16.5% of isolates in 1992 to 8.6% of isolates in 1996 [23]. Proposed strategies to control antimicrobial resistance in the community include use of generic medications, restricted lists of antibiotics, education of physicians, and marketing regulations [2, 5]. Still, the effectiveness of such strategies is not certain.

All of these strategies are in effect in Greece. However, Greece has one of the highest per-person rates of use of antibiotics. Drug expenditure in Greece represents 2.2% of the gross domestic product, a rate that is 3-fold the respective figure for the United Kingdom [26]. Antibiotics accounted for 20% of all drug expenditure in Greece in 1992 (Hellenic Organization for Medicines, data on file). In 1995, use of antibiotics in Greece equaled 26 defined daily doses per 1000 persons per day, which is twice the respective figure for Denmark [27]. Our study documented that means of obtaining antibiotics that circumvent the regular prescription pathway may be important in propagating inappropriate antibiotic use in a European country. Inappropriate dispensing may be even more common when the diagnosis is almost certainly a viral infection. In these situations, antibiotics are considered an easy, trivial solution that requires no second thought. Interventions that try to counter this disturbing phenomenon may need to involve both patients and pharmacists in addition to physicians.

Recently, there has been interest in and ongoing discussions about broadening the numbers and types of medications that might change from prescription status to over-the-counter status [28]. Our study suggests that such changes may need to be evaluated in terms of whether they might increase the inappropriate use of the liberalized prescription medications. For countries that already extensively use over-the-counter drugs, our data suggest that inappropriate use should be closely monitored, and interventions that aim to rationalize drug utilization, including educational and behavioral ones, should be fostered.

  • Received September 28, 2000.
  • Revision received November 13, 2000.
Previous Section
 

References

  1. ↵
    1. Bax RP,
    2. Anderson R,
    3. Crew J,
    4. et al
    . Antibiotic resistance—what can we do? Nat Med 1998;4:545-6.
    CrossRefMedlineWeb of Science
  2. ↵
    1. Carbon C,
    2. Bax RP
    . Regulating the use of antibiotics in the community. BMJ 1998;317:663-5.
    FREE Full Text
  3. ↵
    Drug Utilization Research Group, Latin America. Multicenter study on self-medication and self-prescription in six Latin American countries. Clin Pharmacol Ther 1997;61:488-93.
    CrossRefMedlineWeb of Science
  4. ↵
    1. Chretien JH,
    2. McGarvey M,
    3. deStwolinski A,
    4. Esswein JG
    . Abuse of antibiotics: a study of patients attending a university clinic. Arch Intern Med 1975;135:1063-5.
    Abstract/FREE Full Text
  5. ↵
    1. Butler CC,
    2. Rollnick S,
    3. Pill R,
    4. Maggs-Rapport F,
    5. Stott N
    . Understanding the culture of prescribing: qualitative study of general practitioners' and patients' perceptions of antibiotics for sore throats. BMJ 1998;317:637-42.
    Abstract/FREE Full Text
  6. ↵
    1. Gonzales R,
    2. Steiner JF,
    3. Sande MA
    . Antibiotic prescribing for adults with colds, upper respiratory tract infections, and bronchitis by ambulatory care physicians. JAMA 1997;278:901-4.
    Abstract/FREE Full Text
  7. ↵
    1. Syrogiannopoulos GA,
    2. Grivea IN,
    3. Beratis NG,
    4. et al
    . Resistance patterns of Streptococcus pneumoniae from carriers attending day-care centers in southwestern Greece. Clin Infect Dis 1997;25:188-94.
    Abstract/FREE Full Text
  8. ↵
    1. Paraskakis I,
    2. Lebessi E,
    3. Legakis NJ,
    4. Skandami V,
    5. Kafetzis DA
    . Resistance pattern of Streptococcus pneumoniae in children during a four-year period in Greece. Eur J Clin Microbiol Infect Dis 1998;17:864-6.
    CrossRefMedlineWeb of Science
  9. ↵
    1. De Ferranti SD,
    2. Ioannidis JP,
    3. Lau J,
    4. Anninger WV,
    5. Barza M
    . Are amoxycillin and folate inhibitors as effective as other antibiotics for acute sinusitis? A meta-analysis. BMJ 1998;317:632-7.
    Abstract/FREE Full Text
  10. ↵
    Evidence based report: diagnosis and treatment of acute bacterial rhinosinusitis. Boston: New England Medical Center Evidence Based Practice Center; 1998. document no. AHCPR 290-97-0019.
  11. ↵
    1. Silvestre AJ,
    2. Gehl MB,
    3. Encadela J,
    4. Schelzel G
    . A participant observation study using actors at 30 publicly funded HIV counseling and testing sites in Pennsylvania. Am J Public Health 2000;90:1096-9.
    MedlineWeb of Science
  12. ↵
    1. Peabody JW,
    2. Luck J,
    3. Glassman P,
    4. Dresselhaus TR,
    5. Lee M
    . Comparison of vignettes, standardized patients, and chart abstraction: a prospective validation study of 3 methods for measuring quality. JAMA 2000;283:1715-22.
    Abstract/FREE Full Text
  13. ↵
    1. Guillemot D,
    2. Maison P,
    3. Carbon C,
    4. et al
    . Trends in antimicrobial drug use in the community: France, 1981–1992. J Infect Dis 1998;177:492-7.
    Abstract/FREE Full Text
  14. ↵
    1. Davey PG,
    2. Bax RP,
    3. Newey J,
    4. et al
    . Growth in the use of antibiotics in the community in England and Scotland in 1980–93. BMJ 1996;312:613.
    FREE Full Text
    1. McGavock H,
    2. Webb CH,
    3. Johnston GD,
    4. Milligan E
    . Market penetration of new drugs in one United Kingdom region: implications for general practitioners and administrators. BMJ 1993;307:1118-20.
    Abstract/FREE Full Text
  16. ↵
    1. McCaig LF,
    2. Hughes JM
    . Trends in antimicrobial drug prescribing among office-based physicians in the United States. JAMA 1995;273:214-9.
    Abstract/FREE Full Text
  17. ↵
    1. Levin LS,
    2. Beske F,
    3. Fry J
    . Self-medication in Europe Report on a study of the role of non-prescription medicines. Copenhagen: World Health Organization Regional Office for Europe; 1988.
  18. ↵
    Over-the-counter drugs [editorial]. Lancet 1994;343:1374-5.
    CrossRefMedlineWeb of Science
  19. ↵
    1. Liu YC,
    2. Huang WK,
    3. Huang TS,
    4. Kunin CM
    . Extent of antibiotic use in Taiwan shown by antimicrobial activity in urine. Lancet 1999;354:1360.
    CrossRefMedlineWeb of Science
  20. ↵
    1. McKee MD,
    2. Millis L,
    3. Mainous AG III
    . Antibiotic use for the treatment of upper respiratory infections in a diverse community. J Fam Pract 1999;48:993-6.
    MedlineWeb of Science
  21. ↵
    1. Mainous AG III,
    2. Haeston WJ,
    3. Love MM
    . Antibiotics for colds in children: who are the high prescribers? Arch Pediatr Adolesc Med 1998;152:349-52.
    Abstract/FREE Full Text
  22. ↵
    1. Watson RL,
    2. Dowell SF,
    3. Jayaraman M,
    4. Keyserling H,
    5. Kolczak M,
    6. Schwartz B
    . Antimicrobial use for pediatric upper respiratory infections: reported practice, actual practice, and patient beliefs. Pediatrics 1999;104:1251-7.
    Abstract/FREE Full Text
  23. ↵
    1. Seppala H,
    2. Klaukka T,
    3. Vuopio-Varkila J,
    4. et al
    . The effect of changes in the consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland. N Engl J Med 1997;337:441-6.
    CrossRefMedlineWeb of Science
    1. Baquero F,
    2. Martinez-Beltran J,
    3. Loza E
    . A review of antibiotic resistance patterns of Streptococcus pneumoniae in Europe. J Antimicrob Chemother 1991;28(suppl C):31-8.
    Abstract/FREE Full Text
  25. ↵
    1. Arason VA,
    2. Kristinsson KG,
    3. Sigurdsson JA,
    4. Stefansdottir G,
    5. Molstad S,
    6. Gudmundsson S
    . Do antimicrobials increase the carriage rate of penicillin resistant pneumococci in children? Cross sectional prevalence study. BMJ 1996;313:387-91.
    Abstract/FREE Full Text
  26. ↵
    1. Mossialos E,
    2. Ranos C,
    3. Abel-Smith B
    . Cost containment, pricing and financing of pharmaceuticals in the European Community. Athens: Pharmetrica; 1994.
  27. ↵
    Nordic Council on Medicines. Nordic statistics on medicines. Uppsala, Sweden: NLN; 1996.
  28. ↵
    1. Reeves DS,
    2. Finch RG,
    3. Bax RP,
    4. et al
    . Self-medication of antibacterials without prescription (also called over-the-counter use). A report of a Working Party of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother 1999;44:163-77.
    Abstract/FREE Full Text
| Table of Contents

This Article

  1. Clin Infect Dis. (2001) 33 (1): 76-82. doi: 10.1086/320888
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most