Patient Knowledge and Attitudes about Antiviral Medication and Vaccination for Influenza in an Internal Medicine Clinic

Transcription

1 MAJOR ARTICLE Patient Knowledge and Attitudes about Antiviral Medication and Vaccination for Influenza in an Internal Medicine Clinic Michael A. Gaglia, Jr., 1,2 Robert L. Cook, 3 Kevin L. Kraemer, 1 Michael B. Rothberg 4 1 Division of General Internal Medicine and 2 Division of General Academic Pediatrics, University of Pittsburgh, Pennsylvania; 3 Departments of Epidemiology, Biostatistics, and Medicine, University of Florida, Gainesville; and 4 Division of General Medicine and Geriatrics, Tufts University, Boston, Massachusetts (See the editorial commentary by Linder on pages ) Background. Despite the introduction of Centers for Disease Control and Prevention guidelines for their use, antiviral medications for influenza remain underutilized. Our objective in this study was to describe beliefs, attitudes, and knowledge regarding antiviral medication and vaccination for influenza among patients in an internal medicine clinic. Methods. We conducted a cross-sectional survey of adult patients in an internal medicine clinic from April through June Results. Two-hundred eighty patients completed the survey. Fifty-five percent received influenza vaccination for the most recent influenza season. Overall antiviral knowledge was poor. Of 8 antiviral knowledge questions, the mean percentage of correct answers was 40%; 1 (!1%) of the patients answered all questions correctly, and 47 (18%) answered all questions incorrectly. Only 37 (13%) of the patients reported calling their physician within 48 h after the onset of influenza-like symptoms. Patients with conditions associated with a high risk of complications from influenza were no more likely than other patients to be more knowledgeable about antiviral medication, nor were they more likely to report calling their physician within 48 h after symptom onset or to report receipt of influenza vaccination for the previous influenza season. Only 90 (37%) of the respondents were willing to pay 1$20 for antiviral medication, although 205 (84%) were willing to pay something. Conclusions. Patients are ill-informed about antiviral medication and its benefits, and medication costs may present a barrier to treatment. Physicians should discuss antiviral medication with patients who are at high risk for complications from influenza before the influenza season, and education programs for physicians and patients should be developed. Influenza affects 5% 20% of the population each year [1], and influenza-related illness is responsible for 51,000 deaths and 226,000 hospitalizations each year in the United States [2, 3]. Vaccination remains the cornerstone of influenza control and prevention, but antiviral medications, such as the neuraminidase inhibitors oseltamivir and zanamivir, are a key component of the most recent recommendations by the Ad- Received 11 April 2007; accepted 9 June 2007; electronically published 28 September Reprints or correspondence: Dr. Michael A. Gaglia, Jr., Center for Research on Health Care, University of Pittsburgh, Ste. 600, 230 McKee Place, Pittsburgh, PA (gagliajrma@upmc.edu). Clinical Infectious Diseases 2007; 45: by the Infectious Diseases Society of America. All rights reserved /2007/ $15.00 DOI: / visory Committee on Immunization Practices [4]. Specifically, the recommendations emphasize that treatment of influenza should focus on patients who are at high risk for influenza-related complications. The neuraminidase inhibitors have assumed an even more prominent role for 2 reasons: the emergence of widespread resistance to the M2 ion channel-blocking drugs amantadine and rimantadine [5] and the possible role of neuraminidase inhibitors in helping to control an avian influenza pandemic [6]. Neuraminidase inhibitors shorten the course of acute influenza [7, 8], reduce bacterial complications of influenza [9, 10], and decrease influenza-related hospitalizations [11]. They are also cost-effective for both treatment and prevention of influenza [12 15]. The appropriate use of antiviral drugs for treatment of acute influenza, however, requires education of both physi CID 2007:45 (1 November) Gaglia et al.

2 cians and their patients. Specifically, patients must know to present promptly to their physician, because antiviral drugs are not effective 148 h after onset of influenza symptoms; and physicians must be able to rapidly diagnose influenza. The ability of physicians to diagnose influenza on clinical grounds alone has proven to be inadequate, with physician judgment showing a sensitivity of only 29% [16]. Physician knowledge of antiviral drugs is also poor; only 28% of primary care physicians are aware that antiviral drugs prevent bacterial complications of influenza [17]. As a result, physicians prescribe antiviral medication for!20% of patients who receive a diagnosis of influenza [18, 19]. Patient knowledge and beliefs about antiviral medication are important, because they might influence the likelihood of presenting within the 48-h window in which effective treatment could be administered. There is a paucity of studies, however, that address patient knowledge and attitudes regarding antiviral medications, such as oseltamivir and zanamivir. The objectives of this study were to describe patient knowledge and attitudes regarding antiviral medications and to identify demographic and clinical characteristics associated with higher levels of knowledge regarding antiviral medication. We also sought to examine the proportion of patients who would report calling their physician within 48 h after the onset of influenza-like illness. We hypothesized that patients with conditions associated with a high risk of influenza-related complications would be more knowledgeable regarding antiviral medication and more likely to report calling their physician within 48 h. METHODS Design. We distributed anonymous surveys from April through June 2006 to a convenience sample of 400 patients at an urban, university-affiliated general medicine clinic in Pittsburgh, Pennsylvania. Clinic staff distributed the survey to consecutive patients at check-in, as permitted by clinic patient flow, and individual patients were free to discard the survey if they did not wish to complete it. Information regarding patients who did not return the survey was not available. We offered no incentives to complete the survey. The study protocol was approved by the University of Pittsburgh (Pittsburgh, Pennsylvania) Institutional Review Board. Measures. We selected survey items to represent a broad spectrum of health behaviors and attitudes that might be associated with knowledge regarding antiviral medication. Health attitudes questions included beliefs about influenza infection and vaccination. We selected specific health conditions to represent high-risk groups targeted for seasonal influenza vaccination by the Centers for Disease Control and Prevention: chronic lung disease, diabetes, heart disease or stroke, chronic kidney disease, sickle cell anemia or thalassemia, HIV infection, history of organ transplantation, and history of malignancy [4]. Patients were considered to be at high risk for influenza complications if they had 1 of the above conditions, regardless of age. The survey instrument consisted of the following primary sections: demographic information, medical history, specific health behaviors (e.g., receipt of influenza vaccination before the most recent influenza season), knowledge regarding antiviral medication (with possible responses of true, false, and do not know ), attitudes toward and perceptions of antiviral medication and influenza vaccination (likert-type scale of strongly agree to strongly disagree), and willingness to pay for antiviral medication. Immediately preceding the antiviral section, the survey explained that influenza infection is caused by a virus and that antiviral medications, such as oseltamivir (Tamiflu; Roche) and zanamivir (Relenza; GlaxoSmithKline), are used to treat it. We asked patients When you have a cough and fever, how long do you wait before you call the doctor? and offered the following choice of answers: right away, 1 day, 2 days, 12 days, and I usually don t call. We also asked If a medicine decreased the amount of time that you were sick with a cough, fever, runny nose, and body aches by 1 day, how much would you be willing to pay for this medicine? and offered the patient a choice of $0 or $10 intervals up to $60. We also included questions pertaining to knowledge and attitudes about avian influenza; these results will be reported elsewhere. Clinic staff and physicians reviewed initial versions of the survey for face validity and suggested changes for content and clarity. The final version consisted of 34 items, was written at an eighth grade reading level, and required 5 8 min to complete. The survey was in pen-and-paper format and was available only in English. Patients indicated responses by filling in circles corresponding to each answer. After the surveys were collected, we reviewed them for stray and incomplete marks (e.g., slashes and checkmarks) and corrected them appropriately. We then scanned the surveys into a database using TE- LEform (Cardiff) and confirmed a 10% sample of the surveys manually to validate the data capture process. Data analysis. Analyses and data management were performed with Stata for Windows, version 9.0 (StataCorp). Surveys with incomplete responses were excluded from analyses of that particular incomplete response, but they were not excluded from the entire study. We quantified knowledge of antiviral drugs by computing the percentage of correct responses to the 8 knowledge questions; do not know responses were considered to be incorrect, and questions skipped by the respondent were not counted towards the total number of questions answered. The distribution of the percentage correct was not a normal distribution, so we constructed a categorical variable for level of antiviral knowledge: low (0% 33% correct), moderate (34% 66% cor- Knowledge and Attitudes about Antivirals CID 2007:45 (1 November) 1183

3 rect) and high (67% 100% correct). We then used either the x 2 or Fisher s exact test to examine relationships of demographic characteristics, health characteristics, health behaviors, and attitudes with the level of antiviral knowledge. We used multivariable ordinal logistic regression to determine independent variables associated with higher antiviral knowledge, using a forward regression technique with a P!.15 threshold required for an independent variable to remain in the multivariable model. Only variables with a P value of.20 for the x 2 or Fisher s exact test in univariable analyses were considered for the multivariable model. The final multivariable model only used survey respondents who answered all demographic, clinical, and attitude questions (222 respondents). We also used multivariable logistic regression to examine the association of patient knowledge and beliefs with the following outcomes: willingness to pay for antiviral medication (!$20 vs. 1$20), time to calling a physician when experiencing influenzalike symptoms (!48 h vs. 148 h), and receiving an influenza vaccination before the previous influenza season. We used the same P value thresholds for model entry as were used for the antiviral knowledge level regression. RESULTS Of 400 patients who were offered the survey, 280 (70%) responded. The survey population was similar to the overall clinic population with respect to sex and race and was predominantly white (80%) and college-educated (64%; table 1). Thirty-eight percent of respondents had at least 1 disease associated with a high risk of influenza-related complications, as defined by the Centers for Disease Control and Prevention [4], and 46% were 50 years of age. Only 55% received influenza vaccination for the most recent influenza season. Of interest, 2% of respondents admitted to having a home supply of antiviral medication just in case. In multivariable logistic regression, advancing age, but not high-risk status, was associated with current vaccination status. A composite measure of influenza risk (the number of conditions associated with a high risk of influenza-related complications, including age 65 years) also was not associated with receipt of influenza vaccination. Patients at high risk for influenza-related complications (stratified by age) were slightly more likely to have received influenza vaccination for the previous influenza season (figure 1), but the association was not statistically significant. When asked how long they waited to call their physician when experiencing influenza-like symptoms, only 37 (13%) of respondents reported calling within 48 h after symptom onset. Multivariable logistic regression showed that a history of heart disease (OR, 3.54; 95% CI, ) and taking antibiotics often for viral-like symptoms (OR, 1.64; 95% CI, ) were significantly associated with calling 48 h after onset of Table 1. Demographic characteristics of patients who responded to the influenza survey. Characteristic Patients (n p 280) Age, mean years SD Female sex 190 (69) Race/ethnicity White 219 (80) Black 40 (15) Asian 11 (4) Hispanic 6 (2) Other 5 (2) Education College or higher 177 (64) Some college 53 (19) High school or less 47 (17) Annual income!$20, (21) $20,001 $75, (46) 1$75, (33) Current smoker 39 (14) Comorbidity COPD/chronic lung disease 35 (13) Heart disease 25 (9) Diabetes mellitus 37 (13) 1 Disease associated with high risk of influenza-related complications 105 (38) Received influenza vaccination for most recent influenza season 151 (55) Previously received antiviral medication 66 (25) Home supply of antiviral medication 5 (2) NOTE. Data are no. (%) of patients, unless otherwise indicated. COPD, chronic obstructive pulmonary disease. symptoms. No other high-risk conditions, separately or as a composite, were associated with calling within this time period. When asked how much they were willing to pay for antiviral medication that would shorten the duration of influenza-like symptoms by 1 day (figure 2), 205 (84%) of the respondents were willing to pay something, but only 90 (37%) were willing to pay 1$20. According to multivariable logistic regression analysis, willingness to pay 1$20 for antiviral medication was associated with an annual income 1$75,000 (OR, 1.94; 95% CI, ), belief that antiviral medication are scarce (OR, 2.19; 95% CI, ), calling a physician within 48 h after influenza symptom onset (OR, 2.64; 95% CI, ), and taking leftover antibiotics for influenza-like symptoms (OR, 2.82; 95% CI, ); the belief that one does not need medication to recover from influenza was associated with less willingness to pay (OR, 0.34; 95% CI, ). The mean percentage of correct responses to 8 antiviral knowledge questions was 40% (table 2). One respondent (!1%) answered all questions correctly, and 47 (18%) answered all 1184 CID 2007:45 (1 November) Gaglia et al.

4 Figure 1. Receipt of influenza vaccination for the most recent influenza season, by age and risk group. High-risk group, patients at high risk for influenza and influenza-related complications; low-risk group, patients at low risk for influenza and influenza-related complications. questions incorrectly. Among respondents who answered all 8 knowledge questions (252 respondents), 40 (15%) answered do not know to all 8 questions. Thirty percent knew that antiviral medications only work if they are taken during the first 48 h of symptoms. A significant percentage (96 respondents; 37%) believed that influenza vaccination can cause influenza; more than one-half of the respondents were worried that there is not enough antiviral medication (133 respondents; 52%) or influenza vaccine (131 respondents; 51%) in the United States. Approximately one-third of the respondents believed that they did not need medication to recover from influenza; there was no difference between patients at high risk for complications and other patients in this regard. Multivariable ordinal logistic regression analysis indicated that 3 variables were significantly associated with a higher degree of antiviral knowledge: college education or higher (OR, 1.99; 95% CI, ), the belief that influenza vaccination is effective (OR, 1.94; 95% CI, ), and the belief that one does not need medication to recover from influenza (OR, 2.55; 95% CI, ). Black race (OR, 0.39; 95% CI, ) and the belief that influenza vaccination causes influenza (OR, 0.54; 95% CI, ) were associated with lower antiviral knowledge. CONCLUSIONS We found that overall knowledge regarding antiviral medication was poor. We also found that patients with conditions associated with a high risk of influenza complications were no more likely than patients without such conditions to report receipt of influenza vaccination for the most recent influenza season or to report calling their physician within 48 h after the onset of influenza-like symptoms; in addition, patients with highrisk conditions were not more willing than others to pay 1$20 for antiviral medication. Perhaps most importantly, however, patients with conditions associated with a high risk of complications from influenza were not more knowledgeable regarding antiviral medication; this was contrary to our initial hypothesis. Neuraminidase inhibitors are effective in decreasing the length of illness and in preventing hospitalizations and bacterial complications in patients with influenza who are at high risk for complications [10, 11]. Furthermore, such treatment is endorsed by Centers for Disease Control and Prevention guidelines. Two potential barriers to treatment with antiviral drugs are patient knowledge and the relatively high cost of the medications. Knowledge deficits are important because of the limited window of effectiveness for the neuraminidase inhibitors. Patients must be aware of both the potential benefit of treatment in terms of decreasing morbidity and the need to present for treatment as early as possible. We found that patients at high risk for influenza-related complications are largely unaware of the benefits of antiviral medication. Except for patients with heart disease (for whom the effect was small), patients with high-risk conditions were not more likely than patients Knowledge and Attitudes about Antivirals CID 2007:45 (1 November) 1185

5 Figure 2. Percentage of patients willing to pay for antiviral medication to treat influenza, by cost of medication without high-risk conditions to call within the 48 h window. Indeed, only 13% of all respondents stated they would call within the appropriate time frame. The knowledge deficit in our sample group was not limited to antiviral medication, as evidenced by the 37% of respondents who believed that influenza vaccination causes influenza. This percentage is higher than that reported in a recent Medicare Current Beneficiary Survey [20], conducted in , in which 20% of respondents refused influenza vaccination because of the belief that it would cause influenza. A large proportion of respondents (63%) were unaware that antiviral medications are ineffective against bacteria, demonstrating that a significant number of patients continue to misunderstand the difference between viral and bacterial illness. This is consistent with previous studies, which found a 21% 55% prevalence of the false belief that antibiotics are effective for viral upper respiratory illnesses [21 23]. We also found that patients with high-risk conditions who are!50 years of age continue to be undervaccinated, with only 30% of respondents in our study reporting recent influenza vaccination. This is in agreement with national data, which reveal that only 26% of adults aged years who are at high risk for influenza complications have received a recent influenza vaccination [4]. This is troubling, because antiviral therapy is, at best, an adjunct strategy; influenza vaccination remains the keystone of influenza prevention and control. Even if patients receive a correct diagnosis within 48 h after onset, patients will not benefit from therapy unless they are willing to pay for it. We found that only approximately onethird of patients are willing to pay 1$20 for antiviral medication, which is a common level of copayment for brand-only medications. Even fewer patients (!10%) are willing to pay the retail cost of the medication ( $80 for a 5-day course [24]). It is unclear if the expense of antiviral medication contributes to the dearth of patients at high risk for influenza complications who are treated with antiviral drugs; in our study, patients at high risk for influenza complications were not more willing than other patients to pay for antiviral medication. Physician attitudes are also a likely factor in the relatively low use of antiviral drugs; a recent study showed that 40% of primary care physicians believe that influenza is self-limited and does not require treatment [17]. In addition, our study demonstrates that, regardless of their knowledge level, a majority of patients are worried about shortages of both antiviral medication and influenza vaccine. A small number of patients even reported hoarding antiviral drugs. There is a relative lack of data regarding how patients think and behave regarding such shortages (whether real or perceived). One study found a spike in antiviral medication sales in New York City in the fall of 2006, before the influenza season began [25]; another study found that almost one-half of infectious diseases consultants had been asked by family or friends for a neuraminidase inhibitor prescription specifically for stockpiling purposes [26]. The motivations for such behavior are 1186 CID 2007:45 (1 November) Gaglia et al.

6 Table 2. Knowledge and attitudes regarding antiviral medication. No. (%) of patients Question (correct response) True False Do not know Antiviral medicines work for a cold (false) 37 (14) 116 (44) 109 (42) Antiviral medicines only work if you take them in the first 48 h of the flu (true) 79 (30) 37 (14) 147 (56) Antiviral medicines can also be taken to prevent the flu (true) 63 (24) 89 (35) 106 (41) Antiviral medicines decrease the amount of time you are sick with the flu (true) 147 (57) 18 (7) 95 (37) Antiviral medicines can sometimes cause the flu (false) 49 (19) 102 (39) 111 (42) Antiviral medicine usually has side effects, like nausea or dizziness (false) 79 (30) 23 (9) 158 (61) Antiviral medicines are available over the counter (false) 16 (6) 152 (58) 93 (36) Antiviral medicines also work against bacteria (false) 14 (5) 151 (58) 95 (37) Strongly agree/agree Neutral Strongly disagree/ disagree The flu shot protects you from the flu 179 (69) 47 (18) 34 (13) The flu shot can cause bad side effects, like fever and rash 125 (49) 84 (33) 48 (19) The flu shot can cause the flu 96 (37) 65 (25) 96 (37) My own body can fight off the flu without help from medicine 96 (37) 82 (32) 80 (31) I am worried there is not enough antiviral medicine in this country for everyone 133 (52) 77 (30) 46 (18) I am worried there is not enough flu vaccine in this country for everyone 131 (51) 77 (30) 48 (19) unclear, although fears of a future influenza pandemic are likely to be a key factor. Our survey has several limitations. First, we used an unvalidated instrument, because no validated instruments to measure antiviral knowledge or attitudes were available. We did, however, test the instrument for face validity and content prior to administration. Second, although the questions appear to capture straightforward attitudes and beliefs, patient knowledge and attitudes are not static and could vary temporally on the basis of future events regarding seasonal and pandemic influenza. Indeed, this is evidenced by the recent emergence of neuraminidase-resistant strains of both human and avian influenza [27 29]. Any beliefs or hoarding behavior could certainly change if drug resistance became more widespread. Similarly, the actual behavior of respondents might differ from the behavior reported in the survey. Third, our study was limited to only 1 specific population in a clinic; one would anticipate even lower knowledge scores, however, if this survey were applied to a population with lower education and socioeconomic status. Future studies of this type would benefit from the use of a broader sample of geographic areas and clinic types. Fourth, we did not have data available for patients who were given the survey but did not return it; this could have possibly introduced bias. Our results raise 2 issues: how to identify patients more effectively before the 48-h treatment window expires and how to manage the increased volume of patients with nonspecific viral symptoms that would result. Diagnosis of influenza remains difficult [16, 30], and rapid testing has not gained universal acceptance [17]; therefore, practical strategies for both screening and diagnosis are sorely needed. Future studies should focus on educational interventions for patients and physicians, emphasizing the benefits of antiviral medication for patients at high risk for influenza complications and the importance of early presentation. We also found that the false belief that influenza vaccination causes influenza was associated with lower knowledge of antiviral drugs; this suggests that education should emphasize the importance of influenza vaccination, as well. Studies are also needed to identify the best way to triage patients with influenza-like symptoms who might benefit from antiviral treatment, to avoid overburdening the health care system. Lastly, studies of fears regarding antiviral medication shortages and hoarding of antiviral medications are also lacking. Despite evidence that antiviral medications can decrease the morbidity associated with influenza, physicians have been slow to adopt their use. Our study provides a preliminary understanding of patient perceptions and knowledge regarding antiviral medications and their relation to beliefs regarding influenza and influenza vaccination. We found multiple barriers to optimal detection and treatment of influenza, including poor knowledge of influenza in general, a low percentage of patients reporting that they would call within the first 48 h of illness, and a lack of willingness to pay for antiviral medication. This gap between patient and physician must be bridged with improved patient (and physician) education, more efficient efforts in triage and diagnosis, and a focus on patients who are at high risk for influenza and its complications. The challenge influenza poses to the individual physician demands a paradigm shift in the clinical approach to viral illness. Patients at high risk for Knowledge and Attitudes about Antivirals CID 2007:45 (1 November) 1187

7 influenza and its complications should be viewed with more urgency when they present with influenza-like symptoms, and heath care systems must redouble their efforts to increase influenza vaccination rates in patients who are at high risk for complications from influenza. Acknowledgments We thank Deborah Naglieri-Prescod, for her help with formatting the survey, and Cecelia Stafford, for help distributing the surveys. Financial support. M.A.G. s research is supported in part by a Health Resources and Services Administration training grant in primary care research. Potential conflicts of interest. All authors: no conflicts. References 1. Nicholson KG, Wood JM, Zambon M. Influenza. Lancet 2003; 362: Thompson WW, Shay DK, Weintraub E, et al. Mortality associated with influenza and respiratory syncytial virus in the United States. JAMA 2003; 289: Thompson WW, Shay DK, Weintraub E, et al. Influenza-associated hospitalizations in the United States. JAMA 2004; 292: Centers for Disease Control and Prevention. Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Mortal Wkly Rep 2006; 55: Centers for Disease Control and Prevention. High levels of adamantine resistance among influenza A (H3N2) viruses and interim guidelines for use of antiviral agents United States, influenza season. MMWR Morb Mortal Wkly Rep 2006; 55: The writing committee of the World Health Organization consultation on influenza A/H5. Avian influenza A (H5N1) infections in humans. N Engl J Med 2005; 353: Management of Influenza in the Southern Hemisphere Trialists Study Group. Randomised trial of efficacy and safety of inhaled zanamivir in treatment of influenza A and B virus infections. Lancet 1998; 352: Treanor JJ, Hayden FG, Vrooman PS, et al. Efficacy and safety of the oral neuraminidase inhibitor oseltamivir in treating acute influenza: a randomized controlled trial. JAMA 2000; 283: Kaiser L, Keene ON, Hammond JM, et al. Impact of zanamivir on antibiotic use for respiratory events following acute influenza in adolescents and adults. Arch Intern Med 2000; 160: Lalezari J, Campion K, Keene O, Silagy C. Zanamivir for the treatment of influenza A and B infection in high-risk patients: a pooled analysis of randomized controlled trials. Arch Intern Med 2001; 161: Kaiser L, Wat C, Mills T, et al. Impact of oseltamivir treatment on influenza-related lower respiratory tract complications and hospitalizations. Arch Intern Med 2003; 163: Lee PY, Matchar DB, Clements DA, et al. Economic analysis of influenza vaccination and antiviral treatment for healthy working adults. Ann Intern Med 2002; 137: Smith KJ, Roberts MS. Cost-effectiveness of newer treatment strategies for influenza. Am J Med 2002; 113: Rothberg MB, Bellantonio S, Rose DN. Management of influenza in adults older than 65 years of age: cost-effectiveness of rapid testing and antiviral therapy. Ann Intern Med 2003; 139: Rothberg MB, Rose DN. Vaccination versus treatment of influenza in working adults: a cost-effectiveness analysis. Am J Med 2005; 118: Stein J, Louie J, Flanders S, et al. Performance characteristics of clinical diagnosis, a clinical decision rule, and a rapid influenza test in the detection of influenza infection in a community sample of adults. Ann Emerg Med 2005; 46: Rothberg MB, Bonner AB, Rajab MH, et al. Effects of local variation, specialty, and beliefs on antiviral prescribing for influenza. Clin Infect Dis 2006; 42: Linder JA, Chan JC, Bates DW. Appropriateness of antiviral prescribing for influenza in primary care: a retrospective analysis. J Clin Pharm Ther 2006; 31: Linder JA, Bates DW, Platt R. Antivirals and antibiotics for influenza in the United States, Pharmacoepidemiol Drug Saf 2005; 14: Centers for Disease Control and Prevention. Influenza vaccination and self-reported reasons for not receiving influenza vaccination among Medicare beneficiaries aged 65 years United States, MMWR Morb Mortal Wkly Rep 2004; 53: Belongia EA, Naimi TS, Gale CM, Besser RE. Antibiotic use and upper respiratory infections: a survey of knowledge, attitudes, and experience in Wisconsin and Minnesota. Prev Med 2002; 34: Cummings KC, Rosenberg J, Vugia DJ. Beliefs about appropriate antibacterial therapy, California. Emerg Infect Dis 2005; 11: Wilson AA, Crane LA, Barrett PH, Gonzales R. Public beliefs and use of antibiotics for acute respiratory illness. J Gen Intern Med 1999; 14: Retail price for Tamiflu. Available at: Accessed 21 February Centers for Disease Control and Prevention. Increased antiviral medication sales before the influenza season New York City. MMWR Morb Mortal Wkly Rep 2006; 55: Ortiz JR, Shay DK, Liedtke LA, et al. A national survey of the Infectious Diseases Society of America emerging infections network concerning neuraminidase inhibitor prescription practices and pandemic influenza preparations. Clin Infect Dis 2006; 43: Le QM, Kiso M, Someya K, et al. Isolation of a drug resistant H5N1 virus. Nature 2005; 437: Kiso M, Mitamura K, Sakai-Tagawa Y, et al. Resistant influenza A viruses in children treated with oseltamivir: descriptive study. Lancet 2004; 364: Hatakeyama S, Sugaya N, Ito M, et al. Emergence of influenza B viruses with reduced sensitivity to neuraminidase inhibitors. JAMA 2007; 297: Call SA, Vollenweider MA, Hornung CA, et al. Does this patient have influenza? JAMA 2005; 293: CID 2007:45 (1 November) Gaglia et al.