Oseltamivir is a neuraminidase inhibitor that inhibits influenza virus proliferation, and is used as an antiviral drug against influenza A and B viruses.1 It is known to be effective for reducing the duration of illness in healthy children by about 1 day.2 However, there have been postmarketing reports, mostly from Japan, of delirium and abnormal activities in patients with influenza who took oseltamivir.3,4 These side effects are described in the oseltamivir package insert. Reported side effects have also included sudden death5 leading to widespread concern.6 As a consequence, the US Food and Drug Administration (FDA) warns that influenza patients taking oseltamivir should be closely monitored for signs of abnormal behaviors.3 In March 2007, the Japanese Ministry of Health, Labour and Welfare prohibited the prescription of oseltamivir to children and adolescents aged 10-19 years.7
The Japanese Ministry funded 2 epidemiologic studies in the winters of 2005-2006 and 2006-2007 to assess the relationship between oseltamivir intake and adverse behaviors. The first study included 2846 children, and compared the incidence proportions of adverse behaviors between children who took oseltamivir and those who did not. Oseltamivir was associated with adverse behaviors within the first day of infection.5 However, the authors concluded that “a large study should be conducted next year.”8
Subsequently, a larger observational cohort study including about 10,000 children was conducted in the 2006-2007 winter season. In an interim report released on 10 July 2008, the research group concluded that no positive associations were detected between oseltamivir intake and abnormal behaviors.9 However, the analytic method used in the second study was flawed. Hama pointed out a mistake using intention-to-treat analysis, and in a reanalysis, showed a positive relationship.5,10 Here, we introduce the original study in more detail and describe appropriate analytic methods to account for person-time at risk.
In the second and larger study, sets of 10-20 successive patients infected with influenza virus were recruited from 692 institutions, for a total of 11,661 patients. Data regarding the prescription and intake of the drug, as well as adverse behaviors, were obtained from questionnaires completed by both the doctors and the patients' families.11 Families recorded adverse behaviors during the first 4 days from the onset of fever, and the questionnaires were submitted to their doctors. Doctors evaluated these questionnaires and completed their own questionnaires. The adverse behaviors consisted of hallucinations, delirious speech, frightening episodes, abrupt anger, abnormal activities leading to accidents, and putting anything unusual into the mouth. The accuracy of adverse behaviors in the doctors' questionnaires was checked by pediatricians in the research group who were unaware of prescription status. Patients were excluded if both questionnaires were not obtained, if there were missing data, and if the infection was not confirmed by diagnostic kit. This left 10,316 patients. Further exclusions were patients aged 18 or older (21 patients) and those who lacked information about adverse behaviors (278 patients) leaving 10,017 patients. The number of excluded patients in the treated and untreated groups could not be obtained from the interim report.
Finally, 227 patients (3%) among 7813 patients who were prescribed oseltamivir and 75 patients (3%) among 2204 patients who were not prescribed oseltamivir were excluded because they manifested adverse behaviors before the index doctor visit. Consequently, 9715 patients were analyzed, and odds ratios between treatment with oseltamivir and adverse behaviors were estimated. In this report, we show risk ratios and 95% confidence intervals (CIs) using Epi Info Ver. 3.4.1.
A flow chart of the patients is shown in Figure 1. Oseltamivir was prescribed to 7586 patients and 988 (13%) manifested adverse behaviors. Among the 2129 subjects who did not receive oseltamivir, 187 (9%) manifested adverse behaviors. These data produce a simple risk ratio of 1.48 (95% CI = 1.28-1.72), indicating a positive association between treatment with oseltamivir and adverse behaviors. However, there were 99 patients whose adverse behaviors preceded their treatment with oseltamivir. The research group moved these subjects from the prescribed group to the nonprescribed group. They subtracted these 99 subjects from the numerator and denominator of the prescribed group (leaving 889 and 7487), and added them to the numerator and denominator of the nonprescribed group (making 286 and 2228). The results changed as follows: the incidence of adverse behaviors was 12% among the prescribed group and 13% among the nonprescribed group, leading to a risk ratio of 0.93 (0.82–1.05).
Their analysis overlooked the fact that the 99 incidents of adverse behaviors before starting oseltamivir occurred in a population at risk that—at least for the brief time period before treatment—should also have been included in the untreated group. Since patients were followed for 4 days from the onset of fever, the maximum possible person-time would be 30,344 person-days in the prescribed group and 8516 person-days in the nonprescribed group. However, the person-days in the prescribed group consist of 2 parts, namely the person-days before oseltamivir intake (area A in Fig. 2) and the person-days after oseltamivir intake (area B in Fig. 2). In the nonprescribed group, 187 subjects manifested adverse behaviors from 8516 person-days, which is 0.022 incidences per person-day. Assuming that this incidence is homogeneous over the 4 days; we can assume that the rate of adverse behaviors in the nonprescribed group is equal to the corresponding rate in the prescribed group before oseltamivir intake (A in Fig. 2). From this, we can estimate the number of person-days in A by calculating 99 incidences divided by 0.022, leading to 4500 person-days. This leaves 25,844 person-days after oseltamivir intake (B in Fig. 2).
Using these figures, there are 2 analytic options described by Rothman12: exclusion of the patients manifesting adverse behaviors and their corresponding person-days before oseltamivir intake in the prescribed group, (A in Fig. 2) or reassignment of these to the nonprescribed group. With the first method, the rate of adverse behaviors in the prescribed group is 0.034 (889 incidences divided by 25,844 person-days) and the corresponding rate in the nonprescribed group is 0.022 as described above, leading to a rate ratio of 1.57 (95% CI = 1.34-1.83). With the second method, the rate of adverse behaviors in the prescribed group is 0.034 (as in the first method) and the corresponding rate in the nonprescribed group is still 0.022 (187 plus 99 incidences divided by 8516 plus 4500 person-days), leading again to a rate ratio of 1.57 (1.37-1.79). Thus, using either method, it is clear that when children are infected with influenza, oseltamivir intake increases the rate of adverse behaviors by 1.5-fold compared with nonusers. (The 95% CIs for the rate ratios were estimated using the formula provided in Rothman.12)
The second study was an observational cohort study; we can show only the crude rate ratios from the interim report. Thus, we cannot eliminate a possibility of spurious association due to potential confounders (age, fever, other drugs, etc). However, our reanalysis suggests positive associations between oseltamivir and adverse behaviors consistent with the first study. According to the oseltamivir package insert, the time to maximum plasma concentration is 4-5 hours. Thus, if the analysis were restricted to 1 day or half a day after the first intake, the risk estimates could be even higher.
Overall, 75% of oseltamivir in the world is consumed in Japan,13 making Japan an appropriate place for assessing the side effects of oseltamivir. As described by Weiss,14 randomized trials cannot contribute as much as postmarketing observational studies to our understanding of the adverse effects of a given therapy. Although a side effect of vomiting had been reported in randomized trials of oseltamivir,15 less common side effects may be identifiable only in postmarketing observational studies. The 2 epidemiologic studies conducted in Japan are of great value and the data should be analyzed appropriately, focusing on 1 day or half a day after the first intake.
We thank our colleagues Hiroyuki Doi and Soshi Takao.
2. Matheson NJ, Harnden AR, Perera R, Sheikh A, Symmonds-Abrahams M. Neuraminidase inhibitors for preventing and treating influenza in children. Cochrane Database Syst Rev.
4. Ministry of Health, Labour and Welfare. [Relevant documents for abnormal behavior No. 1-1-1] (in Japanese) Working group for clinical investigation for oseltamivir phosphate Available at: http://www.mhlw.go.jp/shingi/2007/05/s0514-2.html
. Accessed July 23, 2008.
5. Hama R. Fatal neuropsychiatric adverse reactions to oseltamivir: Case series and overview of causal relationships. Int J Risk Safety Med.
10. Hama R. Serious misclassification in Japanese MHLW epidemiologic study on Oseltamivir (August 8, 2008) (response to letter). BMJ.
11. Ministry of Health, Labour and Welfare. Relevant documents for investigations about the frequency of associated symptoms with influenza No. 1-5-2 [in Japanese] Working group for clinical investigation for oseltamivir phosphate Available at: http://www.mhlw.go.jp/shingi/2007/05/s0514-2.html
. Accessed July 23, 2008.
12. Rothman KJ. Epidemiology: An Introduction
. New York: Oxford University Press; 2002.
13. Satoh K, Nonaka R, Ogata A, Nakae D, Uehara S. Effects of oseltamivir phosphate (Tamiflu) and its metabolite (GS4071) on monoamine neurotransmission in the rat brain. Biol Pharm Bull.
14. Weiss NS. Clinical Epidemiology: The Study of the Outcome of Illness
. 3rd ed. New York: Oxford University Press; 2006.
15. Jones M, Del Mar C. Safety of neuraminidase inhibitors for influenza. Expert Opin Drug Saf.