Background In preparation for a global influenza pandemic, there is an urgent need for representative data from populations and settings where the pandemic is most likely to arise. There are no data on oseltamivir efficacy from Asian urban slum populations concerning duration of illness and viral shedding, nor whether efficacy depends on starting treatment < 48 hours or ≥ 48 hours after illness onset. Finally, there are no data on the capacity of the drug, in such settings, to affect household and community transmission rates.

Aims and Objectives This proposal aims to compare the duration of clinical illness among patients treated with oseltamivir vs placebo < 48 hours and ≥ 48 hours after illness onset. It will compare the duration of viral shedding among all treatment groups vs placebo, risk of transmission to household contacts by treatment group and whether neuraminidase inhibitor use creates resistance. Secondarily it aims to measure the effect on influenza.

Design and Methods A double-blind placebo controlled clinical trial design among a population in an urban slum under current influenza disease burden surveillance will be enrolled. Infection status will be confirmed by rRT-PCR. Patients ≥ 1 year old will be randomised to < 48 hour and ≥ 48 hour treatment arms. Family members and neighbours will also be assessed by PCR and a basic reproductive number calculated (R0).

Relevance These findings will address whether oseltamivir can affect illness duration and severity, affect transmission, incidence and resistance in high risk urban Asian settings where a pandemic is most likely to arise.

Purpose Influenza is a disease of global importance, having caused three pandemics in the 20th Century. Although concerns persist about a new pandemic, possibly from an avian influenza A strain, more people died during the 20th Century from seasonal epidemic influenza than from any single pandemic, thus global preparedness must address both epidemic and pandemic influenza. It is generally believed that if a pandemic emerges, an efficacious vaccine will either not be either generally available or broadly protective. Additional strategies are required for effective control. Neuraminidase inhibitors, including oseltamivir, have shown efficacy in limited controlled trials against both human influenza, and have been used in avian influenza cases. Numerous questions, however, persist about the extent of their efficacy. These include whether or not they are effective if given after 48 hours post-symptom onset, whether they reduce the duration and titre of viral shedding, their effect on transmission to household contacts, and how quickly - or even if - resistance will emerge in a high endemic setting during seasonal use. Finally, clinical trials to date have used small samples sizes under controlled settings in industrialized countries. There are no data on the efficacy of neuraminidase inhibitors in over-crowded urban settings with rates of influenza and other respiratory infections, like Dhaka, Bangladesh. The findings from this study will enable better assessment of the performance of neuraminidase inhibitors under the conditions similar to those from which a global pandemic is likely to occur.

This study will evaluate whether oseltamivir is effective at reducing illness and household transmission during the seasonal influenza epidemic in a crowded urban setting in Dhaka, Bangladesh.

Design/Methods This will be a double-blind placebo randomised controlled clinical trial that will identify study subjects in clinic who present with signs/symptoms suggestive of influenza. Patients will be screened using a commercial rapid diagnostic test of high sensitivity and specificity (QuickVue A + B, Quidel, Inc., San Diego, CA, USA), and their status will be confirmed using RT-PCR. Rapid test-positive patients will be randomised to oseltamivir or placebo for the standard twice daily, five-day course. A total of 512 PCR-confirmed patients will be recruited and treated. Patients will provide nasopharyngeal specimens on the day of presentation (day 0), and on days 2, 4 and 7 to determine duration of viral shedding. All patients will be followed up at home daily by field research assistants (FRAs), who will monitor their progress using standardised forms, and refer back to clinic anyone who meets criteria for treatment failure. Once a patient meets criteria for recovery, they will be referred back to clinic for an exit interview with the physician. At the end of the study, comparisons will be made between groups on duration of illness, duration of viral shedding and household and inter-household transmission rates, as well as whether there were differences between those who started oseltamivir < 48 hour or ≥ 48 hours from the onset of illness.

1. Requirements for subject population, rationale for use: The influenza virus, once suspected to primarily attack older persons, is now recognised to cause more illness in children, particularly those < 5 years. Indeed, both hospitalisation rates, serious complications, and even mortality among healthy children < 5 years approach those of older persons with chronic conditions that place them at high risk [1]. Children are also the most potent spreaders of influenza infection in the community [2]. However, in order to model the impact of neuraminidase inhibitors on duration of illness, and especially on intra and inter-household transmission, the entire age spectrum must be examined, as in the early stages of an epidemic, adults may be among the first to show infection signs and seek treatment, even though children will be more important in the spread of the epidemic in its early stages [3].
2. Potential risks, likelihood, seriousness and impact of methodology: The risk is greater than minimal, in that oseltamivir is not routinely prescribed for influenza infection in Bangladesh, and as with any licensed drug, there are potential side effects, the most common being gastrointestinal upset, nausea and vomiting - all of which are rare but reported, as well as sleep or behaviour disturbances, which are even less common and not causally linked to oseltamivir. Likewise, nasopharyngeal washes and swabs are not routinely collected in Bangladesh, and cause minor discomfort during the procedure, however complications, primarily from nasopharyngeal swabs, such as nosebleeds, are also uncommon. There are no alternatives to these procedures for these samples, which are needed to document whether a person had influenza and for how long.
3. Procedures for minimising risks, effectiveness assessments: 1) All physicians have been trained on the procedures for specimen collection, as this has been ongoing in Kamalapur since 2004. A refresher course will be provided to all study physicians prior to the new study to ensure correct procedure. 2) All physicians will be trained about potential side effects, and will relate these to the patients and their caregivers, as they currently do with any medication. 3) FRAs will be trained to look for signs of illness as well as complications to the medication during their daily home visits, and will refer back to clinic any patient with signs suggestive of complication (e.g. skin rash, repeated vomiting).
4. Confidentiality and anonymity: All patient information, which is recorded on a case report form, (CRF) will be kept in a locked cabinet. Only project staff will see it. Patients' name and other potential identifiers will be removed from data shared in public fora. Published data will be aggregated and anonymised.
5. Description of consent procedures: One consent form will be for household consent for participation in weekly home visits for influenza surveillance, and collection of samples for rapid diagnostics (nasal swab) and nasopharyngeal wash or nasopharyngeal swab specimens for index cases, and for subsequent household suspected cases (Household consent, Appendix 1), This form will be obtained by the field research assistant (FRA), as are all surveillance forms used in Kamalapur. This is the same format as the current influenza surveillance for children < 5 years of age. The FRA will read the consent form in its entirety to any responsible adults (those with legal responsibility for the children in the home, or if there are no children, for home itself) in the home, and answer their questions using standardised responses (key points). The second consent form will obtain consent for the index case to receive antiviral or placebo administration, and to provide subsequent NPW or NPS specimens on days 2, 4 and 7 (Oseltamivir Study, Appendix 2). A study physician will obtain this consent form in clinic. If the subject (index case) is an adult (age ≥ 18 years), consent will be obtained from the subject himself/herself; if the subject is a child, consent will be sought from his/her parents or legal guardian. The Oseltamivir study consent form readability score is Flesh-Kincaid Grade level = 7.8. An Assent form will be read to children aged 7-14 years. The Assent form readability score is Flesh-Kincaid Grade level = 5.6. The methodology of obtaining consent for both forms will be standardised. The FRA using the home visits consent form, or the study physician using the antiviral administration form, will read the entire consent form to the patient or parent if the patient is a minor, as this is an illiterate population. If parents have questions, the study physician will provide clarification of the consent form from a standardised set of key points that cover each section of the consent form. Patients, or parents if the patient is a minor, who indicate that they understand and agree to the terms of the study will be asked to provide written consent. If they cannot sign the form, a thumb impression will be taken.
6. Interviews: The only interviews will be those conducted in the home by the FRA to either screen for illness or follow-up ill patients at home, and will focus on illness signs using standardised screening calendar questionnaires or follow-up questionnaires. The other interviews occur in clinic with the study physician, and again focus on history of present illness, past medical history, physical exam, follow-up and an exit interview, as described above. These physician-patient interviews are fundamentally the same as between any physician and patient, except that they are standardised.
7. Potential benefit to individual and society, risk/benefit: Patients may receive direct benefit if they receive the treatment drug, including a mild reduction in the duration of influenza symptoms. In addition, household members of persons treated with oseltamivir may experience fewer influenza illnesses, if transmission is reduced. Patients will receive routine medical care regardless of treatment group, even if they refuse to participate in the study. The daily home follow-up by the trained FRA is another benefit, as she can help to identify any signs indicating the need for medical attention. The information collected will be valuable to inform future pandemic planning in Bangladesh and globally, as there is no information available for this regimen from the developing world. The potential and real benefits thus outweigh the potential and real risks.
8. Drug status: Oseltamivir (Tamiflu®) is not an experimental drug, but is US Federal Drug Administration (FDA) approved, having passed Good Clinical Practices licensing trials, under the terms for use as in this trial. Furthermore, oseltamivir is manufactured by Roche Pharmaceuticals, which has a branch in Bangladesh, which currently supplies local pharmacies with the drug for sale.
9. Experimental new drugs: N/A
10. Placebo, indications for use in the study: In order to objectively demonstrate the efficacy of the drug on duration of illness specifically, but household transmission more generally, both patient and observer have to blinded to the intervention, as knowledge of intervention assignment can affect recorded outcome (observational bias). Although it is known that oseltamivir affects illness duration under controlled settings and in settings with lower disease burden than those documented in Dhaka, Bangladesh, it is not known how effective it will be under these conditions, nor is it known whether it will work for patients who present later in the illness (≥48 hours). For this reason, patients will be allocated to a study arm by pre-allotment using block randomisation, along with the blinding by use of a placebo. No patient will be denied standard of care, as any focal illness, such as pneumonia, bronchitis, bronchiolitis, otitis media, sinusitis, or other common complications of influenza and similar pathogens, will be treated with standard, WHO-recommended or approved regimens.
11. Experimental 'new' drug sponsorship: N/A
12. Use of records, organs, tissues, body fluids, and faetus/abortus: The study will acquire nasopharyngeal wash and/or swab specimens, the objective of which is to collect nasopharyngeal epithelial cells, which will be used in cell culture to isolate influenza virus. The purpose of isolating the influenza virus is to validate whether or not a patient actually had influenza, and if so, for how long was it shed.

Inclusion Criteria:

• Persons at least one year old residing in randomly selected households with at least one major illness sign, or if absent at least two minor illness signs who are rapid test positive for either influenza A or influenza B.

Exclusion Criteria:

• Persons with a history of non-febrile convulsions or
• Persons who are taking anticonvulsive agents, or
• Persons who have a nonrespiratory comorbid condition requiring immediate medical intervention, or
• Persons who are pregnant.