The next influenza pandemic is expected to spread rapidly in resource-poor settings. Influenza viruses spread from human-to-human via large respiratory droplets. Transmission via large-particle respiratory droplets is believed to be mediated by close contact between infected and susceptible persons or contact with droplet-contaminated fomites. Close contact between infected and susceptible persons may consist of skin-to-skin contact (e.g., via hands) or inhalation of respiratory droplets (e.g., due to talking, coughing, or sneezing by the infected person). Airborne transmission, which is expected to result in transmission over long distances (>1 meter) and which would be mediated by ventilation, is believed to be uncommon. Therefore, the greatest risk of transmission from personal contact comes from those people who are closest to an index case, such as contacts living in the same household. There are, to date, no published estimates of the secondary attack ratio of influenza among household contacts of index case-patients in low-income countries. Moreover, the investigators do not have data on the risk factors for secondary transmission of influenza from index case-patients to their household contacts. There is some data for the benefits of promoting handwashing with soap on the risk of all-cause acute respiratory illness among children < 15 years old in a resource-poor setting in Pakistan. But, the investigators do not have evidence that promoting handwashing with soap will acutely reduce the risk of secondary transmission. Therefore, the investigators propose to conduct a study in rural Bangladesh to assess the following:

• The secondary attack ratio of influenza among household contacts of an index case-patient with influenza
• The risk factors for secondary transmission of influenza from an index case-patient to household contacts
• The impact of promoting handwashing with soap on the risk of secondary transmission of influenza from an index case-patient to household contacts
• The impact of handwashing promotion on handwashing behavior six months after intervention
• The impact of handwashing promotion on the prevalence of respiratory infections, diarrhea and influenza

BISTIS builds on hospital-based surveillance for Influenza virus infection, which is ongoing in hospitals around Bangladesh, as part of the Hospital-based Influenza Surveillance (HBIS) and Surveillance for the Epidemiology of Influenza in Bangladesh (SEIB) projects. We intend to recruit patients identified at the Jahurul Islam Medical College Hospital in Kishoregonj, Bangladesh, where both HBIS and SEIB are in place. In this hospital 80% of all the patients who present with influenza-like illness (ILI) to the outpatient departments of Medicine and Pediatrics are from three upazillas of Kishorgonj district: Bajitpur, Kuliar char and Kotiadi. The distances of these three upazillas are within 30 minutes travel time from Jahurul Islam Medical College Hospital (one way) and hence these upazillas will serve as the primary catchment areas for BISTIS. The table below illustrates the number of ILI cases identified through HBIS and SEIB study at Jahurul Islam Medical College Hospital in 2008 and also the number and proportion among them who were tested PCR positive for influenza virus.

n.b., most influenza-positive specimens were collected between May and September in 2007 and 2008

We will also enroll patients who present to two local upazilla health complexes (UHCs), one in Bajitpur and one in Kuliar Char. These local health complexes see numerous patients a day from the rural areas surrounding the clinic. Patients who present to these clinics are more likely to have symptom onset within 24 hours of presentation then those patients who seek care at JIMCH, since patients may only want to go to the hospital if their illness has been severe and prolonged over several days. Some studies on the impact of handwashing promotion have found that the intervention is effective if delivered within 36 hours of the index case-patient's symptom onset; therefore, the UHC sites are appropriate for enrollment of patients for BISTIS in addition to enrollment at JIMCH.

Lastly, we will recruit patients from pharmacies within the catchment areas mentioned above. By collaborating with local pharmacists, we intend on further increasing our enrollment numbers. Patients who present to local pharmacists are likely to have symptom onset within 24 hours of their visit to the pharmacist and are also more likely to be the primary case within their bari.

Specific Aim 1: To measure the secondary attack ratio (SAR) of influenza viruses among household contacts of index cases with influenza, in a rural setting in Bangladesh

Methods for Specific Aim 1

Specimen collection and processing A trained study physician will procure a nasal swab and an oropharyngeal swab from consenting index case-patients meeting the inclusion criteria above using a standardized method. At the Jahurul Islam Medical College Hospital, the nasal and oropharyngeal swabs will both be placed into a single tube containing viral transport media (VTM). An aliquot of the VTM will then be tested for Influenza A and B by the trained physician or nurse using a rapid antigen detection test (QuickVue® Influenza A + B). If the QuickVue® result is positive for Influenza A or Influenza B, the remaining VTM will be kept at 4°C. All VTM will be transported to the ICDDR,B virology laboratory in Dhaka on a weekly basis. At the ICDDR,B virology laboratory, RT-PCR testing for Influenza A (H1N1), Influenza A ( H3N2), and Influenza B will be carried out. If Influenza A H1N1 and A H3N2 are both negative, RT-PCR testing for Influenza A H5N1 will be performed. If the QuickVue® result is negative for both Influenza A and Influenza B, the patient will be informed of the result and thanked for participating in the study. The VTM will be discarded using appropriate infection control procedures, unless the patient is also participating in the ongoing HBIS or SEIB (in which case the remaining specimen will be processed like other surveillance specimens).

Enumeration of bari contacts and questionnaire administration Illness tracking among bari contacts Illness tracking will be carried out on each day for 10 days until after resolution of the index case-patient's illness. Resolution will be defined as the lack of fever, cough, and sore throat for at least 24 hours preceding the FRA's daily illness tracking visit. Thus, if the index case-patient's illness resolves on day 4 after enrollment, illness tracking will continue until day 14 after enrollment. The FRA will visit the patient's home and record information regarding the presence or absence of ILI and SARI symptoms in each household contact using an individual illness tracking form (appendix 9a8a and 9b8b - illness tracking form, ages ≥ 5 and < 5). For the purposes of screening household contacts for influenza, we will alter the age-specific case definition for children, based on data from urban Bangladesh, which demonstrates that fever and rhinorrhea are the most predictive factors for influenza in children < 5 years old (WA Brooks, ICDDR,B, personal communication).

If a household contact meets the age-specific case definition and does NOT have any danger signs, the FRA will obtain written informed consent for specimen collection (appendix 2- specimen collection from household contact for adult > 18 years old, (appendix 3 - specimen collection from household contact for child < 18 years old). The FRA will alert the medical officer, who will visit the home with the FRA no later than the following day in order to collect nasal and oropharyngeal swabs from the ill household contact. She will immediately place both swabs into VTM, which will then be placed into a cool box, containing ice and a thermometer to ensure temperatures < 40 C.

All specimens collected in the field will be placed in a cool box and transported to the ICDDR,B laboratory within 72 hours. At the ICDDR,B virology laboratory, all specimens for household contacts will be tested using RT-PCR for Influenza A (H1N1), A (H3N2), and Influenza B (and A (H5N1) if appropriate).

Illness tracking among household contacts will continue in each household until the 10th full day following the resolution of the index case-patient's symptoms, irrespective of whether any household contact develops illness or not.

Specific Aim 2: To test the efficacy of a handwashing promotion intervention for prevention of intrahousehold transmission of influenza virus

Methods for Specific Aim 2

To address this specific aim, we will conduct a randomized controlled trial. Households of index case-patients with influenza-like illness who are recruited at Jahurul Islam Medical College Hospital the UHCs or the local pharmacies will be randomized to the intervention group or the routine practices group.

The two groups will be defined as:

• Intervention Households: intensive promotion of handwashing with soap, and provision of facilitating tools, to the index-case-patient and all available household contacts
• Routine practices Households: continuation of the household's usual handwashing and respiratory hygiene practices group.

For promotion of handwashing with soap to intervention households, Field Intervention Specialists (FIS) will be trained to carry out a structured intervention that will follow constructs of Social Cognitive Theory (SCT). SCT addresses the reciprocal interaction between individuals, their environment, and health behaviors. Given that intervention will occur at the bari level, group-mediated constructs such as observational learning and reinforcements are highly relevant. FISs will visit the intervention households on a daily basis for 10 days after the resolution of the index case-patient's illness in order to encourage handwashing with soap at the recommended times. Routine practices households will also be exposed to the intervention, but only upon completion of the study. As noted above, under Specific Aim 1, an FRA will visit the home of the index case-patient daily for 10 days after the resolution of the index case-patient's symptoms in order to record age-specific case defining symptoms.

Specific aim 3: To identify risk factors, other than handwashing with soap, for intrahousehold transmission of influenza in a rural setting in Bangladesh.

Methods for Specific Aim 3

To address Specific Aim 3, we will conduct a nested cohort study to assess risk factors for intrahousehold transmission of influenza viruses. Here, the cohort under investigation is the routine practices group, as defined under Specific Aim 2. All data required to address this Specific Aim 3 will have been collected as part of the data collection described above under Specific Aim 1.

A case will be defined as: RT-PCR confirmed Influenza virus infection (A or B) in a household contact of an RT-PCR confirmed Influenza virus infection (A or B) index case-patient during 10 days of follow-up after resolution of the index case-patient's symptoms.

Specific aim 4: To assess whether exposure to the BISTIS intervention results in sustained improvements in handwashing behavior.

Methods for Specific Aim 4

To address Specific Aim 4, we will visit each bari that was enrolled in the intervention study 4 - 7 months after illness tracking is complete. The FRA will measure handwashing behavior at the bari. We will complete a structured observation of the bari's common handwashing behaviors. One or two months after the initial follow-up visit, the FRA will return to the bari and once again collect the same information on the handwashing behavior but will also provide soap to the bari. The FRA will collect the soap two days later. Data from the soap will be used to calculate the number of soap use events in the bari. In total an FRA will visit the household a total of three times, two visits for data collection and one visit to collect the soap.

Specific aim 5: To assess if exposure to the BISTIS intervention results in a reduced risk of respiratory infections, diarrhea, and influenza.

Methods for Specific Aim 5

The measurements of the health outcomes will be done in two different ways. At the first visit, after the handwashing behavior information is collected, the FRA will record whether each member of the bari has had symptoms of fever, cough, sore throat, difficulty breathing, respiratory illness or diarrhea in the previous 48 hours. At the third visit, in April 2010, the FRA will record mobile phone numbers of two or three bari members. The FRA will identify a key informant, who will be able to provide information regarding fever in any bari member. The FRA will phone the bari once each week during the influenza season and speak with the key informant once per week to assess whether any bari member has had fever during the previous 24 hours. If any member is reported to have a fever, we will dispatch an MO or lab/medical technician to the bari to obtain nasopharyngeal swab from that member for flu testing by PCR. A case will be defined as: RT-PCR confirmed Influenza virus infection (A or B).

Inclusion Criteria:

• Persons ≥ 5 years old: Influenza-like illness (ILI), defined as history of fever and either cough or sore throat with fever onset within the previous 24 hours
• Persons < 5 years old: any child with acute fever with onset within the previous 24 hours
• Return to home within 24 hours of presentation to Upazilla Health Complex, Jahurul Islam Medical College Hospital or the local pharmacies; i.e., the index case cannot be admitted for treatment. If admitted, the patient would not be eligible.
• No fever in any bari resident during the 7 days preceding the patient's presentation to hospital (see definition below)
• At least two persons (in addition to the index case-patient) who intend to reside in the bari during the subsequent 20 days
• Residence within 30 minutes travel time (one-way) from the Upazilla Health Complex or Jahurul Islam Medical College Hospital or the local pharmacy.