Evaluating the Use of Oseltamivir for the Treatment of Influenza in Adults

Number of participants who had undetectable values (less than the limit of detection [LOD]), who had values between the LOD and the lower limit of quantification (LLOQ), and who had values ≥LLOQ

Median, 25% and 75% percentile of the value of viral shedding (Results <LOD were imputed as the LOD value, and Results >= LOD, <LLOQ were imputed as the LLOQ value.)

Number of participants with undetectable viral load at both Day 3 and Day 7; detectable at Day 3 and undetectable at Day 7; detectable at Day 7 (irrespective of whether or not detectable at Day 3).

The assessed symptoms were cough, nasal obstruction (stuffy nose), sore throat, fatigue, headache, muscle aches, feverishness, rhinorrhea, nausea, vomiting, diarrhea. Duration of clinical symptoms was defined as the time from Day 0 to the first of two successive measurements at which all clinical symptoms were grade 0 (absent) or 1 (mild). A measurement was considered to be the 8AM or 8PM assessment during Days 0 to 7 (so two measurements were obtained per day) and then the daily assessment thereafter. Time was calculated in half-days through to Day 7. If a subject's first two assessments on (baseline assessment and first subsequent diary card assessment) satisfied this criterion, then the duration was set to zero. For participants who did not have two successive records meeting this criterion, follow-up was censored for analysis purposes at the time of the last but one diary card record with symptoms evaluated.

Fever was considered present based on the diary cards if a subject reported a maximal temperature ≥38.0°C (for the period since the diary card was previously completed) or reported having taken an antipyretic drug (also for the period since the diary card was previously completed). Otherwise, fever was considered not present during the period since the diary card was previously completed, except that the evaluation was considered missing if either the temperature or the antipyretic drug use entry was not completed on the diary card. The duration of fever was defined as the time from Day 0 to the first of two successive assessments (through to Day 7) or to the first assessment (Day 8 onwards) at which no fever was present according to this definition.For participants who did not have two successive records meeting this criterion, follow-up was censored for analysis purposes at the time of the last but one diary card record with fever evaluated.

The assessed symptoms were cough, nasal obstruction (stuffy nose), sore throat, fatigue, headache, muscle aches, feverishness, rhinorrhea, nausea, vomiting, diarrhea. Fever was considered present based on the diary cards if a subject reported a maximal temperature ≥38.0°C (for the period since the diary card was previously completed) or reported having taken an antipyretic drug (also for the period since the diary card was previously completed). Time to resolution of all clinical symptoms and fever is defined as the time from Day 0 to the first of two successive measurements at which all clinical symptoms are grade 0 (absent) or 1(mild) and no fever >=38.0 C or antipyretic drug is reported. For participants who did not have two successive records meeting this criterion, follow-up was censored for analysis purposes at the time of the last but one diary card record with symptoms and fever evaluated.

Time to feeling as good as before influenza is defined as time to the first of two successive 'yes' responses to the question of 'feeling as good as you did before you had the flu'.For participants who did not have two successive records meeting this criterion, follow-up was censored for analysis purposes at the time of the last but one diary card record with question answered.

Time to return to pre-influenza function is defined as the time from Day 0 to the first of two successive 'Yes' answers to the global assessment question 'Are you functioning as well as you were before you had the flu'.For participants who did not have two successive records meeting this criterion, follow-up was censored for analysis purposes at the time of the last but one diary card record with question answered.

Time to return of physical function to pre-illness level was defined as the time from Day 0 to the first of two successive measurements at which the physical function score equals or is better than the pre-illness score (obtained by recall at enrollment). For subjects who did not have two successive records meeting this criterion, follow-up was censored for analysis purposes at the time of the last but one diary card record with physical function evaluated.

For each of the 5 days of treatment, participants were asked whether they took all study drug for that day. All participants were assumed to have taken at least some study drug even if they had zero days with all study drug reported as taken. Missing reports for some or all days were imputed as not having taken all study drug for the days concerned.

The percentage of participants hospitalized by 28 days was constructed by inverting an exact binomial test of the actual percentages (ignoring loss to follow-up).

Participants were assessed for the signs/symptoms suggestive of one of the following complications: Sinusitis, Otitis Media ,Bronchitis / Bronchiolitis, Pneumonia and antibiotic use for reason other than above.

Number of deaths

For participants with a positive influenza test result at Day 0 from qualitative PCR testing on the team collected sample, the laboratory then performed qPCR testing of self-collected samples to quantify viral shedding.

Number of participants who had undetectable values (less than the limit of detection [LOD]), who had values between the LOD and the lower limit of quantification (LLOQ), and who had values ≥LLOQ. Evening samples on Day 0, 1 and 2 were only required under protocol versions 1.0-4.0 and so were only collected for about 20% of participants.

Median, 25% and 75% percentile of the value of viral shedding (Results <LOD were imputed as the LOD value, and Results >= LOD, <LLOQ were imputed as the LLOQ value.). Evening samples on Day 0, 1 and 2 were only required under protocol versions 1.0-4.0 and so were only collected for about 20% of participants.

This AUC was calculated using the trapezoidal rule and the units of measurement are (days*log10 copies/mL) with the fact that it was the level of virus above the LLOQ being considered. The calculation of AUC was undertaken using measurements from Day 0 to Day 3 which were collected under all versions of the protocol (i.e. evening measurements on Days 0, 1 and 2 were not used as these were collected for only about 20% of subjects). Missing values during follow-up were ignored. This is equivalent to imputing a missing value using linear interpolation between the preceding and succeeding available values. For the five subjects with missing values following a last available measurement which was above the LLOQ, the remaining values were assumed to be the mean of preceding value and LLOQ in calculating the AUC.