Governments around the world have spent billions of dollars stockpiling Tamiflu in case of an influenza pandemic. Tamiflu is on the World Health Organization's
list of "essential" drugs. The CDC recommends treatment with Tamiflu be considered in any patient who presents with flu-like symptoms regardless of confirmatory influenza testing. Given such endorsements, it's not surprising that
Tamiflu is widely prescribed in the U.S. during the winter months.
Most of the trials evaluating the effectiveness of Tamiflu are industry-sponsored and have never been published. In 2011, the Cochrane Collaboration requested all the study data pertaining to Tamiflu from Roche, the drug's manufacturer.
Roche balked, and over the next four years, a battle ensued that involved the British Medical Journal
, Roche, and the European Medicines Agency (the European version of the FDA). Finally, in 2013, Roche agreed to release the data
to the Cochrane Collaboration. The Cochrane Collaboration performed a meta-analysis on the data, and in 2014, the analysis was published.
- The Cochrane analysis came to the following conclusions:
- Tamiflu decreased the time to first alleviation of influenza-like illness symptoms in adults by an average of 16.76 hours. Tamiflu had no significant effect in children.
- Tamiflu significantly increased the incidence of nausea and vomiting
- Tamiflu had no significant effect on hospitalizations or incidence of pneumonia
- In prophylactic trials, Tamiflu reduced the incidence of symptomatic influenza. This finding may have been biased by the fact that Tamiflu appears to suppress influenza antibody production
and viral replication on test swabs. These tests were used to confirm influenza infection.
- See Cochrane Tamiflu meta-analysis for more
After the Cochrane findings, Roche decided to fund their own Tamiflu meta-analysis. To do so, they created a "foundation" called the "Multiparty Group for Advice on Science"
(MUGAS). MUGAS was headed by four people, two of whom received fees and funding from Roche. The Roche foundation even has its own website -
. Findings from the MUGAS meta-analysis are presented below.
- Inclusion criteria: all published and unpublished Roche-sponsored randomized placebo-controlled, double-blind trials of oseltamivir treatment in adult influenza
- Primary outcome (intention-to-treat infected): time to alleviation of all symptoms. Seven influenza symptoms (nasal congestion, sore throat, cough, aches and pains, fatigue, headaches,
and chills or sweats) were scored as absent, mild, moderate, or severe. Alleviation was defined to arise when all symptoms scored as absent or mild, and remained so for at least 21.5 h
- The researchers defined two groups for their analysis:
- Intention-to-treat population - entire intention-to-treat population
- Intention-to-treat infected population - defined as the intention-to-treat population in whom influenza infection was confirmed by positive nasal or throat culture or by a four-fold or greater
increase in antibody titer from baseline
- Population sizes (intention-to-treat): Tamiflu - 2402 patients; Placebo - 1926 patients
- Population sizes (intention-to-treat infected): Tamiflu - 1591 patients; Placebo - 1302 patients
- Primary outcome (intention-to-treat infected): The median times to alleviation of symptoms were 97.5 hours for oseltamivir and 122.7 hours for placebo groups (difference -25.2 hours, 95% CI -36.2 to -16.0)
- Primary outcome (intention-to-treat): difference in median time to symptom alleviation was -17.8 hours in favor of Tamiflu (95% CI -27.1 to -9.3)
- Admitted to hospital (intention-to-treat infected) - Tamiflu 0.68%, placebo 2.1% (RR 0.37, 95%CI [0.17 to 0.81])
- Admitted to hospital (intention-to-treat) - Tamiflu 1.0%, placebo 1.84% (RR 0.61, 95%CI [0.36 to 1.03])
- For an outcome called "Lower respiratory tract complications" defined as patients who received prescriptions for antibiotics 48 hours after randomization, the Tamiflu intention-to-treat infected group
had significantly fewer outcomes - Tamiflu - 4.2%, placebo 8.7%, risk difference of -3·8% (95% CI -5.0 to -2.2)
- Tamiflu caused significantly more nausea (9.9% vs 6.2%) and vomiting (8% vs 3.3%)
- Important points about the Roche-sponsored meta-analysis:
- The researchers used antibiotics prescribed 48 hours after randomization as a proxy for "lower respiratory tract complications." Given the lack of proven efficacy and pervasive
overprescribing of antibiotics for respiratory infections, it's almost laughable that a study would use antibiotic prescribing as a legitimate proxy for influenza complications.
- The study focused on the "intention-to-treat infected" population which is a modified intention-to-treat
population. Depending on the individual study design, this may have introduced bias into the study. Issues of multiple comparisons and data dredging also arise.
- The Cochrane meta-analysis found no significant effect of Tamiflu on hospital admissions while the Roche analysis did. Even if one chooses to believe the Roche analysis, the effect was negligible
(absolute difference of 0.84%, intention-to-treat) and not likely meaningful.
- After reviewing both meta-analyses, we came to the following conclusions regarding Tamiflu:
- In the best case scenario, Tamiflu reduces the length of flu symptoms by about 17 hours. This benefit comes at the cost of increased nausea and vomiting.
- In general, Tamiflu studies were not appropriately designed to measure important clinical outcomes like pneumonia and hospitalization. Based on the available information, there is no
evidence that Tamiflu reduces the risk of these outcomes.
- It's unclear if Tamiflu prophylaxis reduces the spread of influenza. Tamiflu suppresses influenza antibody production and influenza growth in a culture. These actions most likely
obscured the outcome measurements in prophylaxis trials.
- In conclusion, after heavy scrutiny and analysis, the benefits of Tamiflu are mostly noise with no clear signal. Most patients should forgo the $150 Tamiflu prescription and reach
for the $2 ibuprofen bottle instead.