Therapeutics Initiative is out of mainstream
It’s time to bring B.C.’s Therapeutics Initiative back into the scientific mainstream. I have written in the B.C. Medical Journal about how the Therapeutics Initiative, which evaluates the effectiveness of drugs for doctors and pharmacists, uses questionable statistics to recommend against the use of medications, and it continues to do so. The TI receives much of its funding from the Ministry of Health and it tells the ministry and Pharmacare what they want to hear: No need to pay for that medication.
Since 2010, across the publication of 27 Therapeutic Letters, TI specifically reviewed 18 drugs or classes of drugs in detail and gave positive recommendations to two.
In its latest publication (posted on Jan. 27) it looked at a drug used for patients with chronic obstructive pulmonary disease. The TI conclusion is that “there are no proven clinically meaningful benefits in terms of reduction in mortality or total serious adverse events for [this drug] in patients with COPD.”
By contrast, the Common Drug Review, the federal body that does directly make recommendations to all the provincial health ministries, says: “The CDEC recommends that [this drug] be listed in a similar manner to other [drugs in its class] for the treatment of COPD.” B.C. Pharmacare does cover this drug, in certain circumstances, after a physician completes a special-authority form.
TI says the drug doesn’t reduce mortality — true. It combined data from studies of almost 7,000 patients and by its calculation, the drug, compared to placebo, reduced death by 50 per cent, but that reduction was not statistically significant — also true.
But because deaths were rare in the studies, only 19, the study needed to statistically prove that a reduction in death would be almost double in size.
But if we are going to recommend only therapies proven to save lives, there’s a whole lot of medicine we can probably stop doing: cataract surgery, replacing arthritic knees. Symptom relief can be important, but it’s apparently not important enough for a TI recommendation.
TI also says the drug doesn’t reduce serious adverse events (SAE). SAE have to be collected in all clinical trials. But most SAE have nothing to do with either the disease being studied or the drug being investigated. The FDA and Health Canada require a full reporting of them — there is full disclosure at clinicaltrials.gov.
What I argued in the medical journal, but which TI refuses to accept, is that adverse events are a safety outcome, not an efficacy outcome. An equal number of SAE in the treatment and in the placebo arms of a study prove the drug is safe — not that it doesn’t work. And yet the TI continue to withhold its recommendation based on its flawed beliefs.
Do you know who else thinks serious adverse events are used to assess safety? The federal review body uses SAE that way. In its review of this drug (which, remember, CDR recommended), it says, in a section titled “Harms (Safety and Tolerability)” that “the incidence of serious adverse events and withdrawal due to adverse events was similar between treatment groups in all trials.” Ergo, the drug is not causing undue side-effects and it is reasonably safe.
These flawed TI recommendations based on flawed statistics are not isolated to this one drug. Other reports from the TI recommended against the use of certain blood thinners after a heart attack, cholesterollowering medications and drugs to prevent or treat osteoporosis because SAE “were not lower.”
Let’s go back to our COPD drug. In all but the smallest of the studies reviewed by TI, there were “clinically significant” improvements in lung function and/or patient symptoms. Those improvements were the primary outcome of interest in each of the studies (while ensuring safety).
The drug reduced symptoms of COPD, which can be very debilitating. It’s safe because it did not cause an excess of any particular adverse outcome or a totality of adverse outcomes. And while there was a strong trend toward a reduction in mortality, the studies were too small to show a “statistically significant” reduction, so TI recommends against its use.
This is not scientific mainstream. This is a group devoid of true peer review who answer only to themselves and the government that funds them. This is a group lauded by editorialists of this and other newspapers.
I support rational, balanced education of physicians, pharmacists and patients. Drug companies are easy to hate. But let’s have an honest, scientifically rational discussion using statistics as they are meant to be used.
And finally, let’s get true experts at the table advising government.