Last fall on a visit to Canada, I spotted a first-rate piece in the Toronto Star by Diana Zlomislic, which chronicled the death of 18-year-old Marit McKenzie who had suffered a catastrophic blood clot blocking blood flow to her lungs. McKenzie had been taking the controversial acne medication Diane-35 for about a year to treat mild acne even though there were questions whether the drug should have been approved for use in Canada. The drug, also prescribed as an off-label use for birth control, had been pulled from the market in France, the Star reported, and Health Canada, the country’s federal health agency, had issued three advisories since 1997 warning doctors that patients who take the drug face an “elevated risk” of blood clots. It had been approved that year as a last resort, short-term treatment for severe acne. Still, four months after McKenzie died in January 2013 and two months after another teenager died from a fatal clot attributed to the use of Diane-35, Health Canada said it had completed a review and concluded the drug’s benefits “continued to outweigh the risks when used as authorized.”
Zlomislic’s October piece, it turned out, was one of more than a dozen stories written by Star reporters in the past two years exposing deaths associated with drugs such as Champix for smoking cessation, antipsychotics for nursing home patients with dementia, antidepressants for kids with ADHD, and antibiotics for acne. The stories revealed serious shortcomings in the Canadian surveillance system for pharmaceuticals, and highlighted Health Canada’s insistence that it would not release crucial drug safety information to the public. In early July, Health Canada changed its mind, and announced it would now tell the public about serious and sometimes fatal side effects that may have been caused by “unapproved off-label prescriptions.” For its dogged reporting and persistence in keeping the pressure on a reluctant government agency, the Star deserves a CJR laurel.
The stories, written by reporters David Bruser and Jesse McLean along with Zlomislic, are fundamentally about government accountability and transparency (themes I’ve examined in recent posts focused on reporting in the US). Off-label uses of drugs (uses for a condition or age group for which they are not approved) became the vehicle for driving home the reporters’ main point that Health Canada was not forthcoming with the public it’s supposed to serve about potentially harmful drugs. In a story detailing deaths attributed to drugs for ADHD, the paper summed up the problem:
The Star has repeatedly asked Health Canada if it has investigated individual reports of suspected side effects. Has Health Canada followed up with those who make the reports–whether a doctor, pharmacist or mother–to learn more about the case? Has the regulator tried to access medical records or conducted interviews with doctors and patients or their family members? Health Canada will not answer.
One particularly compelling story revealed the extraordinarily high use of antipsychotic drugs in nursing homes. Using data given to them by a liberal member of the Ontario government, Bruser and McLean found that more than one-third of all nursing home residents in Ontario have been given risky drugs that can kill elderly patients suffering from dementia–twice the rate in the US, which the US Office of the Inspector General said was too high. Both Bruser and McLean made it clear data were only the beginning of a story. “The data are rarely it,” McLean told me. “They help you get to the it.” After identifying the rates of antipsychotic use, the reporters visited ten nursing facilities in Toronto and contacted 30 more that had the highest use to find out why.
Even though the Ontario government has been aware of what Bruser calls the “staggering” use of antipsychotics, it has hardly rushed to address the problem. So the Star took the lead and constructed a database that allows families to learn how often a particular facility uses antipsychotic drugs to calm residents with dementia. For the model, they looked to a US database called the Minimum Data Set (MDS) maintained by the Centers for Medicare and Medicaid Services. The MDS gives self-reported data on several nursing home characteristics including use of drugs to restrain residents. It’s useful for reporters on this side of the border as well.
The Star found another US database crucial to its work. Health Canada maintains a database listing adverse drug events, but there’s a huge missing link–the clinical indication for a particular drug involved in a bad outcome. “Indication for use was being removed at some point,” Bruser said. So the Star turned to the FDA’s adverse events database, which does give drug indications and allowed the reporters to confirm what they were beginning to suspect–some drugs were being used to treat patients for ailments for which the drugs were not intended. The FDA’s database also contained data from Canada and other countries. “Without that, we wouldn’t have had a story,” said Bruser. Bruser and McLean did tell me the FDA database has one drawback: it is complicated, and the paper’s data analyst, Andrew Bailey, was invaluable in helping reporters work their way through it.
For news outlets that don’t have a data analyst handy, the experience of New York Times business writer Gretchen Morgenson offers a useful tip. Last week, Morgenson reported that Questcor Pharmaceuticals had for years touted the potential benefits of its immune system drug Acthar without noting its side effects. But in a recent regulatory filing, the company disclosed adverse drug events for Acthar had risen from 9.1 percent of prescriptions in 2011 to almost 14 percent in 2013. Morgenson used the FDA database and learned that 20 deaths and six cases of disability may have been associated with Acthar since 2012. In her stories, Morgenson noted she had made a FOIA request to get the data. I asked her why, since the Canadian reporters didn’t need one. She said an FDA press officer warned that the database was hard to synthesize and crunch, and a FOIA request would be more reliable. Because her story was sensitive, she told me, “knowing that the FOIA would mean 100 percent accuracy gave me confidence in my report.”
Both the Star‘s ongoing series and Morgenson’s reporting on Acthar demonstrate there’s a lot of good journalism to be done when it comes to drug safety, and the public databases are the starting point.
Related content:
Investigating a $150 billion ‘black box’
Medicare isn’t doing its own ‘truth checking.’ Why?
Trudy Lieberman is a longtime contributing editor to the Columbia Journalism Review. She is the lead writer for CJR's Covering the Health Care Fight. She also blogs for Health News Review and the Center for Health Journalism. Follow her on Twitter @Trudy_Lieberman.