People who take a prescribed benzodiazepine (such as Xanax) for a long period often build up a tolerance to it, require a higher and higher dose of it, and end up abusing it – right?
That’s a common belief about these sedating drugs, says Dr. Silvia Alessi-Severini.
But when the pharmacy professor led a 2016 study of Manitoba health data, it exposed that belief as largely unfounded. The research was initiated by members of the UM department of psychiatry, who have been long-standing collaborators of Alessi-Severini.
“We showed that when these drugs are prescribed for a diagnosed mood disorder, dose escalation only happens in a very small percentage of individuals,” Alessi-Severini says.
“Many individuals are on this kind of medication for more than two years, and they don’t escalate their dosage. Our study did not find a causal relationship between long-term use and escalation to high doses.”
Alessi-Severini grew up in Italy, where she completed degrees in pharmacy and pharmaceutical chemistry. She immigrated to Canada more than 40 years ago with her husband, physician-microbiologist Dr. Alberto Severini.
She earned her PhD at the University of Alberta. In the mid-1990s, she became a consultant to the Alberta and Manitoba governments, advising them on whether new prescription drugs should be added to their formularies and covered by their drug plans.
The couple moved to Winnipeg and in 2003, Alessi-Severini joined the College of Pharmacy. Her work for governments had lit a fire in her to pursue pharmacoepidemiology – the study of the use, safety and effectiveness of drugs in large populations.
“I was really interested in patterns and trends in prescribing. And I wanted to investigate the effects that medications have in the real world.”
Pharmacoepidemiologists mine databases containing anonymized records of people’s contacts with the health-care system, such as doctor visits, diagnoses, prescriptions and hospital stays.
They can analyze data for large cohorts of patients through years of medication use, detecting red flags such as adverse effects, drug interactions and different responses to drugs based on patients’ characteristics.
The Manitoba Population Research Data Repository, housed at UM’s Manitoba Centre for Health Policy, is one of the most complete data mines in Canada – and the world – for this kind of research.
“There are other provinces with good repositories, but Manitoba’s is absolutely comprehensive,” Alessi-Severini says. “It’s been a fantastic opportunity.”
Artificial intelligence will likely transform data analysis in this kind of research, she says, but human intelligence is still essential. “If you don’t design the study well, if you don’t use the right statistical methodologies, the result will not be reliable.”
The professor, who is nearing retirement, has published more than 80 studies. One of her most-cited, published in 2012 in the Canadian Journal of Psychiatry, examined prescriptions for antipsychotics to children over a 10-year period in Manitoba.
“I really wasn’t prepared to see how much these very potent drugs were increasingly being prescribed to children,” Alessi-Severini says. “That paper generated media attention and stimulated other researchers to study this, which was very positive. When you produce new information, you want it to have impact.
“Sometimes we report good news. In 2014, we documented a decline in Manitoba prescriptions for benzodiazepines in older adults. The message that these drugs can be harmful for people over the age of 65 was getting through to prescribers.”
Alessi-Severini wishes that doctors and other prescribers would consistently follow scientific evidence, but she is philosophical about it.
“There are prescribers who are very enthusiastic – the early adopters. There’s a group that is more moderate and will wait before they change their prescribing pattern. And then there’s a group that is always behind – the last ones to accept what has been shown through research.
“We know it’s impossible for prescribers to read every journal. As scientists, we have to do a better job at sending our message.”
On a positive note, she says, journals are now publishing more “negative” studies that find no link between a medication and a hypothesized effect.
“In the past, journals, and particularly pharmaceutical companies, wanted studies that documented an effect. But if you have one study that shows some effect and 20 that show no effect, and you only publish the positive one, that doesn’t give a clear picture of the truth.”
Alessi-Severini is active on the Manitoba team of the Canadian Network for Observational Drug Effect Studies (CNODES). Health Canada regularly asks the provincial teams to answer a safety question about a drug. The teams agree on a study protocol and analyze their data. Then the “multi-centre” findings are combined.
A CNODES study in 2020 in the Annals of Internal Medicine revealed that SGLT-2 inhibitors (newer drugs to lower blood sugar) were associated with an almost three-fold increased risk for diabetic ketoacidosis – a serious complication – in patients with Type 2 diabetes, compared with DPP-4 inhibitors.
Canada is moving toward a national drug formulary, the professor says. “In the past, sometimes an expensive drug was paid for under the plan in one province, but not in another, and patients had to move. There’s been a lot of work done to create more equitable access to medication across the country.”
Reflecting on her contributions to the College of Pharmacy, Alessi-Severini says she is very proud to have led the introduction four years ago of a new concentration in pharmacoepidemiology, enabling students to earn master’s and PhD degrees in this specialty.
“We have many students now who are stellar researchers,” she says. “Our program has grown significantly, thanks to my young colleagues. I’m leaving this in very good hands.”
BY ALISON MAYES