The problem of assessing the safety of medication
Juding the safety of prescription medication is a lot harder than people think. Randomized trials can tell us a lot about whether a drug works, but not necessarily about how safe it is. One initiative new initiative is the CNODES program put forth by Health Canada which I talk about in my latest piece for the Montreal Gazette.
The Hippocatic oath says “I will use medicine to help the sick … but never with a view to injury.” That poses a unique problem for physicians because all medications have the power to heal and the power to hurt.
In some cases, the trade-off is obvious. If you’re having a heart attack, you should probably take an Aspirin. Yes, Aspirin will increase your risk of bleeding, but the blockage in your heart is the more immediate concern. Sometimes the balance shifts the other way. Arsenic can, and has, been used to treat lymphoma. But an overdose will kill you and much better treatments exist today, so the risks clearly outweigh the benefits.
Sadly, the situation is rarely that clear cut. For example, there were concerns new diabetes drugs, the incretins, could increase the risk of pancreatitis or pancreatic cancer. The new drugs have been well studied in clinical trails and shown to be effective in treating diabetes. But measuring the effectiveness of a new drug isn’t the problem. Measuring safety is.
Assessing new drugs for side effects is much harder than people realize. Common side effects are easy enough. If 1 out every 100 people is allergic to a new medication, that will become obvious in a trial of 10,000 people because you will see 100 allergic reactions. But rare side effects aren’t obvious. If a drug only causes a side effect in 1 of out of every 10,000 people, you aren’t likely to see it in a clinical trial.
The only solution is to conduct a Phase IV trial, also called post-marketing surveillance. The idea is that after you exclude big risks and approve the drug for clinical use, you monitor the population to pick up rarer side effects. The flip side is that you can reassure the public if no risk is found.
The difficulty with side effects research is gathering the data. First, you need data on medication prescriptions. Then you need hospital and physician data about side effects or new diagnoses, like allergic reactions or new kidney failure or new diagnoses of cancer. But most important, you need this on a massive scale. To tease out small but significant risks you need millions of data points. You need national and even international data to get the numbers.
That’s why Health Canada has done something right in this situation. We should protect Canadians from the side effects of prescription medications. However, over-the-counter dietary supplements can also pose significant risks and result in 23,000 ER visits in the United States every year. These natural products can be dangerous and Health Canada should be doing more.
But at least when it comes to prescription medications, Health Canada is moving in the right direction. It recently set up the Canadian Network for Observational Drug Effect Studies (CNODES) to investigate prescription drug safety. The network is headed up by Samy Suissa here in Montreal and involves more than 60 researchers across Canada.
So say the government wants to establish whether incretins really do damage the pancreas. Using CNODES you can answer that question. When CNODES analyzed data on 1.5 million people from Canada, the United States and the United Kingdom, it turns out that incretins do not increase the risk of pancreatitis or pancreatic cancer. They don’t seem to increase the risk of heart failure, either. Fortunately, the early warning signal proved to be untrue.
The only way to answer difficult questions about drug safety is with large projects like this, and that requires government support. In this case, some lingering doubt about these diabetes medications was settled and allowed both doctors and patients to be reassured. Yes, these projects cost money, but given that Health Canada’s mission is to “make this country’s population among the healthiest in the world,” it seems like money well spent.
Christopher Labos is a Montreal doctor who writes about medicine and health issues.