Disease cluster hard to pinpoint when cause of illness is unknown

An Ottawa resident recently raised the alarm over what he saw as a worrying concentration of cases of multiple sclerosis in his neighbourhood, but experts say identifying a disease cluster is more challenging for chronic conditions like MS than infectious diseases like the flu.

Concentration of MS cases in Ottawa could be perfectly normal, epidemiologist says

Jacques Dutrisac, 59, wondered whether the fact that he and more than 10 of his former neighbours in an Ottawa neighbourhood were all diagnosed with MS as adults was merely a coincidence. (Hillary Johnstone/CBC)

An Ottawa resident raised the alarm this week over what he saw as a worrying concentration of cases of multiple sclerosis in his neighbourhood, but experts say identifying a disease cluster is more challenging for chronic conditions like MS than infectious diseases like the flu.

In the case of infectious diseases, the time between infection and symptoms is short — generally, days or weeks, said David Buckeridge, an associate professor of epidemiology and biostatistics at McGill University in Montreal. So, infectious diseases start out as a cluster and can spread to become an outbreak or an epidemic.

But chronic illnesses like MS or cancer might develop over longer time periods, which makes it harder to spot a cluster.

"Even with chronic diseases like cancer, if it were an environmental exposure, which is often difficult to say definitively that was the cause, you're looking at lag times of years or decades," Buckeridge said.

In chronic illnesses whose cause is either unknown or harder to pinpoint, it's more difficult to determine whether a cluster exists because there is no clear common link among cases. 

Diabetes clusters easier to pinpoint

One chronic disease for which definitive clusters are easier to identify is diabetes, Buckeridge said. That's because diabetes clusters are generally found in areas where the risk factors for the disease, such as an unhealthy diet and excess weight, are also clustered.

A cluster is simply more cases than one expects to see statistically. 

The suspicion that a cluster might exist can be useful in some cases even if it doesn't turn out to be one because it can spur research, Buckeridge said. If, for example, data points to a seemingly large number of cases of cancer in one area, researchers will investigate whether or not those cases are linked to an environmental factor.  

David Buckeridge of McGill University says it is difficult to verify whether or not a concentration of cases of a given disease is a true cluster, which is generally defined as more cases than one would expect to see statistically. (Courtesy David Buckeridge)

In the Ottawa case, where 14 people from the same area were found to have been diagnosed with MS as adults, it's possible that people were looking for a trend retroactively where there really wasn't one.

"It's amazing how 14 cases pile up, but they seem to cover long periods of time," Buckeridge said. "If we look at the number of people in and out of that neighbourhood over that period of time, it could be in the thousands or tens of thousands.

"It may be what we expect to see."

Buckeridge used the analogy of someone shooting randomly at a wall and later drawing a bulls-eye around where most of the shots hit. That could make it appear the person is a good shot.

Rama Nair, an epidemiology professor and biostatistician at the University of Ottawa, says that in order to determine whether or not the number of illnesses is unusual, there needs to be a link between the cases. Occurring in the same place and at the same time is not enough, however. Usually, scientists will also look for some commonality in patients' diagnoses.

The Ottawa cases, for example, occurred around the same time and in the same neighbourhood, but there didn't seem to be any other unique factor that grouped the cases together, Nair said. 

Researchers would also need to compare the area in question with similar-size neighbourhoods and measure the incidence of the disease in those areas.

"Most of the time, we're not able to really come up with strong association," Nair said.

Why is MS so prevalent in Canada? 

One of the few certainties in multiple sclerosis is that Canada has the most cases per capita in the world, but researchers are still trying to figure out why.

As of 2013, there were 291 cases of MS for every 100,000 Canadians. Sweden had the second-highest rate, with 189 cases.


Karen Lee, vice-president of research for the Multiple Sclerosis Society of Canada, says there are several hypotheses.

One is that the disease is somehow linked to vitamin D and that on account of Canada's northern climate, its residents are exposed to less sunlight, a source of vitamin D, especially in the winter, Lee said.

Swedes would have similarly limited sun exposure, Lee said, which is why vitamin D is a key area of research.

Clues to why MS is so prevalent in Canada might also emerge from research being done into possible links between the disease and Epstein-Barr virus.

Another avenue of inquiry is examining dietary factors and investigating whether these play a role in the higher prevalence of MS in developed countries.

About the Author

Nicole Riva is a multi-platform writer and social media presenter for CBC News based in Toronto.

With files from Hillary Johnstone