Lies, Damn Lies & Statistics

Man and woman and heartIn Canada, Lyme disease has been nationally reportable since 2009 which means, in essence, that whenever a doctor diagnoses a patient with the illness or a test conducted by a public health laboratory comes up positive, those cases are supposed to be reported to the proper authorities who can then use them to track important information relating to Lyme disease in this country.

Laboratories are doing what they're supposed to, but doctors are not. And since Lyme disease is identified primarily through clinical diagnosis, the failure of doctors to report their cases means that the national reporting system for Lyme disease has been, to put it mildly, ineffective. It's not controversial for me to say that. Everyone from public health officials and researchers to Lyme disease patients and their advocates have known from the beginning that the number of cases reported to the system is nowhere near accurate.

Several key problems

This has led to several key problems, the first one being that we have no idea how many people in Canada are suffering from Lyme disease. In 2015, the total number of cases was reported to be just over 700. We know that number is artificially low. What we don't know is by how much. It's not unusual for the true number of people suffering from a reportable illness - any reportable illness - to be 10 to 12 times the official figure and this is fairly universally the result of underreporting. The case could be made that the true number of Lyme disease cases in Canada is 20 times the official number. It's more difficult to argue for the higher figure, but "more difficult" doesn't mean outside the realm of possibility. That higher figure could be correct. We currently have no way of knowing.

A second problem is that the official statistics are routinely parroted by the media who for the most part don't realize how problematic those numbers are. You'd think they could ask. After all, even Canada's top doctor admits that there's a significant issue with the official tally, so it's not like journalists would be stirring anything up by posing the question. But for the most part, they aren't posing the question. And so it has been widely reported that roughly 700 Canadians were diagnosed with Lyme disease in 2015 when, as I noted above, the real number could be as high as 14,000. This is an important difference. When Canadians hear that 700 people in a country of roughly 36 million contracted a disease, they figure they aren't ever going to get it themselves. When they hear that the number could be in the thousands or tens of thousands, they start to pay attention. But they aren't currently hearing that so, at a point in our history when Lyme disease cases have started to soar, most Canadians continue believe they will never contract the illness.

A third problem revolves around which Lyme disease cases are being recorded and which ones are not. The recorded cases are limited to the ones where the patients came up positive on tests conducted by public health laboratories in Canada. With few exceptions, the official statistics do not record anyone who was diagnosed in Europe or the United States where Lyme disease is more common (and in some regions epidemic). Nor do they record anyone with positive results from commercial labs in the US. And they record virtually no one who was clinically diagnosed with Lyme disease and that's a critical failure since clinical diagnosis is the primary means of diagnosing this illness. It's also the only sanctioned means of diagnosing early Lyme due to a well documented deficit of current Lyme tests that causes them to almost always comes back negative during early infection. As a consequence, most of those early cases are not recorded either.

And yet all of these cases should be recorded. They don't all have to be given the same weight, but their existence should be on file so that public health policy reflects the entirety of the Lyme disease situation in this country and not just a small fragment.

Important information lost

At present time, the majority of Lyme disease cases in Canada are not recorded and that's a problem because important information is gathered about known cases by public health authorities, such as when and where a tick-bite occurred, what treatment the patient received (at least initially), and what that patient's symptoms were. This information can then be used to help formulate effective public health policy. But this information is currently woefully incomplete.

Let's focus for a moment on early symptoms. What percentage of Canadians developed the distinctive rash associated with Lyme disease shortly after they were bitten? There is considerable debate over this issue. Some public health experts put the figure as high as 80 percent. Some advocates put the figure as low as 20 percent. Most researchers guess that the figure is somewhere in between. Note that I say guess. Current statistical data favours none of these positions. With only Bar graph700 cases reported out of a possible 14,000 - remember, this too is a guess - there is no way to say whether the data collected from the reported cases is statistically significant. It may not be. Remember, patients who suffer from early Lyme tend not to test positive for Lyme disease and yet it's these patients who also tend to develop Lyme rashes at a higher rate than sufferers who were diagnosed at later stages. And since existing statistics do not capture information related to those early cases, it's hard to imagine how the data can be anything other than skewed.

Rectifying the situation

The underreporting of Lyme disease cases is not unique to Canada. The US Centers for Disease Control (CDC) knew they had a similar problem that meant the 30,000 annual cases they were reporting could very well be quite a bit on the low side. So they conducted three elegantly designed studies and in 2013, based on the interim findings of those studies, determined that the true number of Lyme cases being diagnosed in that country each year was closer to 300,000. The discrepancy between the two figures may seem shocking, but if you recall what I said above about the real number of people suffering from reportable diseases commonly being 10 to 12 times the official figures, the revised tally is actually on the low side of what the CDC could reasonably have expected to find.

Unfortunately, the differences between the US and Canadian systems of medicine make it so that we can't replicate the design of those studies here in Canada. Several suggestions have been made for how we can overhaul our current reporting system to better align it with reality. Those suggestions include allowing self-reporting by patients, inviting naturopaths to report cases, permitting doctors to report cases anonymously, and loosening up the reporting criteria so that doctors don't feel inhibited about reporting marginal cases.

So far none of these suggestions have been implemented and so the national reporting system continues to limp along, adding very little to our collective understanding of Lyme disease in Canada, and muddying the waters it's supposed to be clarifying.