Getting at those pesky facts

I’m not in the business of sticking up for one region of New Brunswick or the other but a number of posters to this blog have been suggesting that the health of Saint John residents is being severely impacted by industry.

I don’t think the data indicates that. From Statistics Canada’s community health survey, Saint John (Region 2) has a considerably lower rate of Asthma compared to Fredericton or Monton and the people rate themselves high for both overall health and mental health.

A few stats:
Self rated health ranked #2 out of 7 NB health regions
Mental health ranked #2 out of 7 NB health regions
Asthma – ranked #3 out of 7 NB health regions (ahead of Moncton and Fredericton)
Diabetes – well below the provincial average
Has the highest number of disability days of the 7 health regions
Below the national average for injuries

I don’t have the cancer data at my fingertips and we can’t cut the data to areas within the region. The recent Vitality report on Saint John may have better data on this stuff but I don’t see that industry in SJ has had an overly significant negative impact on the population.

Unless industry is subsidizing those juicy hamburgers (they have the second highest rate of obesity in the province).

This entry was posted in Uncategorized. Bookmark the permalink.

0 Responses to Getting at those pesky facts

  1. richard says:

    Cancer survey data are available here: http://www.gnb.ca/0208/pdf/cancer_1997-2001-e.pdf

    If you assume that St John dominates region 2 population-wise, then differences should show up when region 2 is compared to other regions. I don’t see that many differences.

  2. mikel says:

    David david david…and on one of my busiest days-but I just can’t let that go by.

    By that reasoning, since the employment stats went up this month, by definition your whole blog is ‘wrong’ or doesn’t seem to fit the facts. Employment went up, ‘therefore’ jobs are being created and everything is keen.

    I’m sure you see what’s wrong with that picture. This is why science should be taught to EVERYBODY, or at least the scientific method.

    First, as always people, look at the study. Why is it that you find such joy in poking holes in all the various studies that ‘seem’ to contradict you, yet this one is taken at face value? That’s inconsistent.

    The study is SELF analysis. It is statscananda calling up and asking ‘how are you feeling today?’ When somebody asks how are you today do you usually list all the things wrong with you?

    If you look at the study you’ll also notice that a certain percentage didn’t even talk to those they were looking to talk to. That’s right, they used ‘proxies’. And in which province were there the most proxies? You guessed it, at 10% of all surveyed.

    Which leads to the next point. In New Brunswick they spoke to 1000 people. So of 100 of those, they didn’t even talk to the people directly, they essentially asked “hows your mom feeling?” Can you imagine a WORSE way to conduct a health study? It’s more like a health poll.

    So of 1000, there are seven health regions, which means AT MOST they talked to 200 people in not just St. John, but clear to St. Andrews and up past Sussex. Thats a BIG area.

    Meaning, of those IN the city of St. John you are looking at maybe 100 or 120. Out of a city of over 120,000, well, you do the math.

    That’s the study. Now for the pollution. First, most health studies have found that not only are health risks related to cities, but to specific areas WITHIN cities. In Toronto they found the downtown had completely different health concerns than Scarborough. Not surprising.

    And of course air moves (usually). Which means somebody in Quispamsis has a different environment than somebody right downtown.

    Second, let’s talk cancer. The link above is helpful, but only goes to 2001. Here’s the kicker, the claim is that 2% of the population (15,000) has cancer, with ‘about’ 3500 new cases a year. It’s now six years later, which means now 21,000 have cancer. For those in humanities, thats more than DOUBLE in six years.

    So now 4.3% of the province has cancer. At this rate, in ten years almost ten percent of the population will have cancer, and thats INVASIVE cancer. Man, if you think thats GOOD news then you must have been really excited about those job statistics the other day!

    But there is LOTS of different types of pollution, and different pollutants do different things to the body. And some affect at a genetic level which means that the actual ‘disease’ will vary from person to person.

    But also keep in mind other factors. It took them decades to show the link between cigarettes and lung cancer, and another few to actually get that information from the companies who knew it for a long time.

    But there are other variables. New Brunswick is polluted, but the air in Fredericton or Woodstock is NOT the air in Toronto. Thats a different kind of pollution, but its still pollution.

    For health effects though what you say is far from the case. I’ve seen studies that correlate directly between bad air days in St. John and hospital admittances. That’s not unusual, and is the same in southern ontario for different sources of pollutants.

    And of course everything is relative, as I blogged the other day, health canada found much higher than acceptable levels of benzene in soda pop and fruit drinks. How does that compare to what is coming out of a refinery?

    Next, and hopefully finally, if you look at the cancer report, you’ll notice that northern women have a higher incidents of breast cancer. You’ll also notice that incidents of lung cancer is higher for the province than the national average.

    Now, we don’t know WHY, because even scientists don’t know that. More people may smoke, but what if the effect of pollutants from a refinery combined with cigarettes amplifies the probability of cancer? Well, we don’t know.

    The report also says female mortality from cancer is highest in St. John and Campbellton. COULD that be from pulp mills? We don’t know, but can suspect. And if true, it literally means that industrialization KILLS. That’s a big caveat.

    I could go on and on. Non hodgkins lymphoma are increasing, and all forms of cancer together have New Brunswick well above national averages. Keep in mind that industry is well spread out around the province. So its not JUST St. John, heck, that doesn’t even get into Belledune.

    That’s an analysis, and it certainly doesn’t follow that because some statscan poll seemed to find less asthma in St.John than Fredericton than Moncton that industrialization in St. John doesn’t seem to hurt anybody. Far from it.

  3. richard says:

    “You’ll also notice that incidents of lung cancer is higher for the province than the national average.”

    Generally, lung cancer correlates well with smoking and I believe (without checking the data) smoking rates are higher in the Maritimes than nationally. Obesity rates are also higher, which can lead to several health problems. I would guess (but don’t know) that smoking rates might be higher in Bathurst than in Fredericton.

    In addition, cancer rates must be age-adjusted, since aging populations almost always have higher higher cancer rates. NBs population is likely aging faster than in say ON or AB. Thus cancer rates (unadjusted for age) will rise faster in NB.

    It is certainly correct that air pollution leads to increased illness. That has been clearly shown in ON. Not sure how ‘average’ air pollution in NB compares to ON, but one would have to determine the impacts of a number of bad days in Saint John to overall air pollution levels in Saint John to get very far.

  4. mikel says:

    Lung cancer can’t simply be correlated with smoking for the simple reason that most cancer statistics never include the information on how many are smokers.

    Also, age is a factor in many cancers but certainly not all. Several leukemia’s hit adolescants only.

    The point is that there is virtually no way that this report can lead to the conclusion that industry in St. John has no affect on health because several health indicators are higher in other cities according to this fairly lousy study. And to call that ‘getting at those pesky facts’ borders on, dare I say, Al Hoganism. Sorry, had to say it.

  5. richard says:

    “Lung cancer can’t simply be correlated with smoking …”

    Sorry, but that is exactly what has been done. It is not a secret: smoking is the strongest factor associated with lung cancer. Smoking rates are higher in the Maritimes.

    As to age, again that is very well established. Cancer rates zoom with age, especially after 50 years. NB is aging faster than most other provinces.

    If you have data correlating other factors to higher cancer rates, then present it. As it stands, demographic and smoking rate data from NB and other regions can explain much of the cancer incidence in NB. This is not to say that pollution cannot be a factor; just that other factors often out-weigh those effects.

  6. mikel says:

    Nobody said that lung cancer didn’t cause cancer, what was said is that just because something is ‘correlative’ doesn’t mean its ‘causative’. Your final sentence sums it up well…one factor “outweighs” another. Problem is, in statistics you can’t KNOW ‘weight levels’.

    If science can’t predict it, then certainly statistics can’t PROVE it. The study certainly didn’t ask how many SMOKERS there were in St. John and how many smokers had cancer. So smoking while breathing air from a refinery can certainly add another factor, and there is no way to PROVE that. Biology is far from an exact science and statistics certainly doesn’t make it clearer.

  7. richard says:

    “The study certainly didn’t ask how many SMOKERS there were in St. John and how many smokers had cancer.”

    Generally there is a very strong correlation between smoking rates and lung cancer incidence. That has been very well established in many studies and it is accepted that smoking causes cancer. Smoking rates are higher in the Maritimes so lung cancer rates would be expected to be higher as well. Do you really think St John is any different in that regard?

    If you have some data to show that other factors might over-ride smoking or obesity in these cases, then present the data.

    You keep talking about science, but you don’t seem to know much about how it works. Science does not talk about PROOF; science is a set of methods for testing hypotheses. Data tends to lean one way or the other. When repeated surveys or studies lead to the same conclusion, then the case for a hypothesis can be strenghthened, but that is not absolute proof. If you want certainty, try religion.

    Science is data-driven; the data from multiple cancer surveys, when adjusted for age, do not support your hypothesis that industrial pollution in St John or elsewhere in NB is a major contributor to cancer incidence in these regions. So I do not see any reason for great concern over a second refinery in St John, based on cancer incidence.

    That is not to say that man-made and natural products do not cause cancers; they do, and there are some surveys (e.g. pesticide exposure in PEI) that cause concern. But I have not seen any data in several surveys (not just the statscan one) that raise particular red flags for St John.

  8. mikel says:

    Good post, but not great. I WAS talking about certainty, and thats the point-and you made it for me. I’m quite well aware of what science is and the scientific method, but thanks for the lecture.

    BAD science is the kind mentioned above which says ‘I don’t see that many differences’. First, go look at the study, they don’t even mention the number of cancer cases except anecdotally. What they mention is the PERCENT of overall cancer.

    Besides that, there is not just the incidents of cancer, but also the mortality rates. There is also the complications, there is also the severity. None of that is addressed in either of those studies. Plus, national averages also have the same problem-most provinces have some area of heavy industrialization that may skew data.

    The point is simply that, like global warming, you don’t KNOW what the reality is. Like you say, science doesn’t say what we KNOW, but what has at least been proven and seems to follow well established theories.

    I’m going to go back to this on my blog because its a big long issue, however, lets say you want to establish the health effects of a refinery. Well, you don’t simply say ‘lets ask a hundred people around how they’re feeling and see if they have cancer’. You do if you are selling refineries, but not if you actually want to know anything.

    To do that you first need to KNOW some basic facts. You need to know what that refinery is putting out in pollutants, you need to know something about the setup for getting rid of those pollutants.

    You need to know what those pollutants are and their specific effect at various dosages clinically, and at similar places which have similar practices.

    You need to know how people react to those chemicals in various conditions-something that is barely even studied.

    To say that “well, there doesn’t seem to be any more cancer patients in St.John than anywhere else” isn’t even close to ‘scientific’. Even besides that, its well known that certain chemicals may have little effect when at particular levels, but as soon as those levels are increased (such as by adding another refinery) then effects can double, triple or even higher-that’s how safety levels for chemicals are established.

    As you’ve admitted above, there is lots of evidence that on heavily polluted days that hospital admittances go up, so its hardly unreasonable to state that IF those effects are substancially increased then the severity of conditions or the number of admittances will go up as well.

    But nobody is saying ‘refineries cause cancer’ (at least not yet), however, the idea that the health effects of an oil refinery are ‘no cause for concern’ is pretty spurious and, of course, is a judgment call, and not science.

  9. richard says:

    ” First, go look at the study, they don’t even mention the number of cancer cases except anecdotally. What they mention is the PERCENT of overall cancer. “

    First off, Mikel, I was referring to several cancer surveys, not just one in particular. You can go to the cancer society site for some of them. These studies show that there are no significant elevated cancer rates in St John compared to other regions in NB.

    That is how science works (and, yes, you do need a few lectures on that). You either have some data to back up your case, or you do not. You do not, so you are resorting to empty rhetoric. Get some data and we will talk. You do not have a case to support your contention that cancer rates are elevated in St John over other parts of NB due to pollution.

  10. mikel says:

    I didn’t SAY that there was a case, what I said was that YOU don’t have a case that there is none simply because statistics show, whatever (I notice a lack of links there).

    The statistics we DO have shows that NB’s cancer rate is close to doubling. Cancer in old people also has causes, the same as in younger people. Statistics is not science, only a part of it.

    But we are arguing two different conclusions, I didn’t say that the refinery in St. John causes cancer, only that YOU don’t know, and simply quoting cancer statistics doesn’t prove it. I know quite a bit about science and know quite well that the research on dioxins and other pollutants on humans in various environments and under different conditions is very light.

    What we DO know is that a refinery produces a good amount of carcinogenic waste. We KNOW that many of those chemicals cause cancer under specific circumstances. The question is whether those previous assumptions are correct, whether the dosages are prevalent, and whether there are enough conditions that keep the waste away from humans.

    THAT is the science, what you are lecturing on is statistics, not science.