I had a quick look but didn't find a reference to how much each country tests, what exactly they test for, and what constitutes a definition of "cancer" per-country (for example, do countries both classify and report non-malignant neoplasms the same way). Without this info I don't think it's possible to draw any conclusions, except between countries with (presumably) fairly similar reporting and testing regimes thanks to a long shared history, such as the UK and Australia.
The US is such an outrageous outlier that it does seem suspicious. Stereotyping the US health system, my first bet would be that doctors in the US can charge patients slightly more if they identify a cancer.
It could also be due to the EU's stricter pollution, chemical and food regulations compared to the US, the Brussels effect tends to make them apply globally to smaller markets.
I honestly think it could be as simple as being proportional with obesity, sedentary lifestyles, and sugar consumption that exist in the US and Canada.
The US and Canada really stand out specifically in their suburban infrastructure, which may strongly correlate with cancers in very obscure ways, such as simply needing more preservatives, such as sugars in bread, simply because people buy more in bulk because the culture of supermarkets.
I'll be honest, I think the chart needs a full explanation, but when I think of the US and Canada, and what makes them really unique, it's the complete lack of walking anywhere except the few urban centers.
I would really like to see a regional breakdown within the US.
At least by state. If possible also by urban/rural populations.
There are areas which are notorious for greater cancer risk, e.g., "cancer alley" through the Mississippi delta region. And there are regions with far greater access to healthcare, notably the north-east and northern plains states (especially Minnesota), and western states (CA, OR, WA).
Many of the more rural states (WY, ID, NM, ND, SD) would have correspondingly less access to healthcare.
The mix of potential causes (e.g., polluted regions, lifestyle) and available Dx / Rx / Tx) would be useful in distinguishing the cause of reported incidence.
I suspect you can't really get much from that kind of geographic breakdown in practice at least when it comes to pollution or lifestyle, because people move, especially nowadays people move quite a lot from where they are born and exposure to harmful stuff during pregnancy and youth are suspected to be much stronger factors than comparable exposure during adulthood.
- Those who are exposed to a specific area for a prolonged period of time will leave a strong signal.
- Particularly those exposed during pregnancy, infancy, and early childhood, which you note. Even if significant numbers subsequently move you should see a trace amongst those who do not, and the movement will be somewhat arbitrary.
- Those who do move often and readily often move among specific types of areas, mostly large and coastal cities.
- Those who do not move often tend to be concentrated among regions of lower wealth. Where they do move, it's typically among such regions, or often within poorer regions (e.g., from a smaller town to a nearer larger city).
Even then, overall numbers are still ... comparatively small. Looking at California, 6.1 million people left the state from 2010 to 2020, of a total population of just under 40 million, or 15%. Which leaves us with 85% of state residents who did not move out of the state. (They may of course have moved within the state.)
The problem I'm thinking about is that you can end up with as many people having grown in say a pesticide-polluted rural environment later living on big cities as you have people remaining where the problem comes from. The map can then seem to show literally the opposite of the phenomenon occurring.
You can correct for these things, but it's going to require more data, and a more complex causal model describing the phenomenon if you don't want your controls to introduce bias themselves.
It's also a challenge that public-health epidemiologists have been dealing with for a long time, for which there's been a tremendous recent explosion in both data and research methods. And there are ways to test for this, which I'm not fully aware of, though I've some basic familiarity.
I've already addressed some of this above, so with some repetition:
- People simply don't move that much, have been moving less over time within the U.S.,[1] and moreover don't move consistently. So wherever there's an initial strong cause, you'll have a fairly large cohort remaining on that site and showing impacts over time, particularly those who are most susceptible to such influences. Again: neonates, infants, children. Many cancer / disease clusters are found by such mechanisms.
- Where people do move, the end result is something of a "blurring of the signal". You'll get a blob at the origin, and maybe scattered points elsewhere. Those will tend to be at likely points of migrations: nearby neighbourhoods and towns, nearby large cities, regional/national cities of prominence, and (sometimes) locations with established immigrant communities (whether intranational or international). These are ... somewhat ... predictable patterns. The signal will tend to be strongest at or near the source.
- Deeper and extended data. Where topical data (e.g., diagnosis and current residence) don't seem to correlate with a known possible cause, or show a rare-but-below-threshold cluster, epidemiologists will dig for further information. Possibly with patient surveys, possibly other methods. What they're looking for in that case will be recent, or non-recent, movement patterns. Once a probable cluster source is identified, that can be used as a specific clue for further research. This is of more use to an epidemiologist who can conduct such further research than a data scientist who's working off extant databases (partial, limited data capture, etc., etc.), but are possible. And yes, this is one of the fundamental limitations of strictly broadly-captured data research.
There's a lot of medical research, even within healthcare and governmental organisations which relies on fairly low-quality and easily-collected data. The reason is that those data exist and are cheap. The questions are how to maximise utility of such sources, and knowing when to dig deeper.
Again: people moving really isn't the major problem you're making it out to be. Yes, it makes the job somewhat more challenging. But it's still generally tractable.
I'm not saying that moving people are a complete showstopper for epidemiologists, what I'm saying is that it make the map visualization a poor fit for the task because it will induce the casual reader with access to only the map (that is, not epidemiologists with access to the full data) into making wrong conclusions.
Keep in mind this is the incidence of cancer diagnoses. The quality of healthcare, diagnosis standards, and amount of preemptive screening varies heavily between countries.
Even in pretty terrible healthcare systems, I would expect that close to 99% of people dying from cancer will actually be diagnosed with cancer, even if done too late.
You might be able to claim that for some countries, but all European data should be reliable, all countries have reasonably good cause of death recording.
After listening to stories from my Canadian friends about how hard it is to just get basic medicine like antibiotics I'd expect this to be caused by a lack of screening/diagnosis rather than actual differences in cancer rates.
Do they live somewhere that requires a float plane to reach a doctor? I've never heard of a fellow Canadian having a hard time getting antibiotics, what a bizarre anecdote.
The 2020, 2021 dip looks fishy, I would hypothize this is related to the pandemic lockdowns and overburdened hospitals reducing the number of cancer diagnoses. But it's a hypothesis that would need to be confirmed with the relevant numbers.
It would be useful if the data differentiated by type of cancer. For instance, skin cancer diagnosis is straightforward and inexpensive. Visual identification and a biopsy, led by the patient noticing a difference. Contrast that with pancreatic cancer, which requires imaging for diagnosis.
As it is the only meaningful conclusion to draw from this data is that we need more data.
Japan, having ~the highest average age, should be on the top of the list but is not: second-highest bin. Australia is the skin cancer capital of the world ('Approximately 2 in 3 Australians will be diagnosed with skin cancer before the age of 70'), yet is in the third-highest bin. Isn't that interesting!
I'd speculate that a lot of this is down to the kind of diagnostic tests run on the typical patient coming in for treatment. I know from personal experience that in Europe it's a lot less and tends to be a "take an advil, go home and rest" while on my last trip to hospital for broken toe I got every blood test possible done too.
As a doctor friend of mine says, if you apply enough analysis you'll find something wrong somewhere. I think this might be the case in the US and Canada.
I don’t think that’s the main culprit. US has higher incomes than most Euro countries but generally 5 years less of life expectancy. I think there’s at least one major environmental culprit that Americans will eventually be incensed at, like the Sackler situation but much worse.
I wouldn’t bet on construction materials, unless it’s something PFAS-like that accumulates indefinitely and causes problems. But I like the suggestion because we need more out of the box theorizing instead of the obesity trope which clearly isn’t unique to the US.
I'd get this if we were talking about say influenza incidence rates, or cancer survival rates, but not cancer incidence rates.
Given the progressive and terminal nature of cancer, there comes a point where any relatively modern healthcare system will be forced to diagnose it.
That said, non-malignant tumors and other "could be cancer if you squint a bit" like states are relatively common in perfectly healthy people. If you go looking for them on a shoot first and ask questions later, you might very well inflate incidence rates quite significantly. The economic incentives at least in the US healthcare system seems to strongly favor discovering new diagnoses, whether or not they are actually there.
It’s not the only variable but this seems closely correlated with wealth/development. Perhaps cancer is less commonly diagnosed in poorer countries, and perhaps also people in them are more likely to die of something else before cancer gets them.
As someone born and living in Poland, I'm not surprised by high cancer rates. But cancer is still peanuts in comparison to hypertension death rate.
"Polish diet" is basically growth hormone-boosted pork meat, potatoes and some beets or sauerkraut on the side, followed by sugar and flavoring diluted in saturated fat as a dessert. Sweets sold in our stores have different ingredients than the same brands sold in Germany, Austria or Sweden. We breathe coal smoke all winter (I literally moved to seaside for better air; ~35 AQI at the time of writing this) to heat ourselves and export electricity to Germany.
We eat shitton of salt because apart from coal mines, we also have salt mines and government propaganda is that MSG is unhealthy. You can see products marketed as "ZERO GLUTAMATES" on store shelves, and a quick look at nutrients table, shows tons of sodium. Most restaurants oversalt to get you to buy more drinks. To add, Polish people smoke cheap cigatrettes like chimneys. I remember ashtrays even on pharmacies' counters in the 80s - it was unthinkable to extinguish your lit tabacco when entering the pharmacy.
That said, we still have much lower (lower==better) Air Quality Index, usually tens to hundreds times lower than India, and Indian food isn't healthy, too. I'm not sure if the data coming from particular countries is reliable.
I’m certain most people will draw the wrong conclusions from this and use it as evidence of [insert trendy boogieman]. Eg. processed food! Microplastics! Smartphones! Etc!
But this says more about the economic status and healthcare systems of each of these countries than it does anything about actual human cancer rates in said countries.
For example, in the US, the forms of cancer with the highest growing incidence rates (on a population basis) are the hardest to detect and least deadly —- like thyroid cancer, melanoma, etc —- due to advancements in diagnostics. In the past (and in poorer countries today), these things still go unnoticed.
But often not undiagnosed. Many countries in the EU investigate otherwise-unexplained deaths and if they find cancer, it's listed on the death certificate. The data here appears to include that.
So unless there's a problem with over-diagnosis in North America, the EU would appear to have a lower incidence rate.
I agree, the difference is stark and suspicious but we should be a bit more thorough before ignoring data like this. There must be other US vs ... studies out there, right?
Coal pollution and processed pork consumption are possible causes. It's also newly developed, today's GDP per capita is not the same economy Poles have been living in most of their lives.
> Also I suspect most of the countries on the map with low rates are not low with cancer rates but worse public health system. So not very telling
For southern countries, obviously. But the discrepancy between Western Europe (+ Australia and New Zealand) and US+Canada is quite intriguing to say the least.
I can imagine countries with poor healthcare and low life expectancy genuinely have less cancer incidence, because people die before they reach prime cancer age.
Is this controlling for the distribution of the population age?
You should probably computed the incidence, give a persons age in each country. Then, you can take any age distribution, re-weight the incidence in each country with that, and you will get an informative metric.
See just above the map: "This has been age-standardized, assuming a constant age structure of the population for comparisons between countries and over time.". This is what you suggests IIUC?
Of course not, but the average consumption seems to be higher. Particularly (relatively highly processed) dried and cured sausages which are supposedly associated with higher rates of certain kinds of cancers.
Seems like there's absolutely nothing UNIQUELY in common amongst countries with high incidence rates as compared to those with low incidence rates. AI agrees.
I see in Europe Poland is a massive outlier, even compared to much poorer per capita neighbours such as Ukraine or Belarus.
I wonder if (a) it's genuine, (b) if it, is it because some other crucial variables are different (society age, other causes if mortality etc) and finally (c) if it's not a statistical artifact, then why is there more cancer in Poland.
One reason that comes to mind is that Poland very much has a culture of going to see a doctor and getting examined for every little thing. UK for example has an ethos of "take some paracetamol and get on with it". But no idea if that's the actual cause.
I imagine PM2.5 would be a significant cancer contributor in the area.
Poland has somewhere between the worst and second worst air pollution in europe. They are primarily coal-powered (~70% of energy production) and have had issues with private individuals burning trash.
Poland contributes about 8% of the PM2.5 pollution in Denmark despite the wind not being direct - Germany contributing 22% due to a more “favorable” wind, and Denmark itself contributing 20%.
It’s because all other countries fall in that bin. You don’t make a bin for an outlier of that kind. Likely it strips off the crazy outliers at both the top and bottom when making bins.
It's curious to see how, in the face of a higher rate of cancer, many people think that more tests are done in the US and Canada. What about the causes of cancer? Diet, sedentary lifestyle, expensive health system...
This is hacker news, why would people make reasonable inferences? Better to either make it a class issue or better yet make a long winded post about the methodology and condescend everyone into nothing.
Obviously something is wrong, but you’re asking the population of people who for decades has said nothing is wrong. HN is extremely biased in this way. Never come here to learn. Its just to peak into the minds of the rich and autistic.
