3

I'm looking for data/stats related to the number of diagnoses (per year) (per country) (per disease) at least from 2000 up to 2020/2021 for major diseases (like the 5-10 most related to deaths WW). Ideally but not necessarily even (per sex) (per age).

I've had problems finding resources not linked to a single disease or a single country or a single year which aren't just a graph with no underlying data or atomized data. For example regarding cancer with a lot of effort you can gather the data from WHO website(s) https://gco.iarc.fr/projects#registries but overall I couldn't find anything in the form of DB. Any suggestions for sources which wouldn't require thousands of copy-pastes (or about giving up)?

Thanks.

Improve this question
8
  • Is there a reason you need diagnoses specifically, rather than, say, deaths? Diagnoses can be quite a bit trickier and are not typically part of public health records the way that deaths often are. You may have better luck with countries with a centralized health care system but for, say, the US, I don't think these data exist at all, except perhaps by extrapolating from a small sample.
    – Bryan Krause ♦
    Mar 8, 2022 at 15:20
  • Well, (trends of) diagnoses lets you infer/measure/forecast the impact of diseases/healthcare/therapies with more details than the deaths/cause alone. The most basic example is that you can forecast a decrease in deaths/cures (for heart conditions) by a decade or more by noticing that the incidence of obesity decreases. As stated in the question (in that WHO link especially) the data of incidence for cancer are well established though not published in a "good" way for data analisys.
    – Lorenzo
    Mar 8, 2022 at 16:44
  • The US has large datasets (available for purchase) regarding healthcare claims. These datasets have well known inherent biases and obviously they don't apply to the whole world. How do you intend to control for uneven access to diagnostic services among countries?
    – Ian Campbell ♦
    Mar 8, 2022 at 18:51
  • 1
    Does the WHO Global Health Observatory help? They have a data API and a bunch of pages on stats for tons of health conditions.
    – bob1
    Mar 8, 2022 at 21:08
  • @IanCampbell Pay walls for data that would be invaluable for HC research (insurance as well...): sounds like a great idea.. OFC there is uneven access but not only that: out of trend data or new trends require explainations. The sudden deceleration/acceleration in cancer diagnoses before/after 65y in the US (after Medicare) is a major example. Comparing "similiar" countries might mimic "control groups' behaviour". Guessing for impacts due to demogtraphic or organizational changes, conflict of interests, healthcare standards, etc is a good idea as well... 1/2
    – Lorenzo
    Mar 9, 2022 at 9:32

1 Answer 1

Reset to default
1

One possibility for access of data on diseases from the WHO is through their Global Health Observatory data site.

This site contains data on many causes of death, rates of illness etc, and presents data in many ways, including presenting summary pages for a range of conditions by so-called "themes" (lower down on front page). It also has dash-boards near the bottom of the front-page for some WHO initiatives and a COVID dashboard too.

The data behind the information is also accessible via API through their Odata portal, so can be web-scraped. However, the data is divided into datasets, so a world-wide "most common diseases" type search might well require accessing several different datasets.

Improve this answer
2
  • 1
    Spent a good 10 hours and it doesn't seem to have what I was looking for in a "nice" form. I'll parse all the metadata to identify all the relevant/missing tables in the next weeks and let you know in case of success.
    – Lorenzo
    Mar 22, 2022 at 21:15
  • @Lorenzo that's what I feared. Generally when these sorts of data are collected, they are gathered into specific datasets without much overall aggregation, not least because it can make the data unwieldy.
    – bob1
    Mar 22, 2022 at 21:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.