Marijuana use holds three-fold blood pressure death risk: study

I must be misreading that.

Example: Is there a difference between someone who smokes a joint every day for 10 years and someone who eats an edible once every two months for 10 years?

I imagine the answer to this is yes, but it's one of those things you need data to back up. The 'findings' have been so oversimplified in this news article that the portrayed results are likely only unreliable. This makes for great politicized sound-bytes but doesn't really help anyone make a real conclusion or decision.

This seems to be the source for the article, I doubt they even interviewed:

https://www.escardio.org/The-ESC/Press-Office/Press-releases...

The study: http://journals.sagepub.com/doi/10.1177/2047487317723212

Looking at methodology:

We linked participants aged 20 years and above, who responded to questions on marijuana use during the 2005 US National Health and Nutrition Examination Survey to data from the 2011 public-use linked mortality file of the National Center for Health Statistics, Centers for Disease Control and Prevention. Only participants eligible for mortality follow-up were included. We conducted Cox proportional hazards regression analyses to estimate hazard ratios for hypertension, heart disease, and cerebrovascular mortality due to marijuana use. We controlled for cigarette smoking and other relevant variables.

And results:

Of the 1213 eligible participants 72.5% were presumed to be alive. The total follow-up time was 19,569 person-years. Adjusted hazard ratios for death from hypertension among marijuana users compared to non-marijuana users was 3.42 (95% confidence interval: 1.20–9.79) and for each year of marijuana use was 1.04 (95% confidence interval: 1.00–1.07).

My own naive interpretation here is that they drew their conclusions from a sample of ~334 deaths (1213 - 72.5%) by people who filled out a survey in 2005, using only the original survey at the beginning of the 6 year period and their public-use mortality file..

Ratio of sample to population size is mostly irrelevant for statistics. The strength of conclusions from a sample is calculated based on an infinite population size.

It's not even this, it reads as if it was a simple "Have you ever smoked marijuana?" and that's what the conclusion was drawn from.

"Yankey said were limitations in the way marijuana use was assessed -- including that researchers could not be sure whether people had used the drug continuously since they first tried it."

That would be like asking "Have you ever intentionally hurt someone's feelings?" and concluding that people were three times likely to be anti-social.