Major Study Finds Atkins Diet Increases All-Cause Mortality Rate

I already distrust the validity of this article because it says "increased" and "decreased" instead of giving percentages. On further investigation, I am brought to the article in Annals of Internal Medicine.

http://www.annals.org/content/153/5/289.abstract

--Snip--Results: The overall low-carbohydrate score was associated with a modest increase in overall mortality in a pooled analysis (hazard ratio [HR] comparing extreme deciles, 1.12 [95% CI, 1.01 to 1.24]; P for trend = 0.136). The animal low-carbohydrate score was associated with higher all-cause mortality (pooled HR comparing extreme deciles, 1.23 [CI, 1.11 to 1.37]; P for trend = 0.051), cardiovascular mortality (corresponding HR, 1.14 [CI, 1.01 to 1.29]; P for trend = 0.029), and cancer mortality (corresponding HR, 1.28 [CI, 1.02 to 1.60]; P for trend = 0.089). In contrast, a higher vegetable low-carbohydrate score was associated with lower all-cause mortality (HR, 0.80 [CI, 0.75 to 0.85]; P for trend ≤ 0.001) and cardiovascular mortality (HR, 0.77 [CI, 0.68 to 0.87]; P for trend < 0.001). --Snip--

Now, I am by far not a doctor, a chemist, nothing of that nature but, it says "modest increase" which, doesn't sound so threatening. The numbers are a bit confusing. I was trying to read someone's explanation of it but, it's difficult for me to understand.

The original question was essentially what "The relative risk of squamous cell carcinomas among long-term incense users was 1.8 (95% confidence interval [95% CI], 1.2-2.6; P = .004) in the entire respiratory tract." means.

The answer:

--Snip--p stands for probability - in this case, there are only 4 chances in 1000 that the finding of an increased risk is false (happened by chance in the trial)

Relative risk compares the group of interest to some criterion group (here, probably non-users or short-term users). The statistical power of the test (usually based mainly on sample size) was sufficient to be able to say that 19 out of 20 tests (i.e. 95%) would give an estimate of the increased risk among long-term users, compared to the criterion group, as somewhere in the range 1.2 times greater to 2.6 times greater, with 1.8 as the best estimate withn that range. So there is a 5% chance that the risk is outside that range (the 95% confidence interval) - it might even be a reduced risk rather than an increased one. However, a 95% confidence interval is conventionally accepted for many testing purposes - up to you to decide if it's appropriate. --Snip--

I tried figuring it out. I'm dreadful with math. Maybe someone else can come along and give a better explanation.