Self-reported, so *throws hands up in air*

The self-administered questionnaire included information on personal background factors and frequency of ramen consumption. The questionnaire was distributed and collected by mail. The follow-up period was until December 2023. Information on total deaths was obtained from death certificates.

In the question on the frequency of ramen intake on the survey form, the average frequency of ramen intake over the past year was asked in nine levels (<1/month, 1–3 times/month, 1–2 times/week, 3–4 times/week, 5–6 times/week, 1 time/day, 2–3 times/day, 4–6 times/day, and ≥7 times/day). Since the number of respondents in the groups of ≥5–6 times/week was small, they were divided into four groups (<1/month, 1–3 times/month, 1–2 times/week, and ≥3 times/week). The amount of noodle soup intake was determined by the question, “How much soup do you have from ramen, udon, and soba noodles?” with answers on five levels (almost never, one-third, half, two-thirds, and almost all). Here, they were divided into two groups; “≥50%” or “<50%.”

https://www.sciencedirect.com/science/article/pii/S127977072500168X

But I bury the lede.

In the unadjusted analysis, the hazard ratio was significantly higher trend** for the “≥3 times/week” group (hazard ratio [HR]; 1.69, 95% confidence interval [CI]; 0.94–3.03) compared to the “1–2 times/week” group. Multivariate analysis adjusted for age, sex, smoking, alcohol consumption, amount of noodle soup consumed, diabetes, hypertension, and dyslipidaemia, the “≥3 times/week” group showed a non-significantly increased risk for mortality

"If we don't adjust for other factors, it's significant. If we adjust for other factors, it's non-significant." You don't hate journalists enough.