Japan's red yeast rice scandal put kidneys in the spotlight. In Taiwan, 67,876 people on a prescription version reached dialysis far less often, but the comparison group was much sicker

In early 2024, people in Japan who had taken a red yeast rice supplement, Benikoji CholesteHelp, began turning up with damaged kidneys. Suspicion fell on puberulic acid, a by-product of a Penicillium mould that got into the fermentation, rather than on the red yeast itself. That left a wider question hanging: is red yeast rice hard on the kidneys when nothing has gone wrong in the factory? A team from Taipei Medical University and China Medical University went looking in Taiwan's national insurance records, and published what they found in PLOS One on 27 October.

They compared 67,876 adults with high cholesterol who were prescribed a medical-grade capsule, LipoCol Forte, between 2010 and 2014 with 90,836 who were prescribed the statin lovastatin and never the capsule, following both groups through 2019. The outcome was kidney failure needing long-term dialysis. It came at 0.9 cases per 1,000 person-years on the capsule and 3.7 on lovastatin, an adjusted hazard ratio of 0.36. Taken at face value, that is a clean bill of health and then some, with risk about two-thirds lower. Against atorvastatin, rosuvastatin and three other statins the ratio ran from 0.16 to 0.35.

The two groups were not alike to begin with. At the start, 34.6% of the lovastatin group already had chronic kidney disease against 14.4% of capsule users, 52.7% had diabetes against 23.8%, and 56.5% were 60 or older against 24.8%. Claims data hold no creatinine or eGFR, so nobody can say how close anyone's kidneys already were to failing, and no BMI or smoking either. The authors themselves raise a "healthy user" effect, and the subgroups fit it: the ratio was 0.08 among people with no other illnesses on the comorbidity score, but 0.77 among those with the highest scores. Timing points the same way. In a supplementary analysis of 21,881 matched pairs, 496 of the 705 lovastatin cases came in the first two years, about 11.6 per 1,000 person-years against 1.4 on the capsule. By our sums from the paper's table, the gap after that narrows to roughly 1.2 against 0.7. On our reading, that fits some people being started on lovastatin with kidneys already failing, and the matched baseline table does not list chronic kidney disease among the things it balanced.

The paper's other supports are weaker than its abstract suggests. It reports a "dose-dependent reduction", but the ratios are 0.40 for one prescription, usually a 30-day supply, 0.39 for two, 0.31 for three, 0.43 for four and 0.32 for five or more, which is hardly a slope. The check that dropped cases from the first months moved the ratio from 0.36 to 0.72, and it then sat at exactly 0.72, with the same confidence interval, at every cut-off from two months to six. Some person-year counts do not reconcile either (see the note under the data). The authors do call for randomised trials and caution that the findings may not apply to other products. Above all, a capsule holding 5.7 mg of lovastatin was compared with lovastatin, so this is not a test of a supplement against a drug. What the study can fairly say is narrower: it found no sign that this one prescription capsule raised dialysis risk compared with lovastatin. It says nothing about shop-bought supplements, and it does not show that red yeast rice protects kidneys.