That's not what I said, though. There are plenty of reliable biomarkers. They consistently relate with disease in an expected way across populations. They just

That's not what I said, though. There are plenty of reliable biomarkers. They consistently relate with disease in an expected way across populations. They just aren't, individually, powerful. They don't do a great job discriminating who will get disease and who won't - but using them is still much better than not using them.