The formula has a problem dependent on the precision of CRP tests. Because many historic tests had a minimum measureable quantity of 0.6 mg/L then the source data predicts things wrongly with values below that.
5 Likes
As mentioned by @John_Hemming, low CRP is an issue with Levine’s PhenoAge. There are 2 reasons for that: It’s a linear model so it’s only valid around its mean values and worse in the case of CRP they use the log of the CRP rather than the CRP value itself. That’s a problem because it goes to negative infinity at 0 so you can reduce the age by any amount. For instance if I use 1e-30 in the CRP value I get an age reduction of 76 years and a biological age of -15 (minus 15). They should have used log(CRP+0.5) or something like that to avoid that issue with low values.
4 Likes
Life Extension recently added a Phenotypical Age lab test based on statistical models of mortality risk amongst the NHANES III & IV cohorts. Price: $75.
1 Like
This seems to be the same one used in the well-known spreadsheet. I modified this one to have 3 tabs with values of “average” where biological = phenotypic age, plus “healthy” and “unhealthy” where values are physiologically normal but optimal and suboptimal respectively.
Pretty cool to play with IMO
0. Levine Phenoage calculator.xlsx (19.4 KB)
3 Likes
Thanks for this @relaxedmeatball. I missed it. I thought some members might find this $75 price (25% off) might be attractive if they have not made this calculation or want to do it again inexpensively.
1 Like
