This is a manuscript that examines accelerated aging in childhood and the relation to adulthood accelerated aging using two studies, one community sampled and the other a national sample starting in adolescence. My primary critique of this study is that there are no psychological variables that are studied. So I was surprised that this was submitted to Psychological Science instead of Developmental Science or a Gerontology or Aging journal, which seems more appropriate for the topic matter.

My other concern was the use of latent profile analyses. The paper suggests that there are multiple variables that are known to be related to faster aging--why would you need to use a latent profile analyses. Indeed, the supplemental materials suggest that it returns similar findings. This also suggests that this could have been a pre-registered study especially since the Profile Analysis (Table S1) doesn't show a clear 3 class model and low precision. I don't have confidence that the three groups accurately represent the underlying data. Also, there is not enough information provided on the Missing Data Analyses. If using FIML were auxiliary variables used like the Childhood SES and IQ that is noted in the Supplemental Files were related to loss of sample across time in the Dunedin sample.

There is very little information regarding the analyses of the Add Health dataset which is a complex, population data set. Very little of the power of this data set is being used in the study and only to try and demonstrate a replication where the two samples and data are quite different.