Thank you for your comment concerning the following sentence, “However, the number of children for women deceased at ages 70-79 is not 5.6, but 5.1 (Table 3), while that for women deceased at ages 60-69 is 5.6”. It has allowed us to find an error in the text (page 1056, paragraph 5 [...]

View More

Thank you for your comment concerning the following sentence, “However, the number of children for women deceased at ages 70-79 is not 5.6, but 5.1 (Table 3), while that for women deceased at ages 60-69 is 5.6”. It has allowed us to find an error in the text (page 1056, paragraph 5 of the manuscript). The sentence should read “Table 3 shows that centenarian women (excluding childless women - we restricted the analysis to women that had experienced a ‘birth event’) had an average of 5.4 children compared with 5.6 children for women in the same cohorts deceased at ages 60–69 and 5.8 children for women in the 1905–10 cohorts.”

The values reported in Table 3 are correct.

As to the remaining comments, we partly accept your criticisms.

We would like to underline that our manuscript primarily provides an overview of the AKeA2 study, with particular focus on data collection concerning family genealogies of the Sardinian centenarians and their controls, and on the quality of these data. The aim of this work is simply to present the research data and some preliminary results, as specified in the titles of both the manuscript and of paragraph 5. In our descriptive analysis we presented the results that seemed to us most interesting for further investigation, even if some differences are rather small or of low statistical significance.

The wealth of information gathered in the AKeA2 survey can be used to study various hypotheses concerning longevity and centenarian characteristics, especially when the linkage between the demographic variables and the socio-economic and health information collected in the AKeA2 survey is available.

In the near future, when we move from descriptive analyses to explanatory analyses, much more attention will be devoted to the choice of suitable statistical methods, applying multivariate analyses or taking into account your suggestions, depending on the specific hypothesis being investigated.

View Less

Lipsi et al (2015) have published a descriptive study of centenarians of Sardinia in Demographic Research. One of their main conclusions is that “centenarian women have on average fewer children, and at an older age, particularly for their last child”. However, these conclusions are not warranted.