There is a newer version of this record available.

Preprint Open Access

# Controlling for baseline telomere length biases estimates of the rate of telomere attrition.

Bateson, Melissa; Eisenberg, Dan T. A.; Nettle, Daniel

### Dublin Core Export

<?xml version='1.0' encoding='utf-8'?>
<oai_dc:dc xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ http://www.openarchives.org/OAI/2.0/oai_dc.xsd">
<dc:creator>Bateson, Melissa</dc:creator>
<dc:creator>Eisenberg, Dan  T. A.</dc:creator>
<dc:creator>Nettle, Daniel</dc:creator>
<dc:date>2018-07-24</dc:date>
<dc:description>Preprint, associated R scripts and dataset.

Summary: In analyses of longitudinal changes in leukocyte telomere length (LTL) it is common practice to control statistically for baseline LTL. However, theoretical considerations arising from collider bias suggest that this practice could lead to overestimation of the difference in LTL attrition between groups that have experienced different exposures. We used simulated LTL data to explore whether adjusting for baseline LTL results in biased estimates of the true difference in LTL attrition between individuals with different exposures using smokers and non-smokers as an example. We show that if baseline LTL is shorter in smokers than non-smokers and LTL measurement error is non-zero, then adjusting for baseline LTL results in overestimating the true difference in telomere attrition between smokers and non-smokers. The size of this latter bias increases with increasing LTL measurement error. Since it is a robust finding that smokers have shorter baseline LTL than non-smokers and LTL measurement error is substantial, we conclude that the type 1 error rate for reports of effects of smoking on telomere attrition is likely to be above 5%. Using real data from seven longitudinal cohorts we show that in line with our simulation results, the estimated difference in attrition between smokers and non-smokers is greater in models controlling for baseline LTL. Furthermore, as predicted by our simulations, the size of this latter difference is positively associated with signatures of LTL measurement error. On the basis of our analyses we recommend that models of LTL attrition should not control for baseline LTL. Although we have couched our analysis in terms of the effects of smoking, our findings are likely to have general relevance to other factors studied in relation to telomere attrition. Many claims of accelerated LTL attrition in individuals exposed to disease, stress or adversity will need to be re-assessed.</dc:description>
<dc:identifier>https://zenodo.org/record/1320379</dc:identifier>
<dc:identifier>10.5281/zenodo.1320379</dc:identifier>
<dc:identifier>oai:zenodo.org:1320379</dc:identifier>
<dc:language>eng</dc:language>
<dc:relation>info:eu-repo/grantAgreement/EC/H2020/666669/</dc:relation>
<dc:relation>doi:10.5281/zenodo.1009086</dc:relation>
<dc:rights>info:eu-repo/semantics/openAccess</dc:rights>
<dc:subject>telomere attrition</dc:subject>
<dc:subject>telomere length</dc:subject>
<dc:subject>bias</dc:subject>
<dc:subject>measurement error</dc:subject>
<dc:subject>regression to the mean</dc:subject>
<dc:title>Controlling for baseline telomere length biases estimates of the rate of telomere attrition.</dc:title>
<dc:type>info:eu-repo/semantics/preprint</dc:type>
<dc:type>publication-preprint</dc:type>
</oai_dc:dc>

626
486
views