Since the publication of our meta-analysis it has come to light that we inadvertently categorised one study (Gale et al. 2013) as a cross sectional study instead of its true design as a longitudinal study. We have recalculated the analyses affected which is isolated to the cross sectional longitudinal biomarkers of C reactive protein. The updated analyses have not significantly changed our findings regarding CRP or the narrative of the paper in any perceptible way. Specifically, the new OR for CRP longitudinal studies is now 1.08 (95%CI 0.87–1.34) compared to a previous value of 1.06 (95%CI: 0.78–1.44) (please see the new Fig. 1a). Similarly, after removing the study of Gale et al., the cross-sectional findings for CRP are now SMD = 0.38 (95%CI: 0.26–0.50) which was previously 0.25 (95%CI: 0.02–0.49). We wish to apologise to Gale et al. (2013) for this inadvertent mistake and reassure readers that we updated our analyses as soon as this mistake was identified. Moreover, we wish to reiterate that there have been no perceptible changes in any effect sizes regarding CRP and the narrative and meaning of our paper remains unchanged. Fig. 1a Relationship between serum C - reactive protein levels (a) at baseline and incident frailty, adjusted for potential confounders.
|Appare nelle tipologie:||1.1 Articolo su rivista|