The importance of the Biological Variation Data Critical Appraisal Checklist when designing experimental studies on biological variation. Comparison of two papers reporting biological variation results for S100-beta and neuron-specific enolase proteins


Cajano G., Guerra E., Locatelli M., Ceriotti F., Diaz-Garzon Marco J., Fernandez-Calle P., et al.

BIOCHIMICA CLINICA, cilt.43, ss.59-66, 2019 (ESCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 43 Konu: 1
  • Basım Tarihi: 2019
  • Doi Numarası: 10.19186/bc_2019.001
  • Dergi Adı: BIOCHIMICA CLINICA
  • Sayfa Sayısı: ss.59-66

Özet

The Biological Variation Data Critical Appraisal Checklist (BIVAC) has been designed to evaluate biological variation (BV) studies and the reliability of the associated BV estimates. To illustrate its utility, two studies delivering within-subject BV (CVI) data for S100-beta protein and neuron-specific enolase (NSE), markers typically used for melanoma and neuroendocrine tumors, respectively, were appraised using BIVAC. Data from the European Biological Variation Study (EuBIVAS) and the recently published Johnson et al. study (ref n 11) were scored using the 14 BIVAC quality items (QI), with alternatives A, B, C and/or D to verify whether the elements required to obtain reliable BV data, were present and appropriately documented. Grade A indicates compliance with all the QIs and D indicates non compliance. The sizes of the confidence interval (CI) around the CVI estimates were also compared. Johnson's study received a BIVAC grade C, EuBIVAS a grade A. EuBIVAS is a large scale study, with 1609 and 1728 results for NSE and S100-beta, respectively. In Johnson's study, only 40 results were available. The EuBIVAS CVI estimates [NSE, 10.9% (10.3-11.5); S100-beta, 10.2% (9.6-10.7)] were clearly lower than Johnson's CV(I)s [NSE, 22.1% (9.9-34.3); S100-beta, 18.9% (8.5-29.4)]. The overlapping CI between the two estimates are caused by Johnson's CI being about 20 times larger than the corresponding EuBIVAS CI. It is likely that studies that do not comply with all BIVAC QI deliver less reliable, and possibly too high, CVI estimates. Adherence to the BIVAC ensures safe clinical application of BV estimates.