Reporting bias: higher diagnostic accuracy is linked to faster publication in imaging research

ID: 

186

Session: 

Poster session 1

Date: 

Sunday 16 September 2018 - 12:30 to 14:00

All authors in correct order:

Dehmoobad Sharifabadi A1, Korevaar DA2, McGrath T1, van Es N2, Nguyen F1, Alan Frank R1, Stanley C3, Cherpak L1, Dang W1, Salameh J1, McInnes M1
1 University of Ottawa, Canada
2 Academic Medical Centre, The Netherlands
3 Dalhousie University, Canada
Presenting author and contact person

Presenting author:

Anahita Dehmoobad Sharifabadi

Contact person:

Anahita Dehmoobad Sharifabadi
Matthew McInnes
Abstract text
Background:
Time lag bias occurs when various aspects of a research study, such as the strength of the results, impact the timing of its publication. Although this phenomenon has been well studied for therapeutic intervention studies, there is less evidence for diagnostic accuracy research in imaging journals. Prior work has documented delayed publication of studies with inferior test accuracy estimates.

Objectives:
The aim of this study was to evaluate whether higher reported accuracy estimates are associated with a shorter time to publication among published imaging diagnostic accuracy studies.

Methods:
We included primary imaging diagnostic accuracy studies, included in meta-analyses from systematic reviews published in 2015. For each primary study, we extracted accuracy estimates, participant recruitment periods and publication dates. Our primary outcome was the association between Youden's index (sensitivity + specificity -1, a single measure of diagnostic accuracy) and time to publication.

Results:
Our sample included 55 meta-analyses and 781 primary studies, of which 238 (30%) were excluded due to missing study dates Youden's index was negatively correlated with time from completion to publication (rho = -0.11; P = 0.009) (Figure 1). This association remained significant in multivariable Cox regression analyses after adjusting for seven study characteristics: the hazard ratio for publication was 1.09 (95% confidence interval 1.03 to 1.16, P = 0.004) per unit increase for logit-transformed estimates of Youden's index (Table 1).

Conclusion:
Imaging diagnostic accuracy studies with higher accuracy estimates were weakly associated with a shorter time to publication.

Patient or healthcare consumer involvement:
Reporting bias can seriously affect outcomes in systematic reviews, which is particularly problematic given their high regard in the hierarchy of evidence and their role in facilitating clinical decisions. The weak association between accuracy estimates and time to publication identified in imaging research suggests that the impact on diagnostic accuracy systematic review findings may not be as important as it is in evaluations of therapeutic interventions.

Attachments: 

PDF icon Cochrane2018-Table 1.pdf
PDF icon Cochrane2018-Figure 1.pdf

Relevance to patients and consumers: 

Time lag bias occurs when various aspects of a research study, such as the strength of the results, impact the timing of its publication. Our evaluation of time lag bias indicates slower publication of studies with lower accuracy estimates in diagnostic imaging research. Although this delay may influence systematic reviews that synthesize the results of these studies, the association was only weak. As such, clinicians may be less concerned for time lag bias when searching diagnostic imaging literature for clinical questions than they would be for those pertaining to therapeutic interventions.