Skip to main content
Effective Health Care Program

Impact of Contacting Study Authors on Systematic Review Conclusions: Diagnostic Tests for Hepatic Fibrosis

White Paper

People using assistive technology may not be able to fully access information in these files. For additional assistance, please contact us.

Structured Abstract

Background

In 2013, the Pacific Northwest Evidence-based Practice Center conducted a systematic review of screening and diagnostic tests for hepatic fibrosis or cirrhosis in patients with chronic hepatitis C viral infection. However, 17 of the 172 included studies reported diagnostic accuracy results that were discordant from 2 × 2 tables. In addition, 60 studies did not report adequate data to construct 2×2 tables or include in the analysis. This study explores the response rate and impact of contacting authors to provide data that were otherwise missing or discordant.

To determine the efficacy and impact of contacting authors to clarify discordant data or obtain missing data for a systematic review on screening and diagnostic tests for hepatic fibrosis or cirrhosis in patients with chronic hepatitis C viral infection.

Methods

Sixty-six corresponding authors were sent letters requesting additional information or clarification of data from 77 studies that had discrepancies in the data reported or that provided insufficient data to construct 2 × 2 tables. Data received from authors were pooled with data included in the previous review and the diagnostic effect analyzed.

Results

Of the 66 authors, 45 (68%) were successfully contacted and 28 (42%) provided additional data for 29 studies. All authors who provided data did so by the third written request for information. Authors of more recent studies were significantly more likely to be located and provide data compared to authors of older studies. In general, inclusion of the additional 29 studies had only minor effects on the diagnostic accuracy meta-analyses. However, the additional data resulted in reclassification of the utility of three tests.

Conclusions

The results suggest that contacting authors to obtain additional data will likely be successful. However, there was no clear trend in the impact of new data on measures of diagnostic accuracy. As a result, it is unclear whether the time-intensive practice of contacting authors is worth the effort. It would be more effective to require authors of studies to provide 2 × 2 tables within the published manuscript for transparency and to facilitate additional analyses.