Improving search filter development: a study of palliative care literature
2007

Improving Search Filters for Palliative Care Literature

Sample size: 418 publication Evidence: moderate

Author Information

Author(s): Ruth M Sladek, Jennifer Tieman, David C Currow

Primary Institution: Flinders University

Hypothesis

Can the sensitivity of a search filter for palliative care literature be improved by analyzing incorrectly excluded citations?

Conclusion

The sensitivity of the search filter was successfully increased from 46.3% to 64.7%, but this came at the cost of decreased precision.

Supporting Evidence

  • The original search filter had a sensitivity of only 45.4%, meaning many relevant articles were missed.
  • Adding six new MeSH terms improved the sensitivity to 64.7%, but precision dropped significantly.
  • The thematic analysis revealed that none of the themes were intrinsically palliative, indicating a need for better definitions in search strategies.

Takeaway

The study found a way to make it easier to find important palliative care articles by changing the search terms used, but it also made some irrelevant articles show up.

Methodology

The study involved thematic and frequency analysis of MeSH terms from incorrectly excluded citations to improve the search filter's sensitivity.

Potential Biases

The study may have overestimated sensitivity and underestimated precision due to database limitations.

Limitations

The study's findings may be affected by changes in the OVID Medline database over time, which could impact the accuracy of MeSH term frequencies.

Digital Object Identifier (DOI)

10.1186/1472-6947-7-18

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication