Introduction: Clinicians raise at least one question for every two patients they see, but search for an answer to less than half of these questions, and rarely use evidence-based resources. One barrier to evidence-based practice is doubt that the search would yield an answer, and we found insufficient evidence to refute this concern. This study aims to identify what proportion of clinical questions in primary care can be answered with online evidence-based practice resources, and what proportion of these can be answered with pre-appraised evidence.

Materials and methods: Cross-sectional study in two primary care practices. The inclusion criteria were family doctors, generalists and residents working in 2 selected practices. We collected a total of 238 questions from 19 family medicine specialists, 9 family medicine residents and 3 generalist doctors. Doctors were asked to record any clinical question that arose during 4 days of appointments. The primary outcome was the proportion of clinical questions answered with online evidence-based practice resources. The secondary outcome was the level of evidence needed to reach to find the answers (clinical summaries, systematic guidelines, systematic reviews or primary studies), according to Haynes' pre-appraised evidence pyramid model.

Results: 191 of the 206 valid clinical questions could be answered with online evidence-based practice resources (92.7% [95% CI 88.3%-95.9%]). Most of these questions (90.8% CI 95% 85.9%-94.4%) were successfully answered using clinical summaries (BMJ Best Practice, DynaMed or UpToDate), with a median search time of 4 minutes (range 1-16.5).

Conclusions: Contrary to clinician's beliefs, the majority of clinical questions can be answered with online evidence-based practice resources, and most of them with pre-appraised evidence. This study could encourage family doctors to increase the use of clinical summaries. Furthermore, these results highlight the importance of teaching how to search for and apply pre-appraised evidence.