ARS (opens in new window)
Interns - 2009-05
Please open utxr.org in a new tab or window.
Searching and putting it all together. The last session of our year.
There are three types of searching:
- Point of care. Use PIER, UpToDate, and your favorite drug reference.
- Meta-analysis. Get a library or other expert to help you if you are preforming a meta-analysis.
- Searching at the end of the day when you build your expertise.
We are addressing #3 today. Why should you be a good searcher? Answer.
You can probably maintain a good knowledge based by:
- being facile with UpToDate
- keeping up by using emails from FirstWatch and a few favorite journals such as NEJM.org
- using colleagues and consultants as needed
But:
- 20% of primary care visits have a guess being made by the physician (calculated from Ely, 1999)
- Yet UpToDate does not cover everything (PMID 17641755)
- Some of you will become peer leaders (in academics or private practice)
- Hopefully everyone else will want to be expert in a few dieases that they see the most often. For example, even a primary care doc has expertise that rivals specialists for common diseases (PMID 7666878).
Goals for today
- Knowledge of choice in search engines
- Understand the types of articles to search for
- Ability to choose search terms
- Ability to find and use filter/focus/hedges
- Reinforce goals of prior sessions
- Using PICO to compare conflicting articles
- Numeracy
Overview of searching
What search engine for MEDLINE do you want to use?
- Google?
- Try Googling "The treatment of patients with acute coronary syndromes is optimal when the intensity of therapy" (or click here)
- Why did Google retrieve the first link?
- OVID?
- PubMed?
- SUMSearch?
- Many others
What types of articles do you want to find?
- Systematic reviews (this includes practice guidelines if they are based on systematic review of relevant literature)
- Original studies that are more recent than the search date of the systematic reviews
- And possibly classic studies. Even if these are included in the systematic reviews, understanding the quality of the classic study helps determine how well the clinical problem has been studied.
Question 1 - Should you use steroids for septic shock?
- What should your search terms be?
- What filter/focus/hedge should you use?
| Systematic reviews |
Practice guidelines |
Major original studies |
- PMID: 19416302 (Minneci, 2009. Updates earlier review below)
Older reviews that do not include CORTICUS
|
- PMID: 18158437 (Crit Care Med, 2008. Only had access to the abstract of CORTICUS. Does not consider relative adrenal insufficiency.)
|
- PMID: 18184957 (CORTICUS, 2008)**
- PMID: 12186604 (French study, 2002)**
- PMID: 3306374 & 2888017 (classic rcts from 1987 that suggested harm or no benefit from high dose steroids)
|
Evidence table for septic shock
Question 2: What is the role of CT angio in excluding coronary disease?
Who does UpToDate say we can exclude coronary disease in with a negative CT angio?
Now searching MEDLINE:
- What should your search terms be?
X-Ray Computed Tomography AND Coronary Angiography
- What filter/focus/hedge should you use?
- Find a report of the sensitivity and specificity.
- Using http://clinical.uthscsa.edu/calc, is UpToDate correct?