Finding the literature 2

Nikhil Dhawan, MD

10-22-24

Created: 2024-10-22 Tue 13:53

1. Goals of lecture

  • Able to search for narrative reviews
  • Able to search for clinical trials and systematic reviews
  • Be able to use tags and boolean operators (AND, OR)
  • Can use narrative review search forms found here.

2. Questions

  • How have the searches been?
  • Search the literature is a skill that takes practice.
  • You can't learn how to search the literature in a textbook.
  • All of these tips are suggestions and you can break these rules and search anyway you want.

3. Outline

  • Find a review
  • Skim review to get background information and words for PICO search
  • Look at systematic reviews to find more articles and search terms
  • Create synonyms for PICO.
    • Combine synonyms with OR and surround with parentheses
    • For example, (bipolar OR "manic depressive")
  • You can find an outcome measure in the review or use my table
  • Combine the PICO search terms and search for clinical trial by filtering by trial.
  • For example, (bipolar OR "manic depressive") risperidone "young mania rating scale"

4. Finding a review

  1. Google scholar and search "lithium review"
  2. Also find reviews with pubmed looking at high impact journals (JAMA, Lancet, NEJM, biological psychiatry, molecular psychiatry) and searching for recent reviews.
  3. UpToDate

5. Finding narrative reviews in big journals

6. Use this form to search

  • Use quotes for multiple words
  • (""[ti]) ("The New England Journal of Medicine"[Journal] OR JAMA[Journal] OR Lancet[Journal]) review[filter]

Or if you want to add more journals you can add do this search string (""[ti]) ("The New England Journal of Medicine"[Journal] OR JAMA[Journal] OR Lancet[Journal] OR "Molecular Psychiatry"[Journal] OR "Biological Psychiatry"[Journal]) review[filter]

7. Review searches

8. Narrative vs systematic review

  • Narrative review is an expert review with broad information on the topic such as prevalence, diagnosis, treatment options, pathophysiology, epidemiology, natural history of the disease, and risk factors.
  • A systematic review is similar to a study with papers or trials as the participants.

9. Reasons to find a narrative review

  • Get a basic understanding of the topic
  • To use their bibliography to find articles
  • To determine which words are used in the literature so you can search for them
  • Look for population terms, interventions, and outcome measures that are used to help with searches.

10. A faster way

  • You can also just copy this text into the search to search for the journal.
  • I am using pubmed tags to search.
  • ("The New England Journal of Medicine"[Journal] OR Lancet[Journal] OR JAMA[Journal])
  • For a narrative review you can also use the title tag [title]
  • And to filter by review you can use the filter tag with review[filter]

11. My favorite tags

  • Author [AU] or [Author]
  • Entry Date [EDAT]
  • First Author Name [1AU]
  • Journal [TA] or [Journal]
  • Language [LA]
  • Last Author Name [LASTAU]
  • PMID [PMID]
  • Publication Date [DP]
  • Publication Type [PT] = e.g., review, "clinical trial"
  • Title [TI] or [Title]
  • Title/Abstract [TIAB]

12. Finding a review article

(""[ti]) ("The New England Journal of Medicine"[Journal] OR JAMA[Journal] OR Lancet[Journal]) review[filter]

Or if you want to add more journals you can add do this search string (""[ti]) ("The New England Journal of Medicine"[Journal] OR JAMA[Journal] OR Lancet[Journal] OR "Molecular Psychiatry"[Journal] OR "Biological Psychiatry"[Journal]) review[filter]

13. Let's do an example

  • Try to find a narrative review on
    • bipolar disorder

14. Find bipolar disorder review

15. Skimming article we find these two trials

  • Looking through the article you find a great table that compares the different medications and their indications for bipolar disorder
  • You also see this article cited that may be interesting.
  • Tondo L, Abramowicz M, Alda M, et al. Long-term lithium treatment in bipolar disorder: effects on glomerular filtration rate and other metabolic parameters.  Int J Bipolar Disord. 2017;5(1):27. https://doi.org/10.1186/s40345-017-0096-2

16. Looking at Bibliography

  • A network meta-analysis of 101 RCTs that included 20,081 participants treated for 2 to 16 weeks reported that, compared with placebo, the following treatments were associated with significantly better response rates, defined as a 50% reduction in depressive scores: olanzapine plus fluoxetine (56.7% response rate), quetiapine (49% response rate), lurasidone (48.5% response rate), cariprazine (44.5% response rate), lamotrigine (44.4% response rate), olanzapine (44.3% response rate), and lumateperone (42.4% response rate). Other drugs, such as lithium (38.1% response rate), aripiprazole (35.2% response rate), and paroxetine (42.1% response rate), were not associated with significantly better response rates than placebo (34.7% response rate).

Comparative efficacy and tolerability of pharmacological interventions for acute bipolar depression in adults: a systematic review and network meta-analysis

17. Looking at bibliography

18. Systematic Reviews/Meta Analysis

  • Often we are looking for one correct answer, but in reality choosing the correct medication doesn't have one correct answer.
  • Systematic reviews/meta-analyses make one think that there is one best answer, but you need to be able to evaluate the literature.
  • The patients in the trials need to generalize to your patients.
  • Often meta-analyses in psychiatry can group patients that are really different.
  • For example, patient with and without comorbidities or of varying level of severity or inpatient and outpatient.

19. Finding RCT's/PICO search

  • Look for words that could be used in a PICO search
  • P - population or diagnosis
  • I - intervention
  • C - comparator/control (optional)
  • O - outcome (scales)

20. PICO Search basics

  • So each of the categories for PICO can be combined with OR's
  • For the P or population terms, you come up with multiple synonyms and combine them with an OR.
  • So if you are search for the evidence for quetiapine in mania

20.1. OR boolean operators

  • When you use the 'OR' operator between search terms PubMed will look for either of those terms.
  • For example, (children OR pediatric) will search for articles with either "children" or "pediatric"
  • You need to use parentheses so that PubMed knows which terms to use the 'OR' operator on.
  • The search term "children OR pediatric psychosis" will result in far more articles because just having children in the search would be a positive search.
  • (children OR pediatric) psychosis requires psychosis along with either children or pediatric.

20.2. AND operator

  • PubMed searches for articles that have all the words in your search by default.
  • There is an implied "AND" operator between the search terms.
  • However the order of the words is not determined.
  • To search for a phrase you will need to use quotation marks
  • For example, "borderline personality."

21. Bipolar mania

  • The population we are looking for is bipolar disorder patients that are manic.
  • So words used to describe the patients could be "bipolar," "mania," or "manic"
  • We may want to limit this to inpatient's or outpatients, but we can do a search first.
  • If there are too many articles, we can then limit.

22. ICO terms

  • I or intervention is lithium
  • C or comparator I usually leave blank. If there are many articles you can make the search more specific.
  • O or outcome is Young Mania Rating Scale or YMRS. We can look at this table.

23. Outcome table

Construct Scale
MDD PHQ-9,MADRS,HAM-D, Beck Depression Inventory
Psychosis PANSS, CGI-S (used for severity of illness)
Social Phobia Liebowitz-SA-Scale, SPIN and MINI-SPIN
OCD Yale Brown Obsessive Compulsive Scale (Scoring)
Catatonia Bush-Francis Scale
Dementia MMSE, (Montreal Cognitive Assessment)

24. Outcome table

Construct Scale
Tardive Dyskinesia AIMS
EPS Simpson Angus Scale
Mania Young Mania Rating Scale, CARS-M, Clinical Global Impressions-Mania
Irritability Autism Aberrant Behavior Checklist (ABC)
Delirium CAM
Eating Disorder Eating Disorder Examination

25. Putting it together

  • Our search would be
  • P = bipolar OR mania OR manic
  • I = quetiapine
  • C = leave blank
  • O = "young mania rating scale"
  • Surround each of the OR statements with parentheses.
  • (bipolar OR mania OR manic) quetiapine "young mania rating scale"

26. Here is the search

Created by nikhil dhawan,md.