Linked Life Data

12875354

Source:http://linkedlifedata.com/resource/pubmed/id/12875354

Statements in which the resource exists as a subject.
PredicateObject
rdf:type
lifeskim:mentions
pubmed:issue
2
pubmed:dateCreated
2003-7-23
pubmed:abstractText
Four conditions are proposed under which observational research may be adequate to access the efficacy of therapeutic interventions. First, a randomized controlled trail is unethical, not feasible, or not cost-effective. Second, the study design is rigorous, optimal to test the a priori hypothesis, and faithfully executed. The cohort study is the analytical observational design least susceptible to biases and confounding. Even this design, however, may be impractical when the outcome is relatively rare or the interval between the exposure and outcome is relatively long. In these circumstances, the case-control study yields more timely information with less resource expenditure. Third, additional evidence supports the interpretation that the association between the therapeutic intervention and the health outcome is very likely to be causal (ie, the exposure is documented to precede the outcome; the effect size is large; a dose-responses relationship can be demonstrated; the outcome is associated only with the exposure and the exposure produces only the outcome of interest; the relationship is consistently identified in different populations with different experimental designs; or an etiologic mechanism for the association can be proposed). Fourth, certain features of the treatment-health outcome relationship are favorable to drawing inferences from observational research. These features include a large effect size; treatment assignment that is not a function of clinician or patient choice (as in ecologic designs); an outcome that follows the exposure closely in time; and the circumstance in which few other variables affect the risk of the outcome and these variables are well known and easily controllable in the study design or analysis. A useful exercise for large-scale changes in treatment is to examine population-wide health trends for their consistency or inconsistency with the proposition that the new intervention is effective.
pubmed:language
eng
pubmed:journal
pubmed:citationSubset
IM
pubmed:status
MEDLINE
pubmed:month
Jun
pubmed:issn
0095-5108
pubmed:author
pubmed:issnType
Print
pubmed:volume
30
pubmed:owner
NLM
pubmed:authorsComplete
Y
pubmed:pagination
269-83
pubmed:dateRevised
2007-11-15
pubmed:meshHeading
pubmed:year
2003
pubmed:articleTitle
When are observational studies adequate to assess the efficacy of therapeutic interventions?
pubmed:affiliation
Department of Pediatrics, College of Physician and Surgeons, Columbia University and Children's Hospital of New York-Presbyterian, 3959 Broadway, CHS 115, New York, NY 10032, USA. j11084@columbia.edu
pubmed:publicationType
Journal Article, Review