Jump to content
Please Check, and if Necessary, Update Your BB Account Email Address as a Matter of Urgency ×
New Forum: Celebrating 20 Years of Support - Everyone is Invited! ×
  • Please Donate

    Donate with PayPal button

    For nearly 20 years, BenzoBuddies has assisted thousands of people through benzodiazepine withdrawal. Help us reach and support more people in need. More about donations here.

How to Read a Clinical Trial Paper


Recommended Posts

Posted

"How to Read a Clinical Trial Paper - A Lesson in Basic Trial Statistics".  This paper was published in the Gastroenterology & Hepatology Journal but, naturally, the lessons can be applied to all clinical trials.

 

Very informative.  These are the topic headings.

 

  • Abstract
  • Are These Subjects Like My Patients?
  • What Happened to the Subjects? Did They Drop Out? Why?
  • Is the Study Design Biased?
  • Does the Study Include an Intention-to-Treat Analysis?
  • Is This a Test of Superiority? Equivalence? Noninferiority?
  • Does the Measurement Matter? Is It Reproducible? Accurate?
  • Surrogate Outcomes
  • Dichotomous Outcomes, Continuous Outcomes, Correlations, and Time-to-Event Endpoints
  • Multicenter Trials
  • Shifting Target Outcomes
  • How Are Missing Data Addressed?
  • Do the Design and Methods Conform to the Prestudy Guidelines?
  • How Good Are the Results?
  • How Big of an Effect Is It?
  • Checking the Figures
  • Conclusion
    While researchers have learned a great deal from randomized controlled trials, some hypotheses cannot be tested with prospective randomized controlled trials due to cost, duration, or ethical constraints. For example, a recent cross-sectional study of the relationship between fiber intake and diverticular disease is a reminder that other study designs serve an important function in medicine.26 These studies can be important for hypothesis generation or for identifying risk associations. With this point in mind, we emphasize the importance of considering study designs beyond clinical trials, as other types of studies can address important questions that are not suited to a prospective randomized controlled trial design.
     

 

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3380258/

Posted

And this one I posted awhile ago, "How To Read A Medical Study - John Cmar, MD"

http://www.benzobuddies.org/forum/index.php?topic=159671.msg2134159#msg2134159

 

 

"One of the most important aspects of being a healthy skeptic is knowing that just because a scientific study was done on a topic does not mean the study was done well, or that the conclusion the authors reach is supported by what they actually did. But when someone states that a particular study has major flaws or was well-done, what precisely does that mean?

 

In this video, Dr. John Cmar analyzes two different journal articles in detail, focusing on the good, the bad, and the ugly of how studies are done and interpreted.

 

John Cmar, MD, has been long enthralled with horrible infections that could spell doom for humankind, as well as sanity and skepticism in the practice of medicine. He is currently an Instructor of Medicine at the Johns Hopkins University School of Medicine, and the Assistant Director of the Division of Infectious Diseases at Sinai Hospital of Baltimore. He is the lead physician in Sinai's Ryan White initiative, which provides medical care and social assistance to patients with HIV infection who are without medical insurance."

 

 

Excellent lecture!

My main takeaway, unfortunately:  Take most (all?) published studies with a grain of salt.  :(

 

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Who's Online (See full list)

    • [...]
    • [...]
    • [Le...]
    • [ab...]
    • [Ch...]
    • [zo...]
    • [Sc...]
    • [Ki...]
    • [Si...]
    • [No...]
    • [ma...]
    • [ho...]
    • [Ab...]
    • [...]
    • [...]
×
×
  • Create New...