Journal Clubs...

[I Lovely Gator]

---

Members do not see this ad.

Hey All,

I will be starting my rotations soon and just a little nervous about the "journal clubs". Statistics is truly my weakness. 👎 I understand a trial when I see it (observ/RCT/meta-analy/case study etc) but the NUMBERS and % in the study are confusing.
I understand the relative risk etc. But the confidence interval & SD are meaningless to me. I understand that there is a threshold that determine if study is signif or NOT. 😕
I really do not have confidence in this area, maybe because I have not been exposed to a journal club before. I am sure experience with reading one article after another HELPS. 😳

If someone could give me tips on their journal club experience I would greatly appreciate it. Thanks~ 😉

 

[kbv]

don't worry too much about statistics but you should know if the study has enough / adequate / sufficient power, ex. 90% or more or under. review you pharmacoepidemiology book. i don't think SD is included in the trials. most of the times, your preceptors and peers and other staff pharmacists will question you on baseline characteristics, inclusion, exclusion criteria, and endpoints/outcomes of the study. make sure you understand every word stated, ex. what is CHADS2 score in CVD related topics. here is an example you should focus on in JC

GENERALIZABILITY/CRITIQUE/DISCUSSION​

Patient Population
q Are there any major differences in patient characteristics that may confound the results
q Evaluate inclusion and exclusion criteria serve as a guide to the patients that the results may be applicable
Intervention
q Is the intervention being tested representative of current practice or derived from previous well-conducted studies
Endpoints
q Do the endpoints of the trial truly represent what is claimed as being studied
q Is the endpoint used in the trial clinically significant
q If a surrogate endpoint is used, is it validated for correlation to a hard clinical endpoint
Statistics
q What type of data is being assessed (nominal, ordinal, continuous)
q Are the statistical tests used to evaluate the data appropriate1
q Is the effect size clinically relevant
q Evaluate the results in absolute values and calculate number needed to treat (NNT)
q Were subgroups and/or secondary endpoints evaluated? If so what significance can we take from them?

 

[kbv]

e

 

[WVUPharm2007]

Unless you draw the professor who teaches the med lit classes, you'll quickly notice that everyone else hates the intricacies of statistics as much as you do.

For the sake of simplicity...just remember that anything less than p=.05 is "good"..and "significant". P values are more or less a measure of the likelihood of a set of numbers occurring by chance rather than legitimacy (kinda...)

With confidence interval, it's just a range of numbers in which you can say you are 95% sure the true number you are looking for lays within that range.

For instance, say you have something like...Relative Risk Ratio = 3.45, 95% CI = 3.05-3.85.....that would mean that in THIS STUDY, the risk of something happening to one group over the other is 3.45....and there is a 95% chance that if you were to take a study of the entire population of the planet with the same inclusion/exclusion criteria of the study sample that the RRR would be somewhere in between 3.05-3.85.

Don't worry about the stats too much. What you need to be able to do is take applicable info from each study. Know what types of things the authors did that limited the study.....and what they could have done to strengthen the study...notice that the drug company making the test drug funded the study....know what the clinical relevance is....**** like that.

If you REALLY want, I could post an arbitrary recent journal article and everyone on the forum could run through it together...pointing out limitations, what the results mean...how applicable it is to current practice....**** like that.

 

[Priapism321]

Unless you draw the professor who teaches the med lit classes, you'll quickly notice that everyone else hates the intricacies of statistics as much as you do.

If she has a drug information rotation at her home institution, she will draw this professor; he loves the stats.

 

[WVUPharm2007]

If she has a drug information rotation at her home institution, she will draw this professor; he loves the stats.

Yes....all those DI folks love the stats...I lucked into drawing none of them.

 

[SpirivaSunrise]

Lovely Gator~ Here's a checklist I used to go through when I was going through articles for journal club. It's a simple list, but a decent starting point when you're new at it. I posted it a few months ago...but here it is again. Hope it helps & good luck!

 

[SpirivaSunrise]

If you REALLY want, I could post an arbitrary recent journal article and everyone on the forum could run through it together...pointing out limitations, what the results mean...how applicable it is to current practice....**** like that.

Ok...who are you and what did you do with WVU?!? 😉

 

[pinkyrx]

I agree with WVU about most people hating statistics as much as you do... I think 99% of all students feel the same way and for that matter all of my preceptors did too. 👍 I think as long as you understand the big points you'll do fine. Pretty much every journal club I've seen has just skimmed over the statistics part.

Ok...who are you and what did you do with WVU?!? 😉

I was wondering the same thing.

 

[DrZion]

OMG! i did not know this thing is this important. I am in my first year and they're killing me with this journal club thing. My small group leader is a pharmacotherapy specialist(BCPS) and he tears us apart in his group.
Thanks guys for bringing this up, i will make sure i remember what i'm learning now so i wouldn't have problems later.

 

[pinkyrx]

OMG! i did not know this thing is this important. I am in my first year and they're killing me with this journal club thing. My small group leader is a pharmacotherapy specialist(BCPS) and he tears us apart in his group.
Thanks guys for bringing this up, i will make sure i remember what i'm learning now so i wouldn't have problems later.

Our professors were always way more tough on us in the classroom than they or any other preceptor would be on rotation. Hopefully the ones you do on rotation won't be this tough. With that being said, definitely pay attention because it is important later.

 

[tussionex]

depends on the preceptor.

i hate stats and when i sit in on journal clubs...well, i don't even read that stats part of the article. i just skim for the P-value.

 

[WVUPharm2007]

Ok, I was scanning through some recent news item and stumbled upon a good study to shake your fist at menacingly.

Tell me the limitations of this study, a meta-analysis on "Aspirin resistance"....and why one might say it's a bloody pointless *** study to do at this point in time....