I had some issues with the New PubMed and a systematic review search that I
was doing. I also went through all earlier emails by various members on
this list. I note that some issues appear to have been solved (now there is
a csv file saving option and we can change the no of results in a page). I
also noted one new feature that I liked - we can choose to save all
results, or selections (both available earlier) and now - results on the
current page. That is nice!

About my systematic review search

I had gone through 60 steps of searching and search terms. My final step
was #58 AND #59 – so the result no was #60.
In the old PubMed, if I clicked “Create alert” to save the final strategy,
it used to save it with all the search terms with the appropriate Boolean

IN the New PubMed, it saved it as -  *#58 AND #59*!!! How did I figure that
out? When I clicked the name of the search to view results – it said –
“Query not found” and I had a blank result page. I went back to the My NCBI
page and took my cursor close to the saved search name, and I found that it
was saved as #58 AND #59 !! (Luckily I tested this soon after creating the
alert and noticed this problem!)

Next, I went to the Advanced Search page and in the history, for the last
step, I checked under “Details”. I found the entire strategy and pasted it
and saved it. BUT, when I re-ran it from My NCBI, I got only 885 results
instead of the 4000+ results that I originally had. With a message –
“Quoted phrase not found”

Now I had not used quotes for *any* search term. So I asked aloud “What
quoted phrase”? And then found that for my input – deprescr*[tiab], PubMed
had translated my query to "deprescr*"[Title/Abstract]). And so on for
every term

Next…. I removed all quotes (using MS Word’s Find and Replace feature) and
pasted the strategy again. This time I got the 4000 plus results, and saved

Then, I discussed the strategy with the other authors and they suggested
two more terms to be added. I have redone the search

Now I have a new problem:

- Old PubMed gives 7064 with Most recent results and 7075 with Best Match
- New PubMed gives 7073 with either displays

My question – If I take the 7075 results and create an alert via the Old
PubMed, if it goes away in the next few days, will my New PubMed give me
all of them?

I think the BEST thing for me to do just now is to create a PMID list of
all 7075 as a back up

I am going to have more work to do. Which means the authors are going to
have to pay more (increasing number of hours) – thanks to these issues!

I will be writing to the NLM too. Will have to see what they say

Vasumathi Sriganesh
QMed Knowledge Foundation
A-3, Shubham Center, Cardinal Gracious Road
Chakala, Andheri East, Mumbai 400099, India
Tel: 91-22-40054474    Mob: +919867292230
Email: [log in to unmask]