I wonder whether Bio.Entrez's efetch() retrieves all metadata of a PubMed article, given a PMID as input. By all metadata, I mean whether PubMed has any more metadata than what efetch() retrieves.
For example, I see that for the PMID 23954024, efetch() retrieves an abstract that contains a bit less information than the abstract on PubMed's website (http://www.ncbi.nlm.nih.gov/pubmed/23954024):
efetch():
"AbstractText": [
"Rotator cuff tendinopathy is a common source of shoulder pain characterised by persistent and/or recurrent problems for a proportion of sufferers. The aim of this study was to pilot the methods proposed to conduct a substantive study to evaluate the effectiveness of a self-managed loaded exercise programme versus usual physiotherapy treatment for rotator cuff tendinopathy.",
"A single-centre pragmatic unblinded parallel group pilot randomised controlled trial.",
"One private physiotherapy clinic, northern England.",
"Twenty-four participants with rotator cuff tendinopathy.",
"The intervention was a programme of self-managed loaded exercise. The control group received usual physiotherapy treatment.",
"Baseline assessment comprised the Shoulder Pain and Disability Index (SPADI) and the Short-Form 36, repeated three months post randomisation.",
"The recruitment target was met and the majority of participants (98%) were willing to be randomised. 100% retention was attained with all participants completing the SPADI at three months. Exercise adherence rates were excellent (90%). The mean change in SPADI score was -23.7 (95% CI -14.4 to -33.3) points for the self-managed exercise group and -19.0 (95% CI -6.0 to -31.9) points for the usual physiotherapy treatment group. The difference in three month SPADI scores was 0.1 (95% CI -16.6 to 16.9) points in favour of the usual physiotherapy treatment group.",
"In keeping with previous research which indicates the need for further evaluation of self-managed loaded exercise for rotator cuff tendinopathy, these methods and the preliminary evaluation of outcome offer a foundation and stimulus to conduct a substantive study."
],
http://www.ncbi.nlm.nih.gov/pubmed/23954024 : Abstract OBJECTIVES: Rotator cuff tendinopathy is a common source of shoulder pain characterised by persistent and/or recurrent problems for a proportion of sufferers. The aim of this study was to pilot the methods proposed to conduct a substantive study to evaluate the effectiveness of a self-managed loaded exercise programme versus usual physiotherapy treatment for rotator cuff tendinopathy.
DESIGN:
A single-centre pragmatic unblinded parallel group pilot randomised controlled trial.
SETTING:
One private physiotherapy clinic, northern England.
PARTICIPANTS:
Twenty-four participants with rotator cuff tendinopathy.
INTERVENTIONS:
The intervention was a programme of self-managed loaded exercise. The control group received usual physiotherapy treatment.
MAIN OUTCOMES:
Baseline assessment comprised the Shoulder Pain and Disability Index (SPADI) and the Short-Form 36, repeated three months post randomisation.
RESULTS:
The recruitment target was met and the majority of participants (98%) were willing to be randomised. 100% retention was attained with all participants completing the SPADI at three months. Exercise adherence rates were excellent (90%). The mean change in SPADI score was -23.7 (95% CI -14.4 to -33.3) points for the self-managed exercise group and -19.0 (95% CI -6.0 to -31.9) points for the usual physiotherapy treatment group. The difference in three month SPADI scores was 0.1 (95% CI -16.6 to 16.9) points in favour of the usual physiotherapy treatment group.
CONCLUSIONS:
In keeping with previous research which indicates the need for further evaluation of self-managed loaded exercise for rotator cuff tendinopathy, these methods and the preliminary evaluation of outcome offer a foundation and stimulus to conduct a substantive study.
(the OBJECTIVES, DESIGN, SETTING, etc. are missing from efetch()'s abstract.)
What other metadata does efetch() misses, and is there any way to programmatically retrieve the missing information?
The info is not missing:
from Bio import Entrez
Entrez.email = "[email protected]"
handle = Entrez.efetch(db="pubmed", id="23954024", rettype="xml")
print(handle.read())
Part of the output:
<Abstract>
<AbstractText Label="OBJECTIVES" NlmCategory="OBJECTIVE">Rotator cuff tendinopathy is a common source of shoulder pain characterised by persistent and/or recurrent problems for a proportion of sufferers. The aim of this study was to pilot the methods proposed to conduct a substantive study to evaluate the effectiveness of a self-managed loaded exercise programme versus usual physiotherapy treatment for rotator cuff tendinopathy.</AbstractText>
<AbstractText Label="DESIGN" NlmCategory="METHODS">A single-centre pragmatic unblinded parallel group pilot randomised controlled trial.</AbstractText>
<AbstractText Label="SETTING" NlmCategory="METHODS">One private physiotherapy clinic, northern England.</AbstractText>
[...]