User talk:I enjoy sandwiches/dbs
Pubmed search criteria for DBS in PD comparing STN and GPi
[edit]For future editors, this is a helpful search string to filter through articles on Pubmed. This isn't perfect but it's fairly good:
((“Deep brain stimulation”) OR (DBS) OR (Neuromodulation) OR (neurostimulation)) AND ((Parkinson) OR (Parkinson’s) OR (tremor)) AND ((STN) OR (VIM) OR (GPI) OR (“subthalamic nucleus”) OR (“globus pallidus pars interna”) OR (“ventral intermediate nucleus”)) AND ((PDQ) OR (UPDRS) OR (“unified parkinson”) OR (“unified PD”) OR (“adverse event”) OR (“adverse effect”)) AND ((trial) or (study) or (series)) NOT (cZi) NOT (PPN) NOT (“treatment resistant depression”) NOT (epilepsy[MH]) NOT (dystonia[MH]) NOT (Tourette[MH]) NOT (“obsessive compulsive”) NOT (model) NOT (animal) NOT (mouse) NOT (rat) NOT (rodent) NOT (computer) NOT (acute) NOT (microelectrode) NOT (“local field”) NOT (magnetoencephalography) NOT (electromyography) NOT (electroencephalogram) NOT (metaanalysis) NOT (“meta analysis”) NOT (review[PT]) NOT (retrospective) NOT (“case report”) NOT (urinary[MH]) NOT (pilot[text]) NOT (MRI) NOT (“positron emission tomography”)AND English[Lang] AND hasabstract[text] AND "1990/01/01"[PDAT] : "2025/03/06"[PDAT]
This was taken from the Supplemental section of the Nature 2021 review. The dates should be adjusted to whenever you're searching this. This search yielded 256 results in 2021 and 390 results today. Of note, these articles are not appropriate for referencing because meta analyses and position statements have been removed. They are for seeing source material. The search criteria can be fairly easily adapted to focus on different symptoms or meta analyses by removing or adding individual terms. Pubmed and Embase are just Google on steroids and can be learned with just a bit of patience - don't be intimidated, just be patient and thorough. I enjoy sandwiches (talk) 21:00, 6 March 2025 (UTC)
Summary of rewrite
[edit]I have worked on the rewrite extensively over the past several months in an attempt to remove original research, focus on high impact factor meta analyses, and remove publication bias. I also reorganized the sections and updated the history and future directions sections to summarize the entire scope of the procedure and current state of the art. The vast majority of research on DBS to date has been on its use with Parkinson's, which has led to this section being longer than other chronic neurologic diseases, though as research continues to expand and evolve these sections will also be elaborated upon. The article still needs quite a bit of work. This is undoubtedly the most difficult rewrite I've undertaken of the hundreds of medical articles that I've undertaken over my past few decades as a medical editor on this site.
A consistent pattern in the DBS research that I've noticed is that there will be a non-funded study that will publish risks associated with the device. Right behind it, there will inevitably be a device funded study (usually M*dtronic) that uses different exclusion criteria, often without any description of said criteria, and minimizes these risks. For example, the GPi vs STN reviews by Clin Neurol Neurosurg 2021 (government grant) and Nature NPJ 2021 (device funded) use different exclusion criteria, with the government funded article excluding studies that did not directly compare the two targets, while the M*dtronic funded study did, and then the device funded study just arbitrarily added on 5 "extra" studies for "historical context" that followed the devices up to 24 months, while the remainder of articles were only in the first year. Whether intentional or inadvertent, such strategies predispose to p hacking masquerading as Monte Carlo.
I would advise extreme caution with future edits, even from high impact factor journals. Because DBS is massively lucrative, this article at high risk of been brigaded by paid editors or researchers closely associated with the device. This has led to inappropriate selection bias and Type I/II errors throughout that I have attempted to mitigate. Authors' outside funding *must* be researched, not just by looking at their declarations but what they've stated in other articles and through their online internet presence and grant histories. It is difficult to convince a man of something when his wallet depends on his believing the opposite. I enjoy sandwiches (talk) 22:17, 1 April 2025 (UTC)