Jump to content

Dr. Mark Hurt's Blog

  • entries
  • comments
  • views

How do you file articles? The YPA method.

Mark A. Hurt MD


It is now some 24 years ago that I discovered a way to file articles from the literature. You might say, or think, so what? What's the big deal about filling articles from the literature? Here's the big deal: you are, and will continue be, confronted with the problem of how to find and store literature articles you have copied or downloaded as PDF's. How do you go about that task?

When I trained in pathology in the early 1980's, everything -- I mean everything -- was in hard copy. If you wanted a copy of an article or book chapter, you either bought it from the publisher or you copied it from the source material and stuck it in some -- hopefully -- trusted way to find it again.

Most of my colleagues in that period, including me, were overwhelmed immediately by the process of separating and filing these articles.

On or around 1988, I had an idea about filing articles that I have used to this day. Here it is: it's called the “year-first page-first author method” (or YPA method for short). This method has advantages over the PubMed or Doi systems because it gives the user some real and useful information about the article or book (or book chapter) in question.

Let me give some examples:

Here's a journal reference:

[indent=1]Abdelmalek M, Loosemore MP, Hurt MA, Hruza G. Geometric staged excision for the treatment of lentigo maligna and lentigo maligna melanoma: a long-term experience with literature review. Arch Dermatol. 2012 May;148(5):599-604. doi: 10.1001/archdermatol.2011.2155. PubMed PMID: 22782151.[/indent]

In my system this becomes “20120599abde melanoma staged excision”

The first four numbers are the year of publication, the second four are the first page of the article, and the four letters are those of the first four letters of the first author's surname. It is is followed by keywords.

The advantage of this method is that it refers to something intrinsic in the given article, in contradistinction with the unrelated numbers provided by the PubMed and Doi classification schemes.

Here's an example from a book:

[indent=1]Mooi WJ, Krausz T. Malignant transformation of nevi; dysplastic nevi. Chapter 7 in: Pathology of melanocytic disorders, 2nd Ed. London UK: Hodder Arnold, 2007, pp 195-226.[/indent]

In my system, this becomes “20070195mooi dysplastic nevus”

The beauty of my method is that it ties the year and first page of the publication to the filing system, in contrast with PubMed and Doi. It also has the great advantage of creating placeholders for older material that might not have Pubmed or Doi numbers.

For instance, in this article by Norris in 1820 on a “case of fungoide disease” (familial melanoma):

[indent=1]Norris W. Case of Fungoid Disease, Edinburgh Med Surg J 1820; 16:562-565,[/indent]

The YPA number is: “18200562norr melanoma familial”

This kind of system has great advantages when sorting on file names in computers, as it autosorts alphanumerically. PubMed or Doi do not help on that score.

If one uses only a manual system, one need only to write the YPA code on the article and file it sequentially. About 10 years ago, I hired a secretary to go through all of my loose articles and file them with my system. Today, if I need an article from that physical file, I need only to know the year and first page of the article, and in a few seconds, I can discover whether I have the physical article.

In a computer system, I need only go to the file folder that contains the PDF of the article in question and search using the same criteria as in a manual file. I can even enter key words. If the article is there, I'll find it.

Additionally, when searching on PubMed or when looking at reference lists at the end of articles or books, one can identify the YPA codes just be looking at the reference itself; it's embedded in the reference.

Please let me know how you do it. If my system helps you, I'm glad. Happy hunting for those useful articles.


Recommended Comments

Dr. Richard Carr


Hi Mark,

Interesting suggestion but see what you think of my system. When I was a neophyte in pathology circa 1991 I started my slide collection. I was lucky to get my first proper lab top at the same time. An old small [i]compaq[/i]. It had microsoft access in it's basic iterations and I used access to create a database for my slide collection. I started collecting slides from my first year in pathology and it has become a serious (mental) disease every since. I now have over 6,000 slides in the collection and still some of the earliest cases are rare cherised slides. Having it in a database makes finding slides and demographic data very easy. In the comments sections I now paste my referral letters and even sections from other texts and internet resources. The database fields enable me to keep track of the provenance of slides and thank colleagues for cases etc. I recieve regular visitors to look at the slides and always get comments like this is the most well organised collection around. The actual slides are filed numerically in ascending order in 4 categories "Skin: Non melanocytic tumours", "Skin: Melanocytic tumours", "Skin inflammatory / non tumour" and "General slide collection for all non-skin"

I also file my histo and clinical photographs in the same database now. Every case (or image) is linked to the chapter's of Weedon e.g. lichenoid is 03 and vascular tumours is folder 38 in my computer files. By adding the 0 for the single numeric chapters the folders automatically sort on name. It allows me to keep track of literally hundreds of thousands of images (each diagnosis has a sub-folder - you don't want to know how many sub-folders there are for BCC).

Getting back to references it was a shame I only realised in the mid 2000s how to use an access database system for PDF and hard copy references. I had already created a database in which I paste the citations and abstract for references I won't to make a note of. I have added fields with my own key words or comments - like "Nice paper" or e.g. "Basaloid SCC paper" so I can see what papers are for a specific article I am working on. The main citation is in a separate field in the correct format for most journals and can be pasted into any paper. Each new referance generates a unique number (ascending from 001) and is filed under this number in the filing cabinet or PDF folder. So if I need to find a paper I can put any word from the title, authors, abstract etc from my database, find it's unique number in seconds, see if it is PDF or hard copy and pluck it from the cabinet or computer folders in a few seconds. The unique number is the pre-fix in the document name (so the PDFs can be sorted in ascending order) e.g. "448_Spitz tumour grading_Spatz2010". If the paper is hard copy I just write RAC448 on the top right of the paper and stick it in the filing cabinet in position 448. I now need only find a helpful junior or secretary to add all the pre-2000 papers to the system and will be able to die a happy man!

Only problem with all this filing is it takes a lot of time to reap the benefits but at least I don't have problems finding nice cases (or papers) for colleagues etc when collaborating or needing material for a lecture, paper etc. In fact the slide filing system is so quick it is used regularly at weakly teaching multi-header sign-out if a junior makes an incorrect diagnosis we can pull examples for comparison. Recently I pulled all my cellular neurothekeomas to help with a tricky case and left them out for the fellow to take a look and re-file for me. Visitors using the collection can have access to the database while they go through the slides and I am sure it help to see the exact wording in difficult cases like impossible melanocytic referrals!
Link to comment
Mark A. Hurt MD


Hi Richard,

Your slide collection sounds impressive indeed! The way you index and retrieve them using a database program is the way to go in my opinion. Oddly enough, however, I have not been a big slide collector, probably because I was discouraged from doing so in the beginning as well as the institutional bias against doing so. I have met a number of colleagues in the States, however, who have been robust collectors.

It does make some sense to me to use a trusted book as a source of the structure for filing; Weedon's book is a time honored structure, and probably no book will answer every need.

Regarding articles, however, I think my system has more flexibility, in general, because it uses something intrinsic to the article as the sorting function. After I began using it for some years, it became clear to me that it was very easy to file and retrieve articles because of the near universal access to this "intrinsic" information via Pubmed. Thus, even though you won't find a file in my database called "syringocystadenoma papilliferum," it is not difficult to develop a list of those articles from Pubmed followed by journal references or book references. There is also a double check in that the system prevents reference duplication, because the reference number is tied to year and page number. It is true that a reference can duplicate the first 8 characters with another reference. For instance, two articles might have 20130001 as their starting point; however, when the authors' name is added, say 20130001hurt vs 20130001carr, the reference is nearly unique.

In any event, thank you for sharing your experience.


Link to comment
Raul Perret




I was checking the blog and found this interesting thread. It has been a long time since this publication and maybe you already know about this tool but i reccommend the free software Mendeley (https://www.mendeley.com/) for organizing articles that you download from the internet, it is very user friendly, easy to use and saved me a lot of time.


PS Both of your systems look useful 

Link to comment
  • Create New...