MEDLINE® Data Entry by Fiscal Year
Beginning in 1997, the National Library of Medicine® (NLM®) has managed a dramatic shift in how citation data first enter the MEDLINE process: from double keyboarding citations by hand; to scanning and using optical character recognition (OCR); to importing records supplied by publishers in eXtensible Markup Language (XML) format. The data below reflect citations created (not necessarily indexed) for MEDLINE in each fiscal year.
| Fiscal Year | Total Citations Entered | Keyboarded1 Citations | % Keyboarded | OCRed Citations | % OCRed | Citations Received Electronically via XML | % XML |
|---|---|---|---|---|---|---|---|
| 2011 | 845,948 | 0 | 0% | 71,145 | 8% | 774,804 | 92% |
| 2010 | 767,470 | 0 | 0% | 77,309 | 10% | 690,161 | 90% |
| 2009 | 748,479 | 0 | 0% | 87,970 | 12% | 660,509 | 88% |
| 2008 | 720,097 | 0 | 0% | 94,711 | 13% | 625,386 | 86% |
| 2007 | 685,134 | 0 | 0% | 105,841 | 15% | 579,293 | 85% |
| 2006 | 669,124 | 0 | 0% | 122,106 | 18% | 547,018 | 82% |
| 2005 | 639,000 | 0 | 0% | 152,000 | 24% | 487,000 | 76% |
| 2004 | 584,000 | 55,000 | 9% | 98,000 | 17% | 431,000 | 74% |
| 2003 | 535,000 | 101,000 | 19% | 109,000 | 20% | 325,000 | 61% |
| 2002 | 519,000 | 114,000 | 22% | 113,000 | 22% | 292,000 | 56% |
| 2001 | 472,000 | 124,000 | 26% | 129,000 | 27% | 219,000 | 46% |
| 2000 | 488,000 | 156,000 | 32% | 146,000 | 30% | 186,000 | 38% |
| 1999 | 453,000 | 198,000 | 44% | 151,000 | 33% | 104,000 | 23% |
| 1998 | 434,000 | 260,000 | 60% | 130,000 | 30% | 44,000 | 10% |
| 1997 | 496,000 | 434,000 | 88% | 62,000 | 13% | 0 | 0% |
| 1996 | 336,0002 | 336,000 | 100% | 0 | 0% | 0 | 0% |
| 1995 | 410,5593 | 410,559 | 100% | 0 | 0% | 0 | 0% |
The U.S. Government Fiscal Year runs from October 1 - September 30.
Percentage may not add to 100% because of rounding.
1Keyboarding ceased as a method of data entry in June 2004.
2There was a data entry and concomitant indexing backlog resulting from the General Services Administration Board of Contract Appeals' suspension of the NLM editing and data entry contract from late February 1996 to late April 1996. The large backlog was eliminated in early April 1997. This caused fewer records to be added during FY 1996 and a greater number added in FY 1997.
3Number derived from citations present in the database in 2007 that are tagged as being created during FY 1995; not the original processing statistic.
