Restricted: Increasing Access to the Reference Collection 1

: Introduction: The objective of this project was to determine whether or not a hospital library reference collection is still necessary or justified. Two academic hospital libraries moved all reference books to the general collection to see whether increased access to these materials would increase their use. Description: All reference books were updated to circulating status and shelved in the circulating collection. As these items were used, statistics were gathered in the integrated library system (ALMA). Statistics were gathered from August 2014 to January 2015. Circulation statistics for equivalent periods prior to and during the project were compared to determine whether changing access to the collection increased use. Outcomes: Uses of the reference collection items doubled at Seven Oaks General Hospital (SOGH) and more than tripled at Victoria General Hospital (VGH). The percentage of reference titles used tripled at SOGH and doubled at VGH. Discussion: The change to circulating status significantly increased access to and use of the reference collection. This borrowing policy change for the reference collection will be recommended to the other hospital libraries within the University of Manitoba.


Introduction
It is often observed anecdotally that library reference collections are not used as much as they once were, and yet, many libraries continue to keep items on restricted borrowing privileges so that they can only be used in the library. This can be frustrating for modern users, who may not remember the days of card catalogues, no internet, and how massive reference collections used to answer all types of questions. Health librarians in those days could not risk lending reference items for weeks at a time, lest they be unable to answer questions requiring an immediate response. In today's world of incredible internet access, ubiquitous mobile devices, online textbooks, and apps of all types, people are able to answer virtually any query at any time. In light of these technological advances, is a reference collection still justified in a library or is it merely a relic of the past?

Literature review
A review of the literature shows that the popularity of the reference collection in many academic and health sciences libraries has been declining steadily [1Á7]. In response, libraries have been evaluating and assessing their collections, and they report that print reference collections are underused [1Á7]. In King's [5] study of 44 member libraries of the Association of Research Libraries (ARL), 21 libraries assessed the use of their print reference collections.
When evaluating use of print reference collections, the reshelving method, which involves recording use of books on a spreadsheet title list, is the most commonly reported [1Á3, 5Á7]. These studies of reference collections have often resulted in major weeding projects [5]. The University of Alberta Health Sciences Library assessed their reference collection and completed a deselection project in 2011, during which half the collection was merged with their circulating collection. They followed this assessment with a yearlong study tracking the usage of what remained of their reference collection. With only 33% of their collection used, they transferred all of their reference books to the circulating stacks with the most highly used items assigned a four-day loan period [7].
With users trending towards preference for online resources to meet their information needs, some library professionals have focused less on development of print reference collections and more on development of digital environments with access points to online information [4Á7]. Librarians who have evaluated both their print and online reference collections have noted that the use of e-books was significantly higher than their print counterparts, which is another factor that has led some libraries to move most of their reference materials to their circulating collections [5,7]. These trends make print reference collection development increasingly problematic and possibly redundant.
Most of the published literature is based in academic environments, which serve researchers and students. The health facility library context is quite different in the way clients (generally doctors, nurses, and other healthcare staff) access and use information. Healthcare professionals rely heavily on point-of-care tools, which are typically available as downloadable mobile apps that can be accessed immediately on the unit when a question related to patient care arises [8]. According to Boruff and Storie [8] ". . . devices are quickly becoming one of the main tools for accessing clinical information, especially for younger health professionals and trainees". In many cases pointof-care tools are replacing traditional print reference materials, making physical hospital library reference collections even less relevant [9].
Prior studies did not assess use of the reference collection in a healthcare environment where point-ofcare tools and other online resources are becoming the preferred methods for accessing reference information, nor did they assess whether the use of this collection increased after the move to circulating status. The study outlined in this paper seeks to build on the studies discussed in the literature by determining in a hospital library setting, where point-of-care resources are increasingly popular, if books from a print reference collection can be reclassified to circulating status thereby leading to their increased use.

Description
Librarians at Victoria General Hospital Library (VGH) and Seven Oaks General Hospital Library (SOGH), under the umbrella of the University of Manitoba, wanted to increase access to and use of their reference collections without decreasing accessibility to all users. Staff had observed that use of the print reference collection was not proportional to the amounts of resources and time that were spent developing and maintaining it.
Policy dictated that materials in both reference collections were always supposed to be accessible in the library; however, low use had long been observed and special allowances had already been made for these items to be checked out for short term loans that lasted anywhere from one to seven days and, on occasion, for a month or more. Therefore, even in a collection where the books were not supposed to circulate, immediate access could not be guaranteed. This is consistent with King's [5] study of the ARL member libraries; 20 out of 44 libraries allowed reference items to be borrowed, 15 of which determined loan periods on a case-by-case basis. Another common practice at VGH and SOGH was to suggest that patrons photocopy reference materials instead of borrowing the items, but this was periodically problematic as the desired sections were often large and photocopying created copyright concerns. Both of these access issues suggested that moving the reference collection might be possible and beneficial.
A number of other factors indicated that moving the reference collection would be feasible. First it was determined that much of the reference collection was available online. At VGH, 43% of the 95 reference collection titles were available online and at SOGH, 30% of 57 titles were online. These titles were exact matches; more of the reference materials would have had equivalent online access via similar titles, but this was not determined within the scope of this project. Second, physical duplicate copies of most books were available at other locations within the University of Manitoba. Patrons would be able to put one of these duplicate copies on hold and have it brought to their location. Third, there is a recall process in place for all books in the general collection so that if someone requires a book that is checked out, it can be obtained in seven days or less. All of these provide ways for a patron to gain immediate or quick access to an item that may not be instantly available at their location of choice.
Considering the perceived lack of use of the physical reference collection, the availability of items in the online collection, and the other factors outlined previously, it appeared to be a reasonable and enticing possibility that these two hospital libraries could change the access to their physical reference collections. The hope was that improving access to these materials by moving them all from reference to the circulating collection would also increase their use.
To prepare for the project, spreadsheets were created for both locations with individual book information, an indication of equivalent online access for the item, statistics for usage prior to the project, and space for recording usage statistics during the project. Originally it was thought that some books could not be removed from reference, but after extensive discussion, it was determined that all books should be moved to determine more accurately how the collection would be used if circulating. It would be simple to return a book to reference if absolutely necessary. Leaving books on reference that were likely to have high use would have skewed the usage statistics for this project in a negative way. Another consideration was whether the library should have the newest edition of items that were valuable enough to be on reference. Patrons generally want the most current information available, especially in health libraries, so an older edition may not circulate regardless of its location or status. In preparation for the project, newer editions of reference books were purchased when available and usage statistics were then kept for the new item as the "reference item." Immediately prior to the project period, all reference books were reclassified to circulating status, had their reference labels removed, and were shelved in the circulating collection. It was decided that official usage statistics would be taken from the integrated library system (ALMA). The original intent was to count uses manually, but it was determined that human error could not be eradicated. Therefore as the former reference materials were borrowed, sent out to fill holds, or used in-house, these statistics were gathered in ALMA.
SOGH gathered statistics for six months, from the beginning of August 2014 to the end of January 2015. VGH gathered statistics for four months, from the beginning of October 2014 to the end of January 2015. The time periods at both locations were not the same due to a staff shortage at VGH at the outset of the project. In February 2015, the circulation and in-house use of materials during the project were compared with the same statistics for an equivalent period of time prior to the project. SOGH usage was compared with the statistics from January to June 2014 and VGH usage with the statistics from January to April 2014. It would have been ideal to compare the same months in different years, but this was impossible as ALMA was implemented on 1 January 2014, and the statistics from the former system only transferred in aggregate. The aggregate statistics could not be analyzed and more detailed statistics were not accessible through any other means. Because the project took place during an academic term, the comparison months were also taken from an academic term to compare reasonably equivalent usage statistics. Given that the two sites primarily serve hospital staff, who do not work on a term system, use for the fall and winter terms should be fairly equal.

Outcomes
The reference collection items at both locations were used more frequently once they were reclassified to circulating status. At both locations, the percentage of the reference collection items that were used increased (Figure 1). At VGH, the percentage of the collection that was used nearly doubled, from 13% prior to the project to 25% during the study period. At SOGH, the percentage of the collection that was used more than tripled, increasing from 10% to 33%.
Both locations also saw an increase in the total number of reference collection item uses (Figure 2). Use at VGH more than tripled, from 12 item uses prior to the project to 39 item uses during the project. At SOGH, use more than doubled, from 18 item uses to 40.
At both VGH and SOGH, checkouts as a percentage of use increased dramatically (Figure 3). At VGH, checkouts increased from 17% of use to 46% during the project, whereas at SOGH the percentage increased from 13% to 48%. The aggregate average checkout length for both locations prior to the project was five days and checkouts were often discouraged; therefore, most use had been conducted in-house. In contrast, checkout periods during the project increased significantly, as checkout periods for the circulating collection are academic term loans. It is difficult to determine accurate checkout lengths for the items that were used as many were renewed (some multiple times) and continue to be checked out months after the end of the project.

Discussion
The use of both reference collections increased significantly during the project, as books were visible and freely available when placed in the general collection and could now be requested by any user within the University of Manitoba. Anecdotal evidence at both libraries suggests that no complaints or negative feedback were received when resources were taken off reference, and most users did not even notice that a change had taken place. Some users were pleasantly surprised to find that they were now able to borrow materials that they had not been able to borrow before.
Each location is staffed with one librarian and one library assistant. Everyone was consulted informally to find out whether there had been any reference questions that they were unable to answer. With so much information available in online resources, neither location confronted a  This borrowing policy change for the reference collection will be recommended to the other hospital libraries within the University of Manitoba. In times of tightened budgets, libraries should question whether their print reference collections are used and needed with today's widely accessible technology. This study is significant for library practice, particularly in a health library context where so much reference material has moved online to point-of-care tools that are immediately accessible. Other libraries might also witness the same increase in use of their reference materials, and the resources and effort of maintaining a physical reference collection could be transferred to maintenance of an online reference collection.
Based on this project, further study would be recommended in a number of areas. In a hospital library setting, reference material format preferences of different user groups should be investigated. Stated preference of reference material format could then be compared with the actual usage statistics of print, online, and point-of-care resources. Similar studies could also be considered for a hospital library's general collection to maximize cost savings through prevention of print and online duplication.