DOI:10.1111/j.1471-1842.2009.00871.x
The impact of information skills training onindependent literature searching activity and requestsfor mediated literature searchesJohn Addison‡, Steven William Glover* & Christine Thornton†
*Christie Hospital NHS Foundation Trust—Kostoris Library, Wilmslow Road, Withington; †Central Manchester University
Hospitals, NHS Foundation Trust—Trust Library Service, Manchester and ‡Pennine Acute Hospitals, NHS Trust—Education
Centre Library, Oldham, UK
Abstract
Background: Most NHS library services routinely offer both mediated searches and information skills
training sessions to their users. We analyse the impact of these two services on the amount of literature
searching demonstrated by users of hospital- based library services in the north-west of England.
Methods: Data for (1) mediated literature searches, (2) number of library users attending information skills
training sessions, (3) amount of library staff time devoted to information skills training, and (4) number of
Athens-authenticated log-ins to databases were obtained from statistical returns for 2007, and analysed for
significant correlations.
Results: There was evidence of quite strong correlations between the two measures of training activity
and the number of mediated literature searches performed by library staff. There was weaker evidence of
correlation between training activity and total literature searching activity.
Discussion: Attending training sessions may make some library users aware of the difficulty of complex
literature searches and actually reduce their confidence to perform their own complex searches indepen-
dently. The relationships between information skills training, mediated literature searches, and independent
literature searching activity remain complex.
Keywords: Libraries, Hospital, Library Services, Medline, Information Storage and Retrieval
Key Messages
Implications for Practice
C
M
ªH
d Both approaches of training and mediated searching have strengths and weaknesses in their own
right but do not adequately service the user population as efficiently as required.d Training offers hands-on practical skills to users, but these skills decline rapidly when not used regu-
larly and require constant updating.d Mediated searching is more effective for the user but would be a major burden to librarian time if
not resourced adequately.
Implications for Policy
d Strategic decisions about maximising the efficiency and effectiveness of NHS library services in sup-
porting EBP will need to take into account that increasing this number will require diverting staff
time from other duties, or investing in additional skilled staff.d To ensure organisations use the best evidence users must be made aware that library information ser-
vices offer both training and mediated searching services.d Information professionals working in health libraries need the skills to be capable of training users
and providing expert searching services to suit the needs of the client.
orrespondence: Steven William Glover, Christie Hospital NHS Foundation Trust—Kostoris Library, Wilmslow Road, Withington, Manchester
20 4BX, UK. E-mail: [email protected]
2010 The authors. Health Information and Libraries Journal ª 2010 Health Libraries Group
ealth Information and Libraries Journal, 27, pp.191–197 191
Impact of information skills training, S. W. Glover192
Background to Dialog, and again from Dialog to Search
Many NHS library services (23 of 29 Acute Hospi-
tal NHS library services in the NW of England) rou-
tinely offer users both face-to-face information
skills training in how to plan and conduct literature
searches of bibliographic databases, and a mediated
literature searching service, whereby a member of
the library staff conducts a search on behalf of a
user. Indeed, both services are seen as part of the
core business of libraries, to the extent that they are
included as part of the NLH annual statistical
returns, and reports of such activity form part of the
Helicon library accreditation procedures.1 Philo-
sophically, these two services may be considered to
be at opposing ends of a spectrum that runs from
the do-it-yourself approach at one end, to the ask
the information professional to do it for you
approach at the other. In practice, there is no reason
why the two should not complement each other, but
it is reasonable to enquire as to the extent of the
return that investment in each provides. This paper
attempts to analyse evidence of the effect of infor-
mation skills training on subsequent user behaviour.
An effective training service aims to equip users
to conduct high-quality literature searches indepen-
dently, thus removing the necessity for trained
users to seek assistance with future searches. How-
ever, in practice, there are a number of reasons
why this is not always so:d Discussion at the Library and Information for
Health Network North West’s Trainers Group
meetings demonstrate that training sessions
vary in duration, content and delivery.
d Without regular practice all learners tend to loseskills.2 Infrequent database users are likely
to lose or forget information retrieval skills.
d Even those users who regularly undertake theirown literature searches will, by definition
not be as familiar with, or adept at, literature
searching as expert library staff who conduct
literature searches on behalf of many differ-
ent users.
d Time constraints mean that busy clinicians mayprefer to ask the library service to conduct a
search on their behalf.
d The application of NHS procurement regulationsresult in regular changes to database plat-
form as users moved from Webspirs ⁄OVID
ª 2010 The authors. Health Informat
He
2.0 within the relatively short period of
8 years. Consequent changes in appearance
and functionality may disconcert even the
most confident of users.
It is reasonable to expect that the provision of
training would have some impact on user behav-
iour, both in terms of independent literature
searching activity, and in the number of requests
for mediated searches that are made to library
services.
This paper attempts to shed light on four
hypotheses:d Organisations whose library services staff train a
higher proportion of their users will see a
greater amount of literature searching activ-
ity.
d Organisations whose library services staff spenda greater amount of time delivering informa-
tion skills training see a greater amount of
literature searching activity.
d Organisations whose library services staff train ahigher proportion of their users will receive
fewer requests for mediated literature
searches.
d Organisations whose library services staff spenda greater amount of time delivering informa-
tion skills training will receive fewer
requests for mediated literature searches.
This quantitative approach offers a quick snap-
shot of existing practice, made especially manage-
able as the statistics needed are readily available.
Results may well raise questions around reasons
for users’ behaviour, but answers to these are not
attempted here.
What is already known about this topic
The importance of training to support evidence-
based practice (EBP) and continuing professional
development (CPD) and the role of librarians in
providing that training have long been accepted.3,4
The EMPIRIC Project5 looked at views on the
effectiveness of information skills training. The
findings are discussed by Brettle et al. in an evalu-
ation of the qualitative results in 2007.6 The
authors cite three studies7–9 which suggest that
training is valued by those receiving it.
ion and Libraries Journal ª 2010 Health Libraries Group
alth Information and Libraries Journal, 27, pp.191–197
Impact of information skills training, S. W. Glover 193
Despite the value placed on training, there is
some debate about its effectiveness. Schwartz10
reported increased search activity following train-
ing. Ikeda11 and Kronick et al.12 demonstrated that
hands-on training improved end user searching.
Garg and Turtle13 evidenced some impact on the
search skills of medical students. This was borne
out by Robinson and Lawson14 whose study of
training within the primary care setting found that
course attendance was a precursor to increased use
of resources.
However, even where studies suggest that train-
ing does impact on the ability to search more
effectively, other evidence, often within those same
studies, questions the level of effectiveness.7,13
There are also issues around subsequent search fre-
quency15,16 and the long-term impact of train-
ing.17–19
Measures of ‘effectiveness’ are often based on
the subjective self-evaluation carried out at the end
of training sessions. As Harwell et al.20 points out,
this is a weakness and evaluations should be based
on actual use of resources following training rather
than on ‘self-perceived proficiency and intentions
to use online bibliographic databases…’.
Methods
We aimed to obtain statistical data for four usage
indicators for all hospital trusts in the north-west
of England (NW) for the period April 2007 to
March 2008 inclusive. Usage indicators comprised:
log-ins to the Dialog suite of databases, mediated
literature searches performed by library staff, num-
ber of staff receiving information skills training,
and library staff time spent delivering information
skills training.
The NW has 29 hospital trusts. The Isle of Man
service was excluded from this research both for
political ⁄geographical reasons, and because it
offers the same Athens organisational identity to
all its users, which include social care as well as
health staff. An additional five trusts were
excluded because their NLH statistical returns did
not include the required information. This left 23
services, comprising three specialist trusts, five
teaching trusts and 15 general acute trusts which
do offer information skills training. These 23 ser-
vices form the basis of our subsequent analyses.
ª 2010 The authors. Health Information and Libraries Journal ªHealth Information and Libraries Journal, 27, pp.191–197
Access to the Dialog suite of online biblio-
graphic databases was authenticated by the Athens
Access Management System (AMS). Each hospital
trust has its own unique Athens organisation iden-
tifier that is embedded in their users’ Athens
accounts. The AMS records how many times each
individual account is used to log-in to an Athens-
authenticated resource over a given period. Litera-
ture searching activity data was obtained by using
the statistical reports feature on the Athens admin-
istrator interface. Mediated literature searches, and
user training statistics were obtained from the
annual National Library for Health (NLH) statisti-
cal returns of library activity [formerly collected
by the NHS Library and Knowledge Development
Network (LKDN)] submitted by all NHS library
services.
Literature searching activity was defined as
Athens-authenticated log-ins to the Dialog suite
of databases which was provided to NHS users
during this period and includes log-ins made by
library staff as well as by users themselves. In the
case of the specialist and teaching trusts some
clinical staff are University employees and are
entitled to HE Athens-authenticated resources,
which, they may use in preference to NHS
resources. Therefore, literature searching activity
carried out in these organisations may be under-
represented. More generally, some members of
staff submit lists of document requests on print
outs of searches performed on PUBMED, or the
BMA’s own MEDLINE+. These are not Athens-
authenticated resources, and usage of them is not
recorded in the figures.
Two measures of information skills training are
included in the NLH statistical return. Section 10.1
asks for Library staff hours spent on user educa-tion per year, and 10.2 for Number of library
users receiving education per year. No guidance is
offered as to what constitutes user education, in
terms of duration, qualification or expertise of trai-
ner, or even subject matter. (Several library ser-
vices offer training sessions on critical appraisal,
attendance at which is unlikely to improve litera-
ture searching techniques. In addition, several NW
NHS library services were actively involved in
training public library staff in the use of electronic
health resources during this period. Public librari-
ans are not entitled to register for NHS Athens
2010 Health Libraries Group
Impact of information skills training, S. W. Glover194
accounts or to use associated Athens-authenticated
resources, with the result that the measurable
impact of training on subsequent literature search-
ing activity by this group would be invisible for
the purposes of this current research.) Despite
these qualifications, these returns do offer consid-
erable evidence as to the extent and prevalence of
training.
It is clear that not all library services keep
detailed records of the amount of training they
give. Indeed, the NLH guidance recognises this,
and suggests that ‘If regular education…statistics
are unavailable, it is recommended that this infor-
mation be obtained by undertaking a sample study
for one month, and multiplying the resulting figure
by 12’’. Figures submitted by individual library
services do not indicate whether they represent an
exact count, or an estimate based on a ‘sample’
month multiplied by 12. Efforts were made to
ensure that all statistics reported the same activity.
Three library services who submitted outlying
results for numbers of users trained were contacted
Table 1 NHS Trusts, full-time equivalent employees, number of
training, and numbers of users trained, April 2007 to March 2008
NHS Trust WTE
Aintree Hospitals 3518
Blackpool Fylde & Wyre Hospitals 3559
Bolton Hospitals 3173
Cardiothoracic Centre 1180
Christie 1631
Central Manchester University Hospitals 7555
Countess of Chester 2651
East Cheshire 1802
Lancashire Teaching Hospitals 5187
Liverpool Women’s Hospital 1249
North Cheshire Hospitals 3321
North Cumbria Acute Hospitals 3202
Pennine Acute Hospitals 8607
Salford Royal Hospitals 4042
South Manchester University Hospitals 3972
Southport and Ormskirk Hospitals 2547
St Helens & Knowsley Hospitals 3793
Stockport NHS Foundation Trust 3225
Tameside & Glossop Acute 2095
Trafford Healthcare 1730
University Hospitals Morecambe Bay 3972
Wirral University Teaching Hospital 4903
Wrightington, Wigan & Leigh 3773
Total 80 687
ª 2010 The authors. Health Info
and asked to confirm the figures they had submit-
ted. Two of these services subsequently agreed that
they had misinterpreted the instructions at 10.2
and resubmitted revised figures.
Similarly, Mediated literature searches are
included as a separate category (9.1) on the NLH
statistical return. The guidance states that a search
is defined as ‘a distinct question or information
need for which any number of databases or other
resources may be consulted’. Two library services
returned search totals which enumerated each
occasion of rerunning a saved search for alerting
purposes as a separate search, resulting in totals
which immensely exceeded those of other services.
Once contacted, these services supplied revised
totals. Search totals submitted by the remaining
organisations were assumed to accord to the
definition quoted.
In summary: for the reasons described above:
lack of clarity in the guidance for collecting and
submitting statistics, lack of compliance with what
guidance is provided and library users who insist
Dialog log-ins, mediated searches, library staff time spent on
Dialog
Log-ins
Mediated
Searches
Training
(hours)
Training
(users)
1450 115 160 303
1592 68 46 177
3156 260 130 132
801 500 200 250
2483 267 103 286
4208 591 229 483
971 43 35 68
1038 16 120 175
4248 378 90 311
673 65 60 50
944 398 49 228
2135 105 126 336
6514 467 602 767
2313 357 65 63
2640 52 105 198
1540 150 430 950
2505 90 60 160
2161 150 184 316
1038 10 88 67
965 23 104 104
2039 231 165 331
1992 64 107 270
1292 54 50 250
48 698 4454 3466 7031
rmation and Libraries Journal ª 2010 Health Libraries Group
Health Information and Libraries Journal, 27, pp.191–197
Impact of information skills training, S. W. Glover 195
on using non-Athens-authenticated resources, the
accuracy and reliability of the statistics collected is
not absolute. Nevertheless, they offer a basis on
which to analyse any correlations between training
and literature searching.
Once totals for each of the measures of activity
described had been obtained for each of the 23
trusts, an attempt was made to allow for the dif-
fering sizes of each organisation. This was
achieved by dividing each of the measures by the
number of full-time equivalent members of staff
employed by each organisation. This is a fairly
crude measure, as all these organisations employ
a large number of staff who are not involved in
delivering patient care or healthcare management.
One solution to this might have been to restrict
the analysis to numbers of library members, or
numbers of clinicians employed by each organisa-
tion. However, there would be some problems
with this as not all services keep up-to-date mem-
bership records; it is not uncommon for non-
library members and non-clinical staff to attend
Table 2 Dialog log-ins, mediated searches, library staff time spen
equivalent employee by NHS Trust, April 2007 to March 2008
NHS Trust
Dialog
log-ins ⁄ employee
Aintree Hospitals 0.4122
Blackpool Fylde & Wyre Hospitals 0.4473
Bolton Hospitals 0.9946
Cardiothoracic Centre 0.6788
Christie 1.5224
Central Manchester University Hospitals 0.5570
Countess of Chester 0.3663
East Cheshire 0.5760
Lancashire Teaching Hospitals 0.8190
Liverpool Women’s Hospital 0.5388
North Cheshire Hospitals 0.2843
North Cumbria Acute Hospitals 0.6668
Pennine Acute Hospitals 0.7568
Salford Royal Hospitals 0.5722
South Manchester University Hospitals 0.6647
Southport and Ormskirk Hospitals 0.6046
St Helens & Knowsley Hospitals 0.6604
Stockport NHS Foundation Trust 0.6701
Tameside & Glossop Acute 0.4955
Trafford Healthcare 0.5578
University Hospitals Morecambe Bay 0.5133
Wirral University Teaching Hospital 0.4063
Wrightington, Wigan & Leigh 0.3424
Totals 0.6035
ª 2010 The authors. Health Information and Libraries Journal ªHealth Information and Libraries Journal, 27, pp.191–197
training sessions or request mediated literature
searches; and results might be distorted in the
case of services which go out of their way to
encourage membership amongst staff whose
interests in the provisions of the service may be
limited, e.g. use of PCs.
Results
Table 1 lists the name of each trust, and indicates
their number of whole time equivalent employees.
The remaining four columns give the raw scores
for each of the four indicators described.
Table 2 is the result of dividing each of the four
indicators by the number of whole time equivalent
employees working at that trust.
Table 3 presents the degree of correlation (Pear-
son Correlation Co-efficient) between each of the
two measures of training, and the relevant user
behaviour.
We found weak evidence that the number of
users trained correlates with the amount of
t on training, and number of users trained per full-time
Mediated
searches ⁄ employee
Training
hours ⁄ employee
Proportion of
employees trained
0.0327 0.0455 0.0861
0.0191 0.0129 0.0497
0.0819 0.0410 0.0416
0.4237 0.1695 0.2119
0.1637 0.0632 0.1754
0.0782 0.0303 0.0639
0.0162 0.0132 0.0257
0.0089 0.0666 0.0971
0.0729 0.0174 0.0600
0.0520 0.0480 0.0400
0.1198 0.0148 0.0687
0.0328 0.0394 0.1049
0.0543 0.0699 0.0891
0.0883 0.0161 0.0156
0.0131 0.0264 0.0498
0.0589 0.1688 0.3730
0.0237 0.0158 0.0422
0.0465 0.0571 0.0980
0.0048 0.0420 0.0320
0.0133 0.0601 0.0601
0.0582 0.0415 0.0833
0.0131 0.0218 0.0551
0.0143 0.0133 0.0663
0.0552 0.0410 0.0871
2010 Health Libraries Group
Table 3 Pearson’s correlation coefficient r between the two training measures and the two user behaviour outcomes (CI
95%), P-value
Mediated searches Dialog log-ins
Training (users) 0.42 (0.01, 0.71), P = 0.043 0.27 ()0.16, 0.61), P = 0.215
Training (hours) 0.59 (0.24, 0.81), P = 0.003 0.23 ()0.20, 0.59), P = 0.291
Impact of information skills training, S. W. Glover196
organisation-wide literature searching as measured
by log-ins to Dialog.
We also found weak evidence that the amount
of library service staff time spent delivering infor-
mation skills training correlates with the amount of
organisation-wide literature searching as measured
by log-ins to Dialog.
There was quite strong evidence that training
measured in terms of numbers of users correlates
with a demand for mediated literature searches.
We found strong evidence to suggest that the
number of hours devoted to training by library
staff correlates with the number of mediated litera-
ture searches performed by library staff.
Discussion
The results contradicted our expectations, as they
suggest that neither investment of library staff
time in training users nor attendance by users at
information skills training sessions has little impact
on the number of independent literature searches
performed. (Whether the training improved the
quality of users’ literature searching is an interest-
ing question, but one which lies outside the scope
of this current paper.)
The correlation between training hours delivered
by library staff and demand for mediated searches
was particularly unexpected. We had expected pre-
cisely the opposite: that information skills training
would enable users to become more independent
in their literature searching behaviour leading to a
reduced demand for mediated literature searches.
The true picture may, of course, be more compli-
cated than this: the choice between doing searches
oneself, and deciding to refer them to a member of
the library staff does not have to be a once-and-
for-all-time either ⁄or: both may be appropriate in
different circumstances. In addition, attendance at
a training session may also serve to remind users
that library staff provide a mediated literature
ª 2010 The authors. Health Info
searching service, use of which may be preferable
to relying on hard-won, but perhaps shakily
grounded, skills of one’s own.
We speculate that providing more extended
training may make users aware of the complexity
of literature searching, and reduce their confidence
to perform their own searches in situations where
they have a complex question to investigate, or
where they think doing a thorough search really
matters.
One piece of evidence lends some support to
this speculation. In 2008 an unpublished internal
online survey21 was conducted by the library ser-
vice at Pennine Acute in order to gauge user satis-
faction with its mediated literature searching
service. Over 70% of respondents (i.e. staff who
had requested mediated literature searches in the
past year) stated that they had attended literature
searching training at some point in their careers
(16% of them within the past year). Regrettably,
the authors of this questionnaire did not follow up
with the interesting question of why they had
requested a mediated search rather than performing
the search independently.
The number of mediated searches performed
represents less than 9.2% of the total literature
searching activity performed in the services sur-
veyed, and represents a low number of searches
per employee in virtually all cases.
To ensure that the current deployment of library
services offers optimum value, additional research
should be conducted investigating: (a) the reasons
users give for requesting mediated searches rather
than performing the searches themselves, and (b)
comparing the search results of librarians with
those of trained library users.
References
1 Fowler, C. & Trinder, V. 2002 Accreditation of library and
information services in the health sector. HeLicon.
rmation and Libraries Journal ª 2010 Health Libraries Group
Health Information and Libraries Journal, 27, pp.191–197
Impact of information skills training, S. W. Glover 197
Accessible at: http://www.lihnn.nhs.uk/document_uploads/
Accreditation_Reports/accreditation_criteria.pdf.
2 Arthur, W., Bennett, W., Stanush, P. L. & McNelly, T. L.
Factors that influence skill decay and retention. A
quantitative review and analysis. Human Performance 1998,
11, 57–101.
3 NHS Executive. Library and Information Services: Health
Service Guidelines, HSG (97) 47, 1997. Accessible at:
http://www.dh.gov.uk/prod_consum_dh/groups/
dh_digitalassets/@dh/@en/documents/digitalasset/
dh_4012399.pdf (accessed 8 December 2008).
4 Davies, R. & Urquhart, C. J. EVINCE: Establishing the
Value of Information to Nursing Continuing Education.
Report of the EVINCE Project. BLRIC Report 44. Boston
Spa: BLDSC, 1997.
5 Brettle, A., Hulme, C. & Ormandy, P. Effective Methods of
Providing Information for Patient Care: EMPIRIC Project –
Baseline Data Collection Survey. Salford: University of
Salford, 2004.
6 Brettle, A., Hulme, C. & Ormandy, P. Effectiveness of
information skills training and mediated searching:
qualitative results from the EMPIRIC project. Health
Information and Libraries Journal, 2007, 24, 24–33.
7 Brettle, A. Information skills training: a systematic review
of the literature. Health Information and Libraries Journal,
2003, 20(Suppl. 1), 3–9.
8 Urquhart, C. M., Massiter, C., Thomas, R., Sharp, S. &
Smith, J. Getting information to Vocational Trainees: Report
of the GIVTS Project. London: British Library and
Information Commission Research Report, 1999.
9 Wozar, J. A. & Worona, P. C. The use of online information
resources by nurses. Journal of the Medical Library
Association, 2003, 91, 216–221.
10 Schwartz, D. G. Medline training for medical students in the
clinical curriculum. Academic Medicine, 1992, 67, 1040–2446.
11 Ikeda, N. R. Impact of end-user search training on pharmacy
students: a four-year follow-up study. Bulletin of the
Medical Library Association, 1992, 80, 124–130.
12 Kronick, J., Blake, C., Munoz, E., Heilbrunn, L.,
Dunikowski, L. & Milne, W. K. Improving on-line skills and
ª 2010 The authors. Health Information and Libraries Journal ªHealth Information and Libraries Journal, 27, pp.191–197
knowledge; a randomized trial of teaching rural physicians to
use on-line medical information. Canadian Family
Physician, 2003, 49, 312–317.
13 Garg, A. & Turtle, K. M. Effectiveness of training health
professionals in literature search skills using electronic
health databases – a critical appraisal. Health Information
and Libraries Journal, 2003, 20, 33–41.
14 Robinson, S. & Lawson, S. Evaluating the impact
of information skills training within primary care.
Health Information and Libraries Journal, 2005, 22,
63–65.
15 Vogel, E. W., Block, K. R. & Wallingford, K. T. Finding the
evidence: teaching medical residents to search MEDLINE.
Journal of the Medical Library Association, 2002, 90, 327–
330.
16 Erickson, S. & Warner, E. R. The impact of an individual
tutorial session on MEDLINE use among obstetrics and
gynaecology residents in an academic training programme: a
randomized trial. Medical Education, 1998, 32, 269–273.
17 Trinder, V. M., Fleet, G. E. & Gray, A. E. Evaluating the
impact of library user training programmes across Thames
Valley Strategic Health Authority in the UK. Health
Information and Libraries Journal, 2007, 24, 34–40.
18 Booth, A. Sole searching – wasting our time? Health
Information and Libraries Journal, 2005, 22, 70–73.
19 Cheng, G. Y. T. Educational workshop improved
information-seeking skills, knowledge, attitudes and the
search outcome of hospital clinicians: a randomised
controlled trial. Health Information and Libraries Journal,
2003, 20, 22–23.
20 Harwell, T.S., Law, D.G., Ander, J.L. & Helgerson, S.D.
Increasing state public health professionals’ proficiency in
using PubMed. Journal of the Medical Library Association,
2008, 96, 134–137.
21 Pennine Acute Hospitals NHS Trust Library Service.
Literature searching questionnaire (PAT): survey monkey.
Report to Pennine Acute Trust Library Group, November
2008.
Received 18 April 2009; Accepted 22 September 2009
2010 Health Libraries Group
Top Related