Lee et ai.

BMC Primary Core (2024) 25:

172 BMC Primary Care
https://doi.org/10.1186/sl 2875-024-02429-x

RESEARCH

ldentifying primary ca re clinicians'

preferences for, barriers to, and facil
itators of information-seeking in clinical
practice in Singapore: a qualitative study
2 3 4 15
1
Mauricette Moling Lee ·, Wern Ee Tang , Helen Elizabeth and Lorainne Tudor Car • *
Smith

Abstract
Background The growth of medical knowledge and patient care complexity calIs for improved clinician access to evidence-base
resources. This study aimed to explore the primary care clinicians' preferences for, barriers to, and facilitators of information-
seeking in clinicai practice in Singapore.
Methods A convenience sample of ten doctors and ten nurses was recruited. We conducted semi-structured face­
to-face in-depth interviews. The interviews were recorded, transcribed verbatim, and analysed using thematic content analysis
Results Of the 20 participants, eight doctors and ten nurses worked at government-funded polyclinics and two doctors worked
private practice. Most clinicians sought clinicai information daily at the point-of-care.The most searched-for information by
clinicians in practice was less common conditions. Clinicians preferred evidence-based resources such as clinicai practice
guidelines and UpToDate®. Clinicai practice guidelines were mostly used when they were updated or based on memory.
Clinicians also commonly sought answers from their peers. Furthermore, clini­ cians frequently use smartphones to access the
Google search engine and UpToDate® app. The barriers to accessing clinicai information included the lack of time, internet
surfing separation of work computers, limited search functions in the organisation's server, and limited access to medical
literature data bases.The facilitators of accessing clinicai information included convenience, easy access, and trustworthiness
information sources.
Conclusion Most primary care clinicians in our study sought clinicai information at the point-of-care daily
and reported increasing use of smartphones for information-seeking. Future research focusing on interventions to improve
access to credible clinicai information for primary care clinicians at the point-of-care is recommended.
Triai registration This study has been reviewed by NHG Domain Specific Review Board (NHG DSRB) (the central ethics
committee) for ethics approval. NHG DSRB Reference Number: 2018/01355 (3 1/07/2019).
Keywords Evidence-based medicine, lnformation-seeking behaviour, Nurses, Physicians, Smartphone

*Correspondence:
Lorainne Tudor Car
1.tudor.car@imperial.ac.uk
Full list of author information is available at the end of the article

BM
© The Author(s) 2024. Open Access This article is licensed under a (reative Commons Attribution 4.0 lnternational License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as longas you give appropriate credit to
the original authos) and the source, provide a link to the (reative Commons licence, and indicate ifchanges were made.The
images or other third party material inthis article are included inthe article's (reative Commons licence, unless indicated
otherwise in a credit line to the material. lf material is not included in the article's (reative Commons licence and your intended
use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directlyfrom the
copyright holder.To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The (reative Commons Public
Domain Dedication waiver (httpJ/creativecom mons.org/publicdomain/zero/1.0 1) applies to the data made available inthis article,
unless otherwise stated in a credit line to the data.
Lee et ai. BMC Primary (2024) 25: Page 2 of 11
Care 172

Background information [24-27], using them to access websites,

Primary care clinicians provide the bulk of care to mobile apps or communicate with peers [28]. The use of
patients in primary care settings. ln Singapore, there electronic resources improves clinicians' knowledge and
are 23 polyclinics and about 1,800 General Practitioner behaviour as well as patients' outcomes [29]. However,
(GP) clinics with private GPs providing primary care for evidence on how smartphones are used at the point-of
about 80% of the population [1]. The primary care clini care, particularly for evidence-seeking, is limited. Sin
cians provide primary care services at community poly gapore, with a total population of 5.92 million as of the
clinics and private medical clinics around Singapore [1] . end of June 2023 [30], is one of the countries with the
The polyclinics are formed by three healthcare groups - highest smartphone usage among its residents , with
National Healthcare Group, National University Health approximately 5.72 million (97%) users in 2023 [31].
System, and SingHealth [1]. These polyclinics served Cor respondingly, smartphones may be an important
various populations in Singapore's central, northern, infor mation-seeking channel among primary care
north-eastern, western, and eastern parts [1]. Every day, clinicians. However, the increasing cyber threats
clinicians make many clinica! decisions, ranging from worldwide may lead to internet surfing separation as a
diagnosis and prognosis to treatment and patient man common security measure.
agement [2, 3]. However, to provide consistent high Institutional policies limiting access to computers at
quality patient care, such clinica! judgments must be the point-of-care deter clinicians from seeking informa
informed by existing trustworthy medical evidence [4-6]. tion and disrupt their workflow [32]. Due to patient data
To meet their information needs, clinicians seek relevant privacy breaches, the Singapore Ministry of Health intro
information from various sources of information [3]. duced internet surfing separation as a security measure
Searching for and using the information to meet infor in July 2018 in all public healthcare institutions in Sin
mation needs has been described as information-seeking gapore [33]. Internet surfing separation stands for the
behaviour [7-9]. restrictions on internet access and browsing which were
Previous research showed that clinicians often raise enforced in Singapore public healthcare institutions in
questions about patient care in their practice [10]. Half 2018 due to patient data privacy breaches [33]. This has
of those questions are left unanswered. Identifying what limited the internet access of primary care clinicians at
information primary care clinicians need, how they the workplace. Since its introduction, the Internet has
search for required information and how they adopt it not been accessible from any of the clinic's desktop com
into practice is essential in ensuring safe and high-quality puters and has been available through a few work lap
patient care [11, 12]. While there are reports of infor tops with limited availability to the polyclinic staff. At
mation-seeking behaviour in primary care from other the time that this research was conducted, primary care
countries [2, 8, 13, 14], similar reports in Singapore clinicians in the public healthcare sector in Singapore
are limited. did not have access to the internet from their work com
Clinicians may consult severa! sources to support their puters. Clinicians rely on evidence-based information
decisions, including clinica! practice guidelines (CPGs), to make informed decisions about patient care [4-6].
journal articles, peers, and more [3]. However, there is a When access to online resources is restricted, clinicians
wide variation in the adoption of evidence-based prac may struggle to receive current and correct informa tion,
tices across healthcare disciplines, which could lead to thus jeopardising patient safety and the quality of care
poorer primary care outcomes [8, 12, 15-19]. To offered [11, 12]. Therefore, we sought to understand how
mitigate this, a commonly employed approach is the primary care clinicians were addressing their clinica!
development of CPGs, clinica! pathways, or care guides information needs when their work computers were not
[20]. They offer a structured, reliable, and consistent available to access evidence-based resources online. This
approach to health care evidence dissemination and study aimed to explore the primary care clinicians' pref
reduce unnecessary clinica! practice variation [21]. erences for, barriers to, and facilitators of information
However, CPGs are costly to develop and update, seeking in clinica! practice in Singapore.
context-specific, and unevenly adopted across various
healthcare systems [22]. CPG's uptake is affected by
diverse factors such as presentation formats, time Methods
pressures, reputability, and ownership [14, 23]. A qualitative study consisting of semi-structured face-to
Conversely, other sources of clinicai practice-related face in-depth interviews was used to explore the primary
information may not be as valid, credible, or current as care clinicians' preferences for, barriers to, and facilita
CPGs. tors of information-seeking in clinica! practice in Singa
Increasingly, healthcare professionals worldwide use pore. The interviews were conducted between August
their smartphones as an important channel for clinica!
Lee et ai. BMC Primary
Page 3 of 11
(2024) 25: 172
Core

and November 2019 at two polyclinics and two private Data analysis
clinics in Singapore. The qualitative data were analysed using Burnard's
The study was approved by the institutional ethics com method, a structured approach for thematic content
mittee (NHG DSRB Reference Number: 2018/01355). analysis established in 1991 [36]. Burnard's method
All participants read the study information sheet before includes fourteen stages for categorising and coding
providing written consent. This study followed the Con interview transcripts [36] [see Additional file 4]. Types
solidated Criteria for Reporting Qualitative Research of questions were analysed using Ely's classification
guidelines [34] [see Additional file l]. [37]. Burnard's method enhances understanding of the
information-seeking behaviour patterns found by Ely's
approach by doing a comprehensive evaluation. Ely et al.
Participants and recruitment
(2000) developed an approach for categorising clinician
We included primary care doctors and registered nurses queries about patient care [37]. Clinical questions in pri
from the polyclinics and private primary care practices mary care were divided into several main categories. For
aged 2::: 21 years who were fluent in English. We example, the three most common categories of questions
employed convenience sampling in this study. based on Ely's approach were "What is the drug of choice
Prospective partici pants were recruited from various for condition x?", "What is the cause of symptom x?" and
polyclinics through personal contacts and "What test is indicated in situation x?" [37]. Ely et al.
advertisements. Five potential participants were (2000) framework was used by the study team to gain a
contacted but did not respond to the invitation, two better understanding of clinicians' information needs and
potential participants declined participa tion in this to identify the types of questions they had about patient
study and one potential participant resigned before the care. It was used mainly to facilitate the study team's dis
commencement of the study and hence did not cussion. The study team did not adopt the categories.
participate in the study. The analysis was clone independently and in parallel by
two researchers (MML and LTC). First, the researchers
Data collection
familiarised themselves with the transcripts by reading
The interviews were conducted by a female researcher them multiple times. Second, the initial codes were pro
(MML) in designated private meeting rooms or con posed. Third, the themes were derived from the codes.
sultation rooms at various polyclinics or the respective Fourth, the researchers discussed and combined their
consultation rooms of the private practice. MML was themes for comparison. Finally, they reached a consen
provided with sufficient details, resources, and training sus on the themes and how to define them. Apart from
on qualitative research before the study commencement. the initial stages of being acquainted with the transcripts
Before the start of the interview, the researcher intro and recommended initial codes, to streamline our codes,
duced herself, stated the aim of the interview, explained related codes were consolidated into more comprehen
confidentiality, and obtained informed consent and per sive headings. This process allows us to organise them
mission to use a digital voice recorder. The interviewees more effectively under pertinent subthemes. For exam
could pause the interviews due to professional responsi ple, various information sources that were mentioned
bilities at any time. MML conducted the interviews using by the participants such as evidence-based resources,
an interview guide based on a review of the relevant lit non-evidence-based resources, and colleagues have all
erature and team discussions [10] [see Additional file 2]. been merged into a subtheme titled "popular information
The interview topics included the type of questions dur sources" [see Additional file 3]. This process was clone
ing clinical encounters, commonly employed sources iteratively through several rounds. The final list of themes
of clinical information, frequency and timing of infor and subthemes was created by removing repeated or sim
mation-seeking, satisfaction with existing information ilar subthemes. Two other study team members indepen
sources, use of CPGs, barriers to information-seeking, dently created a list of headings without using the first
and reliability of obtained information. All interview ses study team member's list. Three lists were discussed and
sions lasted not more than 60 minutes with a mean inter improved to increase validity and reduce researcher bias.
view time of 25 minutes and were digitally recorded and Finally, we employed abstraction by developing a basic
transcribed. Field notes were taken during the interviews description of the phenomenon under investigation to
for further analysis. Data saturation, defined as no new establish the final subthemes and themes. Tables 1and 2
themes arising after three consecutive interviews [35], illustrate how these stages were conducted.
was achieved after 20 interviews, therefore we stopped Table 1 illustrates that the previous "subtheme" for
recruitment at 20 participants. Participants were com "rare condition" was "most searched information in
pensated with a SGD25 voucher and a meal upon com clinical practice," but it has been revised to "the type of
pletion of the interview.
Lee et ai. BMC Primary
Page 4 of 11
(2024) 25: 172
Core

Table 1 Excerpt of coding memo

Transcript Codes Notes

"Even doing a kind of list search to see what are the recommendations ..we seldom Rare condition Most sea rched information in clinica i
practice see in primary care" Doctor01
"Other ones that 1 would search for would be if the patient comes in with
very...unu sual presentations'.' Doctor07
"what other investigations 1 should do for conditions l'm not that familiar with'.' Doc
tor02
"Conditions, w hich we ...may not be that fam iliar with offhand'.' Doctor04

Table 2 Example of thematic content analysis process

Transcript Codes Subthemes Themes

"doing a kind of list search to see what are Rare condition The type of information needs Access ing information sources
the recommendations...we seldom see in pri-
mary care" Doctorü 1
"pharmacology, or new medicine-related infor- Pharmacology
mation,or w hat's the standard clinicai practice, "New medicine-related
information" most of the time ali this information if this are a li
under our regular clinicai practice"Doctor01 Standard clinica i practice
"teaching pedagogy skills, sometimes...1 willjust Teaching pedagogy s kills
Google or go to any other relevant webs ites
...to look for information that 1 want" DoctorO
1

information needs" to include numerou s codes such as sented in Table 3. Demographic s of clinicians (N=20).
pharmacology and others following additional discussion
with study team members. A third reviewer HES acted as
an arbiter. The coding of transcripts was performed using
a word processor. A predetermined classification system
was not employed since there was insufficient research to
inform the clinicians' perception s of information- seek
ing behaviour in Singapore. ln particular, the dynamic
identification of themes from data was facilitated using
an inductive approach. Burnard 's method was applied
inductively to establish categories and abstraction
through open coding illustrated in Tables 1 and 2. No
single method of analysis is appropriate for every type of
interview data [36]. Burnard's method focuses on a sys
tematic approach to thematic content analysis, which can
improve qualitative research objectivity and transparency
[36]. As descriptive studies can investigate perceived bar
riers to and facilitators of adopting new behaviours [38],
a more descriptive set of themes was appropriate for the
study 's objectives, and it is consistent with Burnard 's
method [36].

Results
A total of 20 clinicians were recruited . Eight doctors and
10 nurses were working in the polyclinic s. All nurses
and three doctors who participated in this study were
females. The demographics of the clinicians is repre
Lee et ai. BMC Primary
Page 5 of 11
(2024) 25: 172
Core

Table 3 Demographics of clinicians (N= 20)

Categories Frequency

Ethnicity n (%)

Chinese 15 (75)
Malay 2 (10)
lndian 3 (15)
Age group (years)
25 -30 3 (15)
3 1-35 8 (40)
36-40 5 (25)
41-45 2 (10)
46-50 2 (10)
Health professional qualificat ions
Bachelor of Medicine and Bachelor of Surgery 10 ( 100)
Master of Medicine in Family Medicine 7 (70)
Fellowship programme, College of Family Physician 2 (20)
Singapore
Graduate Diploma in Family Medicine/Physician 5 (50)
National ITE Certificate in Nursing ( 10)
Diploma in Nursing 5 (50)
Bachelor in Nursing 9 (90)
Adva nced Diploma in Nursing 2 (20)
Specialised Diploma in diabetes management and 2 (20)
cation
Lee et ai. BMC Primary
Page 6 of 11
(2024) 25: 172
Core

Thematic analysis 1) The choice of information sources

Three distinct themes were derived from the analy Is a theme that encompasses different sources clini cians
sis of the interview data, 1) the choice of information in our study used to seek and gather information .
sources, 2) accessing information sources, and 3) the Clinicians' preferred choice of information sources in
role of evidence in information-seeking [see Additional five subthemes: popular information sources, CPGs as
file 3]. This is represented in Fig. 1. Themes and sub an information source, internet as an information source,
themes derived from the interviews . peers as an information source and accessing online
information using smartphones

/
•
.

Fig. 1 Themes and subthemes derived from the interviews

Lee et ai. BMC Primary
Page 7 of 11
(2024) 25: 172
Core

Popular information sources Clinicians mentioned that If nurses had difficulty accessing CPGs, they said that
their first choice point-of-care evidence-based online they tended to seek doctors' opinion:
sources were UpToDate®, an evidence-based resource
that helps clinicians make decisions and informs their
practice [39], CPGs and the Monthly Index of Medical
'1t's very informative. It's quite clear, easy to
Specialties, followed by PubMed (Medline) and continu
refer to...in certain special cases...not stated in the
ing medication education sources. A non-evidence-based book, we will still have to seek...doctor's opinion"
information source, the Google search engine was com Nurse07.
monly mentioned as well. Lastly, clinicians often men Internet as an information source Clinicians mentioned
tioned consulting their colleagues: that the internet provided access to clinical informa tion
for practice. However, clinicians mentioned that it was
'1will Google, look for images and compare...! tell
important to ensure that the information was well
them that I'm looking because I am not sure, and I
grounded and dependable:
want tojust confirm...sometimes even show them the
photo on my phone, to ensure...what they saw, the ''...some...information might not be...so trustwor
rash...might have already disappeared is...what I thy...takes...a little additional filtering process
suspect it is:'Doctor02. before...I can say this is a reliable source or not...
some of the websites...more opinion-based ...very
'1commonly I would search...this app that I have on high...chance of bias...the reference from that writ-
my phone is called UpToDate®, right...because it's ing...written at the bottom where I can do...cross-
the most easiest...easily accessible source of checking...I think the credibility...Jor this...article
informa tion...I'll just type the whole lot into...the written is slightly higher:'DoctorOl.
Lexicomp component of the UpToDate® and
thenfrom there it tells me whether the drugs have "Ijonly you have an internet, you can always show
interactions, what kind of interactions:'Doctor07. it to the patient also. For example, when I search for
some information, I can even help in patient edu
CPGs as an information source Clinicians men
cation...Jor now, I feel it is a bit harder...And then
tioned that CPGs did not apply to all patients. Doctors
I have to rely on my phone to use the UpToDate®:'
described CPGs as evidence-based resources, designed
Doctor03.
to be safe and most relevant to practice as a baseline ref
erence. Doctors considered CPGs lengthy at times and Peers as an information source Clinicians mentioned
there was a need to apply clinical discretion when using approaching peers who were available to seek a second
them. Doctors also mentioned that CPGs focused some opinion on their clinical questions. They also mentioned
times on cost-effectiveness instead of the quality of care: that they tended to approach experts:
'1 think they are useful in summarising the latest
evidence and what...is recommended, especially if ''...it's really a case-to-case basis and it depends if
they are local clinicai practice guidelines, then it's the colleagues around ...Also it depends on the
tailored to our own population ...And keeping in prox imity of the colleague. If the colleague knows
mind perhaps the cost sensitivities, cost a lot but...busy in another roam on another levei
ejfectiveness" Doctor02. then I might approach next doar colleagues
instead." Doc tor06.
Nurses said that they saw CPGs as a standard of prac
tice for clinicians and an easy resource to refer to. How
'1think most of time, if we are going to get our
infor mation immediately, we'll call one of our
ever, some nurses said that they found CPGs difficult to
colleagues
access and outdated:
here...discuss the case...we'll come to a consensus,
what will be the best for our kind of patient ...con
''...but it's not so...easy to access...because you have tribute to the informed decision immediatelY:'
to...enter certain keywords, and sometimes it's not Doc torOl.
that keyword that's going to churn out ali the infor
Accessing online information using smartphones Clini
mation you see...like, try afew times...want to make
cians mentioned that their smartphones were conveni ent
sure that...I'm doing things correctly...Jollowing the
for accessing information for practice. For instance,
guidelines ...just quickly...log into the intranet
accessing the UpToDate® app and Google search engine
and ... searchfor the information:'NurseOl.
using smartphones:
''...commonly I would search...this app that I have
Lee et ai. BMC Primary
Page 8 of 11
(2024) 25: 172
Core

on my phone is called UpToDate®...because it's the The timing and frequency of using CPGs Clinicians said
most easiest...easily accessible source of informa that they commonly use CPGs daily or when there was a
tion...I'll just type the whole lot into...the Lexicomp change or update to the CPGs:
component of...UpToDate® and then from there it
''...day to day, because all these guidelines
tells me whether the drugs have interactions." Doc
I'm famil iar with, it's in my memory...internally we
tor07.
do have guidelines for certain acute
conditions:'Doctor02 .
"J will go on the internet...if I needed information
about...certain medical conditions..Just defini
"Not so many cases...It's quite rare, actually...
tions, just to have an idea of, you know...
Because most of our cases are quite common...we
Correct, pure Google."NurseOl .
still can deal with...Yes...Maybe once afew weeks...
Once a month...When I have concerns or any
doubts...After patient left...yes. M aybe, sometimes
2) Accessing information sources
... And after the doctors consult:' NurseOS .
Is a theme that encompasses different aspects of infor
mation-seeking and access by clinicians in our study. Clinicians discussed convenience, easy access, the
Factors influencing clinicians' utilisation of information trustworthiness of information, having colleagues who
sources in five subthemes: type of information needs, are specialists, and being keen to keep up-to-date as the
the timing and frequency of information needs, the facilitators to seeking clinical information:
timing and frequency of using CPGs, information-seek
ing facilitators and information-seeking barriers . Information-seeking facilitators Clinicians discussed
convenience, easy access, the trustworthiness of infor
The type of information needs Clinicians mentioned mation, having colleagues who are specialists, and being
that they commonly sought information on less common keen to keep up-to-date as the facilitators to seeking clin
health areas such as unusual skin rashes, rare diseases, ical information:
paediatrics, women's health, medications, and at times
'7 find ...clinical practice guidelines quite useful...
concerning all clinical areas:
since it's on our terminal. I do open that up to look
"Drug information...maybe dosing and every at it...it does give us quite a convenient and nofuss
thing...when we are prescribing for paediatric ...we way to be able to access them on our terminal while
also see female patients who are pregnant ...Lac we are seeking information whether during or even
tating, and ali... contraindicated" Doctor03 . after consults:'Doctor06 .

"Other ones that I would search for would be if "work instructions...Policies and protocols ...
the patient comes in with very...unusual presenta Intranet...So I just want to make sure that...I'm
tions."Doctor07 . doing things correctly, that I'm, you know, following
the guidelines. So J'll just quickly enter, you know,
The timing and frequency of information needs Clini
log into the intranet and just search for the
cians explained that they commonly seek clinical infor
informa tion...The information that's on the intranet
mation daily or several times a week . They said that they
has, you know, been validated by an expert, you
either seek information at the point-of-care or at home:
know...So that's why I rely heavily on it."NurseOl.
"J will look at least weekly once...It's of my own
interest...Not during working times, most of the
Information-seeking barriers Clinicians mentioned
time...When we are travelling, in M RT...Sometimes
that internet surfing separation, the lack of time, limited
at home also."NurselO.
access to medical literature databases, and limited search
function in the organisation's server were barriers to
"Not so many cases...It's quite rare, actually...
seeking clinical information:
Because most of our cases are quite common...
we still can deal with...Yes...M aybe once a few "The information I know is there...But it's not so
weeks...Once a month...When I have concerns or easy to search for. ..Not user-friendly, not very
any doubts...After patient left...yes . M aybe, some exhaus tive...Sometimes you just have to...trial-and-
times...And after the doctors consult."NurseOS . error... different keywords:'NurseOl.
Additionally, clinicians frequently mentioned using
smartphones to access clinical information.
Consequently,
Lee et ai. BMC Primary
Page 9 of 11
(2024) 25: 172
Core

doctors said that they were worried that using smart Discussion
phones during a clinical consultation might make them To our knowledge, this is the first study conducted in
seem unprofessional to patients: Singapore to investigate the primary care clinicians' pref
erences for, barriers to, and facilitators of information
'1need to explain to the patient that...I am using my
seeking in clinical practice. Clinicians' mostly researched
phone because I don't have internet access or may
information on conditions such as unusual skin rashes,
appear rude to the patient; I am surfing my phone in
rare diseases, paediatrics, and women 's health. Most
the middle of the consult:' Doctor02 .
clinicians searched clinical information at the point-of
Doctors reported that they were also concerned about care daily for a variety of reasons, including personal
their privacy when they showed their smartphones to interest, clarification of doubts, or self-improvement .
their patients: Sources of information included CPGs, online evidence
based resources, the internet, peers, and smartphones.
''...sometimes ...you don't want to show your phone
Although CPGs were clinicians' preferred sources of
to them(patients) also...Because sometimes you may
information, they did not refer to them regularly and only
have other notifications:'DoctorOS .
did so in memory or when the guidelines were updated.
We also found that using smartphones for seeking clinical
information was commonly reported among clinicians.
3) The role of evidence in information-seeking
The barriers to primary care clinicians' information
Is a theme that explores the role of evidence in clini seeking process were the lack of time, internet surfing
cians' information-seeking in our study. The value of separation of work computers, limited search function of
scientific research for clinicians seeking information in their organisation's server, and limited access to medical
two subthemes: the importance of trustworthy informa literature databases. The facilitators to primary care cli
tion sources and employing evidence-based information nicians' information-seeking process were convenience,
sources. ease of access, and the trustworthine ss of the information
sources.
The importance of trustworthy information sources Like other studies [3, 8, 20, 40, 41], we found that the
Cli nicians agreed that peer-reviewed clinical choice of information sources was affected by the trust
information was reliable. Additionally, doctors expressed worthiness and availability of resources . CPGs were pre
trust in clinical information if there were frequent ferred among clinicians as they were written by experts
updates of the content: or specialists in their field. However, some clinicians felt
''...they(UpToDate ®) do put ...the date of which that CPGs were too lengthy to be used at the point-of
they have updated the articles...it's from multiple care, outdated, and difficult to locate on their organisa
sources...citations and...management ...seems quite tion's server. Additionally, clinicians only referred to
sound:'Doctor06 . CPGs recalled from memory or when they were updated .
This highlights the importance of providing an alterna
"The information that's on the intranet has...been tive evidence-based clinical resource that is succinct and
validated by an expert."NurseOl. easy to refer to at the point-of-care [42]. Using medical
apps for the provision of point-of-care summaries may
Employing evidence-based information sources Clini mitigate the challenges of using CPGs for clinical infor
cians mentioned that emphasising the importance of mation. Correspondingly, clinicians in the polyclinics
evidence in patient care and building an evidence-based commonly referred to the UpToDate® app provided by
culture in the workplace helps to encourage the use of their organisation as a point-of-care resource they could
evidence-based information sources in practice. use on their smartphones. Evidence-based point-of -care
'1don't have any concrete kind of suggestions now resources are commonly presented in key point summa
but...perhaps find some ways to sustain interest...to ries, follow formal categorisation of medical conditions,
remind us that we're doing thisfor best of patients and provide references [43]. Limited research has shown
." Doctor06 . that it was beneficial to integrate UpToDate® searches
into daily clinical practice [42]. Additionally, the Ameri
can Accreditation Commission International's @TRUST
"Ij I have discussions with my peers regarding
programme is one framework designed to encourage
cases then I will, like, refer back to the...to...the
trustworthy online content. It is an invaluable resource
CPG and things like that...I think the
for both individuals looking for health information online
conference...or the... forums they are also a very
and organisations attempting to deliver trustworthy
good source of informa tion:'Nurse03.
Lee et ai. BMC Primary
Page 10 of
(2024) 25: 172
Core

content [44]. However, continual efforts are required to and an open-ended style of questioning. The interactive
encourage its use and ensure that individuals have access nature of our interviews provided richer context and room
to accurate and reliable health information online. There for free responses from the interviewees. We were then able
fore, future studies should investigate the quality of exist to critically scrutinise the conversations and provide insights
ing medical apps in providing point-of-care summaries that were helpful in the final analysis of themes.
and the effects of their use in the primary care setting. There are several limitations. Firstly, we did not explore
We also found that clinicians were seeking clinical the influence of gender and age in the participants' infor
information on their smartphones. This is not surpris mation-seeking behaviour, which has been demonstrated
ing as Singapore's public healthcare institutions enforce in other research in this area [14]. Secondly, the study
internet surfing separation on work computers. Further was limited by environmental factors in the workplace,
more, with the high penetration of smartphones in Singa such as internet and information access. Finally, there
pore [45], these <levices became the next best alternative may be possible social desirability bias, whereby the par
for clinicians to seek online clinical information. Clini ticipants may have presented responses that were more
cians in the polyclinics frequently cited using UpToDate® socially appropriate than their actual thoughts on the
app and the Google search engine on their smartphones. issues explored during the interviews.
Similar to another study [46], we found that doctors often
used Google images on their smartphones to identify less Conclusion
common rashes. Additionally, our study found that clini We found that clinicians frequently sought answers to
cians use Google images to educate patients. However, clinical queries arising from patient care. However, the
clinicians in the polyclinic reported privacy and profes choice of information sources was influenced by the
sionalism concerns as barriers to using smartphones for trustworthiness and availability of the resources. Clini
clinical consultations. These findings were consistent cians in the polyclinic commonly reported using their
with a systematic review assessing the challenges and smartphones for practice. Using UpToDate® app and
opportunities of using mobile <levices by healthcare pro Google search engine was commonly cited as their pre
fessionals [47]. Despite the internet surfing separation in ferred clinical information sources due to its conveni
public healthcare institutions in Singapore and the avail ence and accessibility. While our findings may have been
ability various information sources, we found similar bar reported in other contexts, there are significant and
riers to clinicians seeking clinical information with other novel elements when compareci to healthcare around
studies [3, 20, 48]. Future research may focus on address the world. For example, the implementation of internet
ing specific barriers to using various mobile <levices by surfing separation in public healthcare institutions raises
primary care clinicians at the point-of-care. concerns regarding clinicians' usage of smartphones, as
Finally, smartphones may be an important information well as their privacy and professionalism. This may lead
seeking channel for healthcare professionals, and the hos us to examine the need for some regulation and train
pital or government may be forced to establish legislation ing on the use of smartphones among clinicians, as
to protect healthcare professionals who use smartphones well as the necessity to investigate this further from the
in clinical practice. Compliance with legislation govern patient's perspective. Future studies to improve access
ing smartphone use at work may be examined during the to evidence-based clinical information sources other
evaluation process for healthcare professionals. Guide than CPGs should be explored to address the informa
lines on smartphone use among healthcare professionals tion needs of primary care clinicians. Studies examin
can be tailored to individual conditions, such as patients' ing trustworthiness and effectiveness of using app-based
permission to share medically sensitive information via point-of-care information summaries and exploring the
text. As a result, guidelines could be based on best prac impact of using mobile <levices for information-seeking
tice claims and common actionable statements. Addi by clinicians at the point-of-care will also be useful to
tionally, this study suggests that clinicians have, for the address the information-seeking needs of primary care
most part, been left to navigate information access on clinicians. Furthermore, Large Language Model (LLM)
their responsibility, which may not be the most effec based artificial intelligence (AI) systems, such as Chat
tive. Developing a more robust culture of evidence-based GPT, are increasingly being developed and used. They are
medicine within the organisation is essential and ought used in various disciplines, including healthcare. Some,
to be explicitly promoted moving forward. It could be such as AMIE (Articulate Medical Intelligence Explorer)
beneficial for clinicians to receive organised training on and Pathways Language Model (Med-PaLM 2), have
effective information-seeking strategies and resources. been developed specifically for healthcare [49-51]. More
Our study has several strengths and limitations. Our research into the usage of AI among clinicians is needed
strength is that we employed an in-depth interview approach to assure trust, dependability, and ethical conduct.
Lee et ai. BMC Primary
Page 11 of
(2024) 25: 172
Core

Abbreviations References
GP General Practitioner 1. Ministry of Health. Primary healthcare services Singapore 2022 [updated
CPG Clinicai practice guidelines 31/05/2022]. Available from: https://www.moh.gov.sg/home/our-healthcare
system/healthcare-services-and-facilities/primary-healthcare-services .
2. González-González AI, Dawes M, Sánchez-Mateos J, Riesgo-Fuertes R,
Supplementary lnformation Escortell-Mayor E, Sanz-Cuesta T, et ai. lnformation needs and
The online version contains supplementary material available at informa tion-seeking behavior of primary care physicians. The Annals
https://doi. org/l 0. 1186/sl 2875-024-02429-x . of Family Medicine. 2007;5(4):345-52.
3. Daei A, Soleymani MR, Ashrafi-rizi H, Zargham-Boroujeni A, Kelishadi
R. Clinicai information seeking behavior of physicians: A systematic
Supplementary Material 1 . Supplementary Material 2. review. lnt J Med lnformatics. 2020;139:104144.
Supplementary Material 3. 4. Amiel JM, Andriole DA, Biskobing DM, Brown DR, Cutrer WB, Emery
Supplementary Material 4. MT, et ai. Revisiting the core entrustable professional activities for
entering residency. Acad Med. 2021;96(7S):S14-21 .
5. College of family physicians singapore. Fellowship Programme (FCFPS)
Singapore2022 [updated 2022]. Available from: https://www.cfps.org.sg/
programmes/ fellowship-programme-fc fps/.
Acknowledgements 6. American Library Association . lnformation Literacy Competency Stand
Not applicable . ards for Nursing Unites States of America2013 Available from: https://
www.ala .org/acrl/standards/nursing .
Authors' contributions 7. Braun L, Wiesman F, den Herik Van H, Hasman A Avoiding
Lorainne Tudor Car conceived the idea for this study. Tang Wern Ee litera ture overload in the medical domain. Stud Health
contrib uted to the design of the work and the acquisition of the data. Technol lnform. 2006; 124:497-502.
Mauricette Lee collected the data, analysed it and wrote the manuscript 8. Clarke MA, Belden JL, Koopman RJ, Steege LM, Moore JL, Canfield
with support from Tang Wern Ee, Helen Smith, and Lorainne Tudor Car. SM, et ai. lnformation needs and information-seeking behaviour
Lorainne Tudor Car and Tang Wern Ee supervised the project. analysis of
primary care physicians and nurses: a literature review. Health lnfo
Funding
Libr J. 20 l3;30(3):1 78-90.
This study is funded by Seedcorn Grant Centre for Primary Health Care 9. Ely JW, Burch RJ, Vinson DC. The information needs of family
Research and lnnovation, a joint Lee Kong Chian School of Medicine, and
physicians: case-specific clinicai questions. J Fam Pract.
the National Healthcare Group Polyclinics lnitiative.
l992;35(3):265-9 .
1O. Dei Fiol G, Workman TE, Gorman PN. Clinicai questions raised by
Availability of data and materiais
clinicians at the point of care: a systematic review. JAMA lntern
The datasets generated and/or analysed during the current study are not
Med. 2014;174(5):710-8.
publicly available due the fact that ali data obtained during the course of
11. Al-Dousari E. lnformation Needs and lnformation Seeking Behaviour of
this study is strictly confidential and will be kept by the study team at the
Doctors in Kuwait Government Hospitais: An Exploratory Study :
end of the study for at least 6 years and disposed of according to the
Lough borough University; 2009.
Personal Data Protection Act in Singapore. Data are however available from
12. Young JM, Ward JE. Evidence-based medicine in general practice:
Associate Profes sor Tang Wern Ee (co-author) upon reasonable request
beliefs and barriers among Australian GPs. J Eval Clin Pract. 2001
and with permission of the ethics committee of National Healthcare Group
;7(2):201 -1O.
Domain Specific Review Board (the central ethics committee).
13. Ellsworth MA, Homan JM, Cimino JJ, Peters SG, Pickering BW, Herasevich
V. Point-of-care knowledge-based resource needs of clinicians: a
Declarations survey from a large academic medical center. Appl Clin lnform.
2015;6(2):305-1 7.
Ethics approval and consent to participate 14. Le JV, Pedersen LB, Riisgaard H, Lykkegaard J, Nexoe J, Lemmergaard J,
This study was approved by the National Healthcare Group Domain Specific et ai. Variation in general practitioners' information-seeking behaviour
Review Board (the central ethics committee). National Healthcare Group Domain - a cross-sectional study on the infiuence of gender, age and practice
Specific Review Board Reference Number: 2018/01355 . lnformed consent was form. Scand J Prim Health Care. 2016;34(4):327-35.
obtained from ali participants. All methods were carried out in accordance with 15. Bruin-Huisman L, Abu- Hanna A, van Weert H, Beers E. Potentially inap
relevant guidelines and regulations, in accordance with the Declaration of propriate prescribing to older patients in primary care in the
Helsinki. Netherlands: a retrospective longitudinal study. Age Ageing.
2017;46(4):614-9.
Consent for publication 16. Cahir C, Bennett K, Teljeur C, Fahey T. Potentially inappropriate
Not applicable . prescribing and adverse health outcomes in community dwelling older
patients. Br J Clin Pharmacol. 2014;77(1 ):201-1O.
Competing interests 17. Davies K. The information-seeking behaviour of doctors: a review of
The authors declare no competing interests. the evidence. Health lnfo Libr J. 2007;24(2):78-94.
18. Gill P. Dowell AC, Neal RD, Smith N, Heywood P. Wilson AE . Evidence
Author details based general practice: a retrospective study of interventions in one
1
Lee Kong Chian School of Medicine, Nanyang Technological University, training practice. BMJ. 1996;312(7034):819-21.
Singapore, Novena Campus Clinicai Sciences Building 11 Mandalay Road, 19. Salisbury C, Bosanquet N, Wilkinson E, Bosanquet A, Hasler J. The
Singapore 308232, Singapore. 2Singapore lnstitute ofTechnology, 1O Dover Drive, imple mentation of evidence-based medicine in general practice
Singapore 138683, prescribing. Br J Gen Pract. l998;48(437):1849-52.
Singapore. 3Clinical Research Unit, National Health Group Polyclinics (HQ), 3 20. Aakre CA, Maggio LA, Fiol GD, Cook DA Barriers and facilitators to
Fusio nopolis Link, Nexus @One-North, Singapore 138543, Singapore. 4Family clini cai information seeking: a systematic review. J Am Med lnform
Medicine Assoe. 20 l9;26(1O):1129-40.
and Primary Care, Lee Kong Chian School of Medicine, Nanyang Technological 21. . Scott SD, Grimshaw J, Klassen TP, Nettel-Aguirre A, Johnson DW.
Uni versity. Singapore. Novena Campus Clinicai Sciences. Building 11 Mandalay Under standing implementation processes of clinicai pathways and
Road. Singapore 308232, Singapore. 5Department of Primary Care and Public clinicai practice guidelines in pediatric contexts: a study protocol.
Health, School of Public Health, Imperial College London, London, UK. lmplement Sei. 2011;6(1):133.
22. O'Brien JA, Jacobs LM Jr, Pierce D. Clinicai practice guidelines and
Received: 22 December 2022 Accepted : 12 May 2024 the cost of care: a growing alliance. lnt J Technol Assess Health
Published online: 18 May 2024 Care. 2000;16(04):1077-91
23. Langley C, Faulkner A, Watkins C, Gray S, Harvey 1. Use of guidelines
in primary care-practitioners'perspectives. Fam Pract. 1998;15(2):105-
11.
Lee et ai. BMC Primary
Page 12 of
(2024) 25: 172
Core

24. Al-Ghamdi S. Popularity and impact of using smart devices in 46. Cook DA, Sorensen KJ, Hersh W, Berger RA, Wilkinson JM. Features of
medicine: experiences in Saudi Arabia. BMC Public Health. 2018;18(1 effective medical knowledge resources to support point of care
):531. learning: a focus group study. PLoS ONE. 2013;8(1 l):e803 l8.
25. Ozdalga E, Ozdalga A, Ahuja N. The smartphone in medicine: a review of 47. Gagnon M-P. Ngangue P. Payne-Gagnon J, Desmartis M. m-Health
current and potential use among physicians and students. J Med adop tion by healthcare professionals: a systematic review. J Am
Internet Res. 2012;14(5):el28. Med lnform Assoe. 2015;23(1 ):212-20.
26. Hedhli A, Nsir S, Ouahchi Y. Mjid M. Toujani S, Dhahri B Contribution 48. Brassil E, Gunn B, Shenoy AM. Blanchard R. Unanswered clinicai
of mobile applications to learning and medical practice. Tunis questions: a survey of specialists and primary care providers. J Med Libr
Med. 2021 ;99(12):1134-40. Assoe. 2017;105(1):4-1 1 .
27. Liu Y. Ren W, Qiu Y. Liu J, Yin P. Ren J. The Use of Mobile Phone and 49. Thirunavukarasu AJ, Ting DSJ, Elangovan K, Gutierrez L, Tan TF, Ting
Medi DSW. Large language models in medicine. Nat Med. 2023;29(8):1930-
cal Apps among General Practitioners in Hangzhou City, Eastern 40.
China. JMIR mHealth uHealth. 2016;4(2):e64. 50. Tu T, Palepu A, Schaekermann M. Saab K, Freyberg J, Tanno R, et
28. Ventola CL. Mobile devices and apps for health care professionals: ai. Towards conversational diagnostic ai. arXiv preprint
uses and benefits. P T. 2014;39(5) :356-64. arXiv:240105654 . 2024.
29. Gagnon MP, Pluye P. Desmartis M. Car J, Pagliari C. Labrecque M. 51. . Li J, Dada A, Puladi B, Kleesiek J, Egger J. ChatGPT in healthcare: A
et ai. A systematic review of interventions promoting clinicai taxonomy and systematic review. Comput Methods Programs Biomed.
information retrieval technology (CIRT) adoption by healthcare 2024;245:108013.
professionals. lnt J Med lnformatics. 2010;79(10):669-80.
30. Division NPaT. Population in Brief 2023: Key Trends 2023 [updated 29
Sep 2023]. Available from: https://www.population.gov.sg/media-centre/ Publisher's Note
articles/population-in-brie f-2023-key-trends/# :-:text=Overall%2C% Springer Nature remains neutral with regard to jurisdictional claims in pub
20Singapore's%20tota1%20population%20stood,5.0%25%20increase% lished maps and institutional affiliations.
20from%20June%202022 .
31. Department SR. Number of smartphone users in Singapore from
2019 to 2028 2023 [updated 12 Sep 2023]. Available from:
https://www.stati sta.com/statistics/494598/smartphone-users-in-
singapore/# :-:text=ln% 202022%2C%20the%20number%20o f,over
%206 .l6%20mill ion%20by% 202028.
32. Maggio LA, Aakre CA, Dei Fiol G, Shellum J, Cook DA lmpact of
electronic knowledge resources on clinicai and learning outcomes:
systematic review and meta-analysis. J Med Internet Res.
2019;21(7):el3315.
33. Health Mo. Temporary internet surfacing separation implemented at
all public healthcare clusters 2018 [updated 07/11/2022]. Available
from: https://www.moh.gov.sg/news-h igh1 ights/detai1s/tempora ry-
internet surfacing-separation-impiemented-at-a 11-public-hea lthcare-
clusters.
34. Booth A, Hannes K, Harden A, Noyes J, Harris J, Tong A COREQ
(Consoli dated Criteria for Reporting Qualitative Studies). Guidelines for
Reporting Health Research: A User's Manual2014. p. 214-26.
35. Saunders B, Sim J, Kingstone T, Baker S, Waterfield J, Bartlam B, et
ai. Saturation in qualitative research: exploring its conceptualization
and operationalization. Qual Quant. 2018;52(4):1893-907.
36. Burnard P. A method of analysing interview transcripts in
qualitative research. Nurse Educ Today. 1991;11 (6):461 -6.
37. Ely JW, Osheroff JA, Gorman PN, Ebell MH, Chambliss ML, Pifer EA,
et ai. A taxonomy of generic clinicai questions: classification study.
BMJ. 2000;32 l(7258):429-32.
38. Korstjens 1, Moser A Series: Practical guidance to qualitative
research. Part 2: Context, research questions and designs. Eur J
Gen Pract. 2017;23(1 ):274-9
39. Wolters Kluwer. UpToDate: lndustry-leading clinicai decision support
2023 Available from: https://www.wolterskluwer.com/en/solutions/
uptodate.
40. Dawes M. Sampson U. Knowledge management in clinicai
practice: a systematic review of information seeking behavior in
physicians. lnt J Med lnformatics. 2003;71 (1):9-15 .
41. . Correa VC. Lugo-Agudelo LH, Aguirre-Acevedo DC. Contreras JAP,
Borrero AMP, Patino-Lugo DF, et ai. Individual, health system, and
contextual bar riers and facilitator s for the implementation of
clinicai practice guidelines: a systematic metareview. Health Res
Policy Syst. 2020;18(1 ):74.
42. Low S, Lim T. Utility of the electronic information resource
UpToDate for clinicai decision-making at bedside rounds.
Singapore Med J. 2012;53(2):116-20.
43. Campbell JM. Umapathysivam K, Xue Y, Lockwood C. Evidence-Based
Practice Point-of-Care Resources: A Quantitative Evaluation of Quality,
Rigor, and Content. Worldviews Evid Based Nurs. 2015;12(6):313-27 .
44. American Accreditation Commission lnternational. @TRUST
Certificate 2024 [updated 2024]. Available from:
https://aacihealthcare .com/certi ficates/cl 73-2022-trust-usa/.
45. Statista Research Department. Smartphone market in Singapore-Statis
tics and facts 2022 [updated 30/08/2022]. Available from:
https://www. statista
.com/topics/5842/smartphones-in-singapore/#dossierKeyfigures .