Article Text

Exploring communication preferences and risk thresholds of clinicians and parents of febrile infants under 90 days presenting to the emergency department: a qualitative study
  1. Kathryn Wilson1,2,
  2. Etimbuk Umana1,
  3. David McCleary1,
  4. Thomas Waterfield1,
  5. Kerry Woolfall3
  1. 1Wellcome Wolfson Institute for Experimental Medicine, Queen's University Belfast School of Medicine, Dentistry and Biomedical Sciences, Belfast, UK
  2. 2Royal Belfast Hospital for Sick Children, Belfast, UK
  3. 3University of Liverpool, Liverpool, UK
  1. Correspondence to Dr Kathryn Wilson; kath_wilson{at}


Background Febrile infants under 3 months of age are at higher risk of invasive bacterial illness (IBI) when compared with older children. Increasingly sequential assessment based on age, clinical appearance and biomarkers is used to determine the risk of IBI, and appropriateness of invasive procedures such as lumbar puncture. The purpose of this qualitative study is to report parents and clinicians’ opinions on communication of risks and benefits of sequential assessment and tailored treatment.

Methods 18 parents enrolled in the Febrile Infant Diagnostic Assessment and Outcomes study and seven clinicians from England, Wales and Northern Ireland were purposively selected to participate in virtual qualitative interviews. Data were analysed thematically.

Results Tailored treatment plans were widely supported. Confidence in the clinician was central to parents’ attitude towards management recommendations. Parents’ decision-making preferences change throughout their child’s clinical journey, with an initial preference for clinician-led decisions evolving towards collaborative decision-making as their stress and anxiety reduce. There were widespread differences in preferences for how risk was discussed. Parents self-reported poor retention of information and felt communication adjuncts helped their understanding. Clinicians were generally positive about the use of clinical decision aids as a communication tool, rather than relying on them for decision-making.

Discussion Parents want to feel informed, but their desire to be involved in shared decision-making evolves over time.

Clinicians appear to use their clinical judgement to provide individualised information, evolving their communication in response to perceived parental needs.

Poor information retention highlights the need for repetition of information and use of communication adjuncts.

Trial registration number NCT05259683.

  • Emergency Care
  • Health services research
  • Paediatric Emergency Medicine
  • Sepsis
  • Qualitative research

Data availability statement

Data are available upon reasonable request.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Parents of sick children generally prefer collaborative decision-making. It is unclear how parents of febrile young infants prefer information regarding investigations and treatment to be communicated.


  • Decision-making preferences evolve during the emergency admission from clinician led to collaborative. Acceptance of sequential assessment and tailored care is high but requires guidelines or consensus statement on use. There is variance in parent and clinician preferences for discussing risks and benefits of the clinical management of febrile infants.


  • Creation of a communication framework assists the tailoring of communication to address parents’ needs throughout their child’s admission to hospital.


Febrile infants under 3 months of age are frequently seen in the emergency department (ED) and provide diagnostic challenges.1 This is primarily due to difficulties in differentiating bacterial and viral infections, and clinical concerns regarding higher rates of bacteraemia and meningitis in this cohort. The National Institute for Health and Care Excellence (NICE) guideline NG51 suggests treating all these infants as high risk for serious bacterial illness (SBI) and invasive bacterial illness (IBI).2 This has contributed to significant overinvestigation and treatment of this group, with a 90% admission rate and 80% rate of intravenous antibiotic (IVAB) use, despite an IBI incidence of 2–3%.3 Guidelines from the British Society for Antimicrobial Chemotherapy and NICE guideline NG143 advocate a tailored approach to management based on clinical and laboratory findings.4 5 Research from the American Academy of Pediatrics suggests that ‘we can safely do less’,6 and advocates for the use of clinical decision aids (CDAs) to augment decision-making. Reducing the number of infants having unnecessary investigations has medicoeconomic benefits for hospitals, reduces distress caused by invasive procedures and prolonged hospital admissions and minimises the long-term sequelae of antibiotic use. This includes increased rates of allergies, obesity and changes to the gut biome.7–9

While clinicians routinely balance risk in their decision-making, strict protocols can give an illusion of safety, and deviating from these may be seen as unacceptably high risk. Previous studies suggest that risk thresholds may vary depending on both clinical role and experience.10 11

The time critical management of febrile infants makes effective communication with families crucial, particularly regarding decisions about investigations such as venepuncture and lumbar puncture (LP) and initiation of treatment. To date, there is little evidence reflecting parental communication preferences in this patient group. Existing literature surrounding parental priorities when their child is unwell elucidated themes of varying information needs, the importance of shared decision-making and concerns around the discharge procedure.12–17


This study aimed to explore how clinicians discuss risk with parents of febrile infants, how parents understand these risks and how risk is balanced in decision-making. This included creating a communication framework for discussing sequential assessment and tailored care in highly stressful settings, based on parental and clinician views and experiences.


The Febrile Infant Diagnostic Assessment and Outcomes (FIDO) study protocol has been published and outlines the methodology for the qualitative element of the study.18


Parents/guardians of infants recruited into the FIDO study received a participant information sheet, detailing the purpose of this qualitative study. Those interested in participating in the interviews provided contact details on the FIDO consent form. Purposively selected individuals were contacted by phone with an email link sent to interested parties inviting them to book an interview through an online platform.

Clinicians who had recruited infants to the FIDO study were invited to register interest in being interviewed. Those who did so were provided with a practitioner information sheet prior to the interview.

Participant selection

Participants were purposively selected from FIDO sites across England, Wales and Northern Ireland to maximise sample variance. Clinician selection was based on location, clinical experience and hospital type. Parental selection was based location, parental gender and infant’s diagnosis. All infants had been seen in hospital within 2–6 weeks of the interview and were under 3 months old with a recorded febrile episode at the time of presentation.

Sample size

Based on previous qualitative studies in this setting we anticipated interviewing 15–25 participants to reach the point of information power, which considered sample variance and quality of dialogue in meeting the aims of the study.19–21

Topic guides

Topic guides were created and developed by the research team as part of an interpretive and iterative process as the study progressed. Semistructured interviews were carried out by KWi (MB BCh BAO, MRCPCH) following online and in-person training in qualitative interviewing techniques. Participants were contacted by KWi by phone and email, informed of the study aims and given opportunities to discuss the planned interview prior to their scheduled date. Participants were made aware of KWi’s background as paediatrician who was also a researcher on other aspects of the FIDO study. Two parents had been consented by KWi, and one clinician had previously worked with her. As FIDO is an observational study, infants’ treatment will not have been impacted by inclusion in the trial, therefore a single topic guide focusing on the participants’ experience in hospital was designed.

Iterative topic guides were edited by the research team (KWi, EU, TW) after a pilot interview. Editing was completed after two interviews were coded by KWi, EU and KWo, including addition of additional questions for clinicians about topics that were identified as important by parents in their interviews. Further editing was completed after 10 interviews had been coded and field notes discussed by KWi and KWo. Interviewing was stopped when the research group agreed data saturation had been reached.

Interview platforms

Interviews were carried out remotely via online conferencing platforms and involved a combination of video and audio recordings. The Standards for Reporting Qualitative Research and the Consolidated Criteria for Reporting Qualitative Research checklists are shown in the online supplemental material.


Data were analysed thematically with a focus on the stated objectives of this study. Anonymised interview data were organised and indexed using QSR NVivo V.14 software. Second coding was carried out by an expert member of the team, KWo, who met regularly with KWi to develop the framework. Thematic analysis was informed by the constant comparison approach of grounded theory, with interviews modified to encompass the criterion of catalytic validity.22 Parent and clinician interviews were coded separately, but with a complementary mirroring of the major themes, as seen in figure 1.

Figure 1

NVivo coding tree for qualitative interviews.



15 parent interviews were conducted with 18 parents (three couples interviewed).

Seven clinician interviews were completed. No further or repeat interviews were required as the point of information power was reached.

78% (n=14) were mothers of infants. The remaining 22% (n=4) were fathers. Age range was 22–37 years with a median age of 33 years. 83% (n=15) of parents described themselves as white/white British. The remainder were made up of Asian, Asian American and British Caribbean participants. Interview length was 27–86 min (median 39 min). Infants were recruited from 10 FIDO sites selected to include secondary and quaternary care. Two were from Northern Ireland, one from Wales and the remaining 12 from England.

The infants’ diagnosis was bacterial meningitis in 13% (n=2), urinary tract infection (UTI) in 13% (n=2), one of which also had bacteraemia (7%), skin infection in 7% (n=1) and viral or presumed viral in 60% (n=9), including viral meningitis in 13% (n=2). IBI was present in 13% (n=2), and SBI in 33% (n=5). One child had a serious congenital condition diagnosed during their admission.

93% (n=14) were admitted, 87% (n=13) were treated with IVAB and 67% (n=10) underwent LP.

The majority of clinicians interviewed were trainees (n=6, 86%). This reflects the prevalence of non-consultants treating febrile infants in the prospective FIDO study (83%). Clinicians worked in both tertiary paediatric EDs (42%, n=3) and district general hospitals (68%, n=4). 42% were from a paediatric background (n=3) and 42% (n=3) from emergency medicine, one was dual trained and one interviewee was a general practice trainee. Age range was 28–42 years with a median age of 33 years. Five identified as white British or white Irish and two as Asian or British Asian. Interviews lasted 38–66 min (median 41 min). One clinician was based in Northern Ireland, two in Wales and four in England. The Consolidated Standards of Reporting Trials diagram of patient recruitment can be seen in figure 2.

Figure 2

Consolidated Standards of Reporting Trials (CONSORT) diagram of participant recruitment.


Clinicians described a hospital cultural shift towards more individualised care based on sequential assessment. Consensus on what was an acceptable level of risk would increase clinicians’ confidence in using CDAs. CDAs were believed to be a valuable resource as a communication aid, and to confirm decision not to treat with IVAB or complete invasive investigations such as LPs.

Even if it’s just if you’re using it as a tool to reassure parents if the child is well and your plan is for them to go, if you can give them an actual quantifiable number and say, ‘Actually, we’ve calculated your risk as this.’ (C003)

CDAs would not over-ride clinical judgement if a child was felt to be unwell and were felt to be of more use to junior staff.

Most parents felt well informed during their admission. While a minority expressed a desire to have all information up front, most found the initial time in ED overwhelming and expressed a preference for staggered information provision: ‘For me, that was right to have just gradually drip-fed it… it was definitely explained to me on my terms’ (P003). Repetition of the same information by different clinicians increased confidence in caregivers. Parents who had received a patient information leaflet (PIL) about investigations or discharge advice found this reassuring and expressed that they felt they absorbed the information better having had a chance to read, reflect and ask questions. ‘I remember reading about that [sepsis] and thinking ‘I can’t see that it is something else’’ (P003). Those that did not receive a PIL felt that having one would have been beneficial. ‘If that would be available, then that would be a great resource for parents to have’ (P004).

Parents who were given time to absorb information and ask questions related that this was a positive experience for them, as documented in box 1. Most reported negative experiences centred around communication. This included being dismissive of parental concerns, ‘you shouldn’t actually be here for a temperature’ (P004), not giving information when requested, ‘Unfortunately he failed to answer both my questions and he just kept saying…. If she doesn’t have this test, then she would end up with brain damage’ (P006), poor handover of information between staff and lack of clear management plans, ‘I thought she was coming back to see us but she’d actually left… I obviously then had nobody to ask the question of, ‘Where do we go from here?’’ (P011).

Box 1

Time to reflect.

Parental needs: Time to reflect

P003: They came in and they talked about things, then they gave me the leaflets and then they appropriately gave me the space to digest this. It wasn’t a ‘taking him off my hands and taking him into a room and I didn’t know what was going on’.

P003: So, for us, it always gave opportunity to go away, digest it, ask any questions and appropriately challenge.

P015: You’re kind of taken aback, and for me, it’s just digesting it and then having the space to just try and do that myself and then for someone else to check back in, of going ‘it’s a lot, are there any questions?’

P015: I think if there’s an opportunity to give a moment of space and then come back, I think it’s just respectful.

Unmet needs: Time to reflect

P011: But he kept saying, ‘Are you still following me? Do you still understand?’ But it was very much like a word vomit, telling you loads of stuff and just, well, because you don’t really have the time and then obviously they leave and you have a load of questions.

Most clinicians’ initial communication involved a broad overview of the infant’s clinical condition and differential diagnoses, with a focus on urgent investigations and treatment. This information was expanded as the clinical course progressed. They did not tend to give numerical data and felt that stratifying risk in this way would not be beneficial. Furthermore, they did not have any set numerical risk thresholds they felt they could apply to decisions to patient management, relying instead on clinical judgement and parental preferences.

Parental preferences for the communication of risks of potential management plans varied greatly. While some parents expressed a preference for being fully informed of all risks, including being given statistical data, others felt this was likely to be overwhelming and preferred a broad overview (see table 1). The complexity of delivering information appropriately is further demonstrated by the intrapersonal conflict experienced by interviewee P004 who expressed both a desire to be given statistical data: ‘Stats would always be better’, and relief at not being told all the risks. ‘I’m glad in a way they didn’t go through the risks because I think it was the right decision.

Table 1

Discussion of risk

Most parents expressed a desire to be informed of the rationale behind the clinician’s suggested management but were happy to follow the recommendations. Some would rather not make decisions in the ED, feeling that the clinician was better equipped to do so. ‘I don’t know how quickly I would’ve disagreed to it, because they felt it was necessary, and when your baby is so small, you have to trust the medical professionals around you more than your own emotions’ (P001).

This was echoed by the clinicians, who felt that the burden of decision-making should be theirs but made efforts to adjust their information provision based on perceived parental preference ‘85% of the time…. People are like, ‘Do whatever you need to do, you’re worried, I’m worried, please fix it.’ I think the other 15% of the time, maybe people want a little bit more of a conversation about it and that’s okay’ (C005).

Factors contributing to trust in clinicians included having their concerns validated, clinician confidence in the proposed management plan, ‘the doctor said if it was one of her family members she would be getting the lumbar puncture… I thought, ‘well if she is happy [for a family member to have an LP]… then there’s no reason why I shouldn’t be happy for my son to have it done’’ (P001), being given time to reflect and ask further questions, ‘We challenged them, but we got answers and they were very clear’, repetition of information by various staff, ‘Throughout the stay…it was consistent information given to me’ (P015) and referencing guidelines, ‘They hadn’t just made their own decisions about what tests were needed. [The guideline] was for them to follow, which I think was reassuring to me’ (P005).

While parents were predominantly happy with the information provided during the initial phase of management, some expressed a desire for more information at a later point in their admission, particularly regarding long-term sequelae of antibiotics. Parents valued having the opportunity to talk through the infant’s progress to date, raise questions and be informed of signs of deterioration once they were home. ‘At the point in A+E… then you are definitely happy to accept the appropriate treatment… regardless of side effects… as the admission went on, maybe someone could’ve said ‘here are the risks’, or ‘here’s what to look out for once you go home.’ That might have been helpful’ (P001). Information leaflets, good safety netting and a belief that their child would continue to improve at home were central to parents and clinicians’ confidence in discharge from hospital. Having ‘open access’ or a telephone hotline was a source of significant reassurance after discharge.

As seen in table 2, clinicians were accepting of the concept of a CDA to stratify risk, and recognised that the current overtreatment of febrile infants is not without harm. Acceptance of CDAs would be contingent on national guidelines supporting their use. They felt that CDAs would be used primarily as a communication adjunct to reassure parents and as an extra tool for more junior trainees.

Table 2

Clinician views of CDAs


Previous studies examining priorities of parents of sick children highlighted the importance of good communication and shared decision-making.13 15–17 23 The parents in this study also felt that communication was a key part of reducing stress and improving parental understanding. Clinicians tailored their initial communication based on their assessment of the parents’ information needs and capacity to absorb information at the time. Most parents interviewed felt that the initial information given to them was the right amount, regardless of whether they were given all information at once or staggered. This belief persisted even when the information given was different from the parents’ self-reported belief of how they would like information delivered. This suggests that parental trust in the clinician may be more impactful than the specific information given.

Previous studies have suggested that adult patients retain approximately 50% of information provided.23 24 This cohort of parents recognised their lack of information retention, and widely ascribed it to the stress and anxiety they were experiencing in the ED.

Parents expressed an appreciation of the benefits of having time to reflect and ask further questions prior to the investigation or treatment being commenced. Areas for improvement include giving the parents time to reflect and returning to complete the discussion and answer any questions, giving a clear diagnosis.

In order to provide this time without unduly delaying treatment, it is imperative that clinicians create space for discussion early in the management of these infants. In keeping with statistics on National Health Service complaints, most negative feedback centred around communication.25

While existing literature examining parental views places heavy emphasis on shared decision-making, this is not reflective of the views of this cohort.12–16 26 If they had confidence in their treating clinician and had the opportunity to ask questions, they were happy to consent to the clinician’s suggested management. The high degree of trust was reflected in their belief that had the doctor suggested different management, they felt they would have accepted that. The communication needs of parents are illustrated in figure 3, the febrile infant communication framework.

Figure 3

The febrile infant communication framework. A novel framework demonstrating parental information needs. ED, emergency department.

Parents’ description of information needs throughout admission suggests that as their initial concerns and stress levels reduce, their desire for information increases.13 24 27 28 In addition to safety netting, clinicians should ensure that parents are given the opportunity to ask questions about the management to date, and any other concerns, including long-term impact of antibiotic use. This evolution of parental communication and decision-making preferences is reflected in figure 4.

Figure 4

The evolution of shared decision-making.

Written information was received positively, with parents benefiting from PILs addressing LP and discharge information. This is in keeping with existing literature.24 29–31 PILs about LP may provide a solution to the differing parental preferences for information about risk. Given that many interviewees did not want numerical data, we propose a possible solution would be a QR code on the PIL that would furnish parents with numerical data if desired.

Strengths and limitations

Study strengths include the insight gained from both parent and practitioners with relevant experience from FIDO sites across England, Wales and Northern Ireland. Interviews were conducted until the point of information power. While the research team sought to maximise sample variability purposive selection, most parents came from a professional background. While this may have led to a higher degree of health literacy, common communication themes emerged across all participants, with communication needs varying more across time than between individuals. Differing data protection laws prevented interviews with participants in Scotland and Ireland. As all infants in our study made a full recovery from their feverish illness, this may have skewed views of parents towards a more positive attitude. Purposive selection of infants with a variety of diagnoses has resulted in higher rates of SBI/IBI than are seen in the general population. Despite this, the communication needs of parents and their reported feelings towards more individualised, less invasive management were similar across all groups. Email invitations to interview had poor uptake, with 135 of the 167 invited not booking an interview.


In order to optimise communication with parents, clinicians should engage in early trust building through timely assessment and treatment, validating parental concerns and referencing guidelines. Parents report improved understanding and reduced stress when they receive a verbal information, given time to reflect, and offered the opportunity to ask questions. Adjuncts such as PILs can be used to provide varying levels of details of risks.

Clinician-led decision making is widely accepted in the initial stages of the admission, but acceptability is contingent on open communication and parents feeling informed. Clinicians should continually assess the evolution of individual’s preferences towards collaborative decision-making and tailor communication accordingly.

Clinicians were widely open to the use of CDAs, primarily to confirm their clinical judgement and for use as a communication tool but would require consensus guidelines on risk thresholds before implementing CDAs.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by the Office for Research Ethics Committees (OREC) of Northern Ireland and the Health and Social Care Research Ethics Committee (REC 22/NI/0002). Participants gave informed consent to participate in the study before taking part.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • X @kjwkathryn, @timburgD

  • Contributors EU, DM, KWo and TW were involved in the conception and design of this study. KWi, EU, KWo and TW edited study guides following pilot interviews. KWi performed the qualitative interviews. KWi, EU and KWo coded and thematically analysed the interviews. TW is the guarantor.

  • Funding The FIDO study is funded by the Royal College of Emergency Medicine Doctoral Fellowship. A Clinical Fellowship in Paediatric Research was provided by the Royal Belfast Hospital for Sick Children charitable funds. Study sponsor: Queen's University Belfast.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.