ABSTRACT
Background: ‘Safety netting’ advice allows general practitioners (GPs) to cope with diagnostic uncertainty in primary care. It informs patients on ‘red flag’ features and when and how to seek further help. There is, however, insufficient evidence to support useful choices regarding ‘safety netting’ procedures.
Objectives: To explore how GPs apply ‘safety netting’ in acutely ill children in Flanders.
Methods: We designed a qualitative study consisting of semi-structured interviews with 37 GPs across Flanders. Two researchers performed qualitative analysis based on grounded theory components.
Results: Although unfamiliar with the term, GPs perform ‘safety netting’ in every acutely ill child, guided by their intuition without the use of specific guidelines. They communicate ‘red flag’ features, expected time course of illness and how and when to re-consult and try to tailor their advice to the context, patient and specific illness. Overall, GPs perceive ‘safety netting’ as an important element of the consultation, acknowledging personal and parental limitations, such as parents’ interpretation of their advice. GPs do not feel a need for any form of support in the near future.
Conclusion: GPs apply ‘safety netting’ intuitively and tailor the content. Further research should focus on the impact of ‘safety netting’ on morbidity and how the advice is conveyed to parents.
‘Safety netting’ is a useful instrument to deal with diagnostic uncertainty. GPs apply ‘safety netting’ and perform it intuitively, emphasizing its importance. The effectiveness of ‘safety netting’ resources needs to be examined.
KEY MESSAGE
Introduction
Acute feverish illness is the most common presentation of children to primary care services. Most acutely ill children seen by a general practitioner (GP) suffer from self-limiting viral illnesses (Citation1). A small proportion of these children, however, have a serious infection, with considerable mortality and morbidity. GPs face the challenge of distinguishing these few cases, often seen at an early stage, from the vast majority of non-serious infections to prevent serious complications. Developing appropriate strategies to cope with the resulting diagnostic uncertainty is crucial (Citation1,Citation2).
Heneghan et al. suggested three main strategies: (a) performing further investigations, (b) initiating a treatment to evoke a response, or (c) a ‘wait and see’ policy. If all three fail, GPs can try a combination of the above and share their uncertainty with the patients (Citation3). Buntinx et al. suggest four strategies to deal with diagnostic uncertainty: the GP’s gut feeling, the use of diagnostic algorithms, planning additional tests, and the use of ‘safety netting’ (Citation1–4,Citation5).
Neighbour first introduced the term ‘safety netting’, and considered it as one of the compounds of a good consultation (Citation6). He described ‘safety netting’ as creating a contingency plan and implementing procedures to ensure that the strategy is effective and that the patient is safe in any (un)foreseen eventualities. He defines ‘safety netting’ as encompassing three questions:
If I am right, what do I expect to happen? For example, the fever will be gone three days later.
How will I know if I am wrong? For example, the child still has fever after three days.
What would I do then? For example, make an appointment if the child has fever after three days (Citation6,Citation8).
According to Maguire et al., ‘safety netting’ appears to play an important role in repeated medical help seeking for children with fever. Parents who were not ‘safety netted’ by their doctor were more likely to seek another contact than those who were (Citation7). In 2009, Almond et al. sought clinical consensus about what ‘safety netting’ should include, and what should be recorded, using a modified Delphi approach (Citation8). The NICE guideline on feverish children recommends GPs to use ‘safety netting’ by means of a traffic light system, triaging children at low (green features), intermediate (amber) and high (red) risk of serious infection and in need of further investigations (Citation1).
Though it appears to be an important instrument for clinicians, only few studies examine the actual use of this ‘safety netting’ advice. Covemacker conducted a pilot study on the follow-up of children with acute illness (Citation9). Whenever children got the label ‘seriously ill’ by their GP without being referred to a specialist, most GPs stated they discussed ‘red flag’ features with parents and re-evaluated children at the parents’ discretion. The only other explorative study on ‘safety netting’ behaviour of first contact clinicians was recently conducted by the ‘ASK SNIFF’ (acutely sick kids, safety netting interventions for families) research team. They noticed a range of ‘safety netting’ techniques with many inconsistencies, concerning their relative effectiveness on a variety of outcomes such as referrals, admission rates, re-attendance to healthcare and parental understanding, anxiety and satisfaction (Citation10,Citation11).
In this study, we focus on the ‘safety netting’ advices given by GPs when facing acutely ill children, its application in practice and influencing factors.
Methods
Study design
We designed a qualitative study with semi-structured interviews in Flanders. To direct the interviews, a basic questionnaire was constructed, consisting of questions based on the Britten four types of questions (behaviour, opinion, feelings, knowledge) (Citation12,Citation13). These questions were peer-reviewed by all authors, consisting of clinicians involved in teaching and training in acute paediatric primary care. To explain ‘safety netting’, we used exemplifying cases, such as what advice GPs would give to parents of a child between 0 and 14 years old with 39°C for more than three days. After piloting the interview with two junior doctors, 12 questions were retained, adjusting the questionnaire based on their suggestions (Supplementary File 1, available online).
Ethics
The ethical review board of UZ/KU Leuven under reference ML9287 approved this study.
Recruitment
Recruitment of participants was based on purposive sampling. We invited GPs through personal and professional contacts via email. We tried to obtain a heterogeneous group of GPs (different Flemish counties, practice characteristics, age and gender) gradually throughout the recruitment process, accepting participants accordingly. Recruitment of GPs ended after data saturation was reached, determined as the moment when no additional information emerged from the last two interviews.
Interviews
Two investigators (KB, PD) conducted and audio-recorded all of the interviews. All interviews were anonymized. We used field notes, which included information e.g. on GP’s emotions and circumstances of the interview. Afterwards a verbatim transcription of each interview was performed.
Analysis
We analysed the collected data based on components of grounded theory (Citation14). First, we used deductive coding by composing descriptive themes. Both researchers (KB, PD) developed coding schemes separately using the software QSR NVivo version 10 (QSR International Pty Ltd, Melbourne, Australia). Each researcher coded all interviews performed by the other researcher and vice versa, to avoid bias of coding results. The two researchers (KB, PD) compared the coding schemes after every three interviews. Further interviewing was deemed not suitable due to data saturation. From the content of the codes and the descriptive themes, we introduced emerging themes. Based on these emerging themes we developed our theory consisting of four main quotes. Whenever possible, extracts from the interviews were used to enhance overall clarity.
Results
Participants
Thirty-seven GPs were included in the study. Their baseline characteristics are described in .
Table 1. Demographic characteristics of the 37 GPs.
Results of the interviews
Four emerging themes were isolated from the analysis. ‘All GPs do it intuitively,’ ‘the content of the advice is similar, the approach differs,’ ‘GPs try to provide continuity’ and ‘concerns’ (Summary coding scheme, Supplementary File 2, available online).
’All GPs do it intuitively.’ All interviewees recognized the notion of ‘safety netting’ as part of their medical behaviour, but they were unfamiliar with the term. Most of the interviewees had never read a guideline on ‘safety netting’. When asked about possible support for practice, some GPs thought internet-based support or leaflets might be useful. Only a few felt an actual need for guidelines on this topic.
Many of the interviewees felt that ‘safety netting’ plays a crucial role; some of them even saw it as the most useful part of the GP’s consultation.
Right now I realize that this is normal, it’s something we do every day, it’s part of the job. And I think it plays an important role in healthcare, especially in healthcare of infants and very young children (Source 20 SA).
GPs believe that their relationship with parents improves when ‘safety netting’ is applied. They found ‘safety netting’ to be most useful as a framework for parents. Following the GP’s advice served an educational role: parents learn to recognize ‘red flag’ features and know when to re-consult their physician.
Patient empowerment, that is also my goal, to teach them some skills and how to deal with an ill child and try to make them understand when they have to consult their doctor and when not to. Absolutely, that must be the teacher in me (Source 2B).
GPs considered ‘safety netting’ to be an important instrument to avoid legal issues by informing parents that the disease could potentially deteriorate, empowering parents to manage their child’s illness. According to the interviewees, abiding with the GPs ‘safety netting’ advice can avoid avoidable investigations and urgent visits to the GP, the out-of-hours service or the hospital emergency department. They believed it could reduce overconsumption of medication such as antibiotics.
‘The content of the advice is similar, the approach differs.’ GPs inform parents of an ill child to look out for specific symptoms. The clinical signs most often cited by the GPs were the overall impression of the child’s illness, vomiting, a decrease in appetite and thirst, fever and signs of dehydration. Apart from these signs, parental ‘gut feeling’ was also mentioned as an important element.
Sometimes a child seems perfectly fine, but its parents tell you ‘something’s wrong with my child.’ As a doctor, you see them only once, some children are always quiet, and others are never quiet. That’s why the parents’ feeling is important (Source 09 LE).
According to the GPs, the duration of illness may vary from three to seven days. The time lapse of the disease and the day of the week influence the content of the advice.
Emphasizing the out-of-hours services right before the weekend. So yes, the timing is important (Source 14 ST).
GPs ask parents to get in touch when the condition of the child has not improved after the suggested period and when ‘red flag’ features or additional symptoms arise.
Overall, GPs believed it is extremely important to give ‘safety netting’ advice when confronted with a child suffering from a severe illness that can be managed outside the hospital or when a clear focus is absent. They try to tailor their ‘safety netting’ to the medical history and the age of the child.
When it’s a baby, you do stress the respiratory frequency, the colour, the degree of somnolence, but those are things I wouldn’t emphasize when the child is older, when they’re 5–7 years old (Source 10ME).
The GPs try to give clear and practical advice, either verbal, written, printed or through a leaflet or illustration.
Very often, I note it on paper. In English, or in French, or tailored to the patient (Source 08 LE).
GPs were convinced full understanding of the advice given to parents should be ensured. They check the feasibility and tailor their advice to the individual patient. GPs tend to verify if parents understood the ‘safety netting’ advice, repeat it and try to interpret the body language of parents. In spite of all these measures, GPs found it difficult to estimate whether or not the message was well understood by the parents.
You always have to explain using the patient’s language and adjust to his or her perception. So if someone speaks a local dialect, then I will explain in that same dialect (Source 03HAL).
The following parents’ characteristics are taken into account when applying ‘safety netting’ advice: the degree of concern and the confidence of the parents, their psychosocial and economic background, whether or not they had previous experience with ill children and whether or not the GP knows these parents. GPs differ in opinion whether the parents’ cultural background should be taken into account. However, a language barrier is considered an obstacle by all GPs.
When people come across as rather intelligent, when they have a job or indeed have a better social or economic status, I leave it to them to phone me and check whether I have to see their child again. There is a big difference there (Source 16 LV, reference 1).
‘GPs try to provide continuity.’ Since a follow-up consultation by a colleague is often possible, GPs briefly describe the ‘safety netting’ advice in the electronic medical record and notify their colleagues when an exceptional case presents itself.
I will always write down in the medical record what kind of advice I have given, and then I will write health education and ‘red flag’ features. Of course, I do not write down everything I have said (Source 10 KS, reference 1).
Some GPs always plan a follow-up consultation because they believe it is difficult to evaluate the situation of an ill child over the phone. Other GPs believed a scheduled follow-up consultation is only necessary when their gut feeling tells them something is wrong.
Nevertheless, I find it very hard to give advice on an ill child by phone. I prefer that they return and consult again (Source 03 HV).
GPs believed their availability is very important in the follow-up of an acutely ill child. They do realize however that this is not always feasible. Therefore GPs inform parents on out-of-hours services, e.g. during the weekend. They often use leaflets or display the information of the out-of-hours services in the waiting room or on the answering machine of the GP practice. In order to guarantee the care of an ill child, many GPs make use of a delayed antibiotic prescription or a referral letter, which can be used by the parents if needed.
But when I see a child just before the weekend, in case of doubt, for example a runny ear on day 6, and I see that these people are capable enough to follow my instructions, I would rather give them a prescription for an antibiotic so they can collect it during the weekend if needed (Source 15 HO).
‘Concerns.’ By creating a ‘safety net’, parents might even get more worried than necessary. People can be overwhelmed and miss the essence of your message, or it can be misunderstood. Moreover, it is necessary to allow enough time for proper ‘safety netting’. Some GPs had the impression that parents consult too frequently when they are left in doubt. Patients may interpret ‘safety net’ advice as a sign of diagnostic insecurity of the treating physician.
Because when you focus too much on the ‘red flag’ features, parents can sometimes get the feeling “oops, the doctor does not know what it is” or think “oh, this is so bad” (Source 11 LD).
When asked about circumstantial influence, GPs recognized that ‘safety netting’ is sometimes a bit careless when the waiting room is overcrowded or when the GP is exhausted. In contrast, other GPs stressed that they always take their time when it comes to children.
Right then, I think that, normally with a child, I believe time pressure is somewhat less important… Well, it is something of a routine; it is mostly the same advice that you give (Source 12 SI).
Seeing more than one child during the same consultation is a distracting factor in creating a strong ‘safety net’.
When an entire family comes in with a number of children that start to jump around in here, then I do feel that providing good medicine is more difficult. I try to pay attention to this, but I find these to be very difficult circumstances (Source 14 MS).
Discussion
Main findings
Although GPs are unfamiliar with the term ‘safety netting’, they frequently apply this type of advice using their gut feeling and intuition. Moreover, GPs do not feel a need for guidelines or any other form of support.
The ‘safety net’ advice often contains similar information. Nevertheless, they adjust their explanation to patient’s characteristics and the illness in question.
GPs try to ensure parents of an ill child understand the given advice. However, this proves to be rather difficult and it remains unknown how the parents experience or handle the given advice.
According to these interviewed GPs, offering the parents continuity is an important part of ‘safety netting’, either informing their colleagues verbally or in writing in the electronic medical record, by guaranteeing the availability of a doctor or a referral letter or delayed antibiotic prescription.
‘Safety netting’ can offer a useful framework for the child's parents and improve the patient–doctor relationship. It might fulfil an educational role and aid in reducing overconsumption of investigations, doctor visits and antibiotics. Furthermore, ‘safety netting’ could avoid legal issues.
Strengths and limitations
This study was the first qualitative study to examine how GPs use ‘safety netting’ in Flanders.
We interviewed a heterogeneous group of 37 Flemish GPs. Purposive sampling might have led to bias. However, we attempted to obtain a representative sample, gradually selecting participants throughout the recruitment stage according to their professional, geographical and demographical characteristics.
When invited to participate in our research, GPs intentionally were given little information on the concept of ‘safety netting’, to ensure spontaneous answers during the interview. Since most GPs were not acquainted with the concept of ‘safety netting’, we made use of exemplifying cases at the start of the interview. Nonetheless, this sometimes led to misunderstandings. After interviewing 37 GPs, we reached data saturation.
Although we performed a large number of interviews and reached data saturation, the extent to which these findings are comparable to other settings is not known. Participating GPs may have a stronger interest in providing paediatric care, which could have influenced our results optimistically. The difference in opinions between the doctors, according to their age, sex, practice type and proximity to an ED should be explored in larger samples.
Interpretation of the study results in relation to existing literature
The importance of ‘safety netting’ to deal with diagnostic uncertainty, when complications may occur or when the patient is at higher risk for developing complications is emphasized by different studies (Citation4,Citation8,Citation15,Citation16).
‘All GPs do it intuitively.’ Recently, Jones et al. published a similar study in the UK about the use of ‘safety netting’ by doctors and nurses (Citation10). As in this study, they conclude that GPs provide patients with ‘safety netting’ rather intuitively.
‘The content of the advice is similar, the approach differs.’ Characteristics of how ‘safety netting’ is performed in the Jones et al. study (Citation10) and ours are largely similar and mentioned in the recommendations for good ‘safety netting’ according to Almond et al. (Citation8). The participating GPs in both studies tailor their advice to the specific context.
Almond et al. stressed the importance of parents understanding the message (Citation8). Our research shows that in reality, this turns out to be quite difficult. While GPs always try to pass the message as clearly as possible, it is never entirely clear whether the parents have fully understood the advice and how they deal with it.
GPs in our research provide the advice verbally or in writing. There seems to be no agreement considering the format of delivery amongst the participating GPs in the research by Jones et al. (Citation10), or amongst the doctors that took part in the modified Delphi procedure of Almond et al. (Citation8).
‘GPs try to provide continuity.’ Jones et al. report that only one participating GP noted which advice on ‘safety netting’ he had given (Citation10). In this research, most GPs briefly write down the given advice in the medical record that was also stressed by Almond et al. (Citation8).
The similarity between our results and the study by Jones et al. strengthens the utility of these findings, even though performed in different geographical and healthcare settings.
This conclusion, together with the recognized importance of ‘safety netting’, encourages the development of a guideline on how and when to apply ‘safety netting’ in acutely ill children, to ensure quality ‘safety netting’ in this setting (Citation11).
Implication for research
Further research should focus on the impact of this ‘safety netting’ strategy on a variety of outcomes such as morbidity and mortality of the children, the number of doctor visits in and outside the hospital, and on how parents interpret this advice. This will most likely result in adequate guidance, lowering the threshold for clinicians to apply ‘safety netting’ effectively.
Conclusion
In feverish children, GPs perform ‘safety netting’ intuitively. They tailored their advice to the child and parents. Further research should focus on the impact of ‘safety netting’ on morbidity and the number of re-consultations and how the advice is conveyed to parents.
Supplementary_file_2:_Summary_of_the_coding_scheme
Download PDF (132.5 KB)Supplementary_file_1: Questionnaire
Download PDF (61.4 KB)Acknowledgements
The authors should like to thank all the interviewed GPs for their participation in this study.
Declaration of interest
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References
- NICE. National Institute for Clinical Excellence. Feverish illness in children—assessment and initial management in children younger than 5 years. CG47. London: National Institute for Clinical Excellence, RCOG Press; 2007.
- Van den Bruel A, Aertgeerts B, Bruyninckx R, Aerts M, Buntinx F. Signs and symptoms for diagnosis of serious infections in children: A prospective study in primary care. Br J Gen Pract. 2007;57:538–46.
- Heneghan C, Glasziou P, Thompson M, Rose P, Balla J, Lasserson D, et al. Diagnostic strategies used in primary care. Br Med J. 2009;338:b946.
- Buntinx F, Mant D, Van den Bruel A, Donner-Banzhof N, Dinant GJ. Dealing with low-incidence serious diseases in general practice. Br J Gen Pract. 2011;61:43–6.
- Thompson M, Van den Bruel A, Verbakel JY, Lakhanpaul M, Haj-Hassan T, Stevens R, et al. Systematic review and validation of prediction rules for identifying children with serious infections in emergency departments and urgent-access primary care. Health Technol Assess. 2012;16:1–100.
- Neighbour R. The inner consultation: How to develop an effective and intuitive consulting style. 2nd ed. Oxford: Radcliffe; 2005.
- Maguire S, Ranmal R, Komulainen S, Pearse S, Maconochie I, Lakhanpaul M, et al. Which urgent care services do febrile children use and why? Arch Dis Child 2011;96:810–16.
- Almond S, Mant D, Thompson M. Diagnostic safety-netting. Br J Gen Pract. 2009;59:872–4.
- Covemacker A. Optimizing the monitoring of acutely ill children in general practice. Ghent: Department of Primary Care Ghent University; 2011.
- Jones C, Neill S, Lakhanpaul M, Roland D, Singlehurst-Mooney H, Thompson M. The safety netting behaviour of first contact clinicians: A qualitative study. BMC Fam Pract. 2013;14:140.
- Roland D, Jones C, Neill S, Thompson M, Lakhanpaul M. Safety netting in healthcare settings: What it means, and for whom? Arch Dis Child Educ Pract Ed. 2014;99:48–53.
- Britten N. Qualitative research: Qualitative interviews in medical research. BMJ 1995;311:251–25.
- Patton MQ. How to use qualitative methods in evaluation. Newbury Park, London: Sage; 1987.
- Mortelmans D. Handboek kwalitatieve onderzoeksmethodes. Leuven: Acco uitgeverij; 2007.
- Van den Bruel A, Thompson M. Research into practice—acutely ill children. Br J Gen Pract. 2014;64:311–13.
- Schneider A, Wübken M, Linde K, Bühner M. Communicating and dealing with uncertainty in general practice: The association with neuroticism. PLoS One 2014;9:e102780.