If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Guidance for clinical discussions of risk with people with epilepsy is limited.
•
Timing and frequency: Discussions should ideally be held at initial consultation.
•
Method of communication: Multiple formats via specialists are preferred.
•
How to communicate: Anxiety created and benefit of awareness should be discussed.
•
Discussions should be adapted to individual life contexts and needs.
Abstract
Purpose
Risk communication between clinicians and people with epilepsy (PWE) and their families is under researched. There is limited guidance about when and how to have these discussions. This paper explores the current evidence on quality of risk related conversations in epilepsy and suggests a concept of an evidence-based guideline for person centred structured risk communication.
Methods
A literature search of four electronic database, Ovid Medline, Ovid Embase, PUBMED, and CINAHL, was conducted by two independent reviewers using relevant search terms following the principals of the PRISMA guidance. No limits were applied. Supplementary searches included using backwards and forwards citation searching. A predesigned inclusion and exclusion criteria was administered to the identified results.
Results
From 376 results identified, 17 studies met the final criteria of which ten were quantitative, five qualitative and two mixed methods. Perspectives of PWE and clinicians were represented.
Extracted data was clustered into three domains: communication initiation (e.g. timing, individual tailoring); communication methods (preference for face to face with neurologists); and communication content (acknowledging the anxiety produced by risk communication, the benefits of being self-aware, normalising risk etc.). No papers focused on conversation structure (e.g. helpful phrases), or the best locations to hold conversations.
Conclusion
More research is needed to develop structured communication of risk. An attempt has been made to put current evidence into this format. Clearer guidance will enhance clinicians’ confidence in communicating person centred epilepsy risk with PWE and their families thus improving outcomes.
This paper collates guidance from the literature around how clinicians might best hold conversations about epilepsy risks, without undermining clinician skills of undertaking these conversations. Difficult conversations especially around risk of death are complex and can be misconstrued. There is ongoing debate around risk discussion rationale [
]. The focus of this paper is the communications directly. Greater understanding of communication has significantly affected patient engagement with care, for example, discussing lifestyle risk assessments with GPs [
Deciding if lifestyle is a problem: GP risk assessments or patient evaluations? A conversation analytic study of preventive consultations in general practice.
Engaging terminally ill patients in end of life talk: how experienced palliative medicine doctors navigate the dilemma of promoting discussions about dying.
]. A similar approach would be beneficial in the world of Epilepsy.
1.1 The rationale for discussing epilepsy risk
People with epilepsy (PWE) and their families are concerned that they do not receive enough information about risks related to epilepsy, nor how to manage these [
]. In 2002, 42 % of epilepsy related deaths were identified to be potentially avoidable had PWE and/or their families been more aware of the risks of morbidity and mortality [
]. Discussion of lifestyle behaviours, such as medicine adherence, alcohol and recreational substance intake and how to monitor seizures particularly at night and awareness of outcomes such as Sudden Unexpected Death in Epilepsy (SUDEP) might reduce morbidity and mortality from epilepsy [
NICE Clinical Guideline CG137 - Partial Pharmacological Update of Clinical Guideline 20: the Epilepsies: the diagnosis and management of the epilepsies in adults and children in primary and secondary care.
National Clinical Guideline Centre. London: Royal College of Physicians,
2012
Another rationale for risk conversations is family sequelae following deaths. Families and PWE believe that all those related to or working with a PWE should be informed [
]. A study of a sample of Australian family and friends bereaved by epilepsy death, illustrated that the majority (53 %) had believed that death from a seizure is not possible [
A qualitative study of the reactions of young adults with epilepsy to SUDEP disclosure, perceptions of risks, views on the timing of disclosure, and behavioural change.
]. Thus, some of the issues might relate to how information is communicated, not just whether it is.
Lack of knowledge can be a concern for clinicians, too. General Practitioners (GPs) are often the person that PWE prefer for monitoring Epilepsy, but GPs can feel lacking in confidence to do so [
]. Further, PWE perceive this ‘lack of confidence’ from GPs to avoid discussing the risk and social aspects of the condition, including work, driving and financial effects [
]. Within the paediatric population, one study found that 75.4 % of paediatric care providers were unaware that children with epilepsy were at a risk of SUDEP [
]. Clinicians without a specialist interest in epilepsy particularly struggle with conversations about SUDEP and epilepsy risks and may benefit from further information [
]. However, it is notable that given the limited evidence, more research is needed to be done to verify these findings.
This difference between specialist and non-specialist information giving might relate to assumptions that discussion of SUDEP and epilepsy risk should be with only those at ‘higher risk’, and specialists are more likely to see this group. Though there are factors indicative of cumulative risk [
] there is as yet no validated approach to classify epilepsy risk into ‘high’ or ‘low’. Also, people considered as ‘low risk’ die, and so care needs to be taken not to make false assurances [
]. The preference for only discussing risk with ‘high risk’ groups also occurs in specialist services. Another confounder is that epilepsy is a chronic condition which waxes and vanes with life event and age. Thus, risk rarely remains static and can fluctuate over time [
The incidence of epilepsy death in newly diagnosed PWE is very low (1:10,000). However, if epilepsy is chronic, as in 30–40 %, the risks can increase to significantly high levels such as those with pharmaco-resistant epilepsy that are failed by epilepsy surgery (1:300) [
], that is, that PWE and their families need protecting from knowledge of SUDEP. As of now, SUDEP is only frequently discussed in relationship to increasing adherence to medication only [
Effect of providing sudden unexpected death in epilepsy (SUDEP) information to persons with epilepsy (PWE) and their caregivers-experience from a tertiary care hospital.
], rather than routinely considered. However, SUDEP is only one issue, and evaluation of other factors that might be discussed does not seem widely available.
1.3 Ethics of risk communication
Clinicians may fear distressing PWE or making them uncomfortable in discussing risk [
]. This includes the possibility that clinicians might be giving information that is less relevant to that person where they are perceived as “low risk” (Hamid et al., 2010), to prevent creating unnecessary anxiety [
] noted that the psychological sequelae of risk communications to PWE remain unknown. It was observed that there is considerable debate, including, suggesting that a more compassionate and value based approach to care would recognise that PWE and their families need to be able to make informed decisions, which they cannot do without adequate information, and the clinician should not have the right to remove this choice [
]. They clarify this, stating that PWE might develop better self-management strategies (e.g. not using soft pillows, sleep or medication manage), improving people’s perception of control over their lives and possibly reducing risks directly [
]. Currently, varied aspects of good practice guidance seem spread across different papers possibly leading to key features being missed. These include recommendations that risk conversations should be tailored to individuals, focusing on modifiable risk factors [
]. Other guidance include the need to consider an individual’s readiness to learn, preferred learning styles and expectations, and the range of information sources available [
The current review aimed to extract from the literature available evidence-based recommendations of how clinical conversations about risk and risk factors should occur with PWE, and identify what areas are still under researched.
2. Method
An open-ended search of four databases was conducted in November 2018; Ovid Medline, Ovid Embase, PUBMED, and CINAHL, using the search string: ‘epilepsy death or epilepsy mortality or SUDEP AND patient or individual or disclosure or discuss or inform or communicat* or advice or risk. Where subject headings were available these were searched for ‘Sudden Death’ and ‘Epilepsy’ AND ‘Information processing’ and ‘information dissemination’. Reference list checks were conducted. The PRISMA scoping review checklist was used as a guiding template (Supplementary information). The review protocol is detailed here.
Two researchers screened the papers independently for the pre- designed inclusion and exclusion criteria. A sample of these and all disagreements were discussed with a third researcher. Inclusion and exclusion criteria are detailed in Table 1. These criteria and the search strategy were developed in conjunction with senior clinicians and information specialists.
Table 1Inclusion and exclusion criteria.
Inclusion
Exclusion
Empirical studies
Reviews and opinion pieces
Peer reviewed studies
Papers focused only on identifying risk factors in Epilepsy
Written in English or having an English translation due to limited translation funds
Focuses on PWE under the age of 18 as communication with children may need different sensitivities
Any research method
No mention of SUDEP or epilepsy risks
Evaluations of frequency of conversations and communications
Doesn’t explicitly discuss how to communicate risks
Papers including content of how clinicians discuss risk with PWE
Where there were no recommendations given for practice
Papers including recommendations about how to talk about risk in epilepsy
]. The CASP Critical Appraisal tools were developed as a guide for evaluating research based on features such as clarity of aims and content and were used for qualitative and quantitative studies. The AXIS was developed to explore medical cross-sectional studies based on features relevant to these study designs. The CASP and AXIS critical appraisal tools do not provide an overall score and ratings are subjective; therefore, ratings were given independently by 2 researchers, and discussed in a moderation group with a third. Both tools are designed to give an indication of the credibility of the evidence and potential gaps.
2.1.1 Data extraction method
Data was extracted into a summary table (Results, Table 5). Key questions were applied to each paper, summarising: main points made about communication and risk; identified helpful and difficult factors; gaps identified by paper; guidance recommended. Key phrases offering communication guidance were extracted and grouped.
3. Results
3.1 Description of the papers
There were 376 results, 274 after removal of duplicates. 36 articles remained following title and abstract review and seventeen after application of inclusion/exclusion criteria. Fifteen of these 36 were agreed as suitable by 2 independent reviewers; a further five were disputed and arbitrated by a third reviewer, resulting in 3 further rejections, leaving 17 papers. See Fig. 1.
Of the 17 papers, two used mixed methods, 10 were quantitative and five were qualitative studies. Research quality was assessed using the CASP and AXIS tools, (Table 2, Table 3, Table 4). All studies received either a “good” or a “moderate” rating score. Ratings were derived by assessing potential problems in studies. Ratings were: 3 or less problems, good; 3–6 problems, moderate; 6< problems, poor.
Table 2Quality ratings of mixed methods studies using CASP.
Quantitative data. Moderate quality. Analysis not detailed, limited word count. Qualitative data – Moderate Quality. Limited information aims unclear. More detail for method of analysis needed.
Quantitative data - Moderate quality. Low response rates, possible sample bias as only those with knowledge of SUDEP responded. Clear analysis. Ethics or conflict of interests not discussed. Qualitative - Moderate quality. Clear analysis, two coders used. Relationship of researchers to data not mentioned.
Good quality- recruitment, methods and analysis clearly explained. Sample size justified. Replicable. Codes explained including how they were derived from the interview guide.
Moderate quality- confounding variables possibly not considered e.g. not documenting SUDEP discussion. Follow up time short at only 6 months rather than a year as other research recommends. Acknowledged by researcher.
Good quality. Uses a p value of <0.02 in the logistic regression. 1200 surveys completed, though 9 % response rate, remains substantive. Provides reasons for non-completions. Methods and analysis clear. Replicable. Sample appeared representative.
Good quality. Has a high participation rate (2090), 56.6 being PWE, the remainder, carers or family members answering for the PWE or themselves. Participants likely to be representative of PWE. Statistical analysis clear and thorough. Replicable. Findings and discussions comprehensive.
Moderate quality. A representative sample of epilepsy nurses. High response rate, 103 surveys included in final analysis. Methodology and results difficult to replicate. This could be because of journal word limit.
Good quality. Used newly diagnosed PWE and those with chronic epilepsy over a 20-year period to assess mortality. Methods and analysis clear. Comparisons defined and statistical analysis clear.
Good quality. Methods clear. Data collected over a 9-year period so likely to be representative of SUDEP deaths and reflects national records. Appropriate use of statistics.
Moderate quality. Measures used to assess SUDEP conversation may not be reliable, as they were dependent on accuracy of patient notes. Time constraints and different descriptions of SUDEP conversations e.g. using "serious harm" instead of SUDEP may have led to underestimation of SUDEP conversations. Results and method are very clear.
Seven were rated as moderate quality. Problems were either due to difficulties matching controls in experimental designs, or not enough detail in the analysis to replicate qualitative analyses. These are common concerns in clinical research often related to the research context or journal restrictions. 10 articles were rated as good quality, all quantitative studies.
Data range of all papers collected was between 2001 and 2018; and papers were published between 2006 and 2018. Ten were conducted in the UK, one in Australia, one in the US, two in Canada and the US, one in Canada, Australia and the UK, one in Norway and one in Italy. Four studies elicited the views of PWE and their families, and six studies focused on the views of clinicians, one study focused on bereaved families (see Fig. 2).
There was limited consideration of how gender, ethnicity or age might affect communication. Five papers commented that men were generally considered at higher risk but none discussed this in relation to how communication might be adapted. One paper considered age, gender, duration of epilepsy, level of education and employment as variables that might affect preference for discussions, finding none of these significant predictors.
There was diversity of methods (Fig. 3). Four studies compared risk communication between groups, such as using checklists (n = 3), or reported discussions of risk (n = 1). Two studies were case note reviews. Twelve studies included less than 50 participants, two studies included between 50 and 200 participants, and three studies included over 200 participants.
The papers are summarised in Table 5. Papers were identified based on whether they offered guidance on how to hold conversations about risk with PWE. However, 15 of the 17 papers focused on SUDEP, suggesting a large gap in the literature. None of the papers focused directly on how to communicate about risk. We found limited elaboration around the individual differences that might affect whether, when or how to hold risk conversations; instead papers just reminded clinicians to consider individual differences. Extracted recommendations that might guide clinicians in how and whether to hold risk discussions with PWE are summarised in Table A1 in Appendix A.
Table 5Guidance around how to communicate with PWE about epilepsy risk and risk factors.
Guidance
Paper
Timing and frequency of communications
1. Evaluate whether the clinician has the right experience to engage in discussions
50 42
2. Evaluate the risk of harm or distress in the PWE, for example it might not be appropriate to hold conversations at the time of diagnosis
24 51 46
3. Conversations should be held as early as possible
47, 45, 35
4. The timing of conversations should be individually evaluated
50, 3, 43
5. Conversations should occur at least yearly, more frequently if there are risks, including where people are at low risk, or have an intellectual disability.
4, 3
6. The clinician might also ask PWE whether they want to discuss risk.
42
Method of communication
7. Full information should be provided to PWE and their families.
49, 52
8. People may prefer to be informed by a neurologist.
49, 52, 35
9. There should be more than one method of communication about risk (online, printed, and verbal).
45
10. Having a policy that clinicians must discuss risk and ensure that people are better informed and overcome anxieties from the clinician
44
Content of communication
11. Conversations should be framed in a manner acceptable to a person with epilepsy
48
12. More experienced clinicians should deliver discussions
48, 42
13. Acknowledge both the anxiety of the person with epilepsy and the benefits of increased awareness.
50, 2
14. Normalise the risk of SUDEP / be realistic with actual prevalence and preventative strategies
Despite risk communication being considered important, it remains infrequent or ineffective (47 (US, Canada); 50 (UK); 42 (UK)). The clinician’s direct experience of epilepsy and SUDEP (50 (UK); 42(UK)) and whether they had a specialist interest in epilepsy (46, UK) are important determinants of whether risk is discussed. Communication was also affected by the clinician’s experience of SUDEP (43, Italy). Avoidance of harm to the PWE was the main rationale for not discussing risk (24, US; 43, Italy). It was found in a study that many clinicians still do not believe that SUDEP or other mortality issues should be discussed with PWE (43, Italy). Some consider telling people who are newly diagnosed or who have good seizure control as “lower risk”, thus believing they may cause distress for no reason (51, UK). However, focusing on the experiences of PWE of risk discussions, found that people feel anxious in the short term but then reassured along with an association with subsequent increase in medication use [
]. Other studies report that over 90 % of PWE wanted to be informed of most if not all information through communication with their clinician (49 (Norway); 52 (Australia)). Of the 12 papers considering whether a conversation is valid, Ten supported the position that conversations should be held while two provided more balanced positions, acknowledging that the clinician is best placed to make the ultimate decision, but providing guidance should they feel it clinically appropriate.
3.2.1.2 Timing
Seven of the 17 papers considered the timing of risk discussions. Louik et al. (47, US and Canada) examined the experiences of bereaved families, and their concern that SUDEP should have been discussed with carers prior to death. Of 12 survey and interview studies of PWE, families and clinicians, three postulated that the timing of discussions should be individually evaluated (50 (UK); 3 (UK), 43 (Italy)), but how to do so was not detailed. Two questioned the appropriateness of informing on risk at the time of initial diagnosis (51 (UK); 24 (US)). Three suggested conversations should occur at the first or second clinical meetings (47 (Canada and US); 45 (UK, Australia, Canada); 35 (UK). The remainder did not comment on timing.
3.2.1.3 Frequency of communication
Waddell, McCool, Turner, Norman, Coker, White et al. (2, UK) found that SUDEP conversations were more likely to be held where people are non-compliant with medication. However, despite the emphasis on high risk people being more informed, a study by Ross et al. [
] suggested that those who had had epilepsy for longer than 15 years, had drug resistant epilepsy and/or Learning Disability, were less likely to have been informed.
] also found an increase of case note documented discussions from 4 % in 2009, to 31 % of all referrals a year later, where 240 of 314 were available for the second audit. Young et al. (3, UK) observed that it may be important to repeat discussions to ensure awareness and understanding, particularly for those with ID living in a care setting, or with less recent diagnosis.
Notably, the literature extracted was only that which focused on how to hold conversations. Although the evidence appears to be increasing in support of PWE being informed of risk, one paper discussed the PWE’s right not to know (42, UK).
3.2.2 Method of communication
3.2.2.1 Method of communication
Six studies (52 (Australia)); 35 (UK); 45 (UK, Australia, Canada); Ross et al., 2015 (UK); 47 (US and Canada); 49 (Norway)) focused on the views of PWE and their families directly. Over 90 % of people wanted to be informed of most if not all information (49 (Norway); 52 (Australia)). All studies suggested that they would prefer to be informed face-to-face by an epilepsy specialist. Additional written [
] or electronic material (3 (UK)) might also be helpful.
3.2.2.2 Protocols
Nisbet, Turbull, Mulhern & Razvi (44, UK) evaluated the use of a protocol requiring clinicians to discuss SUDEP with PWE. Based on interviews of six neurologists and four senior medical trainees working in a neurology service, they found that pressure to discuss SUDEP created considerable anxiety in clinicians, who in turn believed this would create anxiety in PWE. PWE in fact responded calmly. Clinicians found the pressure of a forced conversation inhibited risk discussions.
3.2.3 Content of communication
Various points were made about the content of conversations.
] conducted a survey of US neurologists (n = 1200), which suggested that conversation should be framed appropriately for the PWE who is the focus, which can then reduce levels of distress. Clinicians experienced in delivering these discussions, particularly those who had seen PWE for longer, seen more of them, and had more years since qualifying were more confident in reducing distress. However, a demand for further education for clinicians and PWE and care givers was also identified. Morton et al. [
] found that nurses felt that although it can be anxiety provoking, risk conversations increased adherence to medication, and 41 % stated that it enhanced quality of life for people. Guidance was unclear from findings, but the authors argue that clinicians must balance too much information control with the potential benefits of awareness. They also found that nurses preferred face to face communications, however, 36 % (n = 37) of risk conversations were held in nurse led clinics; 27 % (n = 28) in a ward setting; and 36 % (n = 37) over the phone as these were opportunities where risk arose in conversation.
3.2.3.3 Normalising the risk discussion
Morton et al. (42, UK) suggest that normalising the risk of SUDEP is also important, for example, by highlighting that it is less common than being hit by a car. RamachandranNair et al. (35, UK) and RamachandranNair and Jack (45, UK, Australia, Canada) also support this approach, stating that there should be a realistic appraisal of risk with the emphasis on the actual prevalence and preventative strategies.
3.2.3.4 Checklists
Discussions can also be supported by checklists (25; 4, UK), which can improve the safety for PWE (4; 3, UK).
4. Discussion
This paper was designed to extract guidance from the literature around how to communicate risk with PWE, offering additional information for clinicians to draw on in Epilepsy management. How this is appropriately applied will be determined by the clinical judgement of practitioners. However, the complexity of language and interaction deserves careful consideration to enable collaborative care with PWE. In the introduction, two barriers to communications about risk were identified.
4.1 The lack of guidelines for how to hold clinical conversations about risk
This paper extracted from the literature the extent of current guidance on how to hold conversations about epilepsy risks and risk factors with PWE. Findings included: that initiating communications should be moderated on an individual basis, but should generally be held early on, and frequently; preferred methods of communication are face to face with a neurologist; and content of communications should be such that information is framed in an acceptable manner and that risk is communicated, but also normalised. This guidance might provide some direction for clinicians, and improve confidence in holding clinical conversations about risk.
4.2 Ethics of risk communication
All four studies that represented views of PWE and their families found they want increased communication, where those focusing on clinicians’ views (six) were more reserved. The gap seems to be around concern from clinicians about creating more distress than is required. This review identified some methods of reducing anxiety, for example being realistic around effects of risk and normalising the risk of SUDEP; acknowledging anxiety and benefits of awareness; framing conversations appropriately for each individual, and having experienced clinicians available to support discussions. There is also disagreement where PWE recommend conversations are held as early as possible, and neurologists urging caution following recent diagnosis. Lessons might be learnt from areas such as oncology where discussion of person-centred risk and mortality is common practice. Pino and colleagues [
] discuss the importance of learning from the observation of how experienced clinicians interact in order to enhance communication, for example including ‘elaboration solicitations’ in interactions in order to encourage PWE to discuss their end of life concerns, in a way that does not increase anxiety. Observation of experienced epilpetologists to better understand how they work, in order to offer guidance for other professionals would be similarly advantageous.
The reservation evident in practice recommendations may stem from the different participants involved in studies. Six studies focused on the clinician’s views [
]. However, this reflected a higher number of PWE and their families (n = 6641) than clinicians (n = 1638), one study being of (n = 1200) people in Norway [
Deciding if lifestyle is a problem: GP risk assessments or patient evaluations? A conversation analytic study of preventive consultations in general practice.
], again testing researcher driven theories, rather than emergent from the views of PWE and their families. Capture of discussion between clinicians and PWE would be beneficial. Further, clinician’s have responsibility for care, making caution appropriate. There is emerging evidence of the benefits of communication [
] but the long term effects on risk management are not yet fully explored. Further, as expected, where there was caution from clinicians, the papers spent time discussing how to over come conversation difficulties, which seems to be a gap (e.g.
The future development of guidelines may require nuancing with different stages of conversation. It may be appropriate to discuss the type of information that people wish to be informed of, prior to any discussion about risk. This is based on Morton et al’s reminder that PWE and their families also have the ‘the right not to know’ [
], moving beyond medication compliance as the main rationale for engaging with discussions. Services might engage both neurologists and clinical specialists, and PWE and their families in consultation groups to design how best to deliver information, so that the concerns of all stakeholders are represented.
Finally, in terms of communication ethics, the broader social media context should be considered. Whether or not clinicians feel comfortable holding conversations, information is available from other sources such as the media and the internet. The quality of this information is variable, and frequently not designed to support the management of anxiety. PWE can experience this in uncontained ways [
]. They may or may not then raise these discussions with clinicians, and we propose that further research may be conducted to examine the effects of internet information and how this influences PWE. A cautious approach may be preferable, enabling PWE to receive accurate information from a supportive clinician who can refer to relevant support services if required.
4.2.2 Evidence base critique
The evidence for guidance extracted here was limited. Studies have not considered directly how communication works, and most research focuses specifically on communication about SUDEP, rather than other epilepsy risks, such as: such as self-injury, problems with driving, overdose of medication, under-dosing of medication, drug interactions, ongoing seizures, status epilepticus, nor other risks associated with the clinical management of epilepsy. A greater amount of research around communication is required that covers these areas.
Methods and samples were heterogeneous. This heterogeneity captured the views of multiple stakeholders, including neurologists, non-specialist doctors, specialist nurses, PWE, their families, and bereaved families. However, the lack of homogenous studies meant that a meta-analysis or synthesis was not viable, thus reducing analytic power. There was unequal distribution: four studies examined PWE; six studies, clinicians; representing a total of 6,641:1638 participants. There was also no consideration of how PWE’s nor clinician’s individual differences, such as gender, ethnicity, learning disability, age or other demographics, beyond experience of epilepsy, may affect communication, which future research may address.
Methods also varied. Views elicited from PWE were mainly from interviews [
] and one survey. Clinicians were included in four surveys, one interview and one focus group design, so overall with less opportunity for elaboration by clinicians. These surveys were also used to identify predictors of whether conversations were held. Three studies compared groups to test directly for causative relationships. These focused on the use of predesigned checklists [
] and showed evidence of significant improvement in awareness of risk and reduction of risk factors where communication about risk used this structured method. Thus, most guidelines and postulated consequences were based on neurologist perspectives, rather than those of PWE and families, such as that conversations appear to mediate AED adherence, despite there currently being limited evidence for this [
]. There was less opportunity for clinicians to elaborate their responses given the survey methods, though this enabled larger participant numbers. Direct observation of skilled clinicians was also missing, hence people actually engage in risk discussions is not yet well understood.
Seven quantitate surveys were conducted, with Three studies including more than 50 participants and Two more than 200. This seems to offer reasonable confidence in generalisation of findings. These were conducted mostly in the UK [
Engaging terminally ill patients in end of life talk: how experienced palliative medicine doctors navigate the dilemma of promoting discussions about dying.
] and one in Norway, potentially not capturing international variation. All studies, apart from the two group comparison studies, drew on retrospective reports from clinicians and PWE rather than examining how the communications actually took place. Group comparison studies can raise ethical issues, as trials cannot be run where communication is withheld. Alternatively, future research might focus more directly on clinical conversations, following the substantial rise in the use of Conversation Analysis to explore how communication works in practice [
Smart C. Auburn T. Interprofessional care and mental health: a discursive exploration of team meeting practices. 1st ed. Pallgrave MacMillan,
London2019
]. Analysing clinical conversations might enable the identification of best practice for clinicians in engaging people with risk discussions, and provide a helpful method for training clinicians [
], but it is not yet known how they do this, which could be identified and then trained more widely.
4.3 Strengths and Limitations of the review
This review was limited by the low number of published papers available that comment on how clinical communication about risk might be achieved (only 17, with no limiters, all published between 2001 and 2018). None of these papers directly addressed the question of how to have clinical conversations about risk. Six papers focused on the frequency of SUDEP conversations, four on evaluating clinical practices, three on SUDEP awareness, two on use of the safety checklist and one on mortality rates, and one more generally on discussion content. All guidance was extracted from brief mentions of recommendations in the discussion section, not direct empirical data. We argue this is a gap that future research should address.
There were limited inductive studies that might identify new insights for practice. Of those qualitative inductive studies, there was diversity in how themes were extracted, some being driven by the questions asked, again limiting the inductive nature of the research.
Risk discussions with PWE are a complex issue and clinicians will need to make their own best judgement about engaging in them. A research focus on communication in epilepsy contexts may reveal better communication methods, facilitating future collaboration and engagement with people in managing their epilepsy. Thus, some aspects of this paper especially around communication might appear ‘patronising’, especially to clinicians for whom communication is a primary and basic professional need. However, it is important to recognise that lack of structure and communication is highlighted as one of the more serious failings by experts and stakeholder groups particularly those who have lost family and friends to epilepsy. It is important to have consensus on the standards of minimum and basic level of communication to all these existing concerns.
5. Conclusion
There is frequent reference to the limited guidelines provided by NICE or the AAN, about how, when and whom to discuss SUDEP and epilepsy risk with. This paper revealed the limited research available around how conversations with PWE about risk might best be held. It is the first paper to summarise the available guidance, taking a step towards better understanding how conversations about risk with PWE might be held, which has not yet been directly addressed. The guidelines arising here should be used with discretion, as less than optimal empirical evidence is available. Situations should be considered individually, but in general, conversations about risk should be held early on. Future research should better integrate the perspectives of clinicians and PWE, to understand how anxiety on both parts can be overcome. Research should focus directly on clinical practice, to gain direct insight into how experts manage difficult conversations, in order to support and inform other clinicians. Focal areas could include: what it is helpful to and say when conversations are held about risk, how these conversations are received, and where they are best located in a treatment pathway. Research should explore how different service contexts are relevant to the delivery of risk discussions. This will enable clearer and more focused guidance, embedded directly in the experiences of PWE and clinicians, and within the realities of clinical practice.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Declaration of Competing Interest
None.
Appendix A.
Table A1Data extraction from all studies.
Title
Aims to:
Data Collected
Participants
Analysis
Key Findings
Recommendations/ implications
Mohanraj et al, 2006
Analyse mortality data for patients with newly diagnosed and chronic epilepsy who were referred to a single adult service over a 20-year period
Related measures. PWE were assessed for mortality risk on the basis of response to treatment.
Group 1: 890 newly diagnosed PWE;
Keplan-meier survival curves were plotted for the 2 groups and controls. Chi-squared test was used to compare observed and expected deaths.
There were 93 deaths in the newly diagnosed group compared to 64 in age and sex matched controls (p = 0.0007). All excess mortality occurred in patients who did not achieve seizure freedom with treatment. In the chronic group there was more than double the expected number of deaths and incidence of SUDEP compared to newly diagnosed PWE, this being higher for those under 30 yrs. of age.
Routine discussion of mortality at the time of diagnosis not necessary, especially if this does not affect the management of the epilepsy.
Group 2: 2689 PWE with chronic epilepsy and poor seizure control.
Epilepsy risks and mortality should be discussed with PWE who choose not to use AED's, and those who haven’t achieved seizure freedom.
Compared with age and sex matched controls.
42
Present the experiences of neurologists in discussing SUDEP with PWE.
Questionnaires to all practicing neurologists listed on the British neurologist’s database.
387 questionnaires (of 738 sent). Represents 82% of consultant body.
Chi squared test and qualitative analysis.
4.7% discussed SUDEP with all PWE. Years of experience, or level or registration (registrar/neurologist) had no effect. Those with a special interest in epilepsy were more likely to discuss SUDEP. Those discussing SUDEP routinely reported less negative reactions from PWE.
The majority of neurologists are not following the NICE guidelines.
Asked about the circumstances surrounding discussions about SUDEP and PWE’s reactions.
63 were specialist registrars. Remainder were from specialist epilepsy clinics.
Practice effect may make it easier to discuss SUDEP.
NICE give little guidance on how, when and whom to discuss SUDEP with. Epilepsy.
Risks should be contextualised, e.g. likelihood of SUDEP compared to getting hit by a car.
NICE guidelines deny the "right not to know".
50
Examine what, when and how information about SUDEP is disseminated to patients by clinical nurse specialists in epilepsy (CNSEs)
Postal Questionnaires from ESNA members.
146 responses (from 250 sent). 43 excluded as respondents were not CNSE. Time in role ranged from 8 months to 21 years
Descriptive of responses.
6% discussed SUDEP with all patients. 48% stated risk discussions were prompted by discussions of risk factors (e.g. alcohol consumption. Length of service had no significant effect.
CNSE do not discuss risk with all patients. SUDEP discussions need to be addressed carefully, balancing the negative effects of fear in patients, with the benefits of raising awareness. Timing of discussion should be individually determined.
Most discuss SUDEP over the telephone (42%).
62% reported an increase in medication adherence following SUDEP discussions, and subsequent avoidance of risk factors (62%).
Vegni et al, 2011
Explore whether Italian Physicians interested in epilepsy believe that they should discuss SUDEP with patients and or their families or not
Questionnaire.
Recruited through the Italian association against epilepsy (LICE). 195 responses. 49.5% were male. Mean age 45; average practice years 21.
Chi squared test & content analysis
28% of believed that SUDEP should be discussed with most patients. Most believed that PWE be emotional or have a negative reaction (73%).
More research is needed for in-depth understanding. Physicians may disclose risk more often if they have treated someone with SUDEP or following ethical discussions.
Physicians may decide to give risk information based on the patients themselves or for non-clinical reasons such as ethical issues.
Waddell et al, 2013
Conduct an audit of current practices of discussing SUDEP with PWE at a specialist epilepsy clinic.
5 yr. case note analysis.
All patients who attended a specialist epilepsy clinic, and those who had experienced at least 2 unprovoked seizures were included, producing 345 case notes. Mean age 41 yrs. 50.7% were male.
Descriptive statistics and fishers 2 tailed test for comparisons between risk factor groups.
SUDEP is not discussed with all patients. Those with ongoing GTCS and drug resistant seizures were more likely to be informed. There was a trend towards discussing SUDEP with those non-compliant with medication.
An honest and frank debate is required between clinicians, patient advocacy groups and those involved in developing guidelines to allow one to reconcile the disparity between guidelines and clinical practice in regard to SUDEP.
Freidman et al, 2014
Examine SUDEP discussion practices among neurologists in the U.S and Canada.
Online questionnaire-based study.
117, 558 people invited by email. Self-selected if they: were a neurologist who devoted >5% of their time to clinical care and had completed post-grad training.
A composite knowledge score was determined by subtracting the number of incorrect identified risk factors from correctly identified factors & multivariate logistic regression.
82% had incomplete knowledge about SUDEP. 26.1% encountered at least one case of SUDEP in the past 24 months. 6.8% of neurologists discussed SUDEP with all patients.
SUDEP conversations can be framed to minimise distress. Neurologists with more experience and diverse caseloads report less perceived distress to SUDEP discussion. Guidance is needed regarding the best approach to SUDEP discussions for neurologists.
200 completed surveys. 76% adult neurologists; 33% additional training in epilepsy or neurophysiology. 43% saw 100 PWE annually.
Epilepsy training, years in practice, having over 100 patients annually and having a case of SUDEP increased the likelihood of discussion.
62% discussed SUDEP when patients were ‘high risk’. Risk level and additional training was associated with an increased risk of a perceived negative response to SUDEP discussion (p=0.038).
Miller et al, 2014
Describe the practices of epileptologists, neurologists and advanced practice nurses (APNs) regarding discussing SUDEP with their patients, and their rationales for discussing SUDEP
Two focus groups: one with epileptologists and neurologists; one with APN's.
19 epileptologists, 16 neurologists and 8 APNs. All areas of central America were represented
Themes were inductively developed.
Themes:
A standardised approach to discussing SUDEP is needed to facilitate clinicians, and overcome fears that discussions will cause harm, and have no benefit.
1. Reasons for discussing SUDEP: Practical accountability; Moral accountability, proactivity and reactivity, and for nurses only – patient advocacy.
Decisions not to disclose might not be patient centred.
2. Reasons for not discussing SUDEP: to wait for a rapport to be established, being morally accountable, and being out of options.
Clinicians should involve PWE in decision making. The next step is to provide a practical guide for discussing SUDEP. SUDEP education should be a component of epilepsy care.
3. Ways in which SUDE should be discussed: face to face discussion were common with written materials; requirement for standardised protocols that can be adapted to individuals.
Ross et al, 2015
Evaluate current clinical practice and determine whether there had been a change in frequency of SUDEP discussion following prior audit.
A retrospective case note review of a single service.
392 patients were reviewed, 240 case notes were available for review from PWE attending a specialist clinic. 27 new referrals. 213 return patients.
Descriptive statistics and comparisons between risk factor groups using fisher’s two-tailed test.
SUDEP discussions were documented in 67% of new referrals, 30% of return patients.
An inverse relationship between those at greatest risk and those most likely to be informed might reflect perceived difficulties in discussing SUDEP where it seems there are no therapeutic options left to cover. A tailored discussion in clinic with discussion of risk-reducing strategies will be a positive intervention.
These included by neurologists with a specialist epilepsy interest (40%), 30% with epilepsy nurses, 17.4% with GPs. PWE with GTC's were more frequently informed of risks; PWE for longer than 15yrs, drug resistant seizures and LD were less frequently informed.
46% were males. Mean age 36yrs.
Xu et al, 2015
Evaluate awareness and perspectives on SUDEP among adult PWE
Questionnaire.
40 males a 65 females took part in the study, mean age of 41 yrs.
Multivariate logistic regression explored the variables among patients associated with their awareness of SUDEP and willingness to be informed of this.
62% of PWE wished to know all information. 32% wanted a reasonable amount.
The authors encourage health professionals and policy makers to incorporate SUDEP discussions in regular practice and as a quality measure of clinical practice.
52% thought epilepsy was not associated with higher risk of death.
14% had heard about SUDEP before the study.
89% of participants wished to be informed about SUDEP, preferably by their neurologist.
RamachandranNair and Jack, 2016
Understand the range of adult patients' views on discussing SUDEP with PWE; clarify the optimal timing and formulation of information.
23 PWE (7 males and 16 females) recruited through an adult neurology clinic and a community epilepsy agency. Aged 18-65, median, 33.
Directed content analysis
10 (43%) had heard about SUDEP previously; 2 learned through the internet, 4 through the community epilepsy agency, 3 from their neurologist, and one from their mother.
Face to face discussion of SUDEP with the neurologist important. Written information beneficial. Content should include an estimate of individual risk, emphasis on actual prevalence and preventative strategies.
19 participated in interviews. 4 in focus group.
All agreed SUDEP should be discussed at the diagnosis. 3 felt SUDEP disclosure should be individually decided. 19 considered there may be negative effects, such as anxiety.
Neurologist rather than the emergency department, best person to inform.
No participants correctly explained meaning of SUDEP, even when leaflet given. 50% of PWE felt anxious upon learning about SUDEP, short lasting. Some reported then increasing adherence to their medication.
RamachandranNair, Jack and Strohm, 2016
Understand the experiences of relatives of individuals whose deaths were identified as SUDEP and to explore their preferences regarding SUDEP counselling.
In depth semi-structured interviews
Stratified purposeful sampling: people over 18 bereaved by SUDEP.
Thematic analysis using categories from the questionnaire and emergent categories.
1. Experiences at the time of SUDEP: all experienced shock and guilt.
Neurologists should inform patients SUDEP. Optimal timing decided case by case.
27 participants, 21 females and 6 males. 4 were siblings, 5 were spouses and 18 were parents.
2. Awareness of SUDEP: families could accurately define SUDEP, most drawing parallel between SUDEP and SIDS. Only 9 participants knew about SUDEP before the cause of death. Many were angry not to have known earlier, feeling there could have been preventative steps. Awareness of SUDEP after death was valued, giving "peace of mind".
Deliver information face to face at first or second visit. Give printed material.
3. Whether to discuss SUDEP: SUDEP education was overwhelmingly valued, especially for those with specific risk factors. PWE stated it was their right to know about their condition. Possible negative effects included making people over protective of the PWE.
Content should include realistic risk appraisal. Emphasise preventative strategies.
4. Information should be provided by the patient’s neurologist shortly after or at the time of diagnosis.
5. Content of the SUDEP discussion: include an explanation of SUDEP; associated risk factors and preventative methods. Note the rarity of SUDEP.
Shankar et al, 2016
Compare the 17 risk factors to cases of SUDEP deaths, and those living with epilepsy to determine how strongly these factors are associated with SUDEP risk
Comparison of risk factors as identified using the SUDEP and seizure safety checklist were compared between live samples and people who had died from Epilepsy related deaths.
43 SUDEP deaths of 93 Epilepsy deaths occurred between 2004 and 2012. The coroner’s notes were used and compared with the medical notes from 220 live PWE.
Groups were compared and relative risk factors ranked.
9 (of 17) risk factors differed significantly between groups. 2 were not modifiable (duration of epilepsy and GTC's). 7 modifiable risk factors were identified and ranked.
The study supports the use of an evidences-based checklist to discuss potentially modifiable risk factors with patients
Louik et al, 2017
Examine attitudes towards SUDEP discussion among family members
Semi-structured telephone interviews about experiences of SUDEP discussions.
138 family members enrolled on the North American SUDEP Registry. 79% were conducted with at least one parent, 10% with widower and 10% with offspring and 40% with another next of kin or carers
Regression analysis to identify predictors of SUDEP awareness. Factors included: year of death, status epilepticus, GTCS and age.
18.1% recall having discussed SUDEP, initiated by the neurologist in 44% of these cases. 72% appreciated this discussion, 24% were unsure. For those who did not recall a SUDEP discussion 72.3% wish they had, 10% did not and 17% were unsure.
Frequency of SUDEP discussion may not be increasing, contrary to other research. Research should examine why people might not want to learn about SUDEP and whether SUDEP education can mitigate risk factors such as AED adherence. Healthcare should discuss SUDEP when caring for PWE rather than after the person has died.
Common themes they wanted discussed included: incidence of SUDEP, general SUDEP information, importance of AED adherence. No significance was found for factors increasing the likelihood of SUDEP discussion.
Nisbet et al, 2017
To explore the experiences of neurologists in Scotland when discussing SUDEP with the patients.
Qualitative individual interviews
6 consultant neurologists and 4 registrar doctors participated in the research.
Thematic analysis.
Themes:
The FAI in Scotland has increased SUDEP discussion by highlighting possible medico-legal implications.
1) The SUDEP protocol: Clinicians engage in 2 types of SUDEP discussion – with PWE who are newly diagnosed, and uncontrolled. For those with uncontrolled seizures it was used to emphasise the risk and encourage AED adherence. SUDEP was not discussed if the PWE appeared to be distressed or anxious.
Future studies should quantify behaviour changes post SUDEP discussion, as this might facilitate confidence to increase SUDEP conversations.
2) Diffusion of the FAI- The Fatal Accident Inquiry (specific to Scotland) heavily influenced how clinicians’ emphasis on SUDEP discussions although there were some mixed feelings.
3) Breaking good news- there were mixed feeling surrounding discussing SUDEP.
4) Falsely anticipating distress- Clinicians believed they were likely to cause distress and anxiety by discussing SUDEP. Many being surprised when PWE react calmly.
5) Pressure hinders effective communication- Clinicians report that the requirement to discuss SUDEP sometimes inhibited their ability to do it well, due to time pressure.
Henning et al, 2018
Assess how much information PWE and their family members wanted about epilepsy related risks and whether this need was met by health care professionals.
An online questionnaire available through a Norwegian epilepsy association.
1859 participants.
Chi squared test used to analyse group differences and multivariate logistic regression used to determine factors that may predict a wish to be given more information, or people being given more information about increased risk.
90% wanted information about risks and premature death. Having generalised tonic clonic seizures was associated with wanting information about the death risk (p=0.001). Male gender, younger ages and aetiology of epilepsy were significant factors for having been told about SUDEP. 14% of patients felt they had been well informed.
There is a gap between what the patients want to know and what they are told by healthcare providers.
1183 were PWE.
676 were careers answering on behalf of PWE.
Shankar et al, 2018
Determine whether the introduction of a risk communication checklist in an epilepsy clinic leads to a reduction in individual SUDEP epilepsy mortality risk factors.
Repeated measures. Assessment of risk using a checklist was conducted twice as part of routine care in a specialist epilepsy clinic, one year apart.
130 consecutive individuals from the neurology clinic and 129 from the ID clinic. The second application 91 and 93 individuals, respectively.
Paired t-test to compare groups.
Overall reduction in mean risk score for the general population (p = 0.049) but not for the ID population (who had received risk information on numerous occasions previously).
Safety advice had direct implications for reducing risk factors.
There was a risk reduction in the top 25% in both patient groups (p = <0.001).
The introduction of a safety Checklists supported.
Policy might promote use of health promotional checklists in epilepsy to reduce morbidity and mortality.
Young et al, 2018
Assess whether PWE and their families or support workers were SUDEP aware and could recall discussions about SUDEP risk and the role of nocturnal surveillance.
Questionnaire.
121 responses from PWE with LD under active follow up in July 2017. 75% were living in a residential setting
Fisher’s exact test was used to establish differences between those PWE living family or in a residential setting.
PWE with ID placed in a residential setting are more likely to not have person centred risk advice implemented compared to those in a family setting. Those in the family setting were more likely to recall the SUDEP discussion (65%) compared to those in the residential setting [
Deciding if lifestyle is a problem: GP risk assessments or patient evaluations? A conversation analytic study of preventive consultations in general practice.
Engaging terminally ill patients in end of life talk: how experienced palliative medicine doctors navigate the dilemma of promoting discussions about dying.
NICE Clinical Guideline CG137 - Partial Pharmacological Update of Clinical Guideline 20: the Epilepsies: the diagnosis and management of the epilepsies in adults and children in primary and secondary care.
National Clinical Guideline Centre. London: Royal College of Physicians,
2012 (Jan, Available from)
A qualitative study of the reactions of young adults with epilepsy to SUDEP disclosure, perceptions of risks, views on the timing of disclosure, and behavioural change.
Effect of providing sudden unexpected death in epilepsy (SUDEP) information to persons with epilepsy (PWE) and their caregivers-experience from a tertiary care hospital.
Smart C. Auburn T. Interprofessional care and mental health: a discursive exploration of team meeting practices. 1st ed. Pallgrave MacMillan,
London2019 (2018 edition (14th Jan 2019))
Training for enhanced team performance in mental healthcare contests: a workshop and its fit with interprofessional care.
(2019)in: Smart C. Auburn T. Interprofessional care and mental health: a discursive exploration of team meeting practices. Pallgrave MacMillan,
London2019
Risks in epilepsy include those directly related to the occurrence of seizures in vulnerable situations. Some of these include: driving [1]; operating machinery [2,3]; working at heights [3]; working with electricity and electrical appliances [4]; cooking and preparing food with open flames, hot liquids and using knives [5]; bathing or swimming practice [6]; or even just enjoying some recreational sports, such as surfing, sailing or horse riding [7]. There are also situations that may provoke seizures with inherent risks resulting in negative consequences.
30024-8&id=gr1.jpg){kind=link}
30024-8&id=gr2.jpg){kind=link}
30024-8&id=gr3.jpg){kind=link}